Skip to content

Fixes issue with arm intrinsics not compiling in dev profile #1964

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
Nov 27, 2025
Merged

Fixes issue with arm intrinsics not compiling in dev profile #1964

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
5b1fda1
Fixes generator, neon intrinics now build in debug
CrooseGit Oct 24, 2025
51f3dab
Fixes differences found between clang & rust
CrooseGit Nov 6, 2025
f564a01
Provides work-around for vreinterpret inline fail
CrooseGit Nov 24, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion ci/intrinsic-test.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -56,7 +56,7 @@ case ${TARGET} in

aarch64_be-unknown-linux-gnu*)
TEST_CPPFLAGS="-fuse-ld=lld"
TEST_SKIP_INTRINSICS=crates/intrinsic-test/missing_aarch64.txt
TEST_SKIP_INTRINSICS=crates/intrinsic-test/missing_aarch64_be.txt
TEST_CXX_COMPILER="clang++"
TEST_RUNNER="${CARGO_TARGET_AARCH64_BE_UNKNOWN_LINUX_GNU_RUNNER}"
: "${TEST_SAMPLE_INTRINSICS_PERCENTAGE:=100}"
Expand Down
24 changes: 12 additions & 12 deletions crates/core_arch/src/aarch64/neon/generated.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9500,10 +9500,10 @@ pub fn vdivh_f16(a: f16, b: f16) -> f16 {
#[unstable(feature = "stdarch_neon_dotprod", issue = "117224")]
pub fn vdot_laneq_s32<const LANE: i32>(a: int32x2_t, b: int8x8_t, c: int8x16_t) -> int32x2_t {
static_assert_uimm_bits!(LANE, 2);
let c: int32x4_t = vreinterpretq_s32_s8(c);
unsafe {
let c: int32x4_t = transmute(c);
let c: int32x2_t = simd_shuffle!(c, c, [LANE as u32, LANE as u32]);
vdot_s32(a, b, transmute(c))
vdot_s32(a, b, vreinterpret_s8_s32(c))
}
}
#[doc = "Dot product arithmetic (indexed)"]
Expand All @@ -9515,11 +9515,11 @@ pub fn vdot_laneq_s32<const LANE: i32>(a: int32x2_t, b: int8x8_t, c: int8x16_t)
#[unstable(feature = "stdarch_neon_dotprod", issue = "117224")]
pub fn vdotq_laneq_s32<const LANE: i32>(a: int32x4_t, b: int8x16_t, c: int8x16_t) -> int32x4_t {
static_assert_uimm_bits!(LANE, 2);
let c: int32x4_t = vreinterpretq_s32_s8(c);
unsafe {
let c: int32x4_t = transmute(c);
let c: int32x4_t =
simd_shuffle!(c, c, [LANE as u32, LANE as u32, LANE as u32, LANE as u32]);
vdotq_s32(a, b, transmute(c))
vdotq_s32(a, b, vreinterpretq_s8_s32(c))
}
}
#[doc = "Dot product arithmetic (indexed)"]
Expand All @@ -9531,10 +9531,10 @@ pub fn vdotq_laneq_s32<const LANE: i32>(a: int32x4_t, b: int8x16_t, c: int8x16_t
#[unstable(feature = "stdarch_neon_dotprod", issue = "117224")]
pub fn vdot_laneq_u32<const LANE: i32>(a: uint32x2_t, b: uint8x8_t, c: uint8x16_t) -> uint32x2_t {
static_assert_uimm_bits!(LANE, 2);
let c: uint32x4_t = vreinterpretq_u32_u8(c);
unsafe {
let c: uint32x4_t = transmute(c);
let c: uint32x2_t = simd_shuffle!(c, c, [LANE as u32, LANE as u32]);
vdot_u32(a, b, transmute(c))
vdot_u32(a, b, vreinterpret_u8_u32(c))
}
}
#[doc = "Dot product arithmetic (indexed)"]
Expand All @@ -9546,11 +9546,11 @@ pub fn vdot_laneq_u32<const LANE: i32>(a: uint32x2_t, b: uint8x8_t, c: uint8x16_
#[unstable(feature = "stdarch_neon_dotprod", issue = "117224")]
pub fn vdotq_laneq_u32<const LANE: i32>(a: uint32x4_t, b: uint8x16_t, c: uint8x16_t) -> uint32x4_t {
static_assert_uimm_bits!(LANE, 2);
let c: uint32x4_t = vreinterpretq_u32_u8(c);
unsafe {
let c: uint32x4_t = transmute(c);
let c: uint32x4_t =
simd_shuffle!(c, c, [LANE as u32, LANE as u32, LANE as u32, LANE as u32]);
vdotq_u32(a, b, transmute(c))
vdotq_u32(a, b, vreinterpretq_u8_u32(c))
}
}
#[doc = "Set all vector lanes to the same value"]
Expand Down Expand Up @@ -28283,10 +28283,10 @@ pub fn vuqadds_s32(a: i32, b: u32) -> i32 {
#[unstable(feature = "stdarch_neon_i8mm", issue = "117223")]
pub fn vusdot_laneq_s32<const LANE: i32>(a: int32x2_t, b: uint8x8_t, c: int8x16_t) -> int32x2_t {
static_assert_uimm_bits!(LANE, 2);
let c: int32x4_t = vreinterpretq_s32_s8(c);
unsafe {
let c: int32x4_t = transmute(c);
let c: int32x2_t = simd_shuffle!(c, c, [LANE as u32, LANE as u32]);
vusdot_s32(a, b, transmute(c))
vusdot_s32(a, b, vreinterpret_s8_s32(c))
}
}
#[doc = "Dot product index form with unsigned and signed integers"]
Expand All @@ -28298,11 +28298,11 @@ pub fn vusdot_laneq_s32<const LANE: i32>(a: int32x2_t, b: uint8x8_t, c: int8x16_
#[unstable(feature = "stdarch_neon_i8mm", issue = "117223")]
pub fn vusdotq_laneq_s32<const LANE: i32>(a: int32x4_t, b: uint8x16_t, c: int8x16_t) -> int32x4_t {
static_assert_uimm_bits!(LANE, 2);
let c: int32x4_t = vreinterpretq_s32_s8(c);
unsafe {
let c: int32x4_t = transmute(c);
let c: int32x4_t =
simd_shuffle!(c, c, [LANE as u32, LANE as u32, LANE as u32, LANE as u32]);
vusdotq_s32(a, b, transmute(c))
vusdotq_s32(a, b, vreinterpretq_s8_s32(c))
}
}
#[doc = "Unzip vectors"]
Expand Down
Loading