diff --git a/clang/include/clang/Basic/riscv_vector.td b/clang/include/clang/Basic/riscv_vector.td
index 6451e77e77f631..d02a4fffb14115 100644
--- a/clang/include/clang/Basic/riscv_vector.td
+++ b/clang/include/clang/Basic/riscv_vector.td
@@ -1700,7 +1700,6 @@ defm vwmulsu : RVVOutOp0Op1BuiltinSet<"vwmulsu", "csi",
 }
 
 // 12.13. Vector Single-Width Integer Multiply-Add Instructions
-let HasPolicy = false in {
 defm vmacc  : RVVIntTerBuiltinSet;
 defm vnmsac : RVVIntTerBuiltinSet;
 defm vmadd  : RVVIntTerBuiltinSet;
@@ -1721,7 +1720,6 @@ defm vwmaccsu : RVVOutOp1Op2BuiltinSet<"vwmaccsu", "csi",
 defm vwmaccus : RVVOutOp1Op2BuiltinSet<"vwmaccus", "csi",
                                        [["vx", "w", "wwUev"]]>;
 }
-}
 
 // 12.15. Vector Integer Merge Instructions
 // C/C++ Operand: (mask, op1, op2, vl), Intrinsic: (op1, op2, mask, vl)
@@ -1804,7 +1802,6 @@ let Log2LMUL = [-2, -1, 0, 1, 2] in {
 }
 
 // 14.6. Vector Single-Width Floating-Point Fused Multiply-Add Instructions
-let HasPolicy = false in {
 defm vfmacc  : RVVFloatingTerBuiltinSet;
 defm vfnmacc : RVVFloatingTerBuiltinSet;
 defm vfmsac  : RVVFloatingTerBuiltinSet;
@@ -1819,7 +1816,6 @@ defm vfwmacc  : RVVFloatingWidenTerBuiltinSet;
 defm vfwnmacc : RVVFloatingWidenTerBuiltinSet;
 defm vfwmsac  : RVVFloatingWidenTerBuiltinSet;
 defm vfwnmsac : RVVFloatingWidenTerBuiltinSet;
-}
 
 // 14.8. Vector Floating-Point Square-Root Instruction
 def vfsqrt : RVVFloatingUnaryVVBuiltin;
@@ -2003,12 +1999,10 @@ let HasMask = false, HasPolicy = false in {
 }
 
 // 17.3. Vector Slide Instructions
-let HasPolicy = false in {
 // 17.3.1. Vector Slideup Instructions
 defm vslideup   : RVVSlideBuiltinSet;
 // 17.3.2. Vector Slidedown Instructions
 defm vslidedown : RVVSlideBuiltinSet;
-}
 
 // 17.3.3. Vector Slide1up Instructions
 defm vslide1up : RVVSlideOneBuiltinSet;
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfmacc.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfmacc.c
index 4c44b69ca654af..963271d80ca16b 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfmacc.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfmacc.c
@@ -187,7 +187,7 @@ vfloat64m8_t test_vfmacc_vf_f64m8(vfloat64m8_t acc, double op1,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmacc.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmacc.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfmacc_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -198,7 +198,7 @@ vfloat32mf2_t test_vfmacc_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmacc.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmacc.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfmacc_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -208,7 +208,7 @@ vfloat32mf2_t test_vfmacc_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmacc.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmacc.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfmacc_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -219,7 +219,7 @@ vfloat32m1_t test_vfmacc_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmacc.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmacc.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfmacc_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc, float op1,
@@ -229,7 +229,7 @@ vfloat32m1_t test_vfmacc_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmacc.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmacc.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfmacc_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -240,7 +240,7 @@ vfloat32m2_t test_vfmacc_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmacc.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmacc.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfmacc_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc, float op1,
@@ -250,7 +250,7 @@ vfloat32m2_t test_vfmacc_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmacc.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmacc.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfmacc_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
@@ -261,7 +261,7 @@ vfloat32m4_t test_vfmacc_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmacc.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmacc.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfmacc_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
@@ -271,7 +271,7 @@ vfloat32m4_t test_vfmacc_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmacc.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmacc.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfmacc_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
@@ -282,7 +282,7 @@ vfloat32m8_t test_vfmacc_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmacc.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmacc.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfmacc_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
@@ -292,7 +292,7 @@ vfloat32m8_t test_vfmacc_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmacc.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmacc.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfmacc_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -303,7 +303,7 @@ vfloat64m1_t test_vfmacc_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmacc.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmacc.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfmacc_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -313,7 +313,7 @@ vfloat64m1_t test_vfmacc_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmacc.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmacc.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfmacc_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -324,7 +324,7 @@ vfloat64m2_t test_vfmacc_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmacc.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmacc.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfmacc_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -334,7 +334,7 @@ vfloat64m2_t test_vfmacc_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmacc.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmacc.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfmacc_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -345,7 +345,7 @@ vfloat64m4_t test_vfmacc_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmacc.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmacc.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfmacc_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -355,7 +355,7 @@ vfloat64m4_t test_vfmacc_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmacc.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmacc.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfmacc_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -366,7 +366,7 @@ vfloat64m8_t test_vfmacc_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmacc.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmacc.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfmacc_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc, double op1,
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfmadd.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfmadd.c
index 1a81935f884908..09734276482dda 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfmadd.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfmadd.c
@@ -187,7 +187,7 @@ vfloat64m8_t test_vfmadd_vf_f64m8(vfloat64m8_t acc, double op1,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmadd.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmadd.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfmadd_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -198,7 +198,7 @@ vfloat32mf2_t test_vfmadd_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmadd.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmadd.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfmadd_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -208,7 +208,7 @@ vfloat32mf2_t test_vfmadd_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmadd.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmadd.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfmadd_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -219,7 +219,7 @@ vfloat32m1_t test_vfmadd_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmadd.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmadd.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfmadd_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc, float op1,
@@ -229,7 +229,7 @@ vfloat32m1_t test_vfmadd_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmadd.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmadd.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfmadd_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -240,7 +240,7 @@ vfloat32m2_t test_vfmadd_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmadd.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmadd.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfmadd_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc, float op1,
@@ -250,7 +250,7 @@ vfloat32m2_t test_vfmadd_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmadd.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmadd.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfmadd_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
@@ -261,7 +261,7 @@ vfloat32m4_t test_vfmadd_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmadd.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmadd.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfmadd_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
@@ -271,7 +271,7 @@ vfloat32m4_t test_vfmadd_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmadd.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmadd.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfmadd_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
@@ -282,7 +282,7 @@ vfloat32m8_t test_vfmadd_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmadd.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmadd.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfmadd_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
@@ -292,7 +292,7 @@ vfloat32m8_t test_vfmadd_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmadd.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmadd.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfmadd_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -303,7 +303,7 @@ vfloat64m1_t test_vfmadd_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmadd.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmadd.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfmadd_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -313,7 +313,7 @@ vfloat64m1_t test_vfmadd_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmadd.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmadd.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfmadd_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -324,7 +324,7 @@ vfloat64m2_t test_vfmadd_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmadd.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmadd.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfmadd_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -334,7 +334,7 @@ vfloat64m2_t test_vfmadd_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmadd.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmadd.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfmadd_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -345,7 +345,7 @@ vfloat64m4_t test_vfmadd_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmadd.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmadd.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfmadd_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -355,7 +355,7 @@ vfloat64m4_t test_vfmadd_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmadd.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmadd.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfmadd_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -366,7 +366,7 @@ vfloat64m8_t test_vfmadd_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmadd.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmadd.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfmadd_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc, double op1,
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfmsac.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfmsac.c
index e42979c08f3499..36308a35ef815f 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfmsac.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfmsac.c
@@ -187,7 +187,7 @@ vfloat64m8_t test_vfmsac_vf_f64m8(vfloat64m8_t acc, double op1,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmsac.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmsac.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfmsac_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -198,7 +198,7 @@ vfloat32mf2_t test_vfmsac_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmsac.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmsac.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfmsac_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -208,7 +208,7 @@ vfloat32mf2_t test_vfmsac_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmsac.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmsac.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfmsac_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -219,7 +219,7 @@ vfloat32m1_t test_vfmsac_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmsac.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmsac.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfmsac_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc, float op1,
@@ -229,7 +229,7 @@ vfloat32m1_t test_vfmsac_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmsac.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmsac.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfmsac_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -240,7 +240,7 @@ vfloat32m2_t test_vfmsac_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmsac.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmsac.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfmsac_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc, float op1,
@@ -250,7 +250,7 @@ vfloat32m2_t test_vfmsac_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmsac.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmsac.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfmsac_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
@@ -261,7 +261,7 @@ vfloat32m4_t test_vfmsac_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmsac.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmsac.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfmsac_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
@@ -271,7 +271,7 @@ vfloat32m4_t test_vfmsac_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmsac.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmsac.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfmsac_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
@@ -282,7 +282,7 @@ vfloat32m8_t test_vfmsac_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmsac.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmsac.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfmsac_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
@@ -292,7 +292,7 @@ vfloat32m8_t test_vfmsac_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmsac.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmsac.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfmsac_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -303,7 +303,7 @@ vfloat64m1_t test_vfmsac_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmsac.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmsac.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfmsac_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -313,7 +313,7 @@ vfloat64m1_t test_vfmsac_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmsac.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmsac.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfmsac_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -324,7 +324,7 @@ vfloat64m2_t test_vfmsac_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmsac.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmsac.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfmsac_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -334,7 +334,7 @@ vfloat64m2_t test_vfmsac_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmsac.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmsac.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfmsac_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -345,7 +345,7 @@ vfloat64m4_t test_vfmsac_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmsac.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmsac.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfmsac_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -355,7 +355,7 @@ vfloat64m4_t test_vfmsac_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmsac.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmsac.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfmsac_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -366,7 +366,7 @@ vfloat64m8_t test_vfmsac_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmsac.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmsac.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfmsac_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc, double op1,
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfmsub.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfmsub.c
index b3dace5e66bd45..abc60568c3557f 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfmsub.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfmsub.c
@@ -187,7 +187,7 @@ vfloat64m8_t test_vfmsub_vf_f64m8(vfloat64m8_t acc, double op1,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmsub.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmsub.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfmsub_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -198,7 +198,7 @@ vfloat32mf2_t test_vfmsub_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmsub.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmsub.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfmsub_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -208,7 +208,7 @@ vfloat32mf2_t test_vfmsub_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmsub.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmsub.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfmsub_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -219,7 +219,7 @@ vfloat32m1_t test_vfmsub_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmsub.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmsub.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfmsub_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc, float op1,
@@ -229,7 +229,7 @@ vfloat32m1_t test_vfmsub_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmsub.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmsub.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfmsub_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -240,7 +240,7 @@ vfloat32m2_t test_vfmsub_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmsub.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmsub.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfmsub_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc, float op1,
@@ -250,7 +250,7 @@ vfloat32m2_t test_vfmsub_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmsub.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmsub.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfmsub_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
@@ -261,7 +261,7 @@ vfloat32m4_t test_vfmsub_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmsub.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmsub.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfmsub_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
@@ -271,7 +271,7 @@ vfloat32m4_t test_vfmsub_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmsub.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmsub.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfmsub_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
@@ -282,7 +282,7 @@ vfloat32m8_t test_vfmsub_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmsub.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmsub.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfmsub_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
@@ -292,7 +292,7 @@ vfloat32m8_t test_vfmsub_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmsub.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmsub.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfmsub_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -303,7 +303,7 @@ vfloat64m1_t test_vfmsub_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmsub.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmsub.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfmsub_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -313,7 +313,7 @@ vfloat64m1_t test_vfmsub_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmsub.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmsub.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfmsub_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -324,7 +324,7 @@ vfloat64m2_t test_vfmsub_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmsub.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmsub.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfmsub_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -334,7 +334,7 @@ vfloat64m2_t test_vfmsub_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmsub.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmsub.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfmsub_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -345,7 +345,7 @@ vfloat64m4_t test_vfmsub_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmsub.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmsub.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfmsub_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -355,7 +355,7 @@ vfloat64m4_t test_vfmsub_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmsub.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmsub.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfmsub_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -366,7 +366,7 @@ vfloat64m8_t test_vfmsub_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmsub.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmsub.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfmsub_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc, double op1,
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfnmacc.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfnmacc.c
index c4a534d11ad41c..c76d6b88d08694 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfnmacc.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfnmacc.c
@@ -187,7 +187,7 @@ vfloat64m8_t test_vfnmacc_vf_f64m8(vfloat64m8_t acc, double op1,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmacc.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmacc.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfnmacc_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -198,7 +198,7 @@ vfloat32mf2_t test_vfnmacc_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmacc.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmacc.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfnmacc_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -209,7 +209,7 @@ vfloat32mf2_t test_vfnmacc_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmacc.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmacc.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfnmacc_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -220,7 +220,7 @@ vfloat32m1_t test_vfnmacc_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmacc.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmacc.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfnmacc_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -230,7 +230,7 @@ vfloat32m1_t test_vfnmacc_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmacc.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmacc.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfnmacc_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -241,7 +241,7 @@ vfloat32m2_t test_vfnmacc_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmacc.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmacc.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfnmacc_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -251,7 +251,7 @@ vfloat32m2_t test_vfnmacc_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmacc.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmacc.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfnmacc_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
@@ -262,7 +262,7 @@ vfloat32m4_t test_vfnmacc_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmacc.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmacc.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfnmacc_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
@@ -272,7 +272,7 @@ vfloat32m4_t test_vfnmacc_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmacc.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmacc.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfnmacc_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
@@ -283,7 +283,7 @@ vfloat32m8_t test_vfnmacc_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmacc.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmacc.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfnmacc_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
@@ -293,7 +293,7 @@ vfloat32m8_t test_vfnmacc_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmacc.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmacc.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfnmacc_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -304,7 +304,7 @@ vfloat64m1_t test_vfnmacc_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmacc.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmacc.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfnmacc_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -314,7 +314,7 @@ vfloat64m1_t test_vfnmacc_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmacc.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmacc.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfnmacc_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -325,7 +325,7 @@ vfloat64m2_t test_vfnmacc_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmacc.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmacc.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfnmacc_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -335,7 +335,7 @@ vfloat64m2_t test_vfnmacc_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmacc.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmacc.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfnmacc_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -346,7 +346,7 @@ vfloat64m4_t test_vfnmacc_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmacc.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmacc.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfnmacc_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -356,7 +356,7 @@ vfloat64m4_t test_vfnmacc_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmacc.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmacc.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfnmacc_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -367,7 +367,7 @@ vfloat64m8_t test_vfnmacc_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmacc.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmacc.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfnmacc_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfnmadd.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfnmadd.c
index 42a257ad9c0d1c..d1e5246cd29492 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfnmadd.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfnmadd.c
@@ -187,7 +187,7 @@ vfloat64m8_t test_vfnmadd_vf_f64m8(vfloat64m8_t acc, double op1,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmadd.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmadd.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfnmadd_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -198,7 +198,7 @@ vfloat32mf2_t test_vfnmadd_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmadd.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmadd.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfnmadd_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -209,7 +209,7 @@ vfloat32mf2_t test_vfnmadd_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmadd.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmadd.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfnmadd_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -220,7 +220,7 @@ vfloat32m1_t test_vfnmadd_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmadd.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmadd.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfnmadd_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -230,7 +230,7 @@ vfloat32m1_t test_vfnmadd_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmadd.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmadd.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfnmadd_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -241,7 +241,7 @@ vfloat32m2_t test_vfnmadd_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmadd.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmadd.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfnmadd_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -251,7 +251,7 @@ vfloat32m2_t test_vfnmadd_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmadd.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmadd.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfnmadd_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
@@ -262,7 +262,7 @@ vfloat32m4_t test_vfnmadd_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmadd.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmadd.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfnmadd_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
@@ -272,7 +272,7 @@ vfloat32m4_t test_vfnmadd_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmadd.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmadd.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfnmadd_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
@@ -283,7 +283,7 @@ vfloat32m8_t test_vfnmadd_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmadd.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmadd.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfnmadd_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
@@ -293,7 +293,7 @@ vfloat32m8_t test_vfnmadd_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmadd.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmadd.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfnmadd_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -304,7 +304,7 @@ vfloat64m1_t test_vfnmadd_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmadd.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmadd.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfnmadd_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -314,7 +314,7 @@ vfloat64m1_t test_vfnmadd_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmadd.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmadd.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfnmadd_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -325,7 +325,7 @@ vfloat64m2_t test_vfnmadd_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmadd.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmadd.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfnmadd_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -335,7 +335,7 @@ vfloat64m2_t test_vfnmadd_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmadd.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmadd.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfnmadd_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -346,7 +346,7 @@ vfloat64m4_t test_vfnmadd_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmadd.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmadd.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfnmadd_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -356,7 +356,7 @@ vfloat64m4_t test_vfnmadd_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmadd.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmadd.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfnmadd_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -367,7 +367,7 @@ vfloat64m8_t test_vfnmadd_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmadd.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmadd.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfnmadd_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfnmsac.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfnmsac.c
index 8360fa5fa30065..24c2c9abbcb427 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfnmsac.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfnmsac.c
@@ -187,7 +187,7 @@ vfloat64m8_t test_vfnmsac_vf_f64m8(vfloat64m8_t acc, double op1,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmsac.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmsac.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfnmsac_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -198,7 +198,7 @@ vfloat32mf2_t test_vfnmsac_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmsac.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmsac.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfnmsac_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -209,7 +209,7 @@ vfloat32mf2_t test_vfnmsac_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmsac.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmsac.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfnmsac_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -220,7 +220,7 @@ vfloat32m1_t test_vfnmsac_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmsac.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmsac.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfnmsac_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -230,7 +230,7 @@ vfloat32m1_t test_vfnmsac_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmsac.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmsac.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfnmsac_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -241,7 +241,7 @@ vfloat32m2_t test_vfnmsac_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmsac.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmsac.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfnmsac_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -251,7 +251,7 @@ vfloat32m2_t test_vfnmsac_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmsac.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmsac.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfnmsac_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
@@ -262,7 +262,7 @@ vfloat32m4_t test_vfnmsac_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmsac.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmsac.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfnmsac_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
@@ -272,7 +272,7 @@ vfloat32m4_t test_vfnmsac_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmsac.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmsac.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfnmsac_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
@@ -283,7 +283,7 @@ vfloat32m8_t test_vfnmsac_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmsac.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmsac.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfnmsac_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
@@ -293,7 +293,7 @@ vfloat32m8_t test_vfnmsac_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmsac.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmsac.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfnmsac_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -304,7 +304,7 @@ vfloat64m1_t test_vfnmsac_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmsac.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmsac.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfnmsac_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -314,7 +314,7 @@ vfloat64m1_t test_vfnmsac_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmsac.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmsac.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfnmsac_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -325,7 +325,7 @@ vfloat64m2_t test_vfnmsac_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmsac.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmsac.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfnmsac_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -335,7 +335,7 @@ vfloat64m2_t test_vfnmsac_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmsac.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmsac.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfnmsac_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -346,7 +346,7 @@ vfloat64m4_t test_vfnmsac_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmsac.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmsac.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfnmsac_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -356,7 +356,7 @@ vfloat64m4_t test_vfnmsac_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmsac.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmsac.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfnmsac_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -367,7 +367,7 @@ vfloat64m8_t test_vfnmsac_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmsac.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmsac.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfnmsac_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfnmsub.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfnmsub.c
index b75872317f473f..c7c85314b56999 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfnmsub.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfnmsub.c
@@ -187,7 +187,7 @@ vfloat64m8_t test_vfnmsub_vf_f64m8(vfloat64m8_t acc, double op1,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmsub.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmsub.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfnmsub_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -198,7 +198,7 @@ vfloat32mf2_t test_vfnmsub_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmsub.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmsub.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfnmsub_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -209,7 +209,7 @@ vfloat32mf2_t test_vfnmsub_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmsub.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmsub.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfnmsub_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -220,7 +220,7 @@ vfloat32m1_t test_vfnmsub_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmsub.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmsub.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfnmsub_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -230,7 +230,7 @@ vfloat32m1_t test_vfnmsub_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmsub.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmsub.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfnmsub_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -241,7 +241,7 @@ vfloat32m2_t test_vfnmsub_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmsub.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmsub.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfnmsub_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -251,7 +251,7 @@ vfloat32m2_t test_vfnmsub_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmsub.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmsub.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfnmsub_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
@@ -262,7 +262,7 @@ vfloat32m4_t test_vfnmsub_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmsub.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmsub.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfnmsub_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
@@ -272,7 +272,7 @@ vfloat32m4_t test_vfnmsub_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmsub.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmsub.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfnmsub_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
@@ -283,7 +283,7 @@ vfloat32m8_t test_vfnmsub_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmsub.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmsub.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfnmsub_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
@@ -293,7 +293,7 @@ vfloat32m8_t test_vfnmsub_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmsub.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmsub.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfnmsub_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -304,7 +304,7 @@ vfloat64m1_t test_vfnmsub_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmsub.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmsub.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfnmsub_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -314,7 +314,7 @@ vfloat64m1_t test_vfnmsub_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmsub.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmsub.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfnmsub_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -325,7 +325,7 @@ vfloat64m2_t test_vfnmsub_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmsub.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmsub.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfnmsub_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -335,7 +335,7 @@ vfloat64m2_t test_vfnmsub_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmsub.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmsub.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfnmsub_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -346,7 +346,7 @@ vfloat64m4_t test_vfnmsub_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmsub.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmsub.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfnmsub_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -356,7 +356,7 @@ vfloat64m4_t test_vfnmsub_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmsub.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmsub.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfnmsub_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -367,7 +367,7 @@ vfloat64m8_t test_vfnmsub_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmsub.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmsub.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfnmsub_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfwmacc.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfwmacc.c
index 4e55316a3eeca4..462f80cc4afb12 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfwmacc.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfwmacc.c
@@ -87,7 +87,7 @@ vfloat64m8_t test_vfwmacc_vf_f64m8(vfloat64m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwmacc.mask.nxv1f64.nxv1f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwmacc.mask.nxv1f64.nxv1f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfwmacc_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -98,7 +98,7 @@ vfloat64m1_t test_vfwmacc_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwmacc.mask.nxv1f64.f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwmacc.mask.nxv1f64.f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfwmacc_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -108,7 +108,7 @@ vfloat64m1_t test_vfwmacc_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwmacc.mask.nxv2f64.nxv2f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwmacc.mask.nxv2f64.nxv2f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfwmacc_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -119,7 +119,7 @@ vfloat64m2_t test_vfwmacc_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwmacc.mask.nxv2f64.f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwmacc.mask.nxv2f64.f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfwmacc_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -129,7 +129,7 @@ vfloat64m2_t test_vfwmacc_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwmacc.mask.nxv4f64.nxv4f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwmacc.mask.nxv4f64.nxv4f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfwmacc_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -140,7 +140,7 @@ vfloat64m4_t test_vfwmacc_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwmacc.mask.nxv4f64.f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwmacc.mask.nxv4f64.f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfwmacc_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -150,7 +150,7 @@ vfloat64m4_t test_vfwmacc_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwmacc.mask.nxv8f64.nxv8f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwmacc.mask.nxv8f64.nxv8f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfwmacc_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -161,7 +161,7 @@ vfloat64m8_t test_vfwmacc_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwmacc.mask.nxv8f64.f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwmacc.mask.nxv8f64.f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfwmacc_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc, float op1,
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfwmsac.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfwmsac.c
index c4086f06beb531..1a7c37a2984496 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfwmsac.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfwmsac.c
@@ -87,7 +87,7 @@ vfloat64m8_t test_vfwmsac_vf_f64m8(vfloat64m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwmsac.mask.nxv1f64.nxv1f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwmsac.mask.nxv1f64.nxv1f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfwmsac_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -98,7 +98,7 @@ vfloat64m1_t test_vfwmsac_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwmsac.mask.nxv1f64.f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwmsac.mask.nxv1f64.f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfwmsac_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -108,7 +108,7 @@ vfloat64m1_t test_vfwmsac_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwmsac.mask.nxv2f64.nxv2f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwmsac.mask.nxv2f64.nxv2f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfwmsac_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -119,7 +119,7 @@ vfloat64m2_t test_vfwmsac_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwmsac.mask.nxv2f64.f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwmsac.mask.nxv2f64.f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfwmsac_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -129,7 +129,7 @@ vfloat64m2_t test_vfwmsac_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwmsac.mask.nxv4f64.nxv4f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwmsac.mask.nxv4f64.nxv4f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfwmsac_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -140,7 +140,7 @@ vfloat64m4_t test_vfwmsac_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwmsac.mask.nxv4f64.f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwmsac.mask.nxv4f64.f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfwmsac_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -150,7 +150,7 @@ vfloat64m4_t test_vfwmsac_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwmsac.mask.nxv8f64.nxv8f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwmsac.mask.nxv8f64.nxv8f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfwmsac_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -161,7 +161,7 @@ vfloat64m8_t test_vfwmsac_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwmsac.mask.nxv8f64.f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwmsac.mask.nxv8f64.f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfwmsac_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc, float op1,
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfwnmacc.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfwnmacc.c
index 4b5ccba44758fe..9ac61004637aec 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfwnmacc.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfwnmacc.c
@@ -87,7 +87,7 @@ vfloat64m8_t test_vfwnmacc_vf_f64m8(vfloat64m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwnmacc.mask.nxv1f64.nxv1f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwnmacc.mask.nxv1f64.nxv1f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfwnmacc_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -98,7 +98,7 @@ vfloat64m1_t test_vfwnmacc_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwnmacc.mask.nxv1f64.f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwnmacc.mask.nxv1f64.f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfwnmacc_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -108,7 +108,7 @@ vfloat64m1_t test_vfwnmacc_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwnmacc.mask.nxv2f64.nxv2f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwnmacc.mask.nxv2f64.nxv2f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfwnmacc_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -119,7 +119,7 @@ vfloat64m2_t test_vfwnmacc_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwnmacc.mask.nxv2f64.f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwnmacc.mask.nxv2f64.f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfwnmacc_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -129,7 +129,7 @@ vfloat64m2_t test_vfwnmacc_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwnmacc.mask.nxv4f64.nxv4f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwnmacc.mask.nxv4f64.nxv4f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfwnmacc_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -140,7 +140,7 @@ vfloat64m4_t test_vfwnmacc_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwnmacc.mask.nxv4f64.f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwnmacc.mask.nxv4f64.f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfwnmacc_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -150,7 +150,7 @@ vfloat64m4_t test_vfwnmacc_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwnmacc.mask.nxv8f64.nxv8f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwnmacc.mask.nxv8f64.nxv8f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfwnmacc_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -161,7 +161,7 @@ vfloat64m8_t test_vfwnmacc_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwnmacc.mask.nxv8f64.f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwnmacc.mask.nxv8f64.f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfwnmacc_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfwnmsac.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfwnmsac.c
index 16895d5f4b48d3..f237873e74a921 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfwnmsac.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vfwnmsac.c
@@ -87,7 +87,7 @@ vfloat64m8_t test_vfwnmsac_vf_f64m8(vfloat64m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwnmsac.mask.nxv1f64.nxv1f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwnmsac.mask.nxv1f64.nxv1f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfwnmsac_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -98,7 +98,7 @@ vfloat64m1_t test_vfwnmsac_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwnmsac.mask.nxv1f64.f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwnmsac.mask.nxv1f64.f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfwnmsac_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -108,7 +108,7 @@ vfloat64m1_t test_vfwnmsac_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwnmsac.mask.nxv2f64.nxv2f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwnmsac.mask.nxv2f64.nxv2f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfwnmsac_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -119,7 +119,7 @@ vfloat64m2_t test_vfwnmsac_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwnmsac.mask.nxv2f64.f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwnmsac.mask.nxv2f64.f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfwnmsac_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -129,7 +129,7 @@ vfloat64m2_t test_vfwnmsac_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwnmsac.mask.nxv4f64.nxv4f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwnmsac.mask.nxv4f64.nxv4f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfwnmsac_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -140,7 +140,7 @@ vfloat64m4_t test_vfwnmsac_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwnmsac.mask.nxv4f64.f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwnmsac.mask.nxv4f64.f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfwnmsac_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -150,7 +150,7 @@ vfloat64m4_t test_vfwnmsac_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwnmsac.mask.nxv8f64.nxv8f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwnmsac.mask.nxv8f64.nxv8f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfwnmsac_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -161,7 +161,7 @@ vfloat64m8_t test_vfwnmsac_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwnmsac.mask.nxv8f64.f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwnmsac.mask.nxv8f64.f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfwnmsac_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vmacc.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vmacc.c
index b42dd449964f48..569301d120b9b1 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vmacc.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vmacc.c
@@ -798,7 +798,7 @@ vuint64m8_t test_vmacc_vx_u64m8(vuint64m8_t acc, uint64_t op1, vuint64m8_t op2,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vint8mf8_t test_vmacc_vv_i8mf8_m(vbool64_t mask, vint8mf8_t acc, vint8mf8_t op1, vint8mf8_t op2, size_t vl) {
@@ -807,7 +807,7 @@ vint8mf8_t test_vmacc_vv_i8mf8_m(vbool64_t mask, vint8mf8_t acc, vint8mf8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vint8mf8_t test_vmacc_vx_i8mf8_m(vbool64_t mask, vint8mf8_t acc, int8_t op1, vint8mf8_t op2, size_t vl) {
@@ -816,7 +816,7 @@ vint8mf8_t test_vmacc_vx_i8mf8_m(vbool64_t mask, vint8mf8_t acc, int8_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmacc.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmacc.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vint8mf4_t test_vmacc_vv_i8mf4_m(vbool32_t mask, vint8mf4_t acc, vint8mf4_t op1, vint8mf4_t op2, size_t vl) {
@@ -825,7 +825,7 @@ vint8mf4_t test_vmacc_vv_i8mf4_m(vbool32_t mask, vint8mf4_t acc, vint8mf4_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmacc.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmacc.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vint8mf4_t test_vmacc_vx_i8mf4_m(vbool32_t mask, vint8mf4_t acc, int8_t op1, vint8mf4_t op2, size_t vl) {
@@ -834,7 +834,7 @@ vint8mf4_t test_vmacc_vx_i8mf4_m(vbool32_t mask, vint8mf4_t acc, int8_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmacc.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmacc.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vint8mf2_t test_vmacc_vv_i8mf2_m(vbool16_t mask, vint8mf2_t acc, vint8mf2_t op1, vint8mf2_t op2, size_t vl) {
@@ -843,7 +843,7 @@ vint8mf2_t test_vmacc_vv_i8mf2_m(vbool16_t mask, vint8mf2_t acc, vint8mf2_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmacc.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmacc.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vint8mf2_t test_vmacc_vx_i8mf2_m(vbool16_t mask, vint8mf2_t acc, int8_t op1, vint8mf2_t op2, size_t vl) {
@@ -852,7 +852,7 @@ vint8mf2_t test_vmacc_vx_i8mf2_m(vbool16_t mask, vint8mf2_t acc, int8_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmacc.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmacc.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vint8m1_t test_vmacc_vv_i8m1_m(vbool8_t mask, vint8m1_t acc, vint8m1_t op1, vint8m1_t op2, size_t vl) {
@@ -861,7 +861,7 @@ vint8m1_t test_vmacc_vv_i8m1_m(vbool8_t mask, vint8m1_t acc, vint8m1_t op1, vint
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmacc.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmacc.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vint8m1_t test_vmacc_vx_i8m1_m(vbool8_t mask, vint8m1_t acc, int8_t op1, vint8m1_t op2, size_t vl) {
@@ -870,7 +870,7 @@ vint8m1_t test_vmacc_vx_i8m1_m(vbool8_t mask, vint8m1_t acc, int8_t op1, vint8m1
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmacc.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmacc.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vint8m2_t test_vmacc_vv_i8m2_m(vbool4_t mask, vint8m2_t acc, vint8m2_t op1, vint8m2_t op2, size_t vl) {
@@ -879,7 +879,7 @@ vint8m2_t test_vmacc_vv_i8m2_m(vbool4_t mask, vint8m2_t acc, vint8m2_t op1, vint
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmacc.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmacc.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vint8m2_t test_vmacc_vx_i8m2_m(vbool4_t mask, vint8m2_t acc, int8_t op1, vint8m2_t op2, size_t vl) {
@@ -888,7 +888,7 @@ vint8m2_t test_vmacc_vx_i8m2_m(vbool4_t mask, vint8m2_t acc, int8_t op1, vint8m2
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmacc.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmacc.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vint8m4_t test_vmacc_vv_i8m4_m(vbool2_t mask, vint8m4_t acc, vint8m4_t op1, vint8m4_t op2, size_t vl) {
@@ -897,7 +897,7 @@ vint8m4_t test_vmacc_vv_i8m4_m(vbool2_t mask, vint8m4_t acc, vint8m4_t op1, vint
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmacc.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmacc.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vint8m4_t test_vmacc_vx_i8m4_m(vbool2_t mask, vint8m4_t acc, int8_t op1, vint8m4_t op2, size_t vl) {
@@ -906,7 +906,7 @@ vint8m4_t test_vmacc_vx_i8m4_m(vbool2_t mask, vint8m4_t acc, int8_t op1, vint8m4
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmacc.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmacc.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vint8m8_t test_vmacc_vv_i8m8_m(vbool1_t mask, vint8m8_t acc, vint8m8_t op1, vint8m8_t op2, size_t vl) {
@@ -915,7 +915,7 @@ vint8m8_t test_vmacc_vv_i8m8_m(vbool1_t mask, vint8m8_t acc, vint8m8_t op1, vint
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmacc.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmacc.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vint8m8_t test_vmacc_vx_i8m8_m(vbool1_t mask, vint8m8_t acc, int8_t op1, vint8m8_t op2, size_t vl) {
@@ -924,7 +924,7 @@ vint8m8_t test_vmacc_vx_i8m8_m(vbool1_t mask, vint8m8_t acc, int8_t op1, vint8m8
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmacc.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmacc.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vmacc_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc, vint16mf4_t op1, vint16mf4_t op2, size_t vl) {
@@ -933,7 +933,7 @@ vint16mf4_t test_vmacc_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc, vint16mf4_t
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmacc.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmacc.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vmacc_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc, int16_t op1, vint16mf4_t op2, size_t vl) {
@@ -942,7 +942,7 @@ vint16mf4_t test_vmacc_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmacc.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmacc.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vmacc_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc, vint16mf2_t op1, vint16mf2_t op2, size_t vl) {
@@ -951,7 +951,7 @@ vint16mf2_t test_vmacc_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc, vint16mf2_t
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmacc.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmacc.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vmacc_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc, int16_t op1, vint16mf2_t op2, size_t vl) {
@@ -960,7 +960,7 @@ vint16mf2_t test_vmacc_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmacc.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmacc.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vmacc_vv_i16m1_m(vbool16_t mask, vint16m1_t acc, vint16m1_t op1, vint16m1_t op2, size_t vl) {
@@ -969,7 +969,7 @@ vint16m1_t test_vmacc_vv_i16m1_m(vbool16_t mask, vint16m1_t acc, vint16m1_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmacc.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmacc.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vmacc_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int16_t op1, vint16m1_t op2, size_t vl) {
@@ -978,7 +978,7 @@ vint16m1_t test_vmacc_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int16_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmacc.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmacc.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vmacc_vv_i16m2_m(vbool8_t mask, vint16m2_t acc, vint16m2_t op1, vint16m2_t op2, size_t vl) {
@@ -987,7 +987,7 @@ vint16m2_t test_vmacc_vv_i16m2_m(vbool8_t mask, vint16m2_t acc, vint16m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmacc.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmacc.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vmacc_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int16_t op1, vint16m2_t op2, size_t vl) {
@@ -996,7 +996,7 @@ vint16m2_t test_vmacc_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int16_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmacc.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmacc.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vmacc_vv_i16m4_m(vbool4_t mask, vint16m4_t acc, vint16m4_t op1, vint16m4_t op2, size_t vl) {
@@ -1005,7 +1005,7 @@ vint16m4_t test_vmacc_vv_i16m4_m(vbool4_t mask, vint16m4_t acc, vint16m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmacc.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmacc.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vmacc_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int16_t op1, vint16m4_t op2, size_t vl) {
@@ -1014,7 +1014,7 @@ vint16m4_t test_vmacc_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int16_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmacc.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmacc.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vmacc_vv_i16m8_m(vbool2_t mask, vint16m8_t acc, vint16m8_t op1, vint16m8_t op2, size_t vl) {
@@ -1023,7 +1023,7 @@ vint16m8_t test_vmacc_vv_i16m8_m(vbool2_t mask, vint16m8_t acc, vint16m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmacc.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmacc.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vmacc_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int16_t op1, vint16m8_t op2, size_t vl) {
@@ -1032,7 +1032,7 @@ vint16m8_t test_vmacc_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int16_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmacc.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmacc.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vmacc_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc, vint32mf2_t op1, vint32mf2_t op2, size_t vl) {
@@ -1041,7 +1041,7 @@ vint32mf2_t test_vmacc_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc, vint32mf2_t
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmacc.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmacc.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vmacc_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc, int32_t op1, vint32mf2_t op2, size_t vl) {
@@ -1050,7 +1050,7 @@ vint32mf2_t test_vmacc_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc, int32_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmacc.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmacc.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vmacc_vv_i32m1_m(vbool32_t mask, vint32m1_t acc, vint32m1_t op1, vint32m1_t op2, size_t vl) {
@@ -1059,7 +1059,7 @@ vint32m1_t test_vmacc_vv_i32m1_m(vbool32_t mask, vint32m1_t acc, vint32m1_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmacc.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmacc.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vmacc_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int32_t op1, vint32m1_t op2, size_t vl) {
@@ -1068,7 +1068,7 @@ vint32m1_t test_vmacc_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int32_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmacc.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmacc.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vmacc_vv_i32m2_m(vbool16_t mask, vint32m2_t acc, vint32m2_t op1, vint32m2_t op2, size_t vl) {
@@ -1077,7 +1077,7 @@ vint32m2_t test_vmacc_vv_i32m2_m(vbool16_t mask, vint32m2_t acc, vint32m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmacc.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmacc.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vmacc_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int32_t op1, vint32m2_t op2, size_t vl) {
@@ -1086,7 +1086,7 @@ vint32m2_t test_vmacc_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int32_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmacc.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmacc.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vmacc_vv_i32m4_m(vbool8_t mask, vint32m4_t acc, vint32m4_t op1, vint32m4_t op2, size_t vl) {
@@ -1095,7 +1095,7 @@ vint32m4_t test_vmacc_vv_i32m4_m(vbool8_t mask, vint32m4_t acc, vint32m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmacc.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmacc.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vmacc_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int32_t op1, vint32m4_t op2, size_t vl) {
@@ -1104,7 +1104,7 @@ vint32m4_t test_vmacc_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int32_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmacc.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmacc.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vmacc_vv_i32m8_m(vbool4_t mask, vint32m8_t acc, vint32m8_t op1, vint32m8_t op2, size_t vl) {
@@ -1113,7 +1113,7 @@ vint32m8_t test_vmacc_vv_i32m8_m(vbool4_t mask, vint32m8_t acc, vint32m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmacc.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmacc.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vmacc_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int32_t op1, vint32m8_t op2, size_t vl) {
@@ -1122,7 +1122,7 @@ vint32m8_t test_vmacc_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int32_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmacc.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmacc.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vmacc_vv_i64m1_m(vbool64_t mask, vint64m1_t acc, vint64m1_t op1, vint64m1_t op2, size_t vl) {
@@ -1131,7 +1131,7 @@ vint64m1_t test_vmacc_vv_i64m1_m(vbool64_t mask, vint64m1_t acc, vint64m1_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmacc.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmacc.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vmacc_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int64_t op1, vint64m1_t op2, size_t vl) {
@@ -1140,7 +1140,7 @@ vint64m1_t test_vmacc_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int64_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmacc.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmacc.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vmacc_vv_i64m2_m(vbool32_t mask, vint64m2_t acc, vint64m2_t op1, vint64m2_t op2, size_t vl) {
@@ -1149,7 +1149,7 @@ vint64m2_t test_vmacc_vv_i64m2_m(vbool32_t mask, vint64m2_t acc, vint64m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmacc.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmacc.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vmacc_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int64_t op1, vint64m2_t op2, size_t vl) {
@@ -1158,7 +1158,7 @@ vint64m2_t test_vmacc_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int64_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmacc.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmacc.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vmacc_vv_i64m4_m(vbool16_t mask, vint64m4_t acc, vint64m4_t op1, vint64m4_t op2, size_t vl) {
@@ -1167,7 +1167,7 @@ vint64m4_t test_vmacc_vv_i64m4_m(vbool16_t mask, vint64m4_t acc, vint64m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmacc.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmacc.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vmacc_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int64_t op1, vint64m4_t op2, size_t vl) {
@@ -1176,7 +1176,7 @@ vint64m4_t test_vmacc_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int64_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmacc.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmacc.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vmacc_vv_i64m8_m(vbool8_t mask, vint64m8_t acc, vint64m8_t op1, vint64m8_t op2, size_t vl) {
@@ -1185,7 +1185,7 @@ vint64m8_t test_vmacc_vv_i64m8_m(vbool8_t mask, vint64m8_t acc, vint64m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmacc.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmacc.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vmacc_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int64_t op1, vint64m8_t op2, size_t vl) {
@@ -1194,7 +1194,7 @@ vint64m8_t test_vmacc_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int64_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vuint8mf8_t test_vmacc_vv_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, vuint8mf8_t op1, vuint8mf8_t op2, size_t vl) {
@@ -1203,7 +1203,7 @@ vuint8mf8_t test_vmacc_vv_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, vuint8mf8_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vuint8mf8_t test_vmacc_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, uint8_t op1, vuint8mf8_t op2, size_t vl) {
@@ -1212,7 +1212,7 @@ vuint8mf8_t test_vmacc_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmacc.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmacc.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vuint8mf4_t test_vmacc_vv_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, vuint8mf4_t op1, vuint8mf4_t op2, size_t vl) {
@@ -1221,7 +1221,7 @@ vuint8mf4_t test_vmacc_vv_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, vuint8mf4_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmacc.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmacc.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vuint8mf4_t test_vmacc_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, uint8_t op1, vuint8mf4_t op2, size_t vl) {
@@ -1230,7 +1230,7 @@ vuint8mf4_t test_vmacc_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmacc.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmacc.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vuint8mf2_t test_vmacc_vv_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, vuint8mf2_t op1, vuint8mf2_t op2, size_t vl) {
@@ -1239,7 +1239,7 @@ vuint8mf2_t test_vmacc_vv_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, vuint8mf2_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmacc.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmacc.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vuint8mf2_t test_vmacc_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, uint8_t op1, vuint8mf2_t op2, size_t vl) {
@@ -1248,7 +1248,7 @@ vuint8mf2_t test_vmacc_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmacc.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmacc.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vuint8m1_t test_vmacc_vv_u8m1_m(vbool8_t mask, vuint8m1_t acc, vuint8m1_t op1, vuint8m1_t op2, size_t vl) {
@@ -1257,7 +1257,7 @@ vuint8m1_t test_vmacc_vv_u8m1_m(vbool8_t mask, vuint8m1_t acc, vuint8m1_t op1, v
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmacc.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmacc.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vuint8m1_t test_vmacc_vx_u8m1_m(vbool8_t mask, vuint8m1_t acc, uint8_t op1, vuint8m1_t op2, size_t vl) {
@@ -1266,7 +1266,7 @@ vuint8m1_t test_vmacc_vx_u8m1_m(vbool8_t mask, vuint8m1_t acc, uint8_t op1, vuin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmacc.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmacc.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vuint8m2_t test_vmacc_vv_u8m2_m(vbool4_t mask, vuint8m2_t acc, vuint8m2_t op1, vuint8m2_t op2, size_t vl) {
@@ -1275,7 +1275,7 @@ vuint8m2_t test_vmacc_vv_u8m2_m(vbool4_t mask, vuint8m2_t acc, vuint8m2_t op1, v
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmacc.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmacc.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vuint8m2_t test_vmacc_vx_u8m2_m(vbool4_t mask, vuint8m2_t acc, uint8_t op1, vuint8m2_t op2, size_t vl) {
@@ -1284,7 +1284,7 @@ vuint8m2_t test_vmacc_vx_u8m2_m(vbool4_t mask, vuint8m2_t acc, uint8_t op1, vuin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmacc.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmacc.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vuint8m4_t test_vmacc_vv_u8m4_m(vbool2_t mask, vuint8m4_t acc, vuint8m4_t op1, vuint8m4_t op2, size_t vl) {
@@ -1293,7 +1293,7 @@ vuint8m4_t test_vmacc_vv_u8m4_m(vbool2_t mask, vuint8m4_t acc, vuint8m4_t op1, v
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmacc.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmacc.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vuint8m4_t test_vmacc_vx_u8m4_m(vbool2_t mask, vuint8m4_t acc, uint8_t op1, vuint8m4_t op2, size_t vl) {
@@ -1302,7 +1302,7 @@ vuint8m4_t test_vmacc_vx_u8m4_m(vbool2_t mask, vuint8m4_t acc, uint8_t op1, vuin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmacc.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmacc.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vuint8m8_t test_vmacc_vv_u8m8_m(vbool1_t mask, vuint8m8_t acc, vuint8m8_t op1, vuint8m8_t op2, size_t vl) {
@@ -1311,7 +1311,7 @@ vuint8m8_t test_vmacc_vv_u8m8_m(vbool1_t mask, vuint8m8_t acc, vuint8m8_t op1, v
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmacc.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmacc.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vuint8m8_t test_vmacc_vx_u8m8_m(vbool1_t mask, vuint8m8_t acc, uint8_t op1, vuint8m8_t op2, size_t vl) {
@@ -1320,7 +1320,7 @@ vuint8m8_t test_vmacc_vx_u8m8_m(vbool1_t mask, vuint8m8_t acc, uint8_t op1, vuin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmacc.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmacc.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vmacc_vv_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, vuint16mf4_t op1, vuint16mf4_t op2, size_t vl) {
@@ -1329,7 +1329,7 @@ vuint16mf4_t test_vmacc_vv_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, vuint16mf4
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmacc.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmacc.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vmacc_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, uint16_t op1, vuint16mf4_t op2, size_t vl) {
@@ -1338,7 +1338,7 @@ vuint16mf4_t test_vmacc_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, uint16_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmacc.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmacc.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vmacc_vv_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, vuint16mf2_t op1, vuint16mf2_t op2, size_t vl) {
@@ -1347,7 +1347,7 @@ vuint16mf2_t test_vmacc_vv_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, vuint16mf2
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmacc.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmacc.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vmacc_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, uint16_t op1, vuint16mf2_t op2, size_t vl) {
@@ -1356,7 +1356,7 @@ vuint16mf2_t test_vmacc_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, uint16_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmacc.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmacc.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vmacc_vv_u16m1_m(vbool16_t mask, vuint16m1_t acc, vuint16m1_t op1, vuint16m1_t op2, size_t vl) {
@@ -1365,7 +1365,7 @@ vuint16m1_t test_vmacc_vv_u16m1_m(vbool16_t mask, vuint16m1_t acc, vuint16m1_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmacc.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmacc.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vmacc_vx_u16m1_m(vbool16_t mask, vuint16m1_t acc, uint16_t op1, vuint16m1_t op2, size_t vl) {
@@ -1374,7 +1374,7 @@ vuint16m1_t test_vmacc_vx_u16m1_m(vbool16_t mask, vuint16m1_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmacc.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmacc.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vmacc_vv_u16m2_m(vbool8_t mask, vuint16m2_t acc, vuint16m2_t op1, vuint16m2_t op2, size_t vl) {
@@ -1383,7 +1383,7 @@ vuint16m2_t test_vmacc_vv_u16m2_m(vbool8_t mask, vuint16m2_t acc, vuint16m2_t op
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmacc.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmacc.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vmacc_vx_u16m2_m(vbool8_t mask, vuint16m2_t acc, uint16_t op1, vuint16m2_t op2, size_t vl) {
@@ -1392,7 +1392,7 @@ vuint16m2_t test_vmacc_vx_u16m2_m(vbool8_t mask, vuint16m2_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmacc.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmacc.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vmacc_vv_u16m4_m(vbool4_t mask, vuint16m4_t acc, vuint16m4_t op1, vuint16m4_t op2, size_t vl) {
@@ -1401,7 +1401,7 @@ vuint16m4_t test_vmacc_vv_u16m4_m(vbool4_t mask, vuint16m4_t acc, vuint16m4_t op
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmacc.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmacc.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vmacc_vx_u16m4_m(vbool4_t mask, vuint16m4_t acc, uint16_t op1, vuint16m4_t op2, size_t vl) {
@@ -1410,7 +1410,7 @@ vuint16m4_t test_vmacc_vx_u16m4_m(vbool4_t mask, vuint16m4_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmacc.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmacc.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vmacc_vv_u16m8_m(vbool2_t mask, vuint16m8_t acc, vuint16m8_t op1, vuint16m8_t op2, size_t vl) {
@@ -1419,7 +1419,7 @@ vuint16m8_t test_vmacc_vv_u16m8_m(vbool2_t mask, vuint16m8_t acc, vuint16m8_t op
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmacc.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmacc.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vmacc_vx_u16m8_m(vbool2_t mask, vuint16m8_t acc, uint16_t op1, vuint16m8_t op2, size_t vl) {
@@ -1428,7 +1428,7 @@ vuint16m8_t test_vmacc_vx_u16m8_m(vbool2_t mask, vuint16m8_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmacc.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmacc.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vmacc_vv_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, vuint32mf2_t op1, vuint32mf2_t op2, size_t vl) {
@@ -1437,7 +1437,7 @@ vuint32mf2_t test_vmacc_vv_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, vuint32mf2
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmacc.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmacc.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vmacc_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, uint32_t op1, vuint32mf2_t op2, size_t vl) {
@@ -1446,7 +1446,7 @@ vuint32mf2_t test_vmacc_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, uint32_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmacc.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmacc.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vmacc_vv_u32m1_m(vbool32_t mask, vuint32m1_t acc, vuint32m1_t op1, vuint32m1_t op2, size_t vl) {
@@ -1455,7 +1455,7 @@ vuint32m1_t test_vmacc_vv_u32m1_m(vbool32_t mask, vuint32m1_t acc, vuint32m1_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmacc.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmacc.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vmacc_vx_u32m1_m(vbool32_t mask, vuint32m1_t acc, uint32_t op1, vuint32m1_t op2, size_t vl) {
@@ -1464,7 +1464,7 @@ vuint32m1_t test_vmacc_vx_u32m1_m(vbool32_t mask, vuint32m1_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmacc.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmacc.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vmacc_vv_u32m2_m(vbool16_t mask, vuint32m2_t acc, vuint32m2_t op1, vuint32m2_t op2, size_t vl) {
@@ -1473,7 +1473,7 @@ vuint32m2_t test_vmacc_vv_u32m2_m(vbool16_t mask, vuint32m2_t acc, vuint32m2_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmacc.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmacc.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vmacc_vx_u32m2_m(vbool16_t mask, vuint32m2_t acc, uint32_t op1, vuint32m2_t op2, size_t vl) {
@@ -1482,7 +1482,7 @@ vuint32m2_t test_vmacc_vx_u32m2_m(vbool16_t mask, vuint32m2_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmacc.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmacc.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vmacc_vv_u32m4_m(vbool8_t mask, vuint32m4_t acc, vuint32m4_t op1, vuint32m4_t op2, size_t vl) {
@@ -1491,7 +1491,7 @@ vuint32m4_t test_vmacc_vv_u32m4_m(vbool8_t mask, vuint32m4_t acc, vuint32m4_t op
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmacc.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmacc.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vmacc_vx_u32m4_m(vbool8_t mask, vuint32m4_t acc, uint32_t op1, vuint32m4_t op2, size_t vl) {
@@ -1500,7 +1500,7 @@ vuint32m4_t test_vmacc_vx_u32m4_m(vbool8_t mask, vuint32m4_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmacc.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmacc.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vmacc_vv_u32m8_m(vbool4_t mask, vuint32m8_t acc, vuint32m8_t op1, vuint32m8_t op2, size_t vl) {
@@ -1509,7 +1509,7 @@ vuint32m8_t test_vmacc_vv_u32m8_m(vbool4_t mask, vuint32m8_t acc, vuint32m8_t op
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmacc.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmacc.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vmacc_vx_u32m8_m(vbool4_t mask, vuint32m8_t acc, uint32_t op1, vuint32m8_t op2, size_t vl) {
@@ -1518,7 +1518,7 @@ vuint32m8_t test_vmacc_vx_u32m8_m(vbool4_t mask, vuint32m8_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmacc.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmacc.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vmacc_vv_u64m1_m(vbool64_t mask, vuint64m1_t acc, vuint64m1_t op1, vuint64m1_t op2, size_t vl) {
@@ -1527,7 +1527,7 @@ vuint64m1_t test_vmacc_vv_u64m1_m(vbool64_t mask, vuint64m1_t acc, vuint64m1_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmacc.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmacc.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vmacc_vx_u64m1_m(vbool64_t mask, vuint64m1_t acc, uint64_t op1, vuint64m1_t op2, size_t vl) {
@@ -1536,7 +1536,7 @@ vuint64m1_t test_vmacc_vx_u64m1_m(vbool64_t mask, vuint64m1_t acc, uint64_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmacc.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmacc.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vmacc_vv_u64m2_m(vbool32_t mask, vuint64m2_t acc, vuint64m2_t op1, vuint64m2_t op2, size_t vl) {
@@ -1545,7 +1545,7 @@ vuint64m2_t test_vmacc_vv_u64m2_m(vbool32_t mask, vuint64m2_t acc, vuint64m2_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmacc.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmacc.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vmacc_vx_u64m2_m(vbool32_t mask, vuint64m2_t acc, uint64_t op1, vuint64m2_t op2, size_t vl) {
@@ -1554,7 +1554,7 @@ vuint64m2_t test_vmacc_vx_u64m2_m(vbool32_t mask, vuint64m2_t acc, uint64_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmacc.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmacc.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vmacc_vv_u64m4_m(vbool16_t mask, vuint64m4_t acc, vuint64m4_t op1, vuint64m4_t op2, size_t vl) {
@@ -1563,7 +1563,7 @@ vuint64m4_t test_vmacc_vv_u64m4_m(vbool16_t mask, vuint64m4_t acc, vuint64m4_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmacc.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmacc.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vmacc_vx_u64m4_m(vbool16_t mask, vuint64m4_t acc, uint64_t op1, vuint64m4_t op2, size_t vl) {
@@ -1572,7 +1572,7 @@ vuint64m4_t test_vmacc_vx_u64m4_m(vbool16_t mask, vuint64m4_t acc, uint64_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmacc.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmacc.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vmacc_vv_u64m8_m(vbool8_t mask, vuint64m8_t acc, vuint64m8_t op1, vuint64m8_t op2, size_t vl) {
@@ -1581,7 +1581,7 @@ vuint64m8_t test_vmacc_vv_u64m8_m(vbool8_t mask, vuint64m8_t acc, vuint64m8_t op
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmacc.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmacc.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vmacc_vx_u64m8_m(vbool8_t mask, vuint64m8_t acc, uint64_t op1, vuint64m8_t op2, size_t vl) {
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vmadd.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vmadd.c
index 197fdce0d4f119..e6ece6f876b2a3 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vmadd.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vmadd.c
@@ -798,7 +798,7 @@ vuint64m8_t test_vmadd_vx_u64m8(vuint64m8_t acc, uint64_t op1, vuint64m8_t op2,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmadd.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmadd.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vint8mf8_t test_vmadd_vv_i8mf8_m(vbool64_t mask, vint8mf8_t acc, vint8mf8_t op1, vint8mf8_t op2, size_t vl) {
@@ -807,7 +807,7 @@ vint8mf8_t test_vmadd_vv_i8mf8_m(vbool64_t mask, vint8mf8_t acc, vint8mf8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmadd.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmadd.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vint8mf8_t test_vmadd_vx_i8mf8_m(vbool64_t mask, vint8mf8_t acc, int8_t op1, vint8mf8_t op2, size_t vl) {
@@ -816,7 +816,7 @@ vint8mf8_t test_vmadd_vx_i8mf8_m(vbool64_t mask, vint8mf8_t acc, int8_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmadd.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmadd.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vint8mf4_t test_vmadd_vv_i8mf4_m(vbool32_t mask, vint8mf4_t acc, vint8mf4_t op1, vint8mf4_t op2, size_t vl) {
@@ -825,7 +825,7 @@ vint8mf4_t test_vmadd_vv_i8mf4_m(vbool32_t mask, vint8mf4_t acc, vint8mf4_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmadd.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmadd.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vint8mf4_t test_vmadd_vx_i8mf4_m(vbool32_t mask, vint8mf4_t acc, int8_t op1, vint8mf4_t op2, size_t vl) {
@@ -834,7 +834,7 @@ vint8mf4_t test_vmadd_vx_i8mf4_m(vbool32_t mask, vint8mf4_t acc, int8_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmadd.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmadd.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vint8mf2_t test_vmadd_vv_i8mf2_m(vbool16_t mask, vint8mf2_t acc, vint8mf2_t op1, vint8mf2_t op2, size_t vl) {
@@ -843,7 +843,7 @@ vint8mf2_t test_vmadd_vv_i8mf2_m(vbool16_t mask, vint8mf2_t acc, vint8mf2_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmadd.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmadd.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vint8mf2_t test_vmadd_vx_i8mf2_m(vbool16_t mask, vint8mf2_t acc, int8_t op1, vint8mf2_t op2, size_t vl) {
@@ -852,7 +852,7 @@ vint8mf2_t test_vmadd_vx_i8mf2_m(vbool16_t mask, vint8mf2_t acc, int8_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmadd.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmadd.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vint8m1_t test_vmadd_vv_i8m1_m(vbool8_t mask, vint8m1_t acc, vint8m1_t op1, vint8m1_t op2, size_t vl) {
@@ -861,7 +861,7 @@ vint8m1_t test_vmadd_vv_i8m1_m(vbool8_t mask, vint8m1_t acc, vint8m1_t op1, vint
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmadd.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmadd.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vint8m1_t test_vmadd_vx_i8m1_m(vbool8_t mask, vint8m1_t acc, int8_t op1, vint8m1_t op2, size_t vl) {
@@ -870,7 +870,7 @@ vint8m1_t test_vmadd_vx_i8m1_m(vbool8_t mask, vint8m1_t acc, int8_t op1, vint8m1
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmadd.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmadd.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vint8m2_t test_vmadd_vv_i8m2_m(vbool4_t mask, vint8m2_t acc, vint8m2_t op1, vint8m2_t op2, size_t vl) {
@@ -879,7 +879,7 @@ vint8m2_t test_vmadd_vv_i8m2_m(vbool4_t mask, vint8m2_t acc, vint8m2_t op1, vint
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmadd.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmadd.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vint8m2_t test_vmadd_vx_i8m2_m(vbool4_t mask, vint8m2_t acc, int8_t op1, vint8m2_t op2, size_t vl) {
@@ -888,7 +888,7 @@ vint8m2_t test_vmadd_vx_i8m2_m(vbool4_t mask, vint8m2_t acc, int8_t op1, vint8m2
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmadd.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmadd.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vint8m4_t test_vmadd_vv_i8m4_m(vbool2_t mask, vint8m4_t acc, vint8m4_t op1, vint8m4_t op2, size_t vl) {
@@ -897,7 +897,7 @@ vint8m4_t test_vmadd_vv_i8m4_m(vbool2_t mask, vint8m4_t acc, vint8m4_t op1, vint
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmadd.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmadd.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vint8m4_t test_vmadd_vx_i8m4_m(vbool2_t mask, vint8m4_t acc, int8_t op1, vint8m4_t op2, size_t vl) {
@@ -906,7 +906,7 @@ vint8m4_t test_vmadd_vx_i8m4_m(vbool2_t mask, vint8m4_t acc, int8_t op1, vint8m4
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmadd.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmadd.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vint8m8_t test_vmadd_vv_i8m8_m(vbool1_t mask, vint8m8_t acc, vint8m8_t op1, vint8m8_t op2, size_t vl) {
@@ -915,7 +915,7 @@ vint8m8_t test_vmadd_vv_i8m8_m(vbool1_t mask, vint8m8_t acc, vint8m8_t op1, vint
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmadd.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmadd.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vint8m8_t test_vmadd_vx_i8m8_m(vbool1_t mask, vint8m8_t acc, int8_t op1, vint8m8_t op2, size_t vl) {
@@ -924,7 +924,7 @@ vint8m8_t test_vmadd_vx_i8m8_m(vbool1_t mask, vint8m8_t acc, int8_t op1, vint8m8
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmadd.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmadd.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vmadd_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc, vint16mf4_t op1, vint16mf4_t op2, size_t vl) {
@@ -933,7 +933,7 @@ vint16mf4_t test_vmadd_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc, vint16mf4_t
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmadd.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmadd.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vmadd_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc, int16_t op1, vint16mf4_t op2, size_t vl) {
@@ -942,7 +942,7 @@ vint16mf4_t test_vmadd_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmadd.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmadd.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vmadd_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc, vint16mf2_t op1, vint16mf2_t op2, size_t vl) {
@@ -951,7 +951,7 @@ vint16mf2_t test_vmadd_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc, vint16mf2_t
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmadd.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmadd.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vmadd_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc, int16_t op1, vint16mf2_t op2, size_t vl) {
@@ -960,7 +960,7 @@ vint16mf2_t test_vmadd_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmadd.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmadd.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vmadd_vv_i16m1_m(vbool16_t mask, vint16m1_t acc, vint16m1_t op1, vint16m1_t op2, size_t vl) {
@@ -969,7 +969,7 @@ vint16m1_t test_vmadd_vv_i16m1_m(vbool16_t mask, vint16m1_t acc, vint16m1_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmadd.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmadd.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vmadd_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int16_t op1, vint16m1_t op2, size_t vl) {
@@ -978,7 +978,7 @@ vint16m1_t test_vmadd_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int16_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmadd.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmadd.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vmadd_vv_i16m2_m(vbool8_t mask, vint16m2_t acc, vint16m2_t op1, vint16m2_t op2, size_t vl) {
@@ -987,7 +987,7 @@ vint16m2_t test_vmadd_vv_i16m2_m(vbool8_t mask, vint16m2_t acc, vint16m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmadd.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmadd.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vmadd_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int16_t op1, vint16m2_t op2, size_t vl) {
@@ -996,7 +996,7 @@ vint16m2_t test_vmadd_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int16_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmadd.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmadd.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vmadd_vv_i16m4_m(vbool4_t mask, vint16m4_t acc, vint16m4_t op1, vint16m4_t op2, size_t vl) {
@@ -1005,7 +1005,7 @@ vint16m4_t test_vmadd_vv_i16m4_m(vbool4_t mask, vint16m4_t acc, vint16m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmadd.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmadd.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vmadd_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int16_t op1, vint16m4_t op2, size_t vl) {
@@ -1014,7 +1014,7 @@ vint16m4_t test_vmadd_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int16_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmadd.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmadd.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vmadd_vv_i16m8_m(vbool2_t mask, vint16m8_t acc, vint16m8_t op1, vint16m8_t op2, size_t vl) {
@@ -1023,7 +1023,7 @@ vint16m8_t test_vmadd_vv_i16m8_m(vbool2_t mask, vint16m8_t acc, vint16m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmadd.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmadd.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vmadd_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int16_t op1, vint16m8_t op2, size_t vl) {
@@ -1032,7 +1032,7 @@ vint16m8_t test_vmadd_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int16_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmadd.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmadd.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vmadd_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc, vint32mf2_t op1, vint32mf2_t op2, size_t vl) {
@@ -1041,7 +1041,7 @@ vint32mf2_t test_vmadd_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc, vint32mf2_t
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmadd.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmadd.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vmadd_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc, int32_t op1, vint32mf2_t op2, size_t vl) {
@@ -1050,7 +1050,7 @@ vint32mf2_t test_vmadd_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc, int32_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmadd.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmadd.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vmadd_vv_i32m1_m(vbool32_t mask, vint32m1_t acc, vint32m1_t op1, vint32m1_t op2, size_t vl) {
@@ -1059,7 +1059,7 @@ vint32m1_t test_vmadd_vv_i32m1_m(vbool32_t mask, vint32m1_t acc, vint32m1_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmadd.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmadd.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vmadd_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int32_t op1, vint32m1_t op2, size_t vl) {
@@ -1068,7 +1068,7 @@ vint32m1_t test_vmadd_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int32_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmadd.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmadd.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vmadd_vv_i32m2_m(vbool16_t mask, vint32m2_t acc, vint32m2_t op1, vint32m2_t op2, size_t vl) {
@@ -1077,7 +1077,7 @@ vint32m2_t test_vmadd_vv_i32m2_m(vbool16_t mask, vint32m2_t acc, vint32m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmadd.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmadd.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vmadd_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int32_t op1, vint32m2_t op2, size_t vl) {
@@ -1086,7 +1086,7 @@ vint32m2_t test_vmadd_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int32_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmadd.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmadd.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vmadd_vv_i32m4_m(vbool8_t mask, vint32m4_t acc, vint32m4_t op1, vint32m4_t op2, size_t vl) {
@@ -1095,7 +1095,7 @@ vint32m4_t test_vmadd_vv_i32m4_m(vbool8_t mask, vint32m4_t acc, vint32m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmadd.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmadd.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vmadd_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int32_t op1, vint32m4_t op2, size_t vl) {
@@ -1104,7 +1104,7 @@ vint32m4_t test_vmadd_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int32_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmadd.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmadd.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vmadd_vv_i32m8_m(vbool4_t mask, vint32m8_t acc, vint32m8_t op1, vint32m8_t op2, size_t vl) {
@@ -1113,7 +1113,7 @@ vint32m8_t test_vmadd_vv_i32m8_m(vbool4_t mask, vint32m8_t acc, vint32m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmadd.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmadd.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vmadd_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int32_t op1, vint32m8_t op2, size_t vl) {
@@ -1122,7 +1122,7 @@ vint32m8_t test_vmadd_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int32_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmadd.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmadd.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vmadd_vv_i64m1_m(vbool64_t mask, vint64m1_t acc, vint64m1_t op1, vint64m1_t op2, size_t vl) {
@@ -1131,7 +1131,7 @@ vint64m1_t test_vmadd_vv_i64m1_m(vbool64_t mask, vint64m1_t acc, vint64m1_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmadd.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmadd.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vmadd_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int64_t op1, vint64m1_t op2, size_t vl) {
@@ -1140,7 +1140,7 @@ vint64m1_t test_vmadd_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int64_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmadd.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmadd.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vmadd_vv_i64m2_m(vbool32_t mask, vint64m2_t acc, vint64m2_t op1, vint64m2_t op2, size_t vl) {
@@ -1149,7 +1149,7 @@ vint64m2_t test_vmadd_vv_i64m2_m(vbool32_t mask, vint64m2_t acc, vint64m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmadd.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmadd.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vmadd_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int64_t op1, vint64m2_t op2, size_t vl) {
@@ -1158,7 +1158,7 @@ vint64m2_t test_vmadd_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int64_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmadd.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmadd.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vmadd_vv_i64m4_m(vbool16_t mask, vint64m4_t acc, vint64m4_t op1, vint64m4_t op2, size_t vl) {
@@ -1167,7 +1167,7 @@ vint64m4_t test_vmadd_vv_i64m4_m(vbool16_t mask, vint64m4_t acc, vint64m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmadd.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmadd.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vmadd_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int64_t op1, vint64m4_t op2, size_t vl) {
@@ -1176,7 +1176,7 @@ vint64m4_t test_vmadd_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int64_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmadd.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmadd.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vmadd_vv_i64m8_m(vbool8_t mask, vint64m8_t acc, vint64m8_t op1, vint64m8_t op2, size_t vl) {
@@ -1185,7 +1185,7 @@ vint64m8_t test_vmadd_vv_i64m8_m(vbool8_t mask, vint64m8_t acc, vint64m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmadd.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmadd.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vmadd_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int64_t op1, vint64m8_t op2, size_t vl) {
@@ -1194,7 +1194,7 @@ vint64m8_t test_vmadd_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int64_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmadd.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmadd.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vuint8mf8_t test_vmadd_vv_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, vuint8mf8_t op1, vuint8mf8_t op2, size_t vl) {
@@ -1203,7 +1203,7 @@ vuint8mf8_t test_vmadd_vv_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, vuint8mf8_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmadd.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmadd.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vuint8mf8_t test_vmadd_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, uint8_t op1, vuint8mf8_t op2, size_t vl) {
@@ -1212,7 +1212,7 @@ vuint8mf8_t test_vmadd_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmadd.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmadd.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vuint8mf4_t test_vmadd_vv_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, vuint8mf4_t op1, vuint8mf4_t op2, size_t vl) {
@@ -1221,7 +1221,7 @@ vuint8mf4_t test_vmadd_vv_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, vuint8mf4_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmadd.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmadd.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vuint8mf4_t test_vmadd_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, uint8_t op1, vuint8mf4_t op2, size_t vl) {
@@ -1230,7 +1230,7 @@ vuint8mf4_t test_vmadd_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmadd.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmadd.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vuint8mf2_t test_vmadd_vv_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, vuint8mf2_t op1, vuint8mf2_t op2, size_t vl) {
@@ -1239,7 +1239,7 @@ vuint8mf2_t test_vmadd_vv_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, vuint8mf2_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmadd.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmadd.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vuint8mf2_t test_vmadd_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, uint8_t op1, vuint8mf2_t op2, size_t vl) {
@@ -1248,7 +1248,7 @@ vuint8mf2_t test_vmadd_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmadd.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmadd.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vuint8m1_t test_vmadd_vv_u8m1_m(vbool8_t mask, vuint8m1_t acc, vuint8m1_t op1, vuint8m1_t op2, size_t vl) {
@@ -1257,7 +1257,7 @@ vuint8m1_t test_vmadd_vv_u8m1_m(vbool8_t mask, vuint8m1_t acc, vuint8m1_t op1, v
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmadd.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmadd.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vuint8m1_t test_vmadd_vx_u8m1_m(vbool8_t mask, vuint8m1_t acc, uint8_t op1, vuint8m1_t op2, size_t vl) {
@@ -1266,7 +1266,7 @@ vuint8m1_t test_vmadd_vx_u8m1_m(vbool8_t mask, vuint8m1_t acc, uint8_t op1, vuin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmadd.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmadd.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vuint8m2_t test_vmadd_vv_u8m2_m(vbool4_t mask, vuint8m2_t acc, vuint8m2_t op1, vuint8m2_t op2, size_t vl) {
@@ -1275,7 +1275,7 @@ vuint8m2_t test_vmadd_vv_u8m2_m(vbool4_t mask, vuint8m2_t acc, vuint8m2_t op1, v
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmadd.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmadd.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vuint8m2_t test_vmadd_vx_u8m2_m(vbool4_t mask, vuint8m2_t acc, uint8_t op1, vuint8m2_t op2, size_t vl) {
@@ -1284,7 +1284,7 @@ vuint8m2_t test_vmadd_vx_u8m2_m(vbool4_t mask, vuint8m2_t acc, uint8_t op1, vuin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmadd.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmadd.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vuint8m4_t test_vmadd_vv_u8m4_m(vbool2_t mask, vuint8m4_t acc, vuint8m4_t op1, vuint8m4_t op2, size_t vl) {
@@ -1293,7 +1293,7 @@ vuint8m4_t test_vmadd_vv_u8m4_m(vbool2_t mask, vuint8m4_t acc, vuint8m4_t op1, v
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmadd.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmadd.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vuint8m4_t test_vmadd_vx_u8m4_m(vbool2_t mask, vuint8m4_t acc, uint8_t op1, vuint8m4_t op2, size_t vl) {
@@ -1302,7 +1302,7 @@ vuint8m4_t test_vmadd_vx_u8m4_m(vbool2_t mask, vuint8m4_t acc, uint8_t op1, vuin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmadd.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmadd.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vuint8m8_t test_vmadd_vv_u8m8_m(vbool1_t mask, vuint8m8_t acc, vuint8m8_t op1, vuint8m8_t op2, size_t vl) {
@@ -1311,7 +1311,7 @@ vuint8m8_t test_vmadd_vv_u8m8_m(vbool1_t mask, vuint8m8_t acc, vuint8m8_t op1, v
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmadd.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmadd.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vuint8m8_t test_vmadd_vx_u8m8_m(vbool1_t mask, vuint8m8_t acc, uint8_t op1, vuint8m8_t op2, size_t vl) {
@@ -1320,7 +1320,7 @@ vuint8m8_t test_vmadd_vx_u8m8_m(vbool1_t mask, vuint8m8_t acc, uint8_t op1, vuin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmadd.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmadd.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vmadd_vv_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, vuint16mf4_t op1, vuint16mf4_t op2, size_t vl) {
@@ -1329,7 +1329,7 @@ vuint16mf4_t test_vmadd_vv_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, vuint16mf4
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmadd.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmadd.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vmadd_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, uint16_t op1, vuint16mf4_t op2, size_t vl) {
@@ -1338,7 +1338,7 @@ vuint16mf4_t test_vmadd_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, uint16_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmadd.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmadd.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vmadd_vv_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, vuint16mf2_t op1, vuint16mf2_t op2, size_t vl) {
@@ -1347,7 +1347,7 @@ vuint16mf2_t test_vmadd_vv_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, vuint16mf2
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmadd.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmadd.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vmadd_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, uint16_t op1, vuint16mf2_t op2, size_t vl) {
@@ -1356,7 +1356,7 @@ vuint16mf2_t test_vmadd_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, uint16_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmadd.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmadd.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vmadd_vv_u16m1_m(vbool16_t mask, vuint16m1_t acc, vuint16m1_t op1, vuint16m1_t op2, size_t vl) {
@@ -1365,7 +1365,7 @@ vuint16m1_t test_vmadd_vv_u16m1_m(vbool16_t mask, vuint16m1_t acc, vuint16m1_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmadd.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmadd.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vmadd_vx_u16m1_m(vbool16_t mask, vuint16m1_t acc, uint16_t op1, vuint16m1_t op2, size_t vl) {
@@ -1374,7 +1374,7 @@ vuint16m1_t test_vmadd_vx_u16m1_m(vbool16_t mask, vuint16m1_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmadd.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmadd.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vmadd_vv_u16m2_m(vbool8_t mask, vuint16m2_t acc, vuint16m2_t op1, vuint16m2_t op2, size_t vl) {
@@ -1383,7 +1383,7 @@ vuint16m2_t test_vmadd_vv_u16m2_m(vbool8_t mask, vuint16m2_t acc, vuint16m2_t op
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmadd.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmadd.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vmadd_vx_u16m2_m(vbool8_t mask, vuint16m2_t acc, uint16_t op1, vuint16m2_t op2, size_t vl) {
@@ -1392,7 +1392,7 @@ vuint16m2_t test_vmadd_vx_u16m2_m(vbool8_t mask, vuint16m2_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmadd.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmadd.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vmadd_vv_u16m4_m(vbool4_t mask, vuint16m4_t acc, vuint16m4_t op1, vuint16m4_t op2, size_t vl) {
@@ -1401,7 +1401,7 @@ vuint16m4_t test_vmadd_vv_u16m4_m(vbool4_t mask, vuint16m4_t acc, vuint16m4_t op
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmadd.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmadd.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vmadd_vx_u16m4_m(vbool4_t mask, vuint16m4_t acc, uint16_t op1, vuint16m4_t op2, size_t vl) {
@@ -1410,7 +1410,7 @@ vuint16m4_t test_vmadd_vx_u16m4_m(vbool4_t mask, vuint16m4_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmadd.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmadd.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vmadd_vv_u16m8_m(vbool2_t mask, vuint16m8_t acc, vuint16m8_t op1, vuint16m8_t op2, size_t vl) {
@@ -1419,7 +1419,7 @@ vuint16m8_t test_vmadd_vv_u16m8_m(vbool2_t mask, vuint16m8_t acc, vuint16m8_t op
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmadd.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmadd.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vmadd_vx_u16m8_m(vbool2_t mask, vuint16m8_t acc, uint16_t op1, vuint16m8_t op2, size_t vl) {
@@ -1428,7 +1428,7 @@ vuint16m8_t test_vmadd_vx_u16m8_m(vbool2_t mask, vuint16m8_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmadd.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmadd.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vmadd_vv_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, vuint32mf2_t op1, vuint32mf2_t op2, size_t vl) {
@@ -1437,7 +1437,7 @@ vuint32mf2_t test_vmadd_vv_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, vuint32mf2
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmadd.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmadd.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vmadd_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, uint32_t op1, vuint32mf2_t op2, size_t vl) {
@@ -1446,7 +1446,7 @@ vuint32mf2_t test_vmadd_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, uint32_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmadd.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmadd.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vmadd_vv_u32m1_m(vbool32_t mask, vuint32m1_t acc, vuint32m1_t op1, vuint32m1_t op2, size_t vl) {
@@ -1455,7 +1455,7 @@ vuint32m1_t test_vmadd_vv_u32m1_m(vbool32_t mask, vuint32m1_t acc, vuint32m1_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmadd.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmadd.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vmadd_vx_u32m1_m(vbool32_t mask, vuint32m1_t acc, uint32_t op1, vuint32m1_t op2, size_t vl) {
@@ -1464,7 +1464,7 @@ vuint32m1_t test_vmadd_vx_u32m1_m(vbool32_t mask, vuint32m1_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmadd.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmadd.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vmadd_vv_u32m2_m(vbool16_t mask, vuint32m2_t acc, vuint32m2_t op1, vuint32m2_t op2, size_t vl) {
@@ -1473,7 +1473,7 @@ vuint32m2_t test_vmadd_vv_u32m2_m(vbool16_t mask, vuint32m2_t acc, vuint32m2_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmadd.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmadd.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vmadd_vx_u32m2_m(vbool16_t mask, vuint32m2_t acc, uint32_t op1, vuint32m2_t op2, size_t vl) {
@@ -1482,7 +1482,7 @@ vuint32m2_t test_vmadd_vx_u32m2_m(vbool16_t mask, vuint32m2_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmadd.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmadd.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vmadd_vv_u32m4_m(vbool8_t mask, vuint32m4_t acc, vuint32m4_t op1, vuint32m4_t op2, size_t vl) {
@@ -1491,7 +1491,7 @@ vuint32m4_t test_vmadd_vv_u32m4_m(vbool8_t mask, vuint32m4_t acc, vuint32m4_t op
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmadd.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmadd.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vmadd_vx_u32m4_m(vbool8_t mask, vuint32m4_t acc, uint32_t op1, vuint32m4_t op2, size_t vl) {
@@ -1500,7 +1500,7 @@ vuint32m4_t test_vmadd_vx_u32m4_m(vbool8_t mask, vuint32m4_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmadd.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmadd.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vmadd_vv_u32m8_m(vbool4_t mask, vuint32m8_t acc, vuint32m8_t op1, vuint32m8_t op2, size_t vl) {
@@ -1509,7 +1509,7 @@ vuint32m8_t test_vmadd_vv_u32m8_m(vbool4_t mask, vuint32m8_t acc, vuint32m8_t op
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmadd.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmadd.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vmadd_vx_u32m8_m(vbool4_t mask, vuint32m8_t acc, uint32_t op1, vuint32m8_t op2, size_t vl) {
@@ -1518,7 +1518,7 @@ vuint32m8_t test_vmadd_vx_u32m8_m(vbool4_t mask, vuint32m8_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmadd.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmadd.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vmadd_vv_u64m1_m(vbool64_t mask, vuint64m1_t acc, vuint64m1_t op1, vuint64m1_t op2, size_t vl) {
@@ -1527,7 +1527,7 @@ vuint64m1_t test_vmadd_vv_u64m1_m(vbool64_t mask, vuint64m1_t acc, vuint64m1_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmadd.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmadd.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vmadd_vx_u64m1_m(vbool64_t mask, vuint64m1_t acc, uint64_t op1, vuint64m1_t op2, size_t vl) {
@@ -1536,7 +1536,7 @@ vuint64m1_t test_vmadd_vx_u64m1_m(vbool64_t mask, vuint64m1_t acc, uint64_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmadd.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmadd.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vmadd_vv_u64m2_m(vbool32_t mask, vuint64m2_t acc, vuint64m2_t op1, vuint64m2_t op2, size_t vl) {
@@ -1545,7 +1545,7 @@ vuint64m2_t test_vmadd_vv_u64m2_m(vbool32_t mask, vuint64m2_t acc, vuint64m2_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmadd.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmadd.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vmadd_vx_u64m2_m(vbool32_t mask, vuint64m2_t acc, uint64_t op1, vuint64m2_t op2, size_t vl) {
@@ -1554,7 +1554,7 @@ vuint64m2_t test_vmadd_vx_u64m2_m(vbool32_t mask, vuint64m2_t acc, uint64_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmadd.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmadd.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vmadd_vv_u64m4_m(vbool16_t mask, vuint64m4_t acc, vuint64m4_t op1, vuint64m4_t op2, size_t vl) {
@@ -1563,7 +1563,7 @@ vuint64m4_t test_vmadd_vv_u64m4_m(vbool16_t mask, vuint64m4_t acc, vuint64m4_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmadd.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmadd.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vmadd_vx_u64m4_m(vbool16_t mask, vuint64m4_t acc, uint64_t op1, vuint64m4_t op2, size_t vl) {
@@ -1572,7 +1572,7 @@ vuint64m4_t test_vmadd_vx_u64m4_m(vbool16_t mask, vuint64m4_t acc, uint64_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmadd.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmadd.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vmadd_vv_u64m8_m(vbool8_t mask, vuint64m8_t acc, vuint64m8_t op1, vuint64m8_t op2, size_t vl) {
@@ -1581,7 +1581,7 @@ vuint64m8_t test_vmadd_vv_u64m8_m(vbool8_t mask, vuint64m8_t acc, vuint64m8_t op
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmadd.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmadd.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vmadd_vx_u64m8_m(vbool8_t mask, vuint64m8_t acc, uint64_t op1, vuint64m8_t op2, size_t vl) {
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vnmsac.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vnmsac.c
index 9c0683af889dc6..211dcf4d636bac 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vnmsac.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vnmsac.c
@@ -798,7 +798,7 @@ vuint64m8_t test_vnmsac_vx_u64m8(vuint64m8_t acc, uint64_t op1, vuint64m8_t op2,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsac.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsac.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vint8mf8_t test_vnmsac_vv_i8mf8_m(vbool64_t mask, vint8mf8_t acc, vint8mf8_t op1, vint8mf8_t op2, size_t vl) {
@@ -807,7 +807,7 @@ vint8mf8_t test_vnmsac_vv_i8mf8_m(vbool64_t mask, vint8mf8_t acc, vint8mf8_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsac.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsac.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vint8mf8_t test_vnmsac_vx_i8mf8_m(vbool64_t mask, vint8mf8_t acc, int8_t op1, vint8mf8_t op2, size_t vl) {
@@ -816,7 +816,7 @@ vint8mf8_t test_vnmsac_vx_i8mf8_m(vbool64_t mask, vint8mf8_t acc, int8_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsac.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsac.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vint8mf4_t test_vnmsac_vv_i8mf4_m(vbool32_t mask, vint8mf4_t acc, vint8mf4_t op1, vint8mf4_t op2, size_t vl) {
@@ -825,7 +825,7 @@ vint8mf4_t test_vnmsac_vv_i8mf4_m(vbool32_t mask, vint8mf4_t acc, vint8mf4_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsac.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsac.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vint8mf4_t test_vnmsac_vx_i8mf4_m(vbool32_t mask, vint8mf4_t acc, int8_t op1, vint8mf4_t op2, size_t vl) {
@@ -834,7 +834,7 @@ vint8mf4_t test_vnmsac_vx_i8mf4_m(vbool32_t mask, vint8mf4_t acc, int8_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsac.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsac.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vint8mf2_t test_vnmsac_vv_i8mf2_m(vbool16_t mask, vint8mf2_t acc, vint8mf2_t op1, vint8mf2_t op2, size_t vl) {
@@ -843,7 +843,7 @@ vint8mf2_t test_vnmsac_vv_i8mf2_m(vbool16_t mask, vint8mf2_t acc, vint8mf2_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsac.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsac.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vint8mf2_t test_vnmsac_vx_i8mf2_m(vbool16_t mask, vint8mf2_t acc, int8_t op1, vint8mf2_t op2, size_t vl) {
@@ -852,7 +852,7 @@ vint8mf2_t test_vnmsac_vx_i8mf2_m(vbool16_t mask, vint8mf2_t acc, int8_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsac.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsac.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vint8m1_t test_vnmsac_vv_i8m1_m(vbool8_t mask, vint8m1_t acc, vint8m1_t op1, vint8m1_t op2, size_t vl) {
@@ -861,7 +861,7 @@ vint8m1_t test_vnmsac_vv_i8m1_m(vbool8_t mask, vint8m1_t acc, vint8m1_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsac.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsac.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vint8m1_t test_vnmsac_vx_i8m1_m(vbool8_t mask, vint8m1_t acc, int8_t op1, vint8m1_t op2, size_t vl) {
@@ -870,7 +870,7 @@ vint8m1_t test_vnmsac_vx_i8m1_m(vbool8_t mask, vint8m1_t acc, int8_t op1, vint8m
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsac.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsac.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vint8m2_t test_vnmsac_vv_i8m2_m(vbool4_t mask, vint8m2_t acc, vint8m2_t op1, vint8m2_t op2, size_t vl) {
@@ -879,7 +879,7 @@ vint8m2_t test_vnmsac_vv_i8m2_m(vbool4_t mask, vint8m2_t acc, vint8m2_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsac.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsac.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vint8m2_t test_vnmsac_vx_i8m2_m(vbool4_t mask, vint8m2_t acc, int8_t op1, vint8m2_t op2, size_t vl) {
@@ -888,7 +888,7 @@ vint8m2_t test_vnmsac_vx_i8m2_m(vbool4_t mask, vint8m2_t acc, int8_t op1, vint8m
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsac.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsac.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vint8m4_t test_vnmsac_vv_i8m4_m(vbool2_t mask, vint8m4_t acc, vint8m4_t op1, vint8m4_t op2, size_t vl) {
@@ -897,7 +897,7 @@ vint8m4_t test_vnmsac_vv_i8m4_m(vbool2_t mask, vint8m4_t acc, vint8m4_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsac.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsac.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vint8m4_t test_vnmsac_vx_i8m4_m(vbool2_t mask, vint8m4_t acc, int8_t op1, vint8m4_t op2, size_t vl) {
@@ -906,7 +906,7 @@ vint8m4_t test_vnmsac_vx_i8m4_m(vbool2_t mask, vint8m4_t acc, int8_t op1, vint8m
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsac.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsac.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vint8m8_t test_vnmsac_vv_i8m8_m(vbool1_t mask, vint8m8_t acc, vint8m8_t op1, vint8m8_t op2, size_t vl) {
@@ -915,7 +915,7 @@ vint8m8_t test_vnmsac_vv_i8m8_m(vbool1_t mask, vint8m8_t acc, vint8m8_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsac.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsac.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vint8m8_t test_vnmsac_vx_i8m8_m(vbool1_t mask, vint8m8_t acc, int8_t op1, vint8m8_t op2, size_t vl) {
@@ -924,7 +924,7 @@ vint8m8_t test_vnmsac_vx_i8m8_m(vbool1_t mask, vint8m8_t acc, int8_t op1, vint8m
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsac.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsac.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vnmsac_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc, vint16mf4_t op1, vint16mf4_t op2, size_t vl) {
@@ -933,7 +933,7 @@ vint16mf4_t test_vnmsac_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc, vint16mf4_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsac.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsac.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vnmsac_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc, int16_t op1, vint16mf4_t op2, size_t vl) {
@@ -942,7 +942,7 @@ vint16mf4_t test_vnmsac_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc, int16_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsac.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsac.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vnmsac_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc, vint16mf2_t op1, vint16mf2_t op2, size_t vl) {
@@ -951,7 +951,7 @@ vint16mf2_t test_vnmsac_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc, vint16mf2_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsac.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsac.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vnmsac_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc, int16_t op1, vint16mf2_t op2, size_t vl) {
@@ -960,7 +960,7 @@ vint16mf2_t test_vnmsac_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc, int16_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsac.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsac.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vnmsac_vv_i16m1_m(vbool16_t mask, vint16m1_t acc, vint16m1_t op1, vint16m1_t op2, size_t vl) {
@@ -969,7 +969,7 @@ vint16m1_t test_vnmsac_vv_i16m1_m(vbool16_t mask, vint16m1_t acc, vint16m1_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsac.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsac.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vnmsac_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int16_t op1, vint16m1_t op2, size_t vl) {
@@ -978,7 +978,7 @@ vint16m1_t test_vnmsac_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int16_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsac.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsac.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vnmsac_vv_i16m2_m(vbool8_t mask, vint16m2_t acc, vint16m2_t op1, vint16m2_t op2, size_t vl) {
@@ -987,7 +987,7 @@ vint16m2_t test_vnmsac_vv_i16m2_m(vbool8_t mask, vint16m2_t acc, vint16m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsac.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsac.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vnmsac_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int16_t op1, vint16m2_t op2, size_t vl) {
@@ -996,7 +996,7 @@ vint16m2_t test_vnmsac_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int16_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsac.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsac.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vnmsac_vv_i16m4_m(vbool4_t mask, vint16m4_t acc, vint16m4_t op1, vint16m4_t op2, size_t vl) {
@@ -1005,7 +1005,7 @@ vint16m4_t test_vnmsac_vv_i16m4_m(vbool4_t mask, vint16m4_t acc, vint16m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsac.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsac.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vnmsac_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int16_t op1, vint16m4_t op2, size_t vl) {
@@ -1014,7 +1014,7 @@ vint16m4_t test_vnmsac_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int16_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsac.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsac.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vnmsac_vv_i16m8_m(vbool2_t mask, vint16m8_t acc, vint16m8_t op1, vint16m8_t op2, size_t vl) {
@@ -1023,7 +1023,7 @@ vint16m8_t test_vnmsac_vv_i16m8_m(vbool2_t mask, vint16m8_t acc, vint16m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsac.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsac.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vnmsac_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int16_t op1, vint16m8_t op2, size_t vl) {
@@ -1032,7 +1032,7 @@ vint16m8_t test_vnmsac_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int16_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsac.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsac.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vnmsac_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc, vint32mf2_t op1, vint32mf2_t op2, size_t vl) {
@@ -1041,7 +1041,7 @@ vint32mf2_t test_vnmsac_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc, vint32mf2_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsac.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsac.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vnmsac_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc, int32_t op1, vint32mf2_t op2, size_t vl) {
@@ -1050,7 +1050,7 @@ vint32mf2_t test_vnmsac_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc, int32_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsac.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsac.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vnmsac_vv_i32m1_m(vbool32_t mask, vint32m1_t acc, vint32m1_t op1, vint32m1_t op2, size_t vl) {
@@ -1059,7 +1059,7 @@ vint32m1_t test_vnmsac_vv_i32m1_m(vbool32_t mask, vint32m1_t acc, vint32m1_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsac.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsac.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vnmsac_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int32_t op1, vint32m1_t op2, size_t vl) {
@@ -1068,7 +1068,7 @@ vint32m1_t test_vnmsac_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int32_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsac.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsac.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vnmsac_vv_i32m2_m(vbool16_t mask, vint32m2_t acc, vint32m2_t op1, vint32m2_t op2, size_t vl) {
@@ -1077,7 +1077,7 @@ vint32m2_t test_vnmsac_vv_i32m2_m(vbool16_t mask, vint32m2_t acc, vint32m2_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsac.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsac.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vnmsac_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int32_t op1, vint32m2_t op2, size_t vl) {
@@ -1086,7 +1086,7 @@ vint32m2_t test_vnmsac_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int32_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsac.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsac.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vnmsac_vv_i32m4_m(vbool8_t mask, vint32m4_t acc, vint32m4_t op1, vint32m4_t op2, size_t vl) {
@@ -1095,7 +1095,7 @@ vint32m4_t test_vnmsac_vv_i32m4_m(vbool8_t mask, vint32m4_t acc, vint32m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsac.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsac.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vnmsac_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int32_t op1, vint32m4_t op2, size_t vl) {
@@ -1104,7 +1104,7 @@ vint32m4_t test_vnmsac_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int32_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsac.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsac.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vnmsac_vv_i32m8_m(vbool4_t mask, vint32m8_t acc, vint32m8_t op1, vint32m8_t op2, size_t vl) {
@@ -1113,7 +1113,7 @@ vint32m8_t test_vnmsac_vv_i32m8_m(vbool4_t mask, vint32m8_t acc, vint32m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsac.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsac.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vnmsac_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int32_t op1, vint32m8_t op2, size_t vl) {
@@ -1122,7 +1122,7 @@ vint32m8_t test_vnmsac_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int32_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsac.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsac.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vnmsac_vv_i64m1_m(vbool64_t mask, vint64m1_t acc, vint64m1_t op1, vint64m1_t op2, size_t vl) {
@@ -1131,7 +1131,7 @@ vint64m1_t test_vnmsac_vv_i64m1_m(vbool64_t mask, vint64m1_t acc, vint64m1_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsac.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsac.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vnmsac_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int64_t op1, vint64m1_t op2, size_t vl) {
@@ -1140,7 +1140,7 @@ vint64m1_t test_vnmsac_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int64_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsac.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsac.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vnmsac_vv_i64m2_m(vbool32_t mask, vint64m2_t acc, vint64m2_t op1, vint64m2_t op2, size_t vl) {
@@ -1149,7 +1149,7 @@ vint64m2_t test_vnmsac_vv_i64m2_m(vbool32_t mask, vint64m2_t acc, vint64m2_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsac.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsac.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vnmsac_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int64_t op1, vint64m2_t op2, size_t vl) {
@@ -1158,7 +1158,7 @@ vint64m2_t test_vnmsac_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int64_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsac.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsac.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vnmsac_vv_i64m4_m(vbool16_t mask, vint64m4_t acc, vint64m4_t op1, vint64m4_t op2, size_t vl) {
@@ -1167,7 +1167,7 @@ vint64m4_t test_vnmsac_vv_i64m4_m(vbool16_t mask, vint64m4_t acc, vint64m4_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsac.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsac.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vnmsac_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int64_t op1, vint64m4_t op2, size_t vl) {
@@ -1176,7 +1176,7 @@ vint64m4_t test_vnmsac_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int64_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsac.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsac.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vnmsac_vv_i64m8_m(vbool8_t mask, vint64m8_t acc, vint64m8_t op1, vint64m8_t op2, size_t vl) {
@@ -1185,7 +1185,7 @@ vint64m8_t test_vnmsac_vv_i64m8_m(vbool8_t mask, vint64m8_t acc, vint64m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsac.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsac.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vnmsac_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int64_t op1, vint64m8_t op2, size_t vl) {
@@ -1194,7 +1194,7 @@ vint64m8_t test_vnmsac_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int64_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsac.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsac.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vuint8mf8_t test_vnmsac_vv_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, vuint8mf8_t op1, vuint8mf8_t op2, size_t vl) {
@@ -1203,7 +1203,7 @@ vuint8mf8_t test_vnmsac_vv_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, vuint8mf8_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsac.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsac.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vuint8mf8_t test_vnmsac_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, uint8_t op1, vuint8mf8_t op2, size_t vl) {
@@ -1212,7 +1212,7 @@ vuint8mf8_t test_vnmsac_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsac.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsac.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vuint8mf4_t test_vnmsac_vv_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, vuint8mf4_t op1, vuint8mf4_t op2, size_t vl) {
@@ -1221,7 +1221,7 @@ vuint8mf4_t test_vnmsac_vv_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, vuint8mf4_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsac.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsac.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vuint8mf4_t test_vnmsac_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, uint8_t op1, vuint8mf4_t op2, size_t vl) {
@@ -1230,7 +1230,7 @@ vuint8mf4_t test_vnmsac_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsac.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsac.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vuint8mf2_t test_vnmsac_vv_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, vuint8mf2_t op1, vuint8mf2_t op2, size_t vl) {
@@ -1239,7 +1239,7 @@ vuint8mf2_t test_vnmsac_vv_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, vuint8mf2_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsac.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsac.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vuint8mf2_t test_vnmsac_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, uint8_t op1, vuint8mf2_t op2, size_t vl) {
@@ -1248,7 +1248,7 @@ vuint8mf2_t test_vnmsac_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsac.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsac.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vuint8m1_t test_vnmsac_vv_u8m1_m(vbool8_t mask, vuint8m1_t acc, vuint8m1_t op1, vuint8m1_t op2, size_t vl) {
@@ -1257,7 +1257,7 @@ vuint8m1_t test_vnmsac_vv_u8m1_m(vbool8_t mask, vuint8m1_t acc, vuint8m1_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsac.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsac.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vuint8m1_t test_vnmsac_vx_u8m1_m(vbool8_t mask, vuint8m1_t acc, uint8_t op1, vuint8m1_t op2, size_t vl) {
@@ -1266,7 +1266,7 @@ vuint8m1_t test_vnmsac_vx_u8m1_m(vbool8_t mask, vuint8m1_t acc, uint8_t op1, vui
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsac.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsac.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vuint8m2_t test_vnmsac_vv_u8m2_m(vbool4_t mask, vuint8m2_t acc, vuint8m2_t op1, vuint8m2_t op2, size_t vl) {
@@ -1275,7 +1275,7 @@ vuint8m2_t test_vnmsac_vv_u8m2_m(vbool4_t mask, vuint8m2_t acc, vuint8m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsac.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsac.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vuint8m2_t test_vnmsac_vx_u8m2_m(vbool4_t mask, vuint8m2_t acc, uint8_t op1, vuint8m2_t op2, size_t vl) {
@@ -1284,7 +1284,7 @@ vuint8m2_t test_vnmsac_vx_u8m2_m(vbool4_t mask, vuint8m2_t acc, uint8_t op1, vui
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsac.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsac.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vuint8m4_t test_vnmsac_vv_u8m4_m(vbool2_t mask, vuint8m4_t acc, vuint8m4_t op1, vuint8m4_t op2, size_t vl) {
@@ -1293,7 +1293,7 @@ vuint8m4_t test_vnmsac_vv_u8m4_m(vbool2_t mask, vuint8m4_t acc, vuint8m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsac.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsac.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vuint8m4_t test_vnmsac_vx_u8m4_m(vbool2_t mask, vuint8m4_t acc, uint8_t op1, vuint8m4_t op2, size_t vl) {
@@ -1302,7 +1302,7 @@ vuint8m4_t test_vnmsac_vx_u8m4_m(vbool2_t mask, vuint8m4_t acc, uint8_t op1, vui
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsac.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsac.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vuint8m8_t test_vnmsac_vv_u8m8_m(vbool1_t mask, vuint8m8_t acc, vuint8m8_t op1, vuint8m8_t op2, size_t vl) {
@@ -1311,7 +1311,7 @@ vuint8m8_t test_vnmsac_vv_u8m8_m(vbool1_t mask, vuint8m8_t acc, vuint8m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsac.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsac.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vuint8m8_t test_vnmsac_vx_u8m8_m(vbool1_t mask, vuint8m8_t acc, uint8_t op1, vuint8m8_t op2, size_t vl) {
@@ -1320,7 +1320,7 @@ vuint8m8_t test_vnmsac_vx_u8m8_m(vbool1_t mask, vuint8m8_t acc, uint8_t op1, vui
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsac.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsac.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vnmsac_vv_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, vuint16mf4_t op1, vuint16mf4_t op2, size_t vl) {
@@ -1329,7 +1329,7 @@ vuint16mf4_t test_vnmsac_vv_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, vuint16mf
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsac.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsac.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vnmsac_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, uint16_t op1, vuint16mf4_t op2, size_t vl) {
@@ -1338,7 +1338,7 @@ vuint16mf4_t test_vnmsac_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, uint16_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsac.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsac.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vnmsac_vv_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, vuint16mf2_t op1, vuint16mf2_t op2, size_t vl) {
@@ -1347,7 +1347,7 @@ vuint16mf2_t test_vnmsac_vv_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, vuint16mf
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsac.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsac.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vnmsac_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, uint16_t op1, vuint16mf2_t op2, size_t vl) {
@@ -1356,7 +1356,7 @@ vuint16mf2_t test_vnmsac_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, uint16_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsac.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsac.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vnmsac_vv_u16m1_m(vbool16_t mask, vuint16m1_t acc, vuint16m1_t op1, vuint16m1_t op2, size_t vl) {
@@ -1365,7 +1365,7 @@ vuint16m1_t test_vnmsac_vv_u16m1_m(vbool16_t mask, vuint16m1_t acc, vuint16m1_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsac.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsac.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vnmsac_vx_u16m1_m(vbool16_t mask, vuint16m1_t acc, uint16_t op1, vuint16m1_t op2, size_t vl) {
@@ -1374,7 +1374,7 @@ vuint16m1_t test_vnmsac_vx_u16m1_m(vbool16_t mask, vuint16m1_t acc, uint16_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsac.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsac.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vnmsac_vv_u16m2_m(vbool8_t mask, vuint16m2_t acc, vuint16m2_t op1, vuint16m2_t op2, size_t vl) {
@@ -1383,7 +1383,7 @@ vuint16m2_t test_vnmsac_vv_u16m2_m(vbool8_t mask, vuint16m2_t acc, vuint16m2_t o
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsac.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsac.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vnmsac_vx_u16m2_m(vbool8_t mask, vuint16m2_t acc, uint16_t op1, vuint16m2_t op2, size_t vl) {
@@ -1392,7 +1392,7 @@ vuint16m2_t test_vnmsac_vx_u16m2_m(vbool8_t mask, vuint16m2_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsac.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsac.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vnmsac_vv_u16m4_m(vbool4_t mask, vuint16m4_t acc, vuint16m4_t op1, vuint16m4_t op2, size_t vl) {
@@ -1401,7 +1401,7 @@ vuint16m4_t test_vnmsac_vv_u16m4_m(vbool4_t mask, vuint16m4_t acc, vuint16m4_t o
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsac.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsac.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vnmsac_vx_u16m4_m(vbool4_t mask, vuint16m4_t acc, uint16_t op1, vuint16m4_t op2, size_t vl) {
@@ -1410,7 +1410,7 @@ vuint16m4_t test_vnmsac_vx_u16m4_m(vbool4_t mask, vuint16m4_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsac.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsac.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vnmsac_vv_u16m8_m(vbool2_t mask, vuint16m8_t acc, vuint16m8_t op1, vuint16m8_t op2, size_t vl) {
@@ -1419,7 +1419,7 @@ vuint16m8_t test_vnmsac_vv_u16m8_m(vbool2_t mask, vuint16m8_t acc, vuint16m8_t o
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsac.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsac.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vnmsac_vx_u16m8_m(vbool2_t mask, vuint16m8_t acc, uint16_t op1, vuint16m8_t op2, size_t vl) {
@@ -1428,7 +1428,7 @@ vuint16m8_t test_vnmsac_vx_u16m8_m(vbool2_t mask, vuint16m8_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsac.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsac.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vnmsac_vv_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, vuint32mf2_t op1, vuint32mf2_t op2, size_t vl) {
@@ -1437,7 +1437,7 @@ vuint32mf2_t test_vnmsac_vv_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, vuint32mf
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsac.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsac.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vnmsac_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, uint32_t op1, vuint32mf2_t op2, size_t vl) {
@@ -1446,7 +1446,7 @@ vuint32mf2_t test_vnmsac_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, uint32_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsac.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsac.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vnmsac_vv_u32m1_m(vbool32_t mask, vuint32m1_t acc, vuint32m1_t op1, vuint32m1_t op2, size_t vl) {
@@ -1455,7 +1455,7 @@ vuint32m1_t test_vnmsac_vv_u32m1_m(vbool32_t mask, vuint32m1_t acc, vuint32m1_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsac.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsac.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vnmsac_vx_u32m1_m(vbool32_t mask, vuint32m1_t acc, uint32_t op1, vuint32m1_t op2, size_t vl) {
@@ -1464,7 +1464,7 @@ vuint32m1_t test_vnmsac_vx_u32m1_m(vbool32_t mask, vuint32m1_t acc, uint32_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsac.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsac.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vnmsac_vv_u32m2_m(vbool16_t mask, vuint32m2_t acc, vuint32m2_t op1, vuint32m2_t op2, size_t vl) {
@@ -1473,7 +1473,7 @@ vuint32m2_t test_vnmsac_vv_u32m2_m(vbool16_t mask, vuint32m2_t acc, vuint32m2_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsac.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsac.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vnmsac_vx_u32m2_m(vbool16_t mask, vuint32m2_t acc, uint32_t op1, vuint32m2_t op2, size_t vl) {
@@ -1482,7 +1482,7 @@ vuint32m2_t test_vnmsac_vx_u32m2_m(vbool16_t mask, vuint32m2_t acc, uint32_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsac.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsac.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vnmsac_vv_u32m4_m(vbool8_t mask, vuint32m4_t acc, vuint32m4_t op1, vuint32m4_t op2, size_t vl) {
@@ -1491,7 +1491,7 @@ vuint32m4_t test_vnmsac_vv_u32m4_m(vbool8_t mask, vuint32m4_t acc, vuint32m4_t o
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsac.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsac.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vnmsac_vx_u32m4_m(vbool8_t mask, vuint32m4_t acc, uint32_t op1, vuint32m4_t op2, size_t vl) {
@@ -1500,7 +1500,7 @@ vuint32m4_t test_vnmsac_vx_u32m4_m(vbool8_t mask, vuint32m4_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsac.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsac.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vnmsac_vv_u32m8_m(vbool4_t mask, vuint32m8_t acc, vuint32m8_t op1, vuint32m8_t op2, size_t vl) {
@@ -1509,7 +1509,7 @@ vuint32m8_t test_vnmsac_vv_u32m8_m(vbool4_t mask, vuint32m8_t acc, vuint32m8_t o
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsac.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsac.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vnmsac_vx_u32m8_m(vbool4_t mask, vuint32m8_t acc, uint32_t op1, vuint32m8_t op2, size_t vl) {
@@ -1518,7 +1518,7 @@ vuint32m8_t test_vnmsac_vx_u32m8_m(vbool4_t mask, vuint32m8_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsac.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsac.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vnmsac_vv_u64m1_m(vbool64_t mask, vuint64m1_t acc, vuint64m1_t op1, vuint64m1_t op2, size_t vl) {
@@ -1527,7 +1527,7 @@ vuint64m1_t test_vnmsac_vv_u64m1_m(vbool64_t mask, vuint64m1_t acc, vuint64m1_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsac.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsac.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vnmsac_vx_u64m1_m(vbool64_t mask, vuint64m1_t acc, uint64_t op1, vuint64m1_t op2, size_t vl) {
@@ -1536,7 +1536,7 @@ vuint64m1_t test_vnmsac_vx_u64m1_m(vbool64_t mask, vuint64m1_t acc, uint64_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsac.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsac.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vnmsac_vv_u64m2_m(vbool32_t mask, vuint64m2_t acc, vuint64m2_t op1, vuint64m2_t op2, size_t vl) {
@@ -1545,7 +1545,7 @@ vuint64m2_t test_vnmsac_vv_u64m2_m(vbool32_t mask, vuint64m2_t acc, vuint64m2_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsac.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsac.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vnmsac_vx_u64m2_m(vbool32_t mask, vuint64m2_t acc, uint64_t op1, vuint64m2_t op2, size_t vl) {
@@ -1554,7 +1554,7 @@ vuint64m2_t test_vnmsac_vx_u64m2_m(vbool32_t mask, vuint64m2_t acc, uint64_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsac.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsac.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vnmsac_vv_u64m4_m(vbool16_t mask, vuint64m4_t acc, vuint64m4_t op1, vuint64m4_t op2, size_t vl) {
@@ -1563,7 +1563,7 @@ vuint64m4_t test_vnmsac_vv_u64m4_m(vbool16_t mask, vuint64m4_t acc, vuint64m4_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsac.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsac.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vnmsac_vx_u64m4_m(vbool16_t mask, vuint64m4_t acc, uint64_t op1, vuint64m4_t op2, size_t vl) {
@@ -1572,7 +1572,7 @@ vuint64m4_t test_vnmsac_vx_u64m4_m(vbool16_t mask, vuint64m4_t acc, uint64_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsac.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsac.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vnmsac_vv_u64m8_m(vbool8_t mask, vuint64m8_t acc, vuint64m8_t op1, vuint64m8_t op2, size_t vl) {
@@ -1581,7 +1581,7 @@ vuint64m8_t test_vnmsac_vv_u64m8_m(vbool8_t mask, vuint64m8_t acc, vuint64m8_t o
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsac.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsac.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vnmsac_vx_u64m8_m(vbool8_t mask, vuint64m8_t acc, uint64_t op1, vuint64m8_t op2, size_t vl) {
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vnmsub.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vnmsub.c
index 9d3b6a0021b464..a27ef3451d6009 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vnmsub.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vnmsub.c
@@ -798,7 +798,7 @@ vuint64m8_t test_vnmsub_vx_u64m8(vuint64m8_t acc, uint64_t op1, vuint64m8_t op2,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsub.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsub.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vint8mf8_t test_vnmsub_vv_i8mf8_m(vbool64_t mask, vint8mf8_t acc, vint8mf8_t op1, vint8mf8_t op2, size_t vl) {
@@ -807,7 +807,7 @@ vint8mf8_t test_vnmsub_vv_i8mf8_m(vbool64_t mask, vint8mf8_t acc, vint8mf8_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsub.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsub.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vint8mf8_t test_vnmsub_vx_i8mf8_m(vbool64_t mask, vint8mf8_t acc, int8_t op1, vint8mf8_t op2, size_t vl) {
@@ -816,7 +816,7 @@ vint8mf8_t test_vnmsub_vx_i8mf8_m(vbool64_t mask, vint8mf8_t acc, int8_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsub.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsub.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vint8mf4_t test_vnmsub_vv_i8mf4_m(vbool32_t mask, vint8mf4_t acc, vint8mf4_t op1, vint8mf4_t op2, size_t vl) {
@@ -825,7 +825,7 @@ vint8mf4_t test_vnmsub_vv_i8mf4_m(vbool32_t mask, vint8mf4_t acc, vint8mf4_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsub.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsub.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vint8mf4_t test_vnmsub_vx_i8mf4_m(vbool32_t mask, vint8mf4_t acc, int8_t op1, vint8mf4_t op2, size_t vl) {
@@ -834,7 +834,7 @@ vint8mf4_t test_vnmsub_vx_i8mf4_m(vbool32_t mask, vint8mf4_t acc, int8_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsub.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsub.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vint8mf2_t test_vnmsub_vv_i8mf2_m(vbool16_t mask, vint8mf2_t acc, vint8mf2_t op1, vint8mf2_t op2, size_t vl) {
@@ -843,7 +843,7 @@ vint8mf2_t test_vnmsub_vv_i8mf2_m(vbool16_t mask, vint8mf2_t acc, vint8mf2_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsub.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsub.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vint8mf2_t test_vnmsub_vx_i8mf2_m(vbool16_t mask, vint8mf2_t acc, int8_t op1, vint8mf2_t op2, size_t vl) {
@@ -852,7 +852,7 @@ vint8mf2_t test_vnmsub_vx_i8mf2_m(vbool16_t mask, vint8mf2_t acc, int8_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsub.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsub.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vint8m1_t test_vnmsub_vv_i8m1_m(vbool8_t mask, vint8m1_t acc, vint8m1_t op1, vint8m1_t op2, size_t vl) {
@@ -861,7 +861,7 @@ vint8m1_t test_vnmsub_vv_i8m1_m(vbool8_t mask, vint8m1_t acc, vint8m1_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsub.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsub.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vint8m1_t test_vnmsub_vx_i8m1_m(vbool8_t mask, vint8m1_t acc, int8_t op1, vint8m1_t op2, size_t vl) {
@@ -870,7 +870,7 @@ vint8m1_t test_vnmsub_vx_i8m1_m(vbool8_t mask, vint8m1_t acc, int8_t op1, vint8m
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsub.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsub.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vint8m2_t test_vnmsub_vv_i8m2_m(vbool4_t mask, vint8m2_t acc, vint8m2_t op1, vint8m2_t op2, size_t vl) {
@@ -879,7 +879,7 @@ vint8m2_t test_vnmsub_vv_i8m2_m(vbool4_t mask, vint8m2_t acc, vint8m2_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsub.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsub.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vint8m2_t test_vnmsub_vx_i8m2_m(vbool4_t mask, vint8m2_t acc, int8_t op1, vint8m2_t op2, size_t vl) {
@@ -888,7 +888,7 @@ vint8m2_t test_vnmsub_vx_i8m2_m(vbool4_t mask, vint8m2_t acc, int8_t op1, vint8m
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsub.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsub.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vint8m4_t test_vnmsub_vv_i8m4_m(vbool2_t mask, vint8m4_t acc, vint8m4_t op1, vint8m4_t op2, size_t vl) {
@@ -897,7 +897,7 @@ vint8m4_t test_vnmsub_vv_i8m4_m(vbool2_t mask, vint8m4_t acc, vint8m4_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsub.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsub.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vint8m4_t test_vnmsub_vx_i8m4_m(vbool2_t mask, vint8m4_t acc, int8_t op1, vint8m4_t op2, size_t vl) {
@@ -906,7 +906,7 @@ vint8m4_t test_vnmsub_vx_i8m4_m(vbool2_t mask, vint8m4_t acc, int8_t op1, vint8m
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsub.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsub.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vint8m8_t test_vnmsub_vv_i8m8_m(vbool1_t mask, vint8m8_t acc, vint8m8_t op1, vint8m8_t op2, size_t vl) {
@@ -915,7 +915,7 @@ vint8m8_t test_vnmsub_vv_i8m8_m(vbool1_t mask, vint8m8_t acc, vint8m8_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsub.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsub.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vint8m8_t test_vnmsub_vx_i8m8_m(vbool1_t mask, vint8m8_t acc, int8_t op1, vint8m8_t op2, size_t vl) {
@@ -924,7 +924,7 @@ vint8m8_t test_vnmsub_vx_i8m8_m(vbool1_t mask, vint8m8_t acc, int8_t op1, vint8m
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsub.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsub.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vnmsub_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc, vint16mf4_t op1, vint16mf4_t op2, size_t vl) {
@@ -933,7 +933,7 @@ vint16mf4_t test_vnmsub_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc, vint16mf4_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsub.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsub.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vnmsub_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc, int16_t op1, vint16mf4_t op2, size_t vl) {
@@ -942,7 +942,7 @@ vint16mf4_t test_vnmsub_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc, int16_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsub.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsub.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vnmsub_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc, vint16mf2_t op1, vint16mf2_t op2, size_t vl) {
@@ -951,7 +951,7 @@ vint16mf2_t test_vnmsub_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc, vint16mf2_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsub.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsub.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vnmsub_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc, int16_t op1, vint16mf2_t op2, size_t vl) {
@@ -960,7 +960,7 @@ vint16mf2_t test_vnmsub_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc, int16_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsub.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsub.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vnmsub_vv_i16m1_m(vbool16_t mask, vint16m1_t acc, vint16m1_t op1, vint16m1_t op2, size_t vl) {
@@ -969,7 +969,7 @@ vint16m1_t test_vnmsub_vv_i16m1_m(vbool16_t mask, vint16m1_t acc, vint16m1_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsub.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsub.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vnmsub_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int16_t op1, vint16m1_t op2, size_t vl) {
@@ -978,7 +978,7 @@ vint16m1_t test_vnmsub_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int16_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsub.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsub.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vnmsub_vv_i16m2_m(vbool8_t mask, vint16m2_t acc, vint16m2_t op1, vint16m2_t op2, size_t vl) {
@@ -987,7 +987,7 @@ vint16m2_t test_vnmsub_vv_i16m2_m(vbool8_t mask, vint16m2_t acc, vint16m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsub.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsub.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vnmsub_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int16_t op1, vint16m2_t op2, size_t vl) {
@@ -996,7 +996,7 @@ vint16m2_t test_vnmsub_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int16_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsub.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsub.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vnmsub_vv_i16m4_m(vbool4_t mask, vint16m4_t acc, vint16m4_t op1, vint16m4_t op2, size_t vl) {
@@ -1005,7 +1005,7 @@ vint16m4_t test_vnmsub_vv_i16m4_m(vbool4_t mask, vint16m4_t acc, vint16m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsub.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsub.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vnmsub_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int16_t op1, vint16m4_t op2, size_t vl) {
@@ -1014,7 +1014,7 @@ vint16m4_t test_vnmsub_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int16_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsub.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsub.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vnmsub_vv_i16m8_m(vbool2_t mask, vint16m8_t acc, vint16m8_t op1, vint16m8_t op2, size_t vl) {
@@ -1023,7 +1023,7 @@ vint16m8_t test_vnmsub_vv_i16m8_m(vbool2_t mask, vint16m8_t acc, vint16m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsub.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsub.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vnmsub_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int16_t op1, vint16m8_t op2, size_t vl) {
@@ -1032,7 +1032,7 @@ vint16m8_t test_vnmsub_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int16_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsub.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsub.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vnmsub_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc, vint32mf2_t op1, vint32mf2_t op2, size_t vl) {
@@ -1041,7 +1041,7 @@ vint32mf2_t test_vnmsub_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc, vint32mf2_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsub.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsub.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vnmsub_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc, int32_t op1, vint32mf2_t op2, size_t vl) {
@@ -1050,7 +1050,7 @@ vint32mf2_t test_vnmsub_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc, int32_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsub.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsub.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vnmsub_vv_i32m1_m(vbool32_t mask, vint32m1_t acc, vint32m1_t op1, vint32m1_t op2, size_t vl) {
@@ -1059,7 +1059,7 @@ vint32m1_t test_vnmsub_vv_i32m1_m(vbool32_t mask, vint32m1_t acc, vint32m1_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsub.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsub.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vnmsub_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int32_t op1, vint32m1_t op2, size_t vl) {
@@ -1068,7 +1068,7 @@ vint32m1_t test_vnmsub_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int32_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsub.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsub.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vnmsub_vv_i32m2_m(vbool16_t mask, vint32m2_t acc, vint32m2_t op1, vint32m2_t op2, size_t vl) {
@@ -1077,7 +1077,7 @@ vint32m2_t test_vnmsub_vv_i32m2_m(vbool16_t mask, vint32m2_t acc, vint32m2_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsub.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsub.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vnmsub_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int32_t op1, vint32m2_t op2, size_t vl) {
@@ -1086,7 +1086,7 @@ vint32m2_t test_vnmsub_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int32_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsub.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsub.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vnmsub_vv_i32m4_m(vbool8_t mask, vint32m4_t acc, vint32m4_t op1, vint32m4_t op2, size_t vl) {
@@ -1095,7 +1095,7 @@ vint32m4_t test_vnmsub_vv_i32m4_m(vbool8_t mask, vint32m4_t acc, vint32m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsub.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsub.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vnmsub_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int32_t op1, vint32m4_t op2, size_t vl) {
@@ -1104,7 +1104,7 @@ vint32m4_t test_vnmsub_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int32_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsub.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsub.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vnmsub_vv_i32m8_m(vbool4_t mask, vint32m8_t acc, vint32m8_t op1, vint32m8_t op2, size_t vl) {
@@ -1113,7 +1113,7 @@ vint32m8_t test_vnmsub_vv_i32m8_m(vbool4_t mask, vint32m8_t acc, vint32m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsub.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsub.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vnmsub_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int32_t op1, vint32m8_t op2, size_t vl) {
@@ -1122,7 +1122,7 @@ vint32m8_t test_vnmsub_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int32_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsub.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsub.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vnmsub_vv_i64m1_m(vbool64_t mask, vint64m1_t acc, vint64m1_t op1, vint64m1_t op2, size_t vl) {
@@ -1131,7 +1131,7 @@ vint64m1_t test_vnmsub_vv_i64m1_m(vbool64_t mask, vint64m1_t acc, vint64m1_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsub.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsub.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vnmsub_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int64_t op1, vint64m1_t op2, size_t vl) {
@@ -1140,7 +1140,7 @@ vint64m1_t test_vnmsub_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int64_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsub.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsub.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vnmsub_vv_i64m2_m(vbool32_t mask, vint64m2_t acc, vint64m2_t op1, vint64m2_t op2, size_t vl) {
@@ -1149,7 +1149,7 @@ vint64m2_t test_vnmsub_vv_i64m2_m(vbool32_t mask, vint64m2_t acc, vint64m2_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsub.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsub.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vnmsub_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int64_t op1, vint64m2_t op2, size_t vl) {
@@ -1158,7 +1158,7 @@ vint64m2_t test_vnmsub_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int64_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsub.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsub.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vnmsub_vv_i64m4_m(vbool16_t mask, vint64m4_t acc, vint64m4_t op1, vint64m4_t op2, size_t vl) {
@@ -1167,7 +1167,7 @@ vint64m4_t test_vnmsub_vv_i64m4_m(vbool16_t mask, vint64m4_t acc, vint64m4_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsub.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsub.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vnmsub_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int64_t op1, vint64m4_t op2, size_t vl) {
@@ -1176,7 +1176,7 @@ vint64m4_t test_vnmsub_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int64_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsub.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsub.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vnmsub_vv_i64m8_m(vbool8_t mask, vint64m8_t acc, vint64m8_t op1, vint64m8_t op2, size_t vl) {
@@ -1185,7 +1185,7 @@ vint64m8_t test_vnmsub_vv_i64m8_m(vbool8_t mask, vint64m8_t acc, vint64m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsub.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsub.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vnmsub_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int64_t op1, vint64m8_t op2, size_t vl) {
@@ -1194,7 +1194,7 @@ vint64m8_t test_vnmsub_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int64_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsub.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsub.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vuint8mf8_t test_vnmsub_vv_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, vuint8mf8_t op1, vuint8mf8_t op2, size_t vl) {
@@ -1203,7 +1203,7 @@ vuint8mf8_t test_vnmsub_vv_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, vuint8mf8_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsub.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsub.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vuint8mf8_t test_vnmsub_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, uint8_t op1, vuint8mf8_t op2, size_t vl) {
@@ -1212,7 +1212,7 @@ vuint8mf8_t test_vnmsub_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsub.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsub.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vuint8mf4_t test_vnmsub_vv_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, vuint8mf4_t op1, vuint8mf4_t op2, size_t vl) {
@@ -1221,7 +1221,7 @@ vuint8mf4_t test_vnmsub_vv_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, vuint8mf4_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsub.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsub.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vuint8mf4_t test_vnmsub_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, uint8_t op1, vuint8mf4_t op2, size_t vl) {
@@ -1230,7 +1230,7 @@ vuint8mf4_t test_vnmsub_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsub.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsub.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vuint8mf2_t test_vnmsub_vv_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, vuint8mf2_t op1, vuint8mf2_t op2, size_t vl) {
@@ -1239,7 +1239,7 @@ vuint8mf2_t test_vnmsub_vv_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, vuint8mf2_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsub.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsub.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vuint8mf2_t test_vnmsub_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, uint8_t op1, vuint8mf2_t op2, size_t vl) {
@@ -1248,7 +1248,7 @@ vuint8mf2_t test_vnmsub_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsub.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsub.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vuint8m1_t test_vnmsub_vv_u8m1_m(vbool8_t mask, vuint8m1_t acc, vuint8m1_t op1, vuint8m1_t op2, size_t vl) {
@@ -1257,7 +1257,7 @@ vuint8m1_t test_vnmsub_vv_u8m1_m(vbool8_t mask, vuint8m1_t acc, vuint8m1_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsub.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsub.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vuint8m1_t test_vnmsub_vx_u8m1_m(vbool8_t mask, vuint8m1_t acc, uint8_t op1, vuint8m1_t op2, size_t vl) {
@@ -1266,7 +1266,7 @@ vuint8m1_t test_vnmsub_vx_u8m1_m(vbool8_t mask, vuint8m1_t acc, uint8_t op1, vui
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsub.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsub.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vuint8m2_t test_vnmsub_vv_u8m2_m(vbool4_t mask, vuint8m2_t acc, vuint8m2_t op1, vuint8m2_t op2, size_t vl) {
@@ -1275,7 +1275,7 @@ vuint8m2_t test_vnmsub_vv_u8m2_m(vbool4_t mask, vuint8m2_t acc, vuint8m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsub.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsub.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vuint8m2_t test_vnmsub_vx_u8m2_m(vbool4_t mask, vuint8m2_t acc, uint8_t op1, vuint8m2_t op2, size_t vl) {
@@ -1284,7 +1284,7 @@ vuint8m2_t test_vnmsub_vx_u8m2_m(vbool4_t mask, vuint8m2_t acc, uint8_t op1, vui
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsub.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsub.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vuint8m4_t test_vnmsub_vv_u8m4_m(vbool2_t mask, vuint8m4_t acc, vuint8m4_t op1, vuint8m4_t op2, size_t vl) {
@@ -1293,7 +1293,7 @@ vuint8m4_t test_vnmsub_vv_u8m4_m(vbool2_t mask, vuint8m4_t acc, vuint8m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsub.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsub.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vuint8m4_t test_vnmsub_vx_u8m4_m(vbool2_t mask, vuint8m4_t acc, uint8_t op1, vuint8m4_t op2, size_t vl) {
@@ -1302,7 +1302,7 @@ vuint8m4_t test_vnmsub_vx_u8m4_m(vbool2_t mask, vuint8m4_t acc, uint8_t op1, vui
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsub.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsub.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vuint8m8_t test_vnmsub_vv_u8m8_m(vbool1_t mask, vuint8m8_t acc, vuint8m8_t op1, vuint8m8_t op2, size_t vl) {
@@ -1311,7 +1311,7 @@ vuint8m8_t test_vnmsub_vv_u8m8_m(vbool1_t mask, vuint8m8_t acc, vuint8m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsub.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsub.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vuint8m8_t test_vnmsub_vx_u8m8_m(vbool1_t mask, vuint8m8_t acc, uint8_t op1, vuint8m8_t op2, size_t vl) {
@@ -1320,7 +1320,7 @@ vuint8m8_t test_vnmsub_vx_u8m8_m(vbool1_t mask, vuint8m8_t acc, uint8_t op1, vui
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsub.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsub.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vnmsub_vv_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, vuint16mf4_t op1, vuint16mf4_t op2, size_t vl) {
@@ -1329,7 +1329,7 @@ vuint16mf4_t test_vnmsub_vv_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, vuint16mf
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsub.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsub.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vnmsub_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, uint16_t op1, vuint16mf4_t op2, size_t vl) {
@@ -1338,7 +1338,7 @@ vuint16mf4_t test_vnmsub_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, uint16_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsub.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsub.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vnmsub_vv_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, vuint16mf2_t op1, vuint16mf2_t op2, size_t vl) {
@@ -1347,7 +1347,7 @@ vuint16mf2_t test_vnmsub_vv_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, vuint16mf
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsub.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsub.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vnmsub_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, uint16_t op1, vuint16mf2_t op2, size_t vl) {
@@ -1356,7 +1356,7 @@ vuint16mf2_t test_vnmsub_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, uint16_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsub.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsub.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vnmsub_vv_u16m1_m(vbool16_t mask, vuint16m1_t acc, vuint16m1_t op1, vuint16m1_t op2, size_t vl) {
@@ -1365,7 +1365,7 @@ vuint16m1_t test_vnmsub_vv_u16m1_m(vbool16_t mask, vuint16m1_t acc, vuint16m1_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsub.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsub.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vnmsub_vx_u16m1_m(vbool16_t mask, vuint16m1_t acc, uint16_t op1, vuint16m1_t op2, size_t vl) {
@@ -1374,7 +1374,7 @@ vuint16m1_t test_vnmsub_vx_u16m1_m(vbool16_t mask, vuint16m1_t acc, uint16_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsub.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsub.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vnmsub_vv_u16m2_m(vbool8_t mask, vuint16m2_t acc, vuint16m2_t op1, vuint16m2_t op2, size_t vl) {
@@ -1383,7 +1383,7 @@ vuint16m2_t test_vnmsub_vv_u16m2_m(vbool8_t mask, vuint16m2_t acc, vuint16m2_t o
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsub.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsub.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vnmsub_vx_u16m2_m(vbool8_t mask, vuint16m2_t acc, uint16_t op1, vuint16m2_t op2, size_t vl) {
@@ -1392,7 +1392,7 @@ vuint16m2_t test_vnmsub_vx_u16m2_m(vbool8_t mask, vuint16m2_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsub.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsub.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vnmsub_vv_u16m4_m(vbool4_t mask, vuint16m4_t acc, vuint16m4_t op1, vuint16m4_t op2, size_t vl) {
@@ -1401,7 +1401,7 @@ vuint16m4_t test_vnmsub_vv_u16m4_m(vbool4_t mask, vuint16m4_t acc, vuint16m4_t o
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsub.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsub.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vnmsub_vx_u16m4_m(vbool4_t mask, vuint16m4_t acc, uint16_t op1, vuint16m4_t op2, size_t vl) {
@@ -1410,7 +1410,7 @@ vuint16m4_t test_vnmsub_vx_u16m4_m(vbool4_t mask, vuint16m4_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsub.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsub.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vnmsub_vv_u16m8_m(vbool2_t mask, vuint16m8_t acc, vuint16m8_t op1, vuint16m8_t op2, size_t vl) {
@@ -1419,7 +1419,7 @@ vuint16m8_t test_vnmsub_vv_u16m8_m(vbool2_t mask, vuint16m8_t acc, vuint16m8_t o
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsub.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsub.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vnmsub_vx_u16m8_m(vbool2_t mask, vuint16m8_t acc, uint16_t op1, vuint16m8_t op2, size_t vl) {
@@ -1428,7 +1428,7 @@ vuint16m8_t test_vnmsub_vx_u16m8_m(vbool2_t mask, vuint16m8_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsub.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsub.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vnmsub_vv_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, vuint32mf2_t op1, vuint32mf2_t op2, size_t vl) {
@@ -1437,7 +1437,7 @@ vuint32mf2_t test_vnmsub_vv_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, vuint32mf
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsub.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsub.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vnmsub_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, uint32_t op1, vuint32mf2_t op2, size_t vl) {
@@ -1446,7 +1446,7 @@ vuint32mf2_t test_vnmsub_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, uint32_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsub.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsub.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vnmsub_vv_u32m1_m(vbool32_t mask, vuint32m1_t acc, vuint32m1_t op1, vuint32m1_t op2, size_t vl) {
@@ -1455,7 +1455,7 @@ vuint32m1_t test_vnmsub_vv_u32m1_m(vbool32_t mask, vuint32m1_t acc, vuint32m1_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsub.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsub.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vnmsub_vx_u32m1_m(vbool32_t mask, vuint32m1_t acc, uint32_t op1, vuint32m1_t op2, size_t vl) {
@@ -1464,7 +1464,7 @@ vuint32m1_t test_vnmsub_vx_u32m1_m(vbool32_t mask, vuint32m1_t acc, uint32_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsub.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsub.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vnmsub_vv_u32m2_m(vbool16_t mask, vuint32m2_t acc, vuint32m2_t op1, vuint32m2_t op2, size_t vl) {
@@ -1473,7 +1473,7 @@ vuint32m2_t test_vnmsub_vv_u32m2_m(vbool16_t mask, vuint32m2_t acc, vuint32m2_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsub.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsub.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vnmsub_vx_u32m2_m(vbool16_t mask, vuint32m2_t acc, uint32_t op1, vuint32m2_t op2, size_t vl) {
@@ -1482,7 +1482,7 @@ vuint32m2_t test_vnmsub_vx_u32m2_m(vbool16_t mask, vuint32m2_t acc, uint32_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsub.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsub.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vnmsub_vv_u32m4_m(vbool8_t mask, vuint32m4_t acc, vuint32m4_t op1, vuint32m4_t op2, size_t vl) {
@@ -1491,7 +1491,7 @@ vuint32m4_t test_vnmsub_vv_u32m4_m(vbool8_t mask, vuint32m4_t acc, vuint32m4_t o
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsub.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsub.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vnmsub_vx_u32m4_m(vbool8_t mask, vuint32m4_t acc, uint32_t op1, vuint32m4_t op2, size_t vl) {
@@ -1500,7 +1500,7 @@ vuint32m4_t test_vnmsub_vx_u32m4_m(vbool8_t mask, vuint32m4_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsub.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsub.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vnmsub_vv_u32m8_m(vbool4_t mask, vuint32m8_t acc, vuint32m8_t op1, vuint32m8_t op2, size_t vl) {
@@ -1509,7 +1509,7 @@ vuint32m8_t test_vnmsub_vv_u32m8_m(vbool4_t mask, vuint32m8_t acc, vuint32m8_t o
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsub.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsub.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vnmsub_vx_u32m8_m(vbool4_t mask, vuint32m8_t acc, uint32_t op1, vuint32m8_t op2, size_t vl) {
@@ -1518,7 +1518,7 @@ vuint32m8_t test_vnmsub_vx_u32m8_m(vbool4_t mask, vuint32m8_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsub.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsub.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vnmsub_vv_u64m1_m(vbool64_t mask, vuint64m1_t acc, vuint64m1_t op1, vuint64m1_t op2, size_t vl) {
@@ -1527,7 +1527,7 @@ vuint64m1_t test_vnmsub_vv_u64m1_m(vbool64_t mask, vuint64m1_t acc, vuint64m1_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsub.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsub.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vnmsub_vx_u64m1_m(vbool64_t mask, vuint64m1_t acc, uint64_t op1, vuint64m1_t op2, size_t vl) {
@@ -1536,7 +1536,7 @@ vuint64m1_t test_vnmsub_vx_u64m1_m(vbool64_t mask, vuint64m1_t acc, uint64_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsub.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsub.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vnmsub_vv_u64m2_m(vbool32_t mask, vuint64m2_t acc, vuint64m2_t op1, vuint64m2_t op2, size_t vl) {
@@ -1545,7 +1545,7 @@ vuint64m2_t test_vnmsub_vv_u64m2_m(vbool32_t mask, vuint64m2_t acc, vuint64m2_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsub.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsub.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vnmsub_vx_u64m2_m(vbool32_t mask, vuint64m2_t acc, uint64_t op1, vuint64m2_t op2, size_t vl) {
@@ -1554,7 +1554,7 @@ vuint64m2_t test_vnmsub_vx_u64m2_m(vbool32_t mask, vuint64m2_t acc, uint64_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsub.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsub.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vnmsub_vv_u64m4_m(vbool16_t mask, vuint64m4_t acc, vuint64m4_t op1, vuint64m4_t op2, size_t vl) {
@@ -1563,7 +1563,7 @@ vuint64m4_t test_vnmsub_vv_u64m4_m(vbool16_t mask, vuint64m4_t acc, vuint64m4_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsub.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsub.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vnmsub_vx_u64m4_m(vbool16_t mask, vuint64m4_t acc, uint64_t op1, vuint64m4_t op2, size_t vl) {
@@ -1572,7 +1572,7 @@ vuint64m4_t test_vnmsub_vx_u64m4_m(vbool16_t mask, vuint64m4_t acc, uint64_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsub.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsub.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vnmsub_vv_u64m8_m(vbool8_t mask, vuint64m8_t acc, vuint64m8_t op1, vuint64m8_t op2, size_t vl) {
@@ -1581,7 +1581,7 @@ vuint64m8_t test_vnmsub_vv_u64m8_m(vbool8_t mask, vuint64m8_t acc, vuint64m8_t o
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsub.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsub.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vnmsub_vx_u64m8_m(vbool8_t mask, vuint64m8_t acc, uint64_t op1, vuint64m8_t op2, size_t vl) {
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vslidedown.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vslidedown.c
index c07a80a1bc776a..ed45f08c94b04e 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vslidedown.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vslidedown.c
@@ -537,7 +537,7 @@ vfloat64m8_t test_vslidedown_vx_f64m8(vfloat64m8_t dst, vfloat64m8_t src,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vslidedown.mask.nxv1i8.i64(<vscale x 1 x i8> [[DST:%.*]], <vscale x 1 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vslidedown.mask.nxv1i8.i64(<vscale x 1 x i8> [[DST:%.*]], <vscale x 1 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vint8mf8_t test_vslidedown_vx_i8mf8_m(vbool64_t mask, vint8mf8_t dst,
@@ -548,7 +548,7 @@ vint8mf8_t test_vslidedown_vx_i8mf8_m(vbool64_t mask, vint8mf8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vslidedown.mask.nxv2i8.i64(<vscale x 2 x i8> [[DST:%.*]], <vscale x 2 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vslidedown.mask.nxv2i8.i64(<vscale x 2 x i8> [[DST:%.*]], <vscale x 2 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vint8mf4_t test_vslidedown_vx_i8mf4_m(vbool32_t mask, vint8mf4_t dst,
@@ -559,7 +559,7 @@ vint8mf4_t test_vslidedown_vx_i8mf4_m(vbool32_t mask, vint8mf4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vslidedown.mask.nxv4i8.i64(<vscale x 4 x i8> [[DST:%.*]], <vscale x 4 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vslidedown.mask.nxv4i8.i64(<vscale x 4 x i8> [[DST:%.*]], <vscale x 4 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vint8mf2_t test_vslidedown_vx_i8mf2_m(vbool16_t mask, vint8mf2_t dst,
@@ -570,7 +570,7 @@ vint8mf2_t test_vslidedown_vx_i8mf2_m(vbool16_t mask, vint8mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vslidedown.mask.nxv8i8.i64(<vscale x 8 x i8> [[DST:%.*]], <vscale x 8 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vslidedown.mask.nxv8i8.i64(<vscale x 8 x i8> [[DST:%.*]], <vscale x 8 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vint8m1_t test_vslidedown_vx_i8m1_m(vbool8_t mask, vint8m1_t dst, vint8m1_t src,
@@ -580,7 +580,7 @@ vint8m1_t test_vslidedown_vx_i8m1_m(vbool8_t mask, vint8m1_t dst, vint8m1_t src,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vslidedown.mask.nxv16i8.i64(<vscale x 16 x i8> [[DST:%.*]], <vscale x 16 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vslidedown.mask.nxv16i8.i64(<vscale x 16 x i8> [[DST:%.*]], <vscale x 16 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vint8m2_t test_vslidedown_vx_i8m2_m(vbool4_t mask, vint8m2_t dst, vint8m2_t src,
@@ -590,7 +590,7 @@ vint8m2_t test_vslidedown_vx_i8m2_m(vbool4_t mask, vint8m2_t dst, vint8m2_t src,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vslidedown.mask.nxv32i8.i64(<vscale x 32 x i8> [[DST:%.*]], <vscale x 32 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vslidedown.mask.nxv32i8.i64(<vscale x 32 x i8> [[DST:%.*]], <vscale x 32 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vint8m4_t test_vslidedown_vx_i8m4_m(vbool2_t mask, vint8m4_t dst, vint8m4_t src,
@@ -600,7 +600,7 @@ vint8m4_t test_vslidedown_vx_i8m4_m(vbool2_t mask, vint8m4_t dst, vint8m4_t src,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vslidedown.mask.nxv64i8.i64(<vscale x 64 x i8> [[DST:%.*]], <vscale x 64 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vslidedown.mask.nxv64i8.i64(<vscale x 64 x i8> [[DST:%.*]], <vscale x 64 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vint8m8_t test_vslidedown_vx_i8m8_m(vbool1_t mask, vint8m8_t dst, vint8m8_t src,
@@ -610,7 +610,7 @@ vint8m8_t test_vslidedown_vx_i8m8_m(vbool1_t mask, vint8m8_t dst, vint8m8_t src,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vslidedown.mask.nxv1i16.i64(<vscale x 1 x i16> [[DST:%.*]], <vscale x 1 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vslidedown.mask.nxv1i16.i64(<vscale x 1 x i16> [[DST:%.*]], <vscale x 1 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vslidedown_vx_i16mf4_m(vbool64_t mask, vint16mf4_t dst,
@@ -621,7 +621,7 @@ vint16mf4_t test_vslidedown_vx_i16mf4_m(vbool64_t mask, vint16mf4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vslidedown.mask.nxv2i16.i64(<vscale x 2 x i16> [[DST:%.*]], <vscale x 2 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vslidedown.mask.nxv2i16.i64(<vscale x 2 x i16> [[DST:%.*]], <vscale x 2 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vslidedown_vx_i16mf2_m(vbool32_t mask, vint16mf2_t dst,
@@ -632,7 +632,7 @@ vint16mf2_t test_vslidedown_vx_i16mf2_m(vbool32_t mask, vint16mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vslidedown.mask.nxv4i16.i64(<vscale x 4 x i16> [[DST:%.*]], <vscale x 4 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vslidedown.mask.nxv4i16.i64(<vscale x 4 x i16> [[DST:%.*]], <vscale x 4 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vslidedown_vx_i16m1_m(vbool16_t mask, vint16m1_t dst,
@@ -643,7 +643,7 @@ vint16m1_t test_vslidedown_vx_i16m1_m(vbool16_t mask, vint16m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vslidedown.mask.nxv8i16.i64(<vscale x 8 x i16> [[DST:%.*]], <vscale x 8 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vslidedown.mask.nxv8i16.i64(<vscale x 8 x i16> [[DST:%.*]], <vscale x 8 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vslidedown_vx_i16m2_m(vbool8_t mask, vint16m2_t dst,
@@ -654,7 +654,7 @@ vint16m2_t test_vslidedown_vx_i16m2_m(vbool8_t mask, vint16m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vslidedown.mask.nxv16i16.i64(<vscale x 16 x i16> [[DST:%.*]], <vscale x 16 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vslidedown.mask.nxv16i16.i64(<vscale x 16 x i16> [[DST:%.*]], <vscale x 16 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vslidedown_vx_i16m4_m(vbool4_t mask, vint16m4_t dst,
@@ -665,7 +665,7 @@ vint16m4_t test_vslidedown_vx_i16m4_m(vbool4_t mask, vint16m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vslidedown.mask.nxv32i16.i64(<vscale x 32 x i16> [[DST:%.*]], <vscale x 32 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vslidedown.mask.nxv32i16.i64(<vscale x 32 x i16> [[DST:%.*]], <vscale x 32 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vslidedown_vx_i16m8_m(vbool2_t mask, vint16m8_t dst,
@@ -676,7 +676,7 @@ vint16m8_t test_vslidedown_vx_i16m8_m(vbool2_t mask, vint16m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vslidedown.mask.nxv1i32.i64(<vscale x 1 x i32> [[DST:%.*]], <vscale x 1 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vslidedown.mask.nxv1i32.i64(<vscale x 1 x i32> [[DST:%.*]], <vscale x 1 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vslidedown_vx_i32mf2_m(vbool64_t mask, vint32mf2_t dst,
@@ -687,7 +687,7 @@ vint32mf2_t test_vslidedown_vx_i32mf2_m(vbool64_t mask, vint32mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vslidedown.mask.nxv2i32.i64(<vscale x 2 x i32> [[DST:%.*]], <vscale x 2 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vslidedown.mask.nxv2i32.i64(<vscale x 2 x i32> [[DST:%.*]], <vscale x 2 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vslidedown_vx_i32m1_m(vbool32_t mask, vint32m1_t dst,
@@ -698,7 +698,7 @@ vint32m1_t test_vslidedown_vx_i32m1_m(vbool32_t mask, vint32m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vslidedown.mask.nxv4i32.i64(<vscale x 4 x i32> [[DST:%.*]], <vscale x 4 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vslidedown.mask.nxv4i32.i64(<vscale x 4 x i32> [[DST:%.*]], <vscale x 4 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vslidedown_vx_i32m2_m(vbool16_t mask, vint32m2_t dst,
@@ -709,7 +709,7 @@ vint32m2_t test_vslidedown_vx_i32m2_m(vbool16_t mask, vint32m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vslidedown.mask.nxv8i32.i64(<vscale x 8 x i32> [[DST:%.*]], <vscale x 8 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vslidedown.mask.nxv8i32.i64(<vscale x 8 x i32> [[DST:%.*]], <vscale x 8 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vslidedown_vx_i32m4_m(vbool8_t mask, vint32m4_t dst,
@@ -720,7 +720,7 @@ vint32m4_t test_vslidedown_vx_i32m4_m(vbool8_t mask, vint32m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vslidedown.mask.nxv16i32.i64(<vscale x 16 x i32> [[DST:%.*]], <vscale x 16 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vslidedown.mask.nxv16i32.i64(<vscale x 16 x i32> [[DST:%.*]], <vscale x 16 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vslidedown_vx_i32m8_m(vbool4_t mask, vint32m8_t dst,
@@ -731,7 +731,7 @@ vint32m8_t test_vslidedown_vx_i32m8_m(vbool4_t mask, vint32m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vslidedown.mask.nxv1i64.i64(<vscale x 1 x i64> [[DST:%.*]], <vscale x 1 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vslidedown.mask.nxv1i64.i64(<vscale x 1 x i64> [[DST:%.*]], <vscale x 1 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vslidedown_vx_i64m1_m(vbool64_t mask, vint64m1_t dst,
@@ -742,7 +742,7 @@ vint64m1_t test_vslidedown_vx_i64m1_m(vbool64_t mask, vint64m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vslidedown.mask.nxv2i64.i64(<vscale x 2 x i64> [[DST:%.*]], <vscale x 2 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vslidedown.mask.nxv2i64.i64(<vscale x 2 x i64> [[DST:%.*]], <vscale x 2 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vslidedown_vx_i64m2_m(vbool32_t mask, vint64m2_t dst,
@@ -753,7 +753,7 @@ vint64m2_t test_vslidedown_vx_i64m2_m(vbool32_t mask, vint64m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vslidedown.mask.nxv4i64.i64(<vscale x 4 x i64> [[DST:%.*]], <vscale x 4 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vslidedown.mask.nxv4i64.i64(<vscale x 4 x i64> [[DST:%.*]], <vscale x 4 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vslidedown_vx_i64m4_m(vbool16_t mask, vint64m4_t dst,
@@ -764,7 +764,7 @@ vint64m4_t test_vslidedown_vx_i64m4_m(vbool16_t mask, vint64m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vslidedown.mask.nxv8i64.i64(<vscale x 8 x i64> [[DST:%.*]], <vscale x 8 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vslidedown.mask.nxv8i64.i64(<vscale x 8 x i64> [[DST:%.*]], <vscale x 8 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vslidedown_vx_i64m8_m(vbool8_t mask, vint64m8_t dst,
@@ -775,7 +775,7 @@ vint64m8_t test_vslidedown_vx_i64m8_m(vbool8_t mask, vint64m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vslidedown.mask.nxv1i8.i64(<vscale x 1 x i8> [[DST:%.*]], <vscale x 1 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vslidedown.mask.nxv1i8.i64(<vscale x 1 x i8> [[DST:%.*]], <vscale x 1 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vuint8mf8_t test_vslidedown_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t dst,
@@ -786,7 +786,7 @@ vuint8mf8_t test_vslidedown_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vslidedown.mask.nxv2i8.i64(<vscale x 2 x i8> [[DST:%.*]], <vscale x 2 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vslidedown.mask.nxv2i8.i64(<vscale x 2 x i8> [[DST:%.*]], <vscale x 2 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vuint8mf4_t test_vslidedown_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t dst,
@@ -797,7 +797,7 @@ vuint8mf4_t test_vslidedown_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vslidedown.mask.nxv4i8.i64(<vscale x 4 x i8> [[DST:%.*]], <vscale x 4 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vslidedown.mask.nxv4i8.i64(<vscale x 4 x i8> [[DST:%.*]], <vscale x 4 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vuint8mf2_t test_vslidedown_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t dst,
@@ -808,7 +808,7 @@ vuint8mf2_t test_vslidedown_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vslidedown.mask.nxv8i8.i64(<vscale x 8 x i8> [[DST:%.*]], <vscale x 8 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vslidedown.mask.nxv8i8.i64(<vscale x 8 x i8> [[DST:%.*]], <vscale x 8 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vuint8m1_t test_vslidedown_vx_u8m1_m(vbool8_t mask, vuint8m1_t dst,
@@ -818,7 +818,7 @@ vuint8m1_t test_vslidedown_vx_u8m1_m(vbool8_t mask, vuint8m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vslidedown.mask.nxv16i8.i64(<vscale x 16 x i8> [[DST:%.*]], <vscale x 16 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vslidedown.mask.nxv16i8.i64(<vscale x 16 x i8> [[DST:%.*]], <vscale x 16 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vuint8m2_t test_vslidedown_vx_u8m2_m(vbool4_t mask, vuint8m2_t dst,
@@ -828,7 +828,7 @@ vuint8m2_t test_vslidedown_vx_u8m2_m(vbool4_t mask, vuint8m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vslidedown.mask.nxv32i8.i64(<vscale x 32 x i8> [[DST:%.*]], <vscale x 32 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vslidedown.mask.nxv32i8.i64(<vscale x 32 x i8> [[DST:%.*]], <vscale x 32 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vuint8m4_t test_vslidedown_vx_u8m4_m(vbool2_t mask, vuint8m4_t dst,
@@ -838,7 +838,7 @@ vuint8m4_t test_vslidedown_vx_u8m4_m(vbool2_t mask, vuint8m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vslidedown.mask.nxv64i8.i64(<vscale x 64 x i8> [[DST:%.*]], <vscale x 64 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vslidedown.mask.nxv64i8.i64(<vscale x 64 x i8> [[DST:%.*]], <vscale x 64 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vuint8m8_t test_vslidedown_vx_u8m8_m(vbool1_t mask, vuint8m8_t dst,
@@ -848,7 +848,7 @@ vuint8m8_t test_vslidedown_vx_u8m8_m(vbool1_t mask, vuint8m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vslidedown.mask.nxv1i16.i64(<vscale x 1 x i16> [[DST:%.*]], <vscale x 1 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vslidedown.mask.nxv1i16.i64(<vscale x 1 x i16> [[DST:%.*]], <vscale x 1 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vslidedown_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t dst,
@@ -859,7 +859,7 @@ vuint16mf4_t test_vslidedown_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vslidedown.mask.nxv2i16.i64(<vscale x 2 x i16> [[DST:%.*]], <vscale x 2 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vslidedown.mask.nxv2i16.i64(<vscale x 2 x i16> [[DST:%.*]], <vscale x 2 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vslidedown_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t dst,
@@ -870,7 +870,7 @@ vuint16mf2_t test_vslidedown_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vslidedown.mask.nxv4i16.i64(<vscale x 4 x i16> [[DST:%.*]], <vscale x 4 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vslidedown.mask.nxv4i16.i64(<vscale x 4 x i16> [[DST:%.*]], <vscale x 4 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vslidedown_vx_u16m1_m(vbool16_t mask, vuint16m1_t dst,
@@ -881,7 +881,7 @@ vuint16m1_t test_vslidedown_vx_u16m1_m(vbool16_t mask, vuint16m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vslidedown.mask.nxv8i16.i64(<vscale x 8 x i16> [[DST:%.*]], <vscale x 8 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vslidedown.mask.nxv8i16.i64(<vscale x 8 x i16> [[DST:%.*]], <vscale x 8 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vslidedown_vx_u16m2_m(vbool8_t mask, vuint16m2_t dst,
@@ -892,7 +892,7 @@ vuint16m2_t test_vslidedown_vx_u16m2_m(vbool8_t mask, vuint16m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vslidedown.mask.nxv16i16.i64(<vscale x 16 x i16> [[DST:%.*]], <vscale x 16 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vslidedown.mask.nxv16i16.i64(<vscale x 16 x i16> [[DST:%.*]], <vscale x 16 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vslidedown_vx_u16m4_m(vbool4_t mask, vuint16m4_t dst,
@@ -903,7 +903,7 @@ vuint16m4_t test_vslidedown_vx_u16m4_m(vbool4_t mask, vuint16m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vslidedown.mask.nxv32i16.i64(<vscale x 32 x i16> [[DST:%.*]], <vscale x 32 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vslidedown.mask.nxv32i16.i64(<vscale x 32 x i16> [[DST:%.*]], <vscale x 32 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vslidedown_vx_u16m8_m(vbool2_t mask, vuint16m8_t dst,
@@ -914,7 +914,7 @@ vuint16m8_t test_vslidedown_vx_u16m8_m(vbool2_t mask, vuint16m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vslidedown.mask.nxv1i32.i64(<vscale x 1 x i32> [[DST:%.*]], <vscale x 1 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vslidedown.mask.nxv1i32.i64(<vscale x 1 x i32> [[DST:%.*]], <vscale x 1 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vslidedown_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t dst,
@@ -925,7 +925,7 @@ vuint32mf2_t test_vslidedown_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vslidedown.mask.nxv2i32.i64(<vscale x 2 x i32> [[DST:%.*]], <vscale x 2 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vslidedown.mask.nxv2i32.i64(<vscale x 2 x i32> [[DST:%.*]], <vscale x 2 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vslidedown_vx_u32m1_m(vbool32_t mask, vuint32m1_t dst,
@@ -936,7 +936,7 @@ vuint32m1_t test_vslidedown_vx_u32m1_m(vbool32_t mask, vuint32m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vslidedown.mask.nxv4i32.i64(<vscale x 4 x i32> [[DST:%.*]], <vscale x 4 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vslidedown.mask.nxv4i32.i64(<vscale x 4 x i32> [[DST:%.*]], <vscale x 4 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vslidedown_vx_u32m2_m(vbool16_t mask, vuint32m2_t dst,
@@ -947,7 +947,7 @@ vuint32m2_t test_vslidedown_vx_u32m2_m(vbool16_t mask, vuint32m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vslidedown.mask.nxv8i32.i64(<vscale x 8 x i32> [[DST:%.*]], <vscale x 8 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vslidedown.mask.nxv8i32.i64(<vscale x 8 x i32> [[DST:%.*]], <vscale x 8 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vslidedown_vx_u32m4_m(vbool8_t mask, vuint32m4_t dst,
@@ -958,7 +958,7 @@ vuint32m4_t test_vslidedown_vx_u32m4_m(vbool8_t mask, vuint32m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vslidedown.mask.nxv16i32.i64(<vscale x 16 x i32> [[DST:%.*]], <vscale x 16 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vslidedown.mask.nxv16i32.i64(<vscale x 16 x i32> [[DST:%.*]], <vscale x 16 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vslidedown_vx_u32m8_m(vbool4_t mask, vuint32m8_t dst,
@@ -969,7 +969,7 @@ vuint32m8_t test_vslidedown_vx_u32m8_m(vbool4_t mask, vuint32m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vslidedown.mask.nxv1i64.i64(<vscale x 1 x i64> [[DST:%.*]], <vscale x 1 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vslidedown.mask.nxv1i64.i64(<vscale x 1 x i64> [[DST:%.*]], <vscale x 1 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vslidedown_vx_u64m1_m(vbool64_t mask, vuint64m1_t dst,
@@ -980,7 +980,7 @@ vuint64m1_t test_vslidedown_vx_u64m1_m(vbool64_t mask, vuint64m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vslidedown.mask.nxv2i64.i64(<vscale x 2 x i64> [[DST:%.*]], <vscale x 2 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vslidedown.mask.nxv2i64.i64(<vscale x 2 x i64> [[DST:%.*]], <vscale x 2 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vslidedown_vx_u64m2_m(vbool32_t mask, vuint64m2_t dst,
@@ -991,7 +991,7 @@ vuint64m2_t test_vslidedown_vx_u64m2_m(vbool32_t mask, vuint64m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vslidedown.mask.nxv4i64.i64(<vscale x 4 x i64> [[DST:%.*]], <vscale x 4 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vslidedown.mask.nxv4i64.i64(<vscale x 4 x i64> [[DST:%.*]], <vscale x 4 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vslidedown_vx_u64m4_m(vbool16_t mask, vuint64m4_t dst,
@@ -1002,7 +1002,7 @@ vuint64m4_t test_vslidedown_vx_u64m4_m(vbool16_t mask, vuint64m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vslidedown.mask.nxv8i64.i64(<vscale x 8 x i64> [[DST:%.*]], <vscale x 8 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vslidedown.mask.nxv8i64.i64(<vscale x 8 x i64> [[DST:%.*]], <vscale x 8 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vslidedown_vx_u64m8_m(vbool8_t mask, vuint64m8_t dst,
@@ -1013,7 +1013,7 @@ vuint64m8_t test_vslidedown_vx_u64m8_m(vbool8_t mask, vuint64m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vslidedown.mask.nxv1f32.i64(<vscale x 1 x float> [[DST:%.*]], <vscale x 1 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vslidedown.mask.nxv1f32.i64(<vscale x 1 x float> [[DST:%.*]], <vscale x 1 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vslidedown_vx_f32mf2_m(vbool64_t mask, vfloat32mf2_t dst,
@@ -1024,7 +1024,7 @@ vfloat32mf2_t test_vslidedown_vx_f32mf2_m(vbool64_t mask, vfloat32mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vslidedown.mask.nxv2f32.i64(<vscale x 2 x float> [[DST:%.*]], <vscale x 2 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vslidedown.mask.nxv2f32.i64(<vscale x 2 x float> [[DST:%.*]], <vscale x 2 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vslidedown_vx_f32m1_m(vbool32_t mask, vfloat32m1_t dst,
@@ -1035,7 +1035,7 @@ vfloat32m1_t test_vslidedown_vx_f32m1_m(vbool32_t mask, vfloat32m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vslidedown.mask.nxv4f32.i64(<vscale x 4 x float> [[DST:%.*]], <vscale x 4 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vslidedown.mask.nxv4f32.i64(<vscale x 4 x float> [[DST:%.*]], <vscale x 4 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vslidedown_vx_f32m2_m(vbool16_t mask, vfloat32m2_t dst,
@@ -1046,7 +1046,7 @@ vfloat32m2_t test_vslidedown_vx_f32m2_m(vbool16_t mask, vfloat32m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vslidedown.mask.nxv8f32.i64(<vscale x 8 x float> [[DST:%.*]], <vscale x 8 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vslidedown.mask.nxv8f32.i64(<vscale x 8 x float> [[DST:%.*]], <vscale x 8 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vslidedown_vx_f32m4_m(vbool8_t mask, vfloat32m4_t dst,
@@ -1057,7 +1057,7 @@ vfloat32m4_t test_vslidedown_vx_f32m4_m(vbool8_t mask, vfloat32m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vslidedown.mask.nxv16f32.i64(<vscale x 16 x float> [[DST:%.*]], <vscale x 16 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vslidedown.mask.nxv16f32.i64(<vscale x 16 x float> [[DST:%.*]], <vscale x 16 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vslidedown_vx_f32m8_m(vbool4_t mask, vfloat32m8_t dst,
@@ -1068,7 +1068,7 @@ vfloat32m8_t test_vslidedown_vx_f32m8_m(vbool4_t mask, vfloat32m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vslidedown.mask.nxv1f64.i64(<vscale x 1 x double> [[DST:%.*]], <vscale x 1 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vslidedown.mask.nxv1f64.i64(<vscale x 1 x double> [[DST:%.*]], <vscale x 1 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vslidedown_vx_f64m1_m(vbool64_t mask, vfloat64m1_t dst,
@@ -1079,7 +1079,7 @@ vfloat64m1_t test_vslidedown_vx_f64m1_m(vbool64_t mask, vfloat64m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vslidedown.mask.nxv2f64.i64(<vscale x 2 x double> [[DST:%.*]], <vscale x 2 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vslidedown.mask.nxv2f64.i64(<vscale x 2 x double> [[DST:%.*]], <vscale x 2 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vslidedown_vx_f64m2_m(vbool32_t mask, vfloat64m2_t dst,
@@ -1090,7 +1090,7 @@ vfloat64m2_t test_vslidedown_vx_f64m2_m(vbool32_t mask, vfloat64m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vslidedown.mask.nxv4f64.i64(<vscale x 4 x double> [[DST:%.*]], <vscale x 4 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vslidedown.mask.nxv4f64.i64(<vscale x 4 x double> [[DST:%.*]], <vscale x 4 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vslidedown_vx_f64m4_m(vbool16_t mask, vfloat64m4_t dst,
@@ -1101,7 +1101,7 @@ vfloat64m4_t test_vslidedown_vx_f64m4_m(vbool16_t mask, vfloat64m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vslidedown.mask.nxv8f64.i64(<vscale x 8 x double> [[DST:%.*]], <vscale x 8 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vslidedown.mask.nxv8f64.i64(<vscale x 8 x double> [[DST:%.*]], <vscale x 8 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vslidedown_vx_f64m8_m(vbool8_t mask, vfloat64m8_t dst,
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vslideup.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vslideup.c
index 415934c6e2e64c..380296f0e49466 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vslideup.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vslideup.c
@@ -537,7 +537,7 @@ vfloat64m8_t test_vslideup_vx_f64m8(vfloat64m8_t dst, vfloat64m8_t src,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vslideup.mask.nxv1i8.i64(<vscale x 1 x i8> [[DST:%.*]], <vscale x 1 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vslideup.mask.nxv1i8.i64(<vscale x 1 x i8> [[DST:%.*]], <vscale x 1 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vint8mf8_t test_vslideup_vx_i8mf8_m(vbool64_t mask, vint8mf8_t dst,
@@ -547,7 +547,7 @@ vint8mf8_t test_vslideup_vx_i8mf8_m(vbool64_t mask, vint8mf8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vslideup.mask.nxv2i8.i64(<vscale x 2 x i8> [[DST:%.*]], <vscale x 2 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vslideup.mask.nxv2i8.i64(<vscale x 2 x i8> [[DST:%.*]], <vscale x 2 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vint8mf4_t test_vslideup_vx_i8mf4_m(vbool32_t mask, vint8mf4_t dst,
@@ -557,7 +557,7 @@ vint8mf4_t test_vslideup_vx_i8mf4_m(vbool32_t mask, vint8mf4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vslideup.mask.nxv4i8.i64(<vscale x 4 x i8> [[DST:%.*]], <vscale x 4 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vslideup.mask.nxv4i8.i64(<vscale x 4 x i8> [[DST:%.*]], <vscale x 4 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vint8mf2_t test_vslideup_vx_i8mf2_m(vbool16_t mask, vint8mf2_t dst,
@@ -567,7 +567,7 @@ vint8mf2_t test_vslideup_vx_i8mf2_m(vbool16_t mask, vint8mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vslideup.mask.nxv8i8.i64(<vscale x 8 x i8> [[DST:%.*]], <vscale x 8 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vslideup.mask.nxv8i8.i64(<vscale x 8 x i8> [[DST:%.*]], <vscale x 8 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vint8m1_t test_vslideup_vx_i8m1_m(vbool8_t mask, vint8m1_t dst, vint8m1_t src,
@@ -577,7 +577,7 @@ vint8m1_t test_vslideup_vx_i8m1_m(vbool8_t mask, vint8m1_t dst, vint8m1_t src,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vslideup.mask.nxv16i8.i64(<vscale x 16 x i8> [[DST:%.*]], <vscale x 16 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vslideup.mask.nxv16i8.i64(<vscale x 16 x i8> [[DST:%.*]], <vscale x 16 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vint8m2_t test_vslideup_vx_i8m2_m(vbool4_t mask, vint8m2_t dst, vint8m2_t src,
@@ -587,7 +587,7 @@ vint8m2_t test_vslideup_vx_i8m2_m(vbool4_t mask, vint8m2_t dst, vint8m2_t src,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vslideup.mask.nxv32i8.i64(<vscale x 32 x i8> [[DST:%.*]], <vscale x 32 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vslideup.mask.nxv32i8.i64(<vscale x 32 x i8> [[DST:%.*]], <vscale x 32 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vint8m4_t test_vslideup_vx_i8m4_m(vbool2_t mask, vint8m4_t dst, vint8m4_t src,
@@ -597,7 +597,7 @@ vint8m4_t test_vslideup_vx_i8m4_m(vbool2_t mask, vint8m4_t dst, vint8m4_t src,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vslideup.mask.nxv64i8.i64(<vscale x 64 x i8> [[DST:%.*]], <vscale x 64 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vslideup.mask.nxv64i8.i64(<vscale x 64 x i8> [[DST:%.*]], <vscale x 64 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vint8m8_t test_vslideup_vx_i8m8_m(vbool1_t mask, vint8m8_t dst, vint8m8_t src,
@@ -607,7 +607,7 @@ vint8m8_t test_vslideup_vx_i8m8_m(vbool1_t mask, vint8m8_t dst, vint8m8_t src,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vslideup.mask.nxv1i16.i64(<vscale x 1 x i16> [[DST:%.*]], <vscale x 1 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vslideup.mask.nxv1i16.i64(<vscale x 1 x i16> [[DST:%.*]], <vscale x 1 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vslideup_vx_i16mf4_m(vbool64_t mask, vint16mf4_t dst,
@@ -618,7 +618,7 @@ vint16mf4_t test_vslideup_vx_i16mf4_m(vbool64_t mask, vint16mf4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vslideup.mask.nxv2i16.i64(<vscale x 2 x i16> [[DST:%.*]], <vscale x 2 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vslideup.mask.nxv2i16.i64(<vscale x 2 x i16> [[DST:%.*]], <vscale x 2 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vslideup_vx_i16mf2_m(vbool32_t mask, vint16mf2_t dst,
@@ -629,7 +629,7 @@ vint16mf2_t test_vslideup_vx_i16mf2_m(vbool32_t mask, vint16mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vslideup.mask.nxv4i16.i64(<vscale x 4 x i16> [[DST:%.*]], <vscale x 4 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vslideup.mask.nxv4i16.i64(<vscale x 4 x i16> [[DST:%.*]], <vscale x 4 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vslideup_vx_i16m1_m(vbool16_t mask, vint16m1_t dst,
@@ -639,7 +639,7 @@ vint16m1_t test_vslideup_vx_i16m1_m(vbool16_t mask, vint16m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vslideup.mask.nxv8i16.i64(<vscale x 8 x i16> [[DST:%.*]], <vscale x 8 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vslideup.mask.nxv8i16.i64(<vscale x 8 x i16> [[DST:%.*]], <vscale x 8 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vslideup_vx_i16m2_m(vbool8_t mask, vint16m2_t dst,
@@ -649,7 +649,7 @@ vint16m2_t test_vslideup_vx_i16m2_m(vbool8_t mask, vint16m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vslideup.mask.nxv16i16.i64(<vscale x 16 x i16> [[DST:%.*]], <vscale x 16 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vslideup.mask.nxv16i16.i64(<vscale x 16 x i16> [[DST:%.*]], <vscale x 16 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vslideup_vx_i16m4_m(vbool4_t mask, vint16m4_t dst,
@@ -659,7 +659,7 @@ vint16m4_t test_vslideup_vx_i16m4_m(vbool4_t mask, vint16m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vslideup.mask.nxv32i16.i64(<vscale x 32 x i16> [[DST:%.*]], <vscale x 32 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vslideup.mask.nxv32i16.i64(<vscale x 32 x i16> [[DST:%.*]], <vscale x 32 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vslideup_vx_i16m8_m(vbool2_t mask, vint16m8_t dst,
@@ -669,7 +669,7 @@ vint16m8_t test_vslideup_vx_i16m8_m(vbool2_t mask, vint16m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vslideup.mask.nxv1i32.i64(<vscale x 1 x i32> [[DST:%.*]], <vscale x 1 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vslideup.mask.nxv1i32.i64(<vscale x 1 x i32> [[DST:%.*]], <vscale x 1 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vslideup_vx_i32mf2_m(vbool64_t mask, vint32mf2_t dst,
@@ -680,7 +680,7 @@ vint32mf2_t test_vslideup_vx_i32mf2_m(vbool64_t mask, vint32mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vslideup.mask.nxv2i32.i64(<vscale x 2 x i32> [[DST:%.*]], <vscale x 2 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vslideup.mask.nxv2i32.i64(<vscale x 2 x i32> [[DST:%.*]], <vscale x 2 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vslideup_vx_i32m1_m(vbool32_t mask, vint32m1_t dst,
@@ -690,7 +690,7 @@ vint32m1_t test_vslideup_vx_i32m1_m(vbool32_t mask, vint32m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vslideup.mask.nxv4i32.i64(<vscale x 4 x i32> [[DST:%.*]], <vscale x 4 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vslideup.mask.nxv4i32.i64(<vscale x 4 x i32> [[DST:%.*]], <vscale x 4 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vslideup_vx_i32m2_m(vbool16_t mask, vint32m2_t dst,
@@ -700,7 +700,7 @@ vint32m2_t test_vslideup_vx_i32m2_m(vbool16_t mask, vint32m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vslideup.mask.nxv8i32.i64(<vscale x 8 x i32> [[DST:%.*]], <vscale x 8 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vslideup.mask.nxv8i32.i64(<vscale x 8 x i32> [[DST:%.*]], <vscale x 8 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vslideup_vx_i32m4_m(vbool8_t mask, vint32m4_t dst,
@@ -710,7 +710,7 @@ vint32m4_t test_vslideup_vx_i32m4_m(vbool8_t mask, vint32m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vslideup.mask.nxv16i32.i64(<vscale x 16 x i32> [[DST:%.*]], <vscale x 16 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vslideup.mask.nxv16i32.i64(<vscale x 16 x i32> [[DST:%.*]], <vscale x 16 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vslideup_vx_i32m8_m(vbool4_t mask, vint32m8_t dst,
@@ -720,7 +720,7 @@ vint32m8_t test_vslideup_vx_i32m8_m(vbool4_t mask, vint32m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vslideup.mask.nxv1i64.i64(<vscale x 1 x i64> [[DST:%.*]], <vscale x 1 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vslideup.mask.nxv1i64.i64(<vscale x 1 x i64> [[DST:%.*]], <vscale x 1 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vslideup_vx_i64m1_m(vbool64_t mask, vint64m1_t dst,
@@ -730,7 +730,7 @@ vint64m1_t test_vslideup_vx_i64m1_m(vbool64_t mask, vint64m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vslideup.mask.nxv2i64.i64(<vscale x 2 x i64> [[DST:%.*]], <vscale x 2 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vslideup.mask.nxv2i64.i64(<vscale x 2 x i64> [[DST:%.*]], <vscale x 2 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vslideup_vx_i64m2_m(vbool32_t mask, vint64m2_t dst,
@@ -740,7 +740,7 @@ vint64m2_t test_vslideup_vx_i64m2_m(vbool32_t mask, vint64m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vslideup.mask.nxv4i64.i64(<vscale x 4 x i64> [[DST:%.*]], <vscale x 4 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vslideup.mask.nxv4i64.i64(<vscale x 4 x i64> [[DST:%.*]], <vscale x 4 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vslideup_vx_i64m4_m(vbool16_t mask, vint64m4_t dst,
@@ -750,7 +750,7 @@ vint64m4_t test_vslideup_vx_i64m4_m(vbool16_t mask, vint64m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vslideup.mask.nxv8i64.i64(<vscale x 8 x i64> [[DST:%.*]], <vscale x 8 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vslideup.mask.nxv8i64.i64(<vscale x 8 x i64> [[DST:%.*]], <vscale x 8 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vslideup_vx_i64m8_m(vbool8_t mask, vint64m8_t dst,
@@ -760,7 +760,7 @@ vint64m8_t test_vslideup_vx_i64m8_m(vbool8_t mask, vint64m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vslideup.mask.nxv1i8.i64(<vscale x 1 x i8> [[DST:%.*]], <vscale x 1 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vslideup.mask.nxv1i8.i64(<vscale x 1 x i8> [[DST:%.*]], <vscale x 1 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vuint8mf8_t test_vslideup_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t dst,
@@ -771,7 +771,7 @@ vuint8mf8_t test_vslideup_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vslideup.mask.nxv2i8.i64(<vscale x 2 x i8> [[DST:%.*]], <vscale x 2 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vslideup.mask.nxv2i8.i64(<vscale x 2 x i8> [[DST:%.*]], <vscale x 2 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vuint8mf4_t test_vslideup_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t dst,
@@ -782,7 +782,7 @@ vuint8mf4_t test_vslideup_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vslideup.mask.nxv4i8.i64(<vscale x 4 x i8> [[DST:%.*]], <vscale x 4 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vslideup.mask.nxv4i8.i64(<vscale x 4 x i8> [[DST:%.*]], <vscale x 4 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vuint8mf2_t test_vslideup_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t dst,
@@ -793,7 +793,7 @@ vuint8mf2_t test_vslideup_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vslideup.mask.nxv8i8.i64(<vscale x 8 x i8> [[DST:%.*]], <vscale x 8 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vslideup.mask.nxv8i8.i64(<vscale x 8 x i8> [[DST:%.*]], <vscale x 8 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vuint8m1_t test_vslideup_vx_u8m1_m(vbool8_t mask, vuint8m1_t dst,
@@ -803,7 +803,7 @@ vuint8m1_t test_vslideup_vx_u8m1_m(vbool8_t mask, vuint8m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vslideup.mask.nxv16i8.i64(<vscale x 16 x i8> [[DST:%.*]], <vscale x 16 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vslideup.mask.nxv16i8.i64(<vscale x 16 x i8> [[DST:%.*]], <vscale x 16 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vuint8m2_t test_vslideup_vx_u8m2_m(vbool4_t mask, vuint8m2_t dst,
@@ -813,7 +813,7 @@ vuint8m2_t test_vslideup_vx_u8m2_m(vbool4_t mask, vuint8m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vslideup.mask.nxv32i8.i64(<vscale x 32 x i8> [[DST:%.*]], <vscale x 32 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vslideup.mask.nxv32i8.i64(<vscale x 32 x i8> [[DST:%.*]], <vscale x 32 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vuint8m4_t test_vslideup_vx_u8m4_m(vbool2_t mask, vuint8m4_t dst,
@@ -823,7 +823,7 @@ vuint8m4_t test_vslideup_vx_u8m4_m(vbool2_t mask, vuint8m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vslideup.mask.nxv64i8.i64(<vscale x 64 x i8> [[DST:%.*]], <vscale x 64 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vslideup.mask.nxv64i8.i64(<vscale x 64 x i8> [[DST:%.*]], <vscale x 64 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vuint8m8_t test_vslideup_vx_u8m8_m(vbool1_t mask, vuint8m8_t dst,
@@ -833,7 +833,7 @@ vuint8m8_t test_vslideup_vx_u8m8_m(vbool1_t mask, vuint8m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vslideup.mask.nxv1i16.i64(<vscale x 1 x i16> [[DST:%.*]], <vscale x 1 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vslideup.mask.nxv1i16.i64(<vscale x 1 x i16> [[DST:%.*]], <vscale x 1 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vslideup_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t dst,
@@ -844,7 +844,7 @@ vuint16mf4_t test_vslideup_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vslideup.mask.nxv2i16.i64(<vscale x 2 x i16> [[DST:%.*]], <vscale x 2 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vslideup.mask.nxv2i16.i64(<vscale x 2 x i16> [[DST:%.*]], <vscale x 2 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vslideup_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t dst,
@@ -855,7 +855,7 @@ vuint16mf2_t test_vslideup_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vslideup.mask.nxv4i16.i64(<vscale x 4 x i16> [[DST:%.*]], <vscale x 4 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vslideup.mask.nxv4i16.i64(<vscale x 4 x i16> [[DST:%.*]], <vscale x 4 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vslideup_vx_u16m1_m(vbool16_t mask, vuint16m1_t dst,
@@ -866,7 +866,7 @@ vuint16m1_t test_vslideup_vx_u16m1_m(vbool16_t mask, vuint16m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vslideup.mask.nxv8i16.i64(<vscale x 8 x i16> [[DST:%.*]], <vscale x 8 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vslideup.mask.nxv8i16.i64(<vscale x 8 x i16> [[DST:%.*]], <vscale x 8 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vslideup_vx_u16m2_m(vbool8_t mask, vuint16m2_t dst,
@@ -877,7 +877,7 @@ vuint16m2_t test_vslideup_vx_u16m2_m(vbool8_t mask, vuint16m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vslideup.mask.nxv16i16.i64(<vscale x 16 x i16> [[DST:%.*]], <vscale x 16 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vslideup.mask.nxv16i16.i64(<vscale x 16 x i16> [[DST:%.*]], <vscale x 16 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vslideup_vx_u16m4_m(vbool4_t mask, vuint16m4_t dst,
@@ -888,7 +888,7 @@ vuint16m4_t test_vslideup_vx_u16m4_m(vbool4_t mask, vuint16m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vslideup.mask.nxv32i16.i64(<vscale x 32 x i16> [[DST:%.*]], <vscale x 32 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vslideup.mask.nxv32i16.i64(<vscale x 32 x i16> [[DST:%.*]], <vscale x 32 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vslideup_vx_u16m8_m(vbool2_t mask, vuint16m8_t dst,
@@ -899,7 +899,7 @@ vuint16m8_t test_vslideup_vx_u16m8_m(vbool2_t mask, vuint16m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vslideup.mask.nxv1i32.i64(<vscale x 1 x i32> [[DST:%.*]], <vscale x 1 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vslideup.mask.nxv1i32.i64(<vscale x 1 x i32> [[DST:%.*]], <vscale x 1 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vslideup_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t dst,
@@ -910,7 +910,7 @@ vuint32mf2_t test_vslideup_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vslideup.mask.nxv2i32.i64(<vscale x 2 x i32> [[DST:%.*]], <vscale x 2 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vslideup.mask.nxv2i32.i64(<vscale x 2 x i32> [[DST:%.*]], <vscale x 2 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vslideup_vx_u32m1_m(vbool32_t mask, vuint32m1_t dst,
@@ -921,7 +921,7 @@ vuint32m1_t test_vslideup_vx_u32m1_m(vbool32_t mask, vuint32m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vslideup.mask.nxv4i32.i64(<vscale x 4 x i32> [[DST:%.*]], <vscale x 4 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vslideup.mask.nxv4i32.i64(<vscale x 4 x i32> [[DST:%.*]], <vscale x 4 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vslideup_vx_u32m2_m(vbool16_t mask, vuint32m2_t dst,
@@ -932,7 +932,7 @@ vuint32m2_t test_vslideup_vx_u32m2_m(vbool16_t mask, vuint32m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vslideup.mask.nxv8i32.i64(<vscale x 8 x i32> [[DST:%.*]], <vscale x 8 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vslideup.mask.nxv8i32.i64(<vscale x 8 x i32> [[DST:%.*]], <vscale x 8 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vslideup_vx_u32m4_m(vbool8_t mask, vuint32m4_t dst,
@@ -943,7 +943,7 @@ vuint32m4_t test_vslideup_vx_u32m4_m(vbool8_t mask, vuint32m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vslideup.mask.nxv16i32.i64(<vscale x 16 x i32> [[DST:%.*]], <vscale x 16 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vslideup.mask.nxv16i32.i64(<vscale x 16 x i32> [[DST:%.*]], <vscale x 16 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vslideup_vx_u32m8_m(vbool4_t mask, vuint32m8_t dst,
@@ -954,7 +954,7 @@ vuint32m8_t test_vslideup_vx_u32m8_m(vbool4_t mask, vuint32m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vslideup.mask.nxv1i64.i64(<vscale x 1 x i64> [[DST:%.*]], <vscale x 1 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vslideup.mask.nxv1i64.i64(<vscale x 1 x i64> [[DST:%.*]], <vscale x 1 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vslideup_vx_u64m1_m(vbool64_t mask, vuint64m1_t dst,
@@ -965,7 +965,7 @@ vuint64m1_t test_vslideup_vx_u64m1_m(vbool64_t mask, vuint64m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vslideup.mask.nxv2i64.i64(<vscale x 2 x i64> [[DST:%.*]], <vscale x 2 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vslideup.mask.nxv2i64.i64(<vscale x 2 x i64> [[DST:%.*]], <vscale x 2 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vslideup_vx_u64m2_m(vbool32_t mask, vuint64m2_t dst,
@@ -976,7 +976,7 @@ vuint64m2_t test_vslideup_vx_u64m2_m(vbool32_t mask, vuint64m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vslideup.mask.nxv4i64.i64(<vscale x 4 x i64> [[DST:%.*]], <vscale x 4 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vslideup.mask.nxv4i64.i64(<vscale x 4 x i64> [[DST:%.*]], <vscale x 4 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vslideup_vx_u64m4_m(vbool16_t mask, vuint64m4_t dst,
@@ -987,7 +987,7 @@ vuint64m4_t test_vslideup_vx_u64m4_m(vbool16_t mask, vuint64m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vslideup.mask.nxv8i64.i64(<vscale x 8 x i64> [[DST:%.*]], <vscale x 8 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vslideup.mask.nxv8i64.i64(<vscale x 8 x i64> [[DST:%.*]], <vscale x 8 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vslideup_vx_u64m8_m(vbool8_t mask, vuint64m8_t dst,
@@ -998,7 +998,7 @@ vuint64m8_t test_vslideup_vx_u64m8_m(vbool8_t mask, vuint64m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vslideup.mask.nxv1f32.i64(<vscale x 1 x float> [[DST:%.*]], <vscale x 1 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vslideup.mask.nxv1f32.i64(<vscale x 1 x float> [[DST:%.*]], <vscale x 1 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vslideup_vx_f32mf2_m(vbool64_t mask, vfloat32mf2_t dst,
@@ -1009,7 +1009,7 @@ vfloat32mf2_t test_vslideup_vx_f32mf2_m(vbool64_t mask, vfloat32mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vslideup.mask.nxv2f32.i64(<vscale x 2 x float> [[DST:%.*]], <vscale x 2 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vslideup.mask.nxv2f32.i64(<vscale x 2 x float> [[DST:%.*]], <vscale x 2 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vslideup_vx_f32m1_m(vbool32_t mask, vfloat32m1_t dst,
@@ -1020,7 +1020,7 @@ vfloat32m1_t test_vslideup_vx_f32m1_m(vbool32_t mask, vfloat32m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vslideup.mask.nxv4f32.i64(<vscale x 4 x float> [[DST:%.*]], <vscale x 4 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vslideup.mask.nxv4f32.i64(<vscale x 4 x float> [[DST:%.*]], <vscale x 4 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vslideup_vx_f32m2_m(vbool16_t mask, vfloat32m2_t dst,
@@ -1031,7 +1031,7 @@ vfloat32m2_t test_vslideup_vx_f32m2_m(vbool16_t mask, vfloat32m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vslideup.mask.nxv8f32.i64(<vscale x 8 x float> [[DST:%.*]], <vscale x 8 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vslideup.mask.nxv8f32.i64(<vscale x 8 x float> [[DST:%.*]], <vscale x 8 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vslideup_vx_f32m4_m(vbool8_t mask, vfloat32m4_t dst,
@@ -1042,7 +1042,7 @@ vfloat32m4_t test_vslideup_vx_f32m4_m(vbool8_t mask, vfloat32m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vslideup.mask.nxv16f32.i64(<vscale x 16 x float> [[DST:%.*]], <vscale x 16 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vslideup.mask.nxv16f32.i64(<vscale x 16 x float> [[DST:%.*]], <vscale x 16 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vslideup_vx_f32m8_m(vbool4_t mask, vfloat32m8_t dst,
@@ -1053,7 +1053,7 @@ vfloat32m8_t test_vslideup_vx_f32m8_m(vbool4_t mask, vfloat32m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vslideup.mask.nxv1f64.i64(<vscale x 1 x double> [[DST:%.*]], <vscale x 1 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vslideup.mask.nxv1f64.i64(<vscale x 1 x double> [[DST:%.*]], <vscale x 1 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vslideup_vx_f64m1_m(vbool64_t mask, vfloat64m1_t dst,
@@ -1064,7 +1064,7 @@ vfloat64m1_t test_vslideup_vx_f64m1_m(vbool64_t mask, vfloat64m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vslideup.mask.nxv2f64.i64(<vscale x 2 x double> [[DST:%.*]], <vscale x 2 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vslideup.mask.nxv2f64.i64(<vscale x 2 x double> [[DST:%.*]], <vscale x 2 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vslideup_vx_f64m2_m(vbool32_t mask, vfloat64m2_t dst,
@@ -1075,7 +1075,7 @@ vfloat64m2_t test_vslideup_vx_f64m2_m(vbool32_t mask, vfloat64m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vslideup.mask.nxv4f64.i64(<vscale x 4 x double> [[DST:%.*]], <vscale x 4 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vslideup.mask.nxv4f64.i64(<vscale x 4 x double> [[DST:%.*]], <vscale x 4 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vslideup_vx_f64m4_m(vbool16_t mask, vfloat64m4_t dst,
@@ -1086,7 +1086,7 @@ vfloat64m4_t test_vslideup_vx_f64m4_m(vbool16_t mask, vfloat64m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vslideup.mask.nxv8f64.i64(<vscale x 8 x double> [[DST:%.*]], <vscale x 8 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vslideup.mask.nxv8f64.i64(<vscale x 8 x double> [[DST:%.*]], <vscale x 8 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vslideup_vx_f64m8_m(vbool8_t mask, vfloat64m8_t dst,
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vwmacc.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vwmacc.c
index 05edc0718edb3a..916c14745e0d7b 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vwmacc.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vwmacc.c
@@ -1056,7 +1056,7 @@ vint64m8_t test_vwmaccus_vx_i64m8(vint64m8_t acc, uint32_t op1, vint32m4_t op2,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vwmacc_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc,
@@ -1066,7 +1066,7 @@ vint16mf4_t test_vwmacc_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vwmacc_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc, int8_t op1,
@@ -1076,7 +1076,7 @@ vint16mf4_t test_vwmacc_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc, int8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmacc.mask.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmacc.mask.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vwmacc_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc,
@@ -1086,7 +1086,7 @@ vint16mf2_t test_vwmacc_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmacc.mask.nxv2i16.i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmacc.mask.nxv2i16.i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vwmacc_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc, int8_t op1,
@@ -1096,7 +1096,7 @@ vint16mf2_t test_vwmacc_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc, int8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmacc.mask.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmacc.mask.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vwmacc_vv_i16m1_m(vbool16_t mask, vint16m1_t acc,
@@ -1106,7 +1106,7 @@ vint16m1_t test_vwmacc_vv_i16m1_m(vbool16_t mask, vint16m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmacc.mask.nxv4i16.i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmacc.mask.nxv4i16.i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vwmacc_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int8_t op1,
@@ -1116,7 +1116,7 @@ vint16m1_t test_vwmacc_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmacc.mask.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmacc.mask.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vwmacc_vv_i16m2_m(vbool8_t mask, vint16m2_t acc, vint8m1_t op1,
@@ -1126,7 +1126,7 @@ vint16m2_t test_vwmacc_vv_i16m2_m(vbool8_t mask, vint16m2_t acc, vint8m1_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmacc.mask.nxv8i16.i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmacc.mask.nxv8i16.i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vwmacc_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int8_t op1,
@@ -1136,7 +1136,7 @@ vint16m2_t test_vwmacc_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmacc.mask.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmacc.mask.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vwmacc_vv_i16m4_m(vbool4_t mask, vint16m4_t acc, vint8m2_t op1,
@@ -1146,7 +1146,7 @@ vint16m4_t test_vwmacc_vv_i16m4_m(vbool4_t mask, vint16m4_t acc, vint8m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmacc.mask.nxv16i16.i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmacc.mask.nxv16i16.i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vwmacc_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int8_t op1,
@@ -1156,7 +1156,7 @@ vint16m4_t test_vwmacc_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmacc.mask.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmacc.mask.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vwmacc_vv_i16m8_m(vbool2_t mask, vint16m8_t acc, vint8m4_t op1,
@@ -1166,7 +1166,7 @@ vint16m8_t test_vwmacc_vv_i16m8_m(vbool2_t mask, vint16m8_t acc, vint8m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmacc.mask.nxv32i16.i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmacc.mask.nxv32i16.i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vwmacc_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int8_t op1,
@@ -1176,7 +1176,7 @@ vint16m8_t test_vwmacc_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmacc.mask.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmacc.mask.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vwmacc_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc,
@@ -1187,7 +1187,7 @@ vint32mf2_t test_vwmacc_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmacc.mask.nxv1i32.i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmacc.mask.nxv1i32.i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vwmacc_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc,
@@ -1197,7 +1197,7 @@ vint32mf2_t test_vwmacc_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmacc.mask.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmacc.mask.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vwmacc_vv_i32m1_m(vbool32_t mask, vint32m1_t acc,
@@ -1207,7 +1207,7 @@ vint32m1_t test_vwmacc_vv_i32m1_m(vbool32_t mask, vint32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmacc.mask.nxv2i32.i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmacc.mask.nxv2i32.i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vwmacc_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int16_t op1,
@@ -1217,7 +1217,7 @@ vint32m1_t test_vwmacc_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmacc.mask.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmacc.mask.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vwmacc_vv_i32m2_m(vbool16_t mask, vint32m2_t acc,
@@ -1227,7 +1227,7 @@ vint32m2_t test_vwmacc_vv_i32m2_m(vbool16_t mask, vint32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmacc.mask.nxv4i32.i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmacc.mask.nxv4i32.i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vwmacc_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int16_t op1,
@@ -1237,7 +1237,7 @@ vint32m2_t test_vwmacc_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmacc.mask.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmacc.mask.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vwmacc_vv_i32m4_m(vbool8_t mask, vint32m4_t acc, vint16m2_t op1,
@@ -1247,7 +1247,7 @@ vint32m4_t test_vwmacc_vv_i32m4_m(vbool8_t mask, vint32m4_t acc, vint16m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmacc.mask.nxv8i32.i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmacc.mask.nxv8i32.i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vwmacc_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int16_t op1,
@@ -1257,7 +1257,7 @@ vint32m4_t test_vwmacc_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmacc.mask.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmacc.mask.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vwmacc_vv_i32m8_m(vbool4_t mask, vint32m8_t acc, vint16m4_t op1,
@@ -1267,7 +1267,7 @@ vint32m8_t test_vwmacc_vv_i32m8_m(vbool4_t mask, vint32m8_t acc, vint16m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmacc.mask.nxv16i32.i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmacc.mask.nxv16i32.i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vwmacc_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int16_t op1,
@@ -1277,7 +1277,7 @@ vint32m8_t test_vwmacc_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmacc.mask.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmacc.mask.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vwmacc_vv_i64m1_m(vbool64_t mask, vint64m1_t acc,
@@ -1287,7 +1287,7 @@ vint64m1_t test_vwmacc_vv_i64m1_m(vbool64_t mask, vint64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmacc.mask.nxv1i64.i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmacc.mask.nxv1i64.i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vwmacc_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int32_t op1,
@@ -1297,7 +1297,7 @@ vint64m1_t test_vwmacc_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int32_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmacc.mask.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmacc.mask.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vwmacc_vv_i64m2_m(vbool32_t mask, vint64m2_t acc,
@@ -1307,7 +1307,7 @@ vint64m2_t test_vwmacc_vv_i64m2_m(vbool32_t mask, vint64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmacc.mask.nxv2i64.i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmacc.mask.nxv2i64.i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vwmacc_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int32_t op1,
@@ -1317,7 +1317,7 @@ vint64m2_t test_vwmacc_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int32_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmacc.mask.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmacc.mask.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vwmacc_vv_i64m4_m(vbool16_t mask, vint64m4_t acc,
@@ -1327,7 +1327,7 @@ vint64m4_t test_vwmacc_vv_i64m4_m(vbool16_t mask, vint64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmacc.mask.nxv4i64.i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmacc.mask.nxv4i64.i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vwmacc_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int32_t op1,
@@ -1337,7 +1337,7 @@ vint64m4_t test_vwmacc_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int32_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmacc.mask.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmacc.mask.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vwmacc_vv_i64m8_m(vbool8_t mask, vint64m8_t acc, vint32m4_t op1,
@@ -1347,7 +1347,7 @@ vint64m8_t test_vwmacc_vv_i64m8_m(vbool8_t mask, vint64m8_t acc, vint32m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmacc.mask.nxv8i64.i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmacc.mask.nxv8i64.i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vwmacc_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int32_t op1,
@@ -1357,7 +1357,7 @@ vint64m8_t test_vwmacc_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int32_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmaccu.mask.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmaccu.mask.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vwmaccu_vv_u16mf4_m(vbool64_t mask, vuint16mf4_t acc,
@@ -1368,7 +1368,7 @@ vuint16mf4_t test_vwmaccu_vv_u16mf4_m(vbool64_t mask, vuint16mf4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmaccu.mask.nxv1i16.i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmaccu.mask.nxv1i16.i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vwmaccu_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t acc,
@@ -1378,7 +1378,7 @@ vuint16mf4_t test_vwmaccu_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmaccu.mask.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmaccu.mask.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vwmaccu_vv_u16mf2_m(vbool32_t mask, vuint16mf2_t acc,
@@ -1389,7 +1389,7 @@ vuint16mf2_t test_vwmaccu_vv_u16mf2_m(vbool32_t mask, vuint16mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmaccu.mask.nxv2i16.i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmaccu.mask.nxv2i16.i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vwmaccu_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t acc,
@@ -1399,7 +1399,7 @@ vuint16mf2_t test_vwmaccu_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmaccu.mask.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmaccu.mask.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vwmaccu_vv_u16m1_m(vbool16_t mask, vuint16m1_t acc,
@@ -1410,7 +1410,7 @@ vuint16m1_t test_vwmaccu_vv_u16m1_m(vbool16_t mask, vuint16m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmaccu.mask.nxv4i16.i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmaccu.mask.nxv4i16.i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vwmaccu_vx_u16m1_m(vbool16_t mask, vuint16m1_t acc,
@@ -1420,7 +1420,7 @@ vuint16m1_t test_vwmaccu_vx_u16m1_m(vbool16_t mask, vuint16m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmaccu.mask.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmaccu.mask.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vwmaccu_vv_u16m2_m(vbool8_t mask, vuint16m2_t acc,
@@ -1430,7 +1430,7 @@ vuint16m2_t test_vwmaccu_vv_u16m2_m(vbool8_t mask, vuint16m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmaccu.mask.nxv8i16.i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmaccu.mask.nxv8i16.i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vwmaccu_vx_u16m2_m(vbool8_t mask, vuint16m2_t acc, uint8_t op1,
@@ -1440,7 +1440,7 @@ vuint16m2_t test_vwmaccu_vx_u16m2_m(vbool8_t mask, vuint16m2_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmaccu.mask.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmaccu.mask.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vwmaccu_vv_u16m4_m(vbool4_t mask, vuint16m4_t acc,
@@ -1450,7 +1450,7 @@ vuint16m4_t test_vwmaccu_vv_u16m4_m(vbool4_t mask, vuint16m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmaccu.mask.nxv16i16.i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmaccu.mask.nxv16i16.i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vwmaccu_vx_u16m4_m(vbool4_t mask, vuint16m4_t acc, uint8_t op1,
@@ -1460,7 +1460,7 @@ vuint16m4_t test_vwmaccu_vx_u16m4_m(vbool4_t mask, vuint16m4_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmaccu.mask.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmaccu.mask.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vwmaccu_vv_u16m8_m(vbool2_t mask, vuint16m8_t acc,
@@ -1470,7 +1470,7 @@ vuint16m8_t test_vwmaccu_vv_u16m8_m(vbool2_t mask, vuint16m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmaccu.mask.nxv32i16.i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmaccu.mask.nxv32i16.i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vwmaccu_vx_u16m8_m(vbool2_t mask, vuint16m8_t acc, uint8_t op1,
@@ -1480,7 +1480,7 @@ vuint16m8_t test_vwmaccu_vx_u16m8_m(vbool2_t mask, vuint16m8_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmaccu.mask.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmaccu.mask.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vwmaccu_vv_u32mf2_m(vbool64_t mask, vuint32mf2_t acc,
@@ -1491,7 +1491,7 @@ vuint32mf2_t test_vwmaccu_vv_u32mf2_m(vbool64_t mask, vuint32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmaccu.mask.nxv1i32.i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmaccu.mask.nxv1i32.i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vwmaccu_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t acc,
@@ -1502,7 +1502,7 @@ vuint32mf2_t test_vwmaccu_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmaccu.mask.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmaccu.mask.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vwmaccu_vv_u32m1_m(vbool32_t mask, vuint32m1_t acc,
@@ -1513,7 +1513,7 @@ vuint32m1_t test_vwmaccu_vv_u32m1_m(vbool32_t mask, vuint32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmaccu.mask.nxv2i32.i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmaccu.mask.nxv2i32.i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vwmaccu_vx_u32m1_m(vbool32_t mask, vuint32m1_t acc,
@@ -1523,7 +1523,7 @@ vuint32m1_t test_vwmaccu_vx_u32m1_m(vbool32_t mask, vuint32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmaccu.mask.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmaccu.mask.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vwmaccu_vv_u32m2_m(vbool16_t mask, vuint32m2_t acc,
@@ -1534,7 +1534,7 @@ vuint32m2_t test_vwmaccu_vv_u32m2_m(vbool16_t mask, vuint32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmaccu.mask.nxv4i32.i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmaccu.mask.nxv4i32.i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vwmaccu_vx_u32m2_m(vbool16_t mask, vuint32m2_t acc,
@@ -1544,7 +1544,7 @@ vuint32m2_t test_vwmaccu_vx_u32m2_m(vbool16_t mask, vuint32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmaccu.mask.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmaccu.mask.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vwmaccu_vv_u32m4_m(vbool8_t mask, vuint32m4_t acc,
@@ -1555,7 +1555,7 @@ vuint32m4_t test_vwmaccu_vv_u32m4_m(vbool8_t mask, vuint32m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmaccu.mask.nxv8i32.i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmaccu.mask.nxv8i32.i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vwmaccu_vx_u32m4_m(vbool8_t mask, vuint32m4_t acc,
@@ -1565,7 +1565,7 @@ vuint32m4_t test_vwmaccu_vx_u32m4_m(vbool8_t mask, vuint32m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmaccu.mask.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmaccu.mask.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vwmaccu_vv_u32m8_m(vbool4_t mask, vuint32m8_t acc,
@@ -1576,7 +1576,7 @@ vuint32m8_t test_vwmaccu_vv_u32m8_m(vbool4_t mask, vuint32m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmaccu.mask.nxv16i32.i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmaccu.mask.nxv16i32.i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vwmaccu_vx_u32m8_m(vbool4_t mask, vuint32m8_t acc,
@@ -1586,7 +1586,7 @@ vuint32m8_t test_vwmaccu_vx_u32m8_m(vbool4_t mask, vuint32m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmaccu.mask.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmaccu.mask.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vwmaccu_vv_u64m1_m(vbool64_t mask, vuint64m1_t acc,
@@ -1597,7 +1597,7 @@ vuint64m1_t test_vwmaccu_vv_u64m1_m(vbool64_t mask, vuint64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmaccu.mask.nxv1i64.i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmaccu.mask.nxv1i64.i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vwmaccu_vx_u64m1_m(vbool64_t mask, vuint64m1_t acc,
@@ -1607,7 +1607,7 @@ vuint64m1_t test_vwmaccu_vx_u64m1_m(vbool64_t mask, vuint64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmaccu.mask.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmaccu.mask.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vwmaccu_vv_u64m2_m(vbool32_t mask, vuint64m2_t acc,
@@ -1618,7 +1618,7 @@ vuint64m2_t test_vwmaccu_vv_u64m2_m(vbool32_t mask, vuint64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmaccu.mask.nxv2i64.i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmaccu.mask.nxv2i64.i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vwmaccu_vx_u64m2_m(vbool32_t mask, vuint64m2_t acc,
@@ -1628,7 +1628,7 @@ vuint64m2_t test_vwmaccu_vx_u64m2_m(vbool32_t mask, vuint64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmaccu.mask.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmaccu.mask.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vwmaccu_vv_u64m4_m(vbool16_t mask, vuint64m4_t acc,
@@ -1639,7 +1639,7 @@ vuint64m4_t test_vwmaccu_vv_u64m4_m(vbool16_t mask, vuint64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmaccu.mask.nxv4i64.i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmaccu.mask.nxv4i64.i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vwmaccu_vx_u64m4_m(vbool16_t mask, vuint64m4_t acc,
@@ -1649,7 +1649,7 @@ vuint64m4_t test_vwmaccu_vx_u64m4_m(vbool16_t mask, vuint64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmaccu.mask.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmaccu.mask.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vwmaccu_vv_u64m8_m(vbool8_t mask, vuint64m8_t acc,
@@ -1660,7 +1660,7 @@ vuint64m8_t test_vwmaccu_vv_u64m8_m(vbool8_t mask, vuint64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmaccu.mask.nxv8i64.i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmaccu.mask.nxv8i64.i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vwmaccu_vx_u64m8_m(vbool8_t mask, vuint64m8_t acc,
@@ -1670,7 +1670,7 @@ vuint64m8_t test_vwmaccu_vx_u64m8_m(vbool8_t mask, vuint64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmaccsu.mask.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmaccsu.mask.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vwmaccsu_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc,
@@ -1681,7 +1681,7 @@ vint16mf4_t test_vwmaccsu_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmaccsu.mask.nxv1i16.i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmaccsu.mask.nxv1i16.i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vwmaccsu_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc,
@@ -1691,7 +1691,7 @@ vint16mf4_t test_vwmaccsu_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmaccsu.mask.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmaccsu.mask.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vwmaccsu_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc,
@@ -1702,7 +1702,7 @@ vint16mf2_t test_vwmaccsu_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmaccsu.mask.nxv2i16.i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmaccsu.mask.nxv2i16.i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vwmaccsu_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc,
@@ -1712,7 +1712,7 @@ vint16mf2_t test_vwmaccsu_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmaccsu.mask.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmaccsu.mask.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vwmaccsu_vv_i16m1_m(vbool16_t mask, vint16m1_t acc,
@@ -1723,7 +1723,7 @@ vint16m1_t test_vwmaccsu_vv_i16m1_m(vbool16_t mask, vint16m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmaccsu.mask.nxv4i16.i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmaccsu.mask.nxv4i16.i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vwmaccsu_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int8_t op1,
@@ -1733,7 +1733,7 @@ vint16m1_t test_vwmaccsu_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmaccsu.mask.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmaccsu.mask.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vwmaccsu_vv_i16m2_m(vbool8_t mask, vint16m2_t acc,
@@ -1743,7 +1743,7 @@ vint16m2_t test_vwmaccsu_vv_i16m2_m(vbool8_t mask, vint16m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmaccsu.mask.nxv8i16.i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmaccsu.mask.nxv8i16.i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vwmaccsu_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int8_t op1,
@@ -1753,7 +1753,7 @@ vint16m2_t test_vwmaccsu_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmaccsu.mask.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmaccsu.mask.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vwmaccsu_vv_i16m4_m(vbool4_t mask, vint16m4_t acc,
@@ -1763,7 +1763,7 @@ vint16m4_t test_vwmaccsu_vv_i16m4_m(vbool4_t mask, vint16m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmaccsu.mask.nxv16i16.i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmaccsu.mask.nxv16i16.i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vwmaccsu_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int8_t op1,
@@ -1773,7 +1773,7 @@ vint16m4_t test_vwmaccsu_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmaccsu.mask.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmaccsu.mask.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vwmaccsu_vv_i16m8_m(vbool2_t mask, vint16m8_t acc,
@@ -1783,7 +1783,7 @@ vint16m8_t test_vwmaccsu_vv_i16m8_m(vbool2_t mask, vint16m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmaccsu.mask.nxv32i16.i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmaccsu.mask.nxv32i16.i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vwmaccsu_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int8_t op1,
@@ -1793,7 +1793,7 @@ vint16m8_t test_vwmaccsu_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmaccsu.mask.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmaccsu.mask.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vwmaccsu_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc,
@@ -1804,7 +1804,7 @@ vint32mf2_t test_vwmaccsu_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmaccsu.mask.nxv1i32.i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmaccsu.mask.nxv1i32.i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vwmaccsu_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc,
@@ -1815,7 +1815,7 @@ vint32mf2_t test_vwmaccsu_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmaccsu.mask.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmaccsu.mask.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vwmaccsu_vv_i32m1_m(vbool32_t mask, vint32m1_t acc,
@@ -1826,7 +1826,7 @@ vint32m1_t test_vwmaccsu_vv_i32m1_m(vbool32_t mask, vint32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmaccsu.mask.nxv2i32.i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmaccsu.mask.nxv2i32.i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vwmaccsu_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int16_t op1,
@@ -1836,7 +1836,7 @@ vint32m1_t test_vwmaccsu_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmaccsu.mask.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmaccsu.mask.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vwmaccsu_vv_i32m2_m(vbool16_t mask, vint32m2_t acc,
@@ -1847,7 +1847,7 @@ vint32m2_t test_vwmaccsu_vv_i32m2_m(vbool16_t mask, vint32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmaccsu.mask.nxv4i32.i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmaccsu.mask.nxv4i32.i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vwmaccsu_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int16_t op1,
@@ -1857,7 +1857,7 @@ vint32m2_t test_vwmaccsu_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmaccsu.mask.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmaccsu.mask.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vwmaccsu_vv_i32m4_m(vbool8_t mask, vint32m4_t acc,
@@ -1868,7 +1868,7 @@ vint32m4_t test_vwmaccsu_vv_i32m4_m(vbool8_t mask, vint32m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmaccsu.mask.nxv8i32.i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmaccsu.mask.nxv8i32.i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vwmaccsu_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int16_t op1,
@@ -1878,7 +1878,7 @@ vint32m4_t test_vwmaccsu_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmaccsu.mask.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmaccsu.mask.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vwmaccsu_vv_i32m8_m(vbool4_t mask, vint32m8_t acc,
@@ -1889,7 +1889,7 @@ vint32m8_t test_vwmaccsu_vv_i32m8_m(vbool4_t mask, vint32m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmaccsu.mask.nxv16i32.i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmaccsu.mask.nxv16i32.i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vwmaccsu_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int16_t op1,
@@ -1899,7 +1899,7 @@ vint32m8_t test_vwmaccsu_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmaccsu.mask.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmaccsu.mask.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vwmaccsu_vv_i64m1_m(vbool64_t mask, vint64m1_t acc,
@@ -1910,7 +1910,7 @@ vint64m1_t test_vwmaccsu_vv_i64m1_m(vbool64_t mask, vint64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmaccsu.mask.nxv1i64.i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmaccsu.mask.nxv1i64.i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vwmaccsu_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int32_t op1,
@@ -1920,7 +1920,7 @@ vint64m1_t test_vwmaccsu_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int32_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmaccsu.mask.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmaccsu.mask.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vwmaccsu_vv_i64m2_m(vbool32_t mask, vint64m2_t acc,
@@ -1931,7 +1931,7 @@ vint64m2_t test_vwmaccsu_vv_i64m2_m(vbool32_t mask, vint64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmaccsu.mask.nxv2i64.i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmaccsu.mask.nxv2i64.i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vwmaccsu_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int32_t op1,
@@ -1941,7 +1941,7 @@ vint64m2_t test_vwmaccsu_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int32_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmaccsu.mask.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmaccsu.mask.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vwmaccsu_vv_i64m4_m(vbool16_t mask, vint64m4_t acc,
@@ -1952,7 +1952,7 @@ vint64m4_t test_vwmaccsu_vv_i64m4_m(vbool16_t mask, vint64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmaccsu.mask.nxv4i64.i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmaccsu.mask.nxv4i64.i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vwmaccsu_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int32_t op1,
@@ -1962,7 +1962,7 @@ vint64m4_t test_vwmaccsu_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int32_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmaccsu.mask.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmaccsu.mask.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vwmaccsu_vv_i64m8_m(vbool8_t mask, vint64m8_t acc,
@@ -1973,7 +1973,7 @@ vint64m8_t test_vwmaccsu_vv_i64m8_m(vbool8_t mask, vint64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmaccsu.mask.nxv8i64.i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmaccsu.mask.nxv8i64.i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vwmaccsu_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int32_t op1,
@@ -1983,7 +1983,7 @@ vint64m8_t test_vwmaccsu_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int32_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmaccus.mask.nxv1i16.i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmaccus.mask.nxv1i16.i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vwmaccus_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc,
@@ -1993,7 +1993,7 @@ vint16mf4_t test_vwmaccus_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmaccus.mask.nxv2i16.i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmaccus.mask.nxv2i16.i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vwmaccus_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc,
@@ -2003,7 +2003,7 @@ vint16mf2_t test_vwmaccus_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmaccus.mask.nxv4i16.i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmaccus.mask.nxv4i16.i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vwmaccus_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, uint8_t op1,
@@ -2013,7 +2013,7 @@ vint16m1_t test_vwmaccus_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmaccus.mask.nxv8i16.i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmaccus.mask.nxv8i16.i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vwmaccus_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, uint8_t op1,
@@ -2023,7 +2023,7 @@ vint16m2_t test_vwmaccus_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmaccus.mask.nxv16i16.i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmaccus.mask.nxv16i16.i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vwmaccus_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, uint8_t op1,
@@ -2033,7 +2033,7 @@ vint16m4_t test_vwmaccus_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmaccus.mask.nxv32i16.i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmaccus.mask.nxv32i16.i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vwmaccus_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, uint8_t op1,
@@ -2043,7 +2043,7 @@ vint16m8_t test_vwmaccus_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmaccus.mask.nxv1i32.i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmaccus.mask.nxv1i32.i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vwmaccus_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc,
@@ -2054,7 +2054,7 @@ vint32mf2_t test_vwmaccus_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmaccus.mask.nxv2i32.i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmaccus.mask.nxv2i32.i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vwmaccus_vx_i32m1_m(vbool32_t mask, vint32m1_t acc,
@@ -2064,7 +2064,7 @@ vint32m1_t test_vwmaccus_vx_i32m1_m(vbool32_t mask, vint32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmaccus.mask.nxv4i32.i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmaccus.mask.nxv4i32.i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vwmaccus_vx_i32m2_m(vbool16_t mask, vint32m2_t acc,
@@ -2074,7 +2074,7 @@ vint32m2_t test_vwmaccus_vx_i32m2_m(vbool16_t mask, vint32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmaccus.mask.nxv8i32.i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmaccus.mask.nxv8i32.i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vwmaccus_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, uint16_t op1,
@@ -2084,7 +2084,7 @@ vint32m4_t test_vwmaccus_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmaccus.mask.nxv16i32.i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmaccus.mask.nxv16i32.i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vwmaccus_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, uint16_t op1,
@@ -2094,7 +2094,7 @@ vint32m8_t test_vwmaccus_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmaccus.mask.nxv1i64.i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmaccus.mask.nxv1i64.i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vwmaccus_vx_i64m1_m(vbool64_t mask, vint64m1_t acc,
@@ -2104,7 +2104,7 @@ vint64m1_t test_vwmaccus_vx_i64m1_m(vbool64_t mask, vint64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmaccus.mask.nxv2i64.i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmaccus.mask.nxv2i64.i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vwmaccus_vx_i64m2_m(vbool32_t mask, vint64m2_t acc,
@@ -2114,7 +2114,7 @@ vint64m2_t test_vwmaccus_vx_i64m2_m(vbool32_t mask, vint64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmaccus.mask.nxv4i64.i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmaccus.mask.nxv4i64.i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vwmaccus_vx_i64m4_m(vbool16_t mask, vint64m4_t acc,
@@ -2124,7 +2124,7 @@ vint64m4_t test_vwmaccus_vx_i64m4_m(vbool16_t mask, vint64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmaccus.mask.nxv8i64.i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmaccus.mask.nxv8i64.i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vwmaccus_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, uint32_t op1,
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfmacc.c b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfmacc.c
index 5f01c595361453..ab4ef84ccc2572 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfmacc.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfmacc.c
@@ -188,7 +188,7 @@ vfloat64m8_t test_vfmacc_vf_f64m8(vfloat64m8_t acc, double op1,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmacc.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmacc.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfmacc_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -199,7 +199,7 @@ vfloat32mf2_t test_vfmacc_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmacc.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmacc.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfmacc_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -209,7 +209,7 @@ vfloat32mf2_t test_vfmacc_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmacc.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmacc.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfmacc_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -220,7 +220,7 @@ vfloat32m1_t test_vfmacc_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmacc.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmacc.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfmacc_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc, float op1,
@@ -230,7 +230,7 @@ vfloat32m1_t test_vfmacc_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmacc.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmacc.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfmacc_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -241,7 +241,7 @@ vfloat32m2_t test_vfmacc_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmacc.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmacc.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfmacc_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc, float op1,
@@ -251,7 +251,7 @@ vfloat32m2_t test_vfmacc_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmacc.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmacc.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfmacc_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
@@ -262,7 +262,7 @@ vfloat32m4_t test_vfmacc_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmacc.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmacc.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfmacc_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
@@ -272,7 +272,7 @@ vfloat32m4_t test_vfmacc_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmacc.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmacc.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfmacc_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
@@ -283,7 +283,7 @@ vfloat32m8_t test_vfmacc_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmacc.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmacc.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfmacc_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
@@ -293,7 +293,7 @@ vfloat32m8_t test_vfmacc_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmacc.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmacc.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfmacc_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -304,7 +304,7 @@ vfloat64m1_t test_vfmacc_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmacc.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmacc.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfmacc_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -314,7 +314,7 @@ vfloat64m1_t test_vfmacc_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmacc.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmacc.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfmacc_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -325,7 +325,7 @@ vfloat64m2_t test_vfmacc_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmacc.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmacc.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfmacc_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -335,7 +335,7 @@ vfloat64m2_t test_vfmacc_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmacc.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmacc.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfmacc_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -346,7 +346,7 @@ vfloat64m4_t test_vfmacc_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmacc.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmacc.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfmacc_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -356,7 +356,7 @@ vfloat64m4_t test_vfmacc_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmacc.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmacc.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfmacc_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -367,7 +367,7 @@ vfloat64m8_t test_vfmacc_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmacc.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmacc.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfmacc_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc, double op1,
@@ -485,7 +485,7 @@ vfloat16m8_t test_vfmacc_vf_f16m8 (vfloat16m8_t vd, _Float16 rs1, vfloat16m8_t v
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfmacc.mask.nxv1f16.nxv1f16.i64(<vscale x 1 x half> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfmacc.mask.nxv1f16.nxv1f16.i64(<vscale x 1 x half> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x half> [[TMP0]]
 //
 vfloat16mf4_t test_vfmacc_vv_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, vfloat16mf4_t vs1, vfloat16mf4_t vs2, size_t vl) {
@@ -494,7 +494,7 @@ vfloat16mf4_t test_vfmacc_vv_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, vfloat1
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfmacc.mask.nxv1f16.f16.i64(<vscale x 1 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfmacc.mask.nxv1f16.f16.i64(<vscale x 1 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x half> [[TMP0]]
 //
 vfloat16mf4_t test_vfmacc_vf_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, _Float16 rs1, vfloat16mf4_t vs2, size_t vl) {
@@ -503,7 +503,7 @@ vfloat16mf4_t test_vfmacc_vf_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, _Float1
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfmacc.mask.nxv2f16.nxv2f16.i64(<vscale x 2 x half> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfmacc.mask.nxv2f16.nxv2f16.i64(<vscale x 2 x half> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x half> [[TMP0]]
 //
 vfloat16mf2_t test_vfmacc_vv_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, vfloat16mf2_t vs1, vfloat16mf2_t vs2, size_t vl) {
@@ -512,7 +512,7 @@ vfloat16mf2_t test_vfmacc_vv_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, vfloat1
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfmacc.mask.nxv2f16.f16.i64(<vscale x 2 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfmacc.mask.nxv2f16.f16.i64(<vscale x 2 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x half> [[TMP0]]
 //
 vfloat16mf2_t test_vfmacc_vf_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, _Float16 rs1, vfloat16mf2_t vs2, size_t vl) {
@@ -521,7 +521,7 @@ vfloat16mf2_t test_vfmacc_vf_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, _Float1
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfmacc.mask.nxv4f16.nxv4f16.i64(<vscale x 4 x half> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfmacc.mask.nxv4f16.nxv4f16.i64(<vscale x 4 x half> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x half> [[TMP0]]
 //
 vfloat16m1_t test_vfmacc_vv_f16m1_m (vbool16_t mask, vfloat16m1_t vd, vfloat16m1_t vs1, vfloat16m1_t vs2, size_t vl) {
@@ -530,7 +530,7 @@ vfloat16m1_t test_vfmacc_vv_f16m1_m (vbool16_t mask, vfloat16m1_t vd, vfloat16m1
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfmacc.mask.nxv4f16.f16.i64(<vscale x 4 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfmacc.mask.nxv4f16.f16.i64(<vscale x 4 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x half> [[TMP0]]
 //
 vfloat16m1_t test_vfmacc_vf_f16m1_m (vbool16_t mask, vfloat16m1_t vd, _Float16 rs1, vfloat16m1_t vs2, size_t vl) {
@@ -539,7 +539,7 @@ vfloat16m1_t test_vfmacc_vf_f16m1_m (vbool16_t mask, vfloat16m1_t vd, _Float16 r
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfmacc.mask.nxv8f16.nxv8f16.i64(<vscale x 8 x half> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfmacc.mask.nxv8f16.nxv8f16.i64(<vscale x 8 x half> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x half> [[TMP0]]
 //
 vfloat16m2_t test_vfmacc_vv_f16m2_m (vbool8_t mask, vfloat16m2_t vd, vfloat16m2_t vs1, vfloat16m2_t vs2, size_t vl) {
@@ -548,7 +548,7 @@ vfloat16m2_t test_vfmacc_vv_f16m2_m (vbool8_t mask, vfloat16m2_t vd, vfloat16m2_
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfmacc.mask.nxv8f16.f16.i64(<vscale x 8 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfmacc.mask.nxv8f16.f16.i64(<vscale x 8 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x half> [[TMP0]]
 //
 vfloat16m2_t test_vfmacc_vf_f16m2_m (vbool8_t mask, vfloat16m2_t vd, _Float16 rs1, vfloat16m2_t vs2, size_t vl) {
@@ -557,7 +557,7 @@ vfloat16m2_t test_vfmacc_vf_f16m2_m (vbool8_t mask, vfloat16m2_t vd, _Float16 rs
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfmacc.mask.nxv16f16.nxv16f16.i64(<vscale x 16 x half> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfmacc.mask.nxv16f16.nxv16f16.i64(<vscale x 16 x half> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x half> [[TMP0]]
 //
 vfloat16m4_t test_vfmacc_vv_f16m4_m (vbool4_t mask, vfloat16m4_t vd, vfloat16m4_t vs1, vfloat16m4_t vs2, size_t vl) {
@@ -566,7 +566,7 @@ vfloat16m4_t test_vfmacc_vv_f16m4_m (vbool4_t mask, vfloat16m4_t vd, vfloat16m4_
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfmacc.mask.nxv16f16.f16.i64(<vscale x 16 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfmacc.mask.nxv16f16.f16.i64(<vscale x 16 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x half> [[TMP0]]
 //
 vfloat16m4_t test_vfmacc_vf_f16m4_m (vbool4_t mask, vfloat16m4_t vd, _Float16 rs1, vfloat16m4_t vs2, size_t vl) {
@@ -575,7 +575,7 @@ vfloat16m4_t test_vfmacc_vf_f16m4_m (vbool4_t mask, vfloat16m4_t vd, _Float16 rs
 
 // CHECK-RV64-LABEL: @test_vfmacc_vv_f16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfmacc.mask.nxv32f16.nxv32f16.i64(<vscale x 32 x half> [[VD:%.*]], <vscale x 32 x half> [[VS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfmacc.mask.nxv32f16.nxv32f16.i64(<vscale x 32 x half> [[VD:%.*]], <vscale x 32 x half> [[VS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x half> [[TMP0]]
 //
 vfloat16m8_t test_vfmacc_vv_f16m8_m (vbool2_t mask, vfloat16m8_t vd, vfloat16m8_t vs1, vfloat16m8_t vs2, size_t vl) {
@@ -584,7 +584,7 @@ vfloat16m8_t test_vfmacc_vv_f16m8_m (vbool2_t mask, vfloat16m8_t vd, vfloat16m8_
 
 // CHECK-RV64-LABEL: @test_vfmacc_vf_f16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfmacc.mask.nxv32f16.f16.i64(<vscale x 32 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfmacc.mask.nxv32f16.f16.i64(<vscale x 32 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x half> [[TMP0]]
 //
 vfloat16m8_t test_vfmacc_vf_f16m8_m (vbool2_t mask, vfloat16m8_t vd, _Float16 rs1, vfloat16m8_t vs2, size_t vl) {
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfmadd.c b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfmadd.c
index 79e8d32e85d9fa..29351df8302b9e 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfmadd.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfmadd.c
@@ -188,7 +188,7 @@ vfloat64m8_t test_vfmadd_vf_f64m8(vfloat64m8_t acc, double op1,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmadd.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmadd.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfmadd_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -199,7 +199,7 @@ vfloat32mf2_t test_vfmadd_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmadd.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmadd.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfmadd_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -209,7 +209,7 @@ vfloat32mf2_t test_vfmadd_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmadd.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmadd.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfmadd_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -220,7 +220,7 @@ vfloat32m1_t test_vfmadd_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmadd.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmadd.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfmadd_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc, float op1,
@@ -230,7 +230,7 @@ vfloat32m1_t test_vfmadd_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmadd.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmadd.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfmadd_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -241,7 +241,7 @@ vfloat32m2_t test_vfmadd_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmadd.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmadd.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfmadd_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc, float op1,
@@ -251,7 +251,7 @@ vfloat32m2_t test_vfmadd_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmadd.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmadd.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfmadd_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
@@ -262,7 +262,7 @@ vfloat32m4_t test_vfmadd_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmadd.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmadd.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfmadd_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
@@ -272,7 +272,7 @@ vfloat32m4_t test_vfmadd_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmadd.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmadd.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfmadd_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
@@ -283,7 +283,7 @@ vfloat32m8_t test_vfmadd_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmadd.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmadd.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfmadd_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
@@ -293,7 +293,7 @@ vfloat32m8_t test_vfmadd_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmadd.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmadd.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfmadd_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -304,7 +304,7 @@ vfloat64m1_t test_vfmadd_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmadd.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmadd.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfmadd_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -314,7 +314,7 @@ vfloat64m1_t test_vfmadd_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmadd.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmadd.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfmadd_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -325,7 +325,7 @@ vfloat64m2_t test_vfmadd_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmadd.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmadd.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfmadd_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -335,7 +335,7 @@ vfloat64m2_t test_vfmadd_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmadd.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmadd.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfmadd_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -346,7 +346,7 @@ vfloat64m4_t test_vfmadd_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmadd.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmadd.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfmadd_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -356,7 +356,7 @@ vfloat64m4_t test_vfmadd_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmadd.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmadd.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfmadd_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -367,7 +367,7 @@ vfloat64m8_t test_vfmadd_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmadd.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmadd.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfmadd_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc, double op1,
@@ -485,7 +485,7 @@ vfloat16m8_t test_vfmadd_vf_f16m8 (vfloat16m8_t vd, _Float16 rs1, vfloat16m8_t v
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfmadd.mask.nxv1f16.nxv1f16.i64(<vscale x 1 x half> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfmadd.mask.nxv1f16.nxv1f16.i64(<vscale x 1 x half> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x half> [[TMP0]]
 //
 vfloat16mf4_t test_vfmadd_vv_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, vfloat16mf4_t vs1, vfloat16mf4_t vs2, size_t vl) {
@@ -494,7 +494,7 @@ vfloat16mf4_t test_vfmadd_vv_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, vfloat1
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfmadd.mask.nxv1f16.f16.i64(<vscale x 1 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfmadd.mask.nxv1f16.f16.i64(<vscale x 1 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x half> [[TMP0]]
 //
 vfloat16mf4_t test_vfmadd_vf_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, _Float16 rs1, vfloat16mf4_t vs2, size_t vl) {
@@ -503,7 +503,7 @@ vfloat16mf4_t test_vfmadd_vf_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, _Float1
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfmadd.mask.nxv2f16.nxv2f16.i64(<vscale x 2 x half> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfmadd.mask.nxv2f16.nxv2f16.i64(<vscale x 2 x half> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x half> [[TMP0]]
 //
 vfloat16mf2_t test_vfmadd_vv_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, vfloat16mf2_t vs1, vfloat16mf2_t vs2, size_t vl) {
@@ -512,7 +512,7 @@ vfloat16mf2_t test_vfmadd_vv_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, vfloat1
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfmadd.mask.nxv2f16.f16.i64(<vscale x 2 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfmadd.mask.nxv2f16.f16.i64(<vscale x 2 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x half> [[TMP0]]
 //
 vfloat16mf2_t test_vfmadd_vf_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, _Float16 rs1, vfloat16mf2_t vs2, size_t vl) {
@@ -521,7 +521,7 @@ vfloat16mf2_t test_vfmadd_vf_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, _Float1
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfmadd.mask.nxv4f16.nxv4f16.i64(<vscale x 4 x half> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfmadd.mask.nxv4f16.nxv4f16.i64(<vscale x 4 x half> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x half> [[TMP0]]
 //
 vfloat16m1_t test_vfmadd_vv_f16m1_m (vbool16_t mask, vfloat16m1_t vd, vfloat16m1_t vs1, vfloat16m1_t vs2, size_t vl) {
@@ -530,7 +530,7 @@ vfloat16m1_t test_vfmadd_vv_f16m1_m (vbool16_t mask, vfloat16m1_t vd, vfloat16m1
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfmadd.mask.nxv4f16.f16.i64(<vscale x 4 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfmadd.mask.nxv4f16.f16.i64(<vscale x 4 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x half> [[TMP0]]
 //
 vfloat16m1_t test_vfmadd_vf_f16m1_m (vbool16_t mask, vfloat16m1_t vd, _Float16 rs1, vfloat16m1_t vs2, size_t vl) {
@@ -539,7 +539,7 @@ vfloat16m1_t test_vfmadd_vf_f16m1_m (vbool16_t mask, vfloat16m1_t vd, _Float16 r
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfmadd.mask.nxv8f16.nxv8f16.i64(<vscale x 8 x half> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfmadd.mask.nxv8f16.nxv8f16.i64(<vscale x 8 x half> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x half> [[TMP0]]
 //
 vfloat16m2_t test_vfmadd_vv_f16m2_m (vbool8_t mask, vfloat16m2_t vd, vfloat16m2_t vs1, vfloat16m2_t vs2, size_t vl) {
@@ -548,7 +548,7 @@ vfloat16m2_t test_vfmadd_vv_f16m2_m (vbool8_t mask, vfloat16m2_t vd, vfloat16m2_
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfmadd.mask.nxv8f16.f16.i64(<vscale x 8 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfmadd.mask.nxv8f16.f16.i64(<vscale x 8 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x half> [[TMP0]]
 //
 vfloat16m2_t test_vfmadd_vf_f16m2_m (vbool8_t mask, vfloat16m2_t vd, _Float16 rs1, vfloat16m2_t vs2, size_t vl) {
@@ -557,7 +557,7 @@ vfloat16m2_t test_vfmadd_vf_f16m2_m (vbool8_t mask, vfloat16m2_t vd, _Float16 rs
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfmadd.mask.nxv16f16.nxv16f16.i64(<vscale x 16 x half> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfmadd.mask.nxv16f16.nxv16f16.i64(<vscale x 16 x half> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x half> [[TMP0]]
 //
 vfloat16m4_t test_vfmadd_vv_f16m4_m (vbool4_t mask, vfloat16m4_t vd, vfloat16m4_t vs1, vfloat16m4_t vs2, size_t vl) {
@@ -566,7 +566,7 @@ vfloat16m4_t test_vfmadd_vv_f16m4_m (vbool4_t mask, vfloat16m4_t vd, vfloat16m4_
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfmadd.mask.nxv16f16.f16.i64(<vscale x 16 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfmadd.mask.nxv16f16.f16.i64(<vscale x 16 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x half> [[TMP0]]
 //
 vfloat16m4_t test_vfmadd_vf_f16m4_m (vbool4_t mask, vfloat16m4_t vd, _Float16 rs1, vfloat16m4_t vs2, size_t vl) {
@@ -575,7 +575,7 @@ vfloat16m4_t test_vfmadd_vf_f16m4_m (vbool4_t mask, vfloat16m4_t vd, _Float16 rs
 
 // CHECK-RV64-LABEL: @test_vfmadd_vv_f16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfmadd.mask.nxv32f16.nxv32f16.i64(<vscale x 32 x half> [[VD:%.*]], <vscale x 32 x half> [[VS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfmadd.mask.nxv32f16.nxv32f16.i64(<vscale x 32 x half> [[VD:%.*]], <vscale x 32 x half> [[VS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x half> [[TMP0]]
 //
 vfloat16m8_t test_vfmadd_vv_f16m8_m (vbool2_t mask, vfloat16m8_t vd, vfloat16m8_t vs1, vfloat16m8_t vs2, size_t vl) {
@@ -584,7 +584,7 @@ vfloat16m8_t test_vfmadd_vv_f16m8_m (vbool2_t mask, vfloat16m8_t vd, vfloat16m8_
 
 // CHECK-RV64-LABEL: @test_vfmadd_vf_f16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfmadd.mask.nxv32f16.f16.i64(<vscale x 32 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfmadd.mask.nxv32f16.f16.i64(<vscale x 32 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x half> [[TMP0]]
 //
 vfloat16m8_t test_vfmadd_vf_f16m8_m (vbool2_t mask, vfloat16m8_t vd, _Float16 rs1, vfloat16m8_t vs2, size_t vl) {
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfmsac.c b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfmsac.c
index d8e380ec8b20e2..39458d19f069e6 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfmsac.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfmsac.c
@@ -188,7 +188,7 @@ vfloat64m8_t test_vfmsac_vf_f64m8(vfloat64m8_t acc, double op1,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmsac.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmsac.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfmsac_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -199,7 +199,7 @@ vfloat32mf2_t test_vfmsac_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmsac.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmsac.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfmsac_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -209,7 +209,7 @@ vfloat32mf2_t test_vfmsac_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmsac.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmsac.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfmsac_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -220,7 +220,7 @@ vfloat32m1_t test_vfmsac_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmsac.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmsac.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfmsac_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc, float op1,
@@ -230,7 +230,7 @@ vfloat32m1_t test_vfmsac_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmsac.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmsac.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfmsac_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -241,7 +241,7 @@ vfloat32m2_t test_vfmsac_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmsac.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmsac.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfmsac_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc, float op1,
@@ -251,7 +251,7 @@ vfloat32m2_t test_vfmsac_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmsac.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmsac.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfmsac_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
@@ -262,7 +262,7 @@ vfloat32m4_t test_vfmsac_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmsac.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmsac.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfmsac_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
@@ -272,7 +272,7 @@ vfloat32m4_t test_vfmsac_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmsac.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmsac.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfmsac_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
@@ -283,7 +283,7 @@ vfloat32m8_t test_vfmsac_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmsac.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmsac.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfmsac_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
@@ -293,7 +293,7 @@ vfloat32m8_t test_vfmsac_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmsac.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmsac.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfmsac_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -304,7 +304,7 @@ vfloat64m1_t test_vfmsac_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmsac.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmsac.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfmsac_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -314,7 +314,7 @@ vfloat64m1_t test_vfmsac_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmsac.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmsac.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfmsac_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -325,7 +325,7 @@ vfloat64m2_t test_vfmsac_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmsac.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmsac.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfmsac_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -335,7 +335,7 @@ vfloat64m2_t test_vfmsac_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmsac.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmsac.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfmsac_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -346,7 +346,7 @@ vfloat64m4_t test_vfmsac_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmsac.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmsac.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfmsac_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -356,7 +356,7 @@ vfloat64m4_t test_vfmsac_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmsac.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmsac.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfmsac_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -367,7 +367,7 @@ vfloat64m8_t test_vfmsac_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmsac.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmsac.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfmsac_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc, double op1,
@@ -485,7 +485,7 @@ vfloat16m8_t test_vfmsac_vf_f16m8 (vfloat16m8_t vd, _Float16 rs1, vfloat16m8_t v
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfmsac.mask.nxv1f16.nxv1f16.i64(<vscale x 1 x half> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfmsac.mask.nxv1f16.nxv1f16.i64(<vscale x 1 x half> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x half> [[TMP0]]
 //
 vfloat16mf4_t test_vfmsac_vv_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, vfloat16mf4_t vs1, vfloat16mf4_t vs2, size_t vl) {
@@ -494,7 +494,7 @@ vfloat16mf4_t test_vfmsac_vv_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, vfloat1
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfmsac.mask.nxv1f16.f16.i64(<vscale x 1 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfmsac.mask.nxv1f16.f16.i64(<vscale x 1 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x half> [[TMP0]]
 //
 vfloat16mf4_t test_vfmsac_vf_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, _Float16 rs1, vfloat16mf4_t vs2, size_t vl) {
@@ -503,7 +503,7 @@ vfloat16mf4_t test_vfmsac_vf_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, _Float1
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfmsac.mask.nxv2f16.nxv2f16.i64(<vscale x 2 x half> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfmsac.mask.nxv2f16.nxv2f16.i64(<vscale x 2 x half> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x half> [[TMP0]]
 //
 vfloat16mf2_t test_vfmsac_vv_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, vfloat16mf2_t vs1, vfloat16mf2_t vs2, size_t vl) {
@@ -512,7 +512,7 @@ vfloat16mf2_t test_vfmsac_vv_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, vfloat1
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfmsac.mask.nxv2f16.f16.i64(<vscale x 2 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfmsac.mask.nxv2f16.f16.i64(<vscale x 2 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x half> [[TMP0]]
 //
 vfloat16mf2_t test_vfmsac_vf_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, _Float16 rs1, vfloat16mf2_t vs2, size_t vl) {
@@ -521,7 +521,7 @@ vfloat16mf2_t test_vfmsac_vf_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, _Float1
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfmsac.mask.nxv4f16.nxv4f16.i64(<vscale x 4 x half> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfmsac.mask.nxv4f16.nxv4f16.i64(<vscale x 4 x half> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x half> [[TMP0]]
 //
 vfloat16m1_t test_vfmsac_vv_f16m1_m (vbool16_t mask, vfloat16m1_t vd, vfloat16m1_t vs1, vfloat16m1_t vs2, size_t vl) {
@@ -530,7 +530,7 @@ vfloat16m1_t test_vfmsac_vv_f16m1_m (vbool16_t mask, vfloat16m1_t vd, vfloat16m1
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfmsac.mask.nxv4f16.f16.i64(<vscale x 4 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfmsac.mask.nxv4f16.f16.i64(<vscale x 4 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x half> [[TMP0]]
 //
 vfloat16m1_t test_vfmsac_vf_f16m1_m (vbool16_t mask, vfloat16m1_t vd, _Float16 rs1, vfloat16m1_t vs2, size_t vl) {
@@ -539,7 +539,7 @@ vfloat16m1_t test_vfmsac_vf_f16m1_m (vbool16_t mask, vfloat16m1_t vd, _Float16 r
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfmsac.mask.nxv8f16.nxv8f16.i64(<vscale x 8 x half> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfmsac.mask.nxv8f16.nxv8f16.i64(<vscale x 8 x half> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x half> [[TMP0]]
 //
 vfloat16m2_t test_vfmsac_vv_f16m2_m (vbool8_t mask, vfloat16m2_t vd, vfloat16m2_t vs1, vfloat16m2_t vs2, size_t vl) {
@@ -548,7 +548,7 @@ vfloat16m2_t test_vfmsac_vv_f16m2_m (vbool8_t mask, vfloat16m2_t vd, vfloat16m2_
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfmsac.mask.nxv8f16.f16.i64(<vscale x 8 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfmsac.mask.nxv8f16.f16.i64(<vscale x 8 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x half> [[TMP0]]
 //
 vfloat16m2_t test_vfmsac_vf_f16m2_m (vbool8_t mask, vfloat16m2_t vd, _Float16 rs1, vfloat16m2_t vs2, size_t vl) {
@@ -557,7 +557,7 @@ vfloat16m2_t test_vfmsac_vf_f16m2_m (vbool8_t mask, vfloat16m2_t vd, _Float16 rs
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfmsac.mask.nxv16f16.nxv16f16.i64(<vscale x 16 x half> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfmsac.mask.nxv16f16.nxv16f16.i64(<vscale x 16 x half> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x half> [[TMP0]]
 //
 vfloat16m4_t test_vfmsac_vv_f16m4_m (vbool4_t mask, vfloat16m4_t vd, vfloat16m4_t vs1, vfloat16m4_t vs2, size_t vl) {
@@ -566,7 +566,7 @@ vfloat16m4_t test_vfmsac_vv_f16m4_m (vbool4_t mask, vfloat16m4_t vd, vfloat16m4_
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfmsac.mask.nxv16f16.f16.i64(<vscale x 16 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfmsac.mask.nxv16f16.f16.i64(<vscale x 16 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x half> [[TMP0]]
 //
 vfloat16m4_t test_vfmsac_vf_f16m4_m (vbool4_t mask, vfloat16m4_t vd, _Float16 rs1, vfloat16m4_t vs2, size_t vl) {
@@ -575,7 +575,7 @@ vfloat16m4_t test_vfmsac_vf_f16m4_m (vbool4_t mask, vfloat16m4_t vd, _Float16 rs
 
 // CHECK-RV64-LABEL: @test_vfmsac_vv_f16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfmsac.mask.nxv32f16.nxv32f16.i64(<vscale x 32 x half> [[VD:%.*]], <vscale x 32 x half> [[VS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfmsac.mask.nxv32f16.nxv32f16.i64(<vscale x 32 x half> [[VD:%.*]], <vscale x 32 x half> [[VS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x half> [[TMP0]]
 //
 vfloat16m8_t test_vfmsac_vv_f16m8_m (vbool2_t mask, vfloat16m8_t vd, vfloat16m8_t vs1, vfloat16m8_t vs2, size_t vl) {
@@ -584,7 +584,7 @@ vfloat16m8_t test_vfmsac_vv_f16m8_m (vbool2_t mask, vfloat16m8_t vd, vfloat16m8_
 
 // CHECK-RV64-LABEL: @test_vfmsac_vf_f16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfmsac.mask.nxv32f16.f16.i64(<vscale x 32 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfmsac.mask.nxv32f16.f16.i64(<vscale x 32 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x half> [[TMP0]]
 //
 vfloat16m8_t test_vfmsac_vf_f16m8_m (vbool2_t mask, vfloat16m8_t vd, _Float16 rs1, vfloat16m8_t vs2, size_t vl) {
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfmsub.c b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfmsub.c
index 529cb80d3b8f8d..9bd7a681514602 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfmsub.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfmsub.c
@@ -188,7 +188,7 @@ vfloat64m8_t test_vfmsub_vf_f64m8(vfloat64m8_t acc, double op1,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmsub.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmsub.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfmsub_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -199,7 +199,7 @@ vfloat32mf2_t test_vfmsub_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmsub.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfmsub.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfmsub_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -209,7 +209,7 @@ vfloat32mf2_t test_vfmsub_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmsub.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmsub.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfmsub_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -220,7 +220,7 @@ vfloat32m1_t test_vfmsub_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmsub.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfmsub.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfmsub_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc, float op1,
@@ -230,7 +230,7 @@ vfloat32m1_t test_vfmsub_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmsub.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmsub.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfmsub_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -241,7 +241,7 @@ vfloat32m2_t test_vfmsub_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmsub.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfmsub.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfmsub_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc, float op1,
@@ -251,7 +251,7 @@ vfloat32m2_t test_vfmsub_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmsub.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmsub.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfmsub_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
@@ -262,7 +262,7 @@ vfloat32m4_t test_vfmsub_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmsub.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfmsub.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfmsub_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
@@ -272,7 +272,7 @@ vfloat32m4_t test_vfmsub_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmsub.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmsub.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfmsub_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
@@ -283,7 +283,7 @@ vfloat32m8_t test_vfmsub_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmsub.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfmsub.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfmsub_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
@@ -293,7 +293,7 @@ vfloat32m8_t test_vfmsub_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmsub.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmsub.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfmsub_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -304,7 +304,7 @@ vfloat64m1_t test_vfmsub_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmsub.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfmsub.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfmsub_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -314,7 +314,7 @@ vfloat64m1_t test_vfmsub_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmsub.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmsub.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfmsub_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -325,7 +325,7 @@ vfloat64m2_t test_vfmsub_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmsub.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfmsub.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfmsub_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -335,7 +335,7 @@ vfloat64m2_t test_vfmsub_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmsub.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmsub.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfmsub_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -346,7 +346,7 @@ vfloat64m4_t test_vfmsub_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmsub.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfmsub.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfmsub_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -356,7 +356,7 @@ vfloat64m4_t test_vfmsub_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmsub.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmsub.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfmsub_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -367,7 +367,7 @@ vfloat64m8_t test_vfmsub_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmsub.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfmsub.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfmsub_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc, double op1,
@@ -485,7 +485,7 @@ vfloat16m8_t test_vfmsub_vf_f16m8 (vfloat16m8_t vd, _Float16 rs1, vfloat16m8_t v
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfmsub.mask.nxv1f16.nxv1f16.i64(<vscale x 1 x half> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfmsub.mask.nxv1f16.nxv1f16.i64(<vscale x 1 x half> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x half> [[TMP0]]
 //
 vfloat16mf4_t test_vfmsub_vv_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, vfloat16mf4_t vs1, vfloat16mf4_t vs2, size_t vl) {
@@ -494,7 +494,7 @@ vfloat16mf4_t test_vfmsub_vv_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, vfloat1
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfmsub.mask.nxv1f16.f16.i64(<vscale x 1 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfmsub.mask.nxv1f16.f16.i64(<vscale x 1 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x half> [[TMP0]]
 //
 vfloat16mf4_t test_vfmsub_vf_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, _Float16 rs1, vfloat16mf4_t vs2, size_t vl) {
@@ -503,7 +503,7 @@ vfloat16mf4_t test_vfmsub_vf_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, _Float1
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfmsub.mask.nxv2f16.nxv2f16.i64(<vscale x 2 x half> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfmsub.mask.nxv2f16.nxv2f16.i64(<vscale x 2 x half> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x half> [[TMP0]]
 //
 vfloat16mf2_t test_vfmsub_vv_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, vfloat16mf2_t vs1, vfloat16mf2_t vs2, size_t vl) {
@@ -512,7 +512,7 @@ vfloat16mf2_t test_vfmsub_vv_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, vfloat1
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfmsub.mask.nxv2f16.f16.i64(<vscale x 2 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfmsub.mask.nxv2f16.f16.i64(<vscale x 2 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x half> [[TMP0]]
 //
 vfloat16mf2_t test_vfmsub_vf_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, _Float16 rs1, vfloat16mf2_t vs2, size_t vl) {
@@ -521,7 +521,7 @@ vfloat16mf2_t test_vfmsub_vf_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, _Float1
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfmsub.mask.nxv4f16.nxv4f16.i64(<vscale x 4 x half> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfmsub.mask.nxv4f16.nxv4f16.i64(<vscale x 4 x half> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x half> [[TMP0]]
 //
 vfloat16m1_t test_vfmsub_vv_f16m1_m (vbool16_t mask, vfloat16m1_t vd, vfloat16m1_t vs1, vfloat16m1_t vs2, size_t vl) {
@@ -530,7 +530,7 @@ vfloat16m1_t test_vfmsub_vv_f16m1_m (vbool16_t mask, vfloat16m1_t vd, vfloat16m1
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfmsub.mask.nxv4f16.f16.i64(<vscale x 4 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfmsub.mask.nxv4f16.f16.i64(<vscale x 4 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x half> [[TMP0]]
 //
 vfloat16m1_t test_vfmsub_vf_f16m1_m (vbool16_t mask, vfloat16m1_t vd, _Float16 rs1, vfloat16m1_t vs2, size_t vl) {
@@ -539,7 +539,7 @@ vfloat16m1_t test_vfmsub_vf_f16m1_m (vbool16_t mask, vfloat16m1_t vd, _Float16 r
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfmsub.mask.nxv8f16.nxv8f16.i64(<vscale x 8 x half> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfmsub.mask.nxv8f16.nxv8f16.i64(<vscale x 8 x half> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x half> [[TMP0]]
 //
 vfloat16m2_t test_vfmsub_vv_f16m2_m (vbool8_t mask, vfloat16m2_t vd, vfloat16m2_t vs1, vfloat16m2_t vs2, size_t vl) {
@@ -548,7 +548,7 @@ vfloat16m2_t test_vfmsub_vv_f16m2_m (vbool8_t mask, vfloat16m2_t vd, vfloat16m2_
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfmsub.mask.nxv8f16.f16.i64(<vscale x 8 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfmsub.mask.nxv8f16.f16.i64(<vscale x 8 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x half> [[TMP0]]
 //
 vfloat16m2_t test_vfmsub_vf_f16m2_m (vbool8_t mask, vfloat16m2_t vd, _Float16 rs1, vfloat16m2_t vs2, size_t vl) {
@@ -557,7 +557,7 @@ vfloat16m2_t test_vfmsub_vf_f16m2_m (vbool8_t mask, vfloat16m2_t vd, _Float16 rs
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfmsub.mask.nxv16f16.nxv16f16.i64(<vscale x 16 x half> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfmsub.mask.nxv16f16.nxv16f16.i64(<vscale x 16 x half> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x half> [[TMP0]]
 //
 vfloat16m4_t test_vfmsub_vv_f16m4_m (vbool4_t mask, vfloat16m4_t vd, vfloat16m4_t vs1, vfloat16m4_t vs2, size_t vl) {
@@ -566,7 +566,7 @@ vfloat16m4_t test_vfmsub_vv_f16m4_m (vbool4_t mask, vfloat16m4_t vd, vfloat16m4_
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfmsub.mask.nxv16f16.f16.i64(<vscale x 16 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfmsub.mask.nxv16f16.f16.i64(<vscale x 16 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x half> [[TMP0]]
 //
 vfloat16m4_t test_vfmsub_vf_f16m4_m (vbool4_t mask, vfloat16m4_t vd, _Float16 rs1, vfloat16m4_t vs2, size_t vl) {
@@ -575,7 +575,7 @@ vfloat16m4_t test_vfmsub_vf_f16m4_m (vbool4_t mask, vfloat16m4_t vd, _Float16 rs
 
 // CHECK-RV64-LABEL: @test_vfmsub_vv_f16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfmsub.mask.nxv32f16.nxv32f16.i64(<vscale x 32 x half> [[VD:%.*]], <vscale x 32 x half> [[VS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfmsub.mask.nxv32f16.nxv32f16.i64(<vscale x 32 x half> [[VD:%.*]], <vscale x 32 x half> [[VS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x half> [[TMP0]]
 //
 vfloat16m8_t test_vfmsub_vv_f16m8_m (vbool2_t mask, vfloat16m8_t vd, vfloat16m8_t vs1, vfloat16m8_t vs2, size_t vl) {
@@ -584,7 +584,7 @@ vfloat16m8_t test_vfmsub_vv_f16m8_m (vbool2_t mask, vfloat16m8_t vd, vfloat16m8_
 
 // CHECK-RV64-LABEL: @test_vfmsub_vf_f16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfmsub.mask.nxv32f16.f16.i64(<vscale x 32 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfmsub.mask.nxv32f16.f16.i64(<vscale x 32 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x half> [[TMP0]]
 //
 vfloat16m8_t test_vfmsub_vf_f16m8_m (vbool2_t mask, vfloat16m8_t vd, _Float16 rs1, vfloat16m8_t vs2, size_t vl) {
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfnmacc.c b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfnmacc.c
index 192a5619332f22..703d7aa097f7cf 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfnmacc.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfnmacc.c
@@ -188,7 +188,7 @@ vfloat64m8_t test_vfnmacc_vf_f64m8(vfloat64m8_t acc, double op1,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmacc.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmacc.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfnmacc_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -199,7 +199,7 @@ vfloat32mf2_t test_vfnmacc_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmacc.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmacc.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfnmacc_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -210,7 +210,7 @@ vfloat32mf2_t test_vfnmacc_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmacc.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmacc.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfnmacc_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -221,7 +221,7 @@ vfloat32m1_t test_vfnmacc_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmacc.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmacc.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfnmacc_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -231,7 +231,7 @@ vfloat32m1_t test_vfnmacc_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmacc.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmacc.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfnmacc_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -242,7 +242,7 @@ vfloat32m2_t test_vfnmacc_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmacc.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmacc.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfnmacc_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -252,7 +252,7 @@ vfloat32m2_t test_vfnmacc_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmacc.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmacc.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfnmacc_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
@@ -263,7 +263,7 @@ vfloat32m4_t test_vfnmacc_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmacc.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmacc.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfnmacc_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
@@ -273,7 +273,7 @@ vfloat32m4_t test_vfnmacc_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmacc.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmacc.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfnmacc_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
@@ -284,7 +284,7 @@ vfloat32m8_t test_vfnmacc_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmacc.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmacc.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfnmacc_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
@@ -294,7 +294,7 @@ vfloat32m8_t test_vfnmacc_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmacc.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmacc.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfnmacc_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -305,7 +305,7 @@ vfloat64m1_t test_vfnmacc_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmacc.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmacc.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfnmacc_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -315,7 +315,7 @@ vfloat64m1_t test_vfnmacc_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmacc.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmacc.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfnmacc_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -326,7 +326,7 @@ vfloat64m2_t test_vfnmacc_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmacc.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmacc.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfnmacc_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -336,7 +336,7 @@ vfloat64m2_t test_vfnmacc_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmacc.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmacc.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfnmacc_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -347,7 +347,7 @@ vfloat64m4_t test_vfnmacc_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmacc.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmacc.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfnmacc_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -357,7 +357,7 @@ vfloat64m4_t test_vfnmacc_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmacc.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmacc.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfnmacc_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -368,7 +368,7 @@ vfloat64m8_t test_vfnmacc_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmacc.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmacc.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfnmacc_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -486,7 +486,7 @@ vfloat16m8_t test_vfnmacc_vf_f16m8 (vfloat16m8_t vd, _Float16 rs1, vfloat16m8_t
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfnmacc.mask.nxv1f16.nxv1f16.i64(<vscale x 1 x half> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfnmacc.mask.nxv1f16.nxv1f16.i64(<vscale x 1 x half> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x half> [[TMP0]]
 //
 vfloat16mf4_t test_vfnmacc_vv_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, vfloat16mf4_t vs1, vfloat16mf4_t vs2, size_t vl) {
@@ -495,7 +495,7 @@ vfloat16mf4_t test_vfnmacc_vv_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, vfloat
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfnmacc.mask.nxv1f16.f16.i64(<vscale x 1 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfnmacc.mask.nxv1f16.f16.i64(<vscale x 1 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x half> [[TMP0]]
 //
 vfloat16mf4_t test_vfnmacc_vf_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, _Float16 rs1, vfloat16mf4_t vs2, size_t vl) {
@@ -504,7 +504,7 @@ vfloat16mf4_t test_vfnmacc_vf_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, _Float
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfnmacc.mask.nxv2f16.nxv2f16.i64(<vscale x 2 x half> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfnmacc.mask.nxv2f16.nxv2f16.i64(<vscale x 2 x half> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x half> [[TMP0]]
 //
 vfloat16mf2_t test_vfnmacc_vv_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, vfloat16mf2_t vs1, vfloat16mf2_t vs2, size_t vl) {
@@ -513,7 +513,7 @@ vfloat16mf2_t test_vfnmacc_vv_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, vfloat
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfnmacc.mask.nxv2f16.f16.i64(<vscale x 2 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfnmacc.mask.nxv2f16.f16.i64(<vscale x 2 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x half> [[TMP0]]
 //
 vfloat16mf2_t test_vfnmacc_vf_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, _Float16 rs1, vfloat16mf2_t vs2, size_t vl) {
@@ -522,7 +522,7 @@ vfloat16mf2_t test_vfnmacc_vf_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, _Float
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfnmacc.mask.nxv4f16.nxv4f16.i64(<vscale x 4 x half> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfnmacc.mask.nxv4f16.nxv4f16.i64(<vscale x 4 x half> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x half> [[TMP0]]
 //
 vfloat16m1_t test_vfnmacc_vv_f16m1_m (vbool16_t mask, vfloat16m1_t vd, vfloat16m1_t vs1, vfloat16m1_t vs2, size_t vl) {
@@ -531,7 +531,7 @@ vfloat16m1_t test_vfnmacc_vv_f16m1_m (vbool16_t mask, vfloat16m1_t vd, vfloat16m
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfnmacc.mask.nxv4f16.f16.i64(<vscale x 4 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfnmacc.mask.nxv4f16.f16.i64(<vscale x 4 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x half> [[TMP0]]
 //
 vfloat16m1_t test_vfnmacc_vf_f16m1_m (vbool16_t mask, vfloat16m1_t vd, _Float16 rs1, vfloat16m1_t vs2, size_t vl) {
@@ -540,7 +540,7 @@ vfloat16m1_t test_vfnmacc_vf_f16m1_m (vbool16_t mask, vfloat16m1_t vd, _Float16
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfnmacc.mask.nxv8f16.nxv8f16.i64(<vscale x 8 x half> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfnmacc.mask.nxv8f16.nxv8f16.i64(<vscale x 8 x half> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x half> [[TMP0]]
 //
 vfloat16m2_t test_vfnmacc_vv_f16m2_m (vbool8_t mask, vfloat16m2_t vd, vfloat16m2_t vs1, vfloat16m2_t vs2, size_t vl) {
@@ -549,7 +549,7 @@ vfloat16m2_t test_vfnmacc_vv_f16m2_m (vbool8_t mask, vfloat16m2_t vd, vfloat16m2
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfnmacc.mask.nxv8f16.f16.i64(<vscale x 8 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfnmacc.mask.nxv8f16.f16.i64(<vscale x 8 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x half> [[TMP0]]
 //
 vfloat16m2_t test_vfnmacc_vf_f16m2_m (vbool8_t mask, vfloat16m2_t vd, _Float16 rs1, vfloat16m2_t vs2, size_t vl) {
@@ -558,7 +558,7 @@ vfloat16m2_t test_vfnmacc_vf_f16m2_m (vbool8_t mask, vfloat16m2_t vd, _Float16 r
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfnmacc.mask.nxv16f16.nxv16f16.i64(<vscale x 16 x half> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfnmacc.mask.nxv16f16.nxv16f16.i64(<vscale x 16 x half> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x half> [[TMP0]]
 //
 vfloat16m4_t test_vfnmacc_vv_f16m4_m (vbool4_t mask, vfloat16m4_t vd, vfloat16m4_t vs1, vfloat16m4_t vs2, size_t vl) {
@@ -567,7 +567,7 @@ vfloat16m4_t test_vfnmacc_vv_f16m4_m (vbool4_t mask, vfloat16m4_t vd, vfloat16m4
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfnmacc.mask.nxv16f16.f16.i64(<vscale x 16 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfnmacc.mask.nxv16f16.f16.i64(<vscale x 16 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x half> [[TMP0]]
 //
 vfloat16m4_t test_vfnmacc_vf_f16m4_m (vbool4_t mask, vfloat16m4_t vd, _Float16 rs1, vfloat16m4_t vs2, size_t vl) {
@@ -576,7 +576,7 @@ vfloat16m4_t test_vfnmacc_vf_f16m4_m (vbool4_t mask, vfloat16m4_t vd, _Float16 r
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vv_f16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfnmacc.mask.nxv32f16.nxv32f16.i64(<vscale x 32 x half> [[VD:%.*]], <vscale x 32 x half> [[VS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfnmacc.mask.nxv32f16.nxv32f16.i64(<vscale x 32 x half> [[VD:%.*]], <vscale x 32 x half> [[VS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x half> [[TMP0]]
 //
 vfloat16m8_t test_vfnmacc_vv_f16m8_m (vbool2_t mask, vfloat16m8_t vd, vfloat16m8_t vs1, vfloat16m8_t vs2, size_t vl) {
@@ -585,7 +585,7 @@ vfloat16m8_t test_vfnmacc_vv_f16m8_m (vbool2_t mask, vfloat16m8_t vd, vfloat16m8
 
 // CHECK-RV64-LABEL: @test_vfnmacc_vf_f16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfnmacc.mask.nxv32f16.f16.i64(<vscale x 32 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfnmacc.mask.nxv32f16.f16.i64(<vscale x 32 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x half> [[TMP0]]
 //
 vfloat16m8_t test_vfnmacc_vf_f16m8_m (vbool2_t mask, vfloat16m8_t vd, _Float16 rs1, vfloat16m8_t vs2, size_t vl) {
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfnmadd.c b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfnmadd.c
index 0e0ae186b87dd7..adefa17bb4f371 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfnmadd.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfnmadd.c
@@ -188,7 +188,7 @@ vfloat64m8_t test_vfnmadd_vf_f64m8(vfloat64m8_t acc, double op1,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmadd.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmadd.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfnmadd_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -199,7 +199,7 @@ vfloat32mf2_t test_vfnmadd_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmadd.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmadd.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfnmadd_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -210,7 +210,7 @@ vfloat32mf2_t test_vfnmadd_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmadd.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmadd.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfnmadd_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -221,7 +221,7 @@ vfloat32m1_t test_vfnmadd_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmadd.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmadd.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfnmadd_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -231,7 +231,7 @@ vfloat32m1_t test_vfnmadd_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmadd.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmadd.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfnmadd_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -242,7 +242,7 @@ vfloat32m2_t test_vfnmadd_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmadd.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmadd.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfnmadd_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -252,7 +252,7 @@ vfloat32m2_t test_vfnmadd_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmadd.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmadd.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfnmadd_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
@@ -263,7 +263,7 @@ vfloat32m4_t test_vfnmadd_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmadd.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmadd.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfnmadd_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
@@ -273,7 +273,7 @@ vfloat32m4_t test_vfnmadd_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmadd.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmadd.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfnmadd_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
@@ -284,7 +284,7 @@ vfloat32m8_t test_vfnmadd_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmadd.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmadd.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfnmadd_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
@@ -294,7 +294,7 @@ vfloat32m8_t test_vfnmadd_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmadd.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmadd.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfnmadd_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -305,7 +305,7 @@ vfloat64m1_t test_vfnmadd_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmadd.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmadd.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfnmadd_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -315,7 +315,7 @@ vfloat64m1_t test_vfnmadd_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmadd.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmadd.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfnmadd_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -326,7 +326,7 @@ vfloat64m2_t test_vfnmadd_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmadd.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmadd.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfnmadd_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -336,7 +336,7 @@ vfloat64m2_t test_vfnmadd_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmadd.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmadd.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfnmadd_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -347,7 +347,7 @@ vfloat64m4_t test_vfnmadd_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmadd.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmadd.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfnmadd_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -357,7 +357,7 @@ vfloat64m4_t test_vfnmadd_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmadd.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmadd.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfnmadd_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -368,7 +368,7 @@ vfloat64m8_t test_vfnmadd_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmadd.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmadd.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfnmadd_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -486,7 +486,7 @@ vfloat16m8_t test_vfnmadd_vf_f16m8 (vfloat16m8_t vd, _Float16 rs1, vfloat16m8_t
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfnmadd.mask.nxv1f16.nxv1f16.i64(<vscale x 1 x half> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfnmadd.mask.nxv1f16.nxv1f16.i64(<vscale x 1 x half> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x half> [[TMP0]]
 //
 vfloat16mf4_t test_vfnmadd_vv_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, vfloat16mf4_t vs1, vfloat16mf4_t vs2, size_t vl) {
@@ -495,7 +495,7 @@ vfloat16mf4_t test_vfnmadd_vv_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, vfloat
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfnmadd.mask.nxv1f16.f16.i64(<vscale x 1 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfnmadd.mask.nxv1f16.f16.i64(<vscale x 1 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x half> [[TMP0]]
 //
 vfloat16mf4_t test_vfnmadd_vf_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, _Float16 rs1, vfloat16mf4_t vs2, size_t vl) {
@@ -504,7 +504,7 @@ vfloat16mf4_t test_vfnmadd_vf_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, _Float
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfnmadd.mask.nxv2f16.nxv2f16.i64(<vscale x 2 x half> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfnmadd.mask.nxv2f16.nxv2f16.i64(<vscale x 2 x half> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x half> [[TMP0]]
 //
 vfloat16mf2_t test_vfnmadd_vv_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, vfloat16mf2_t vs1, vfloat16mf2_t vs2, size_t vl) {
@@ -513,7 +513,7 @@ vfloat16mf2_t test_vfnmadd_vv_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, vfloat
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfnmadd.mask.nxv2f16.f16.i64(<vscale x 2 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfnmadd.mask.nxv2f16.f16.i64(<vscale x 2 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x half> [[TMP0]]
 //
 vfloat16mf2_t test_vfnmadd_vf_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, _Float16 rs1, vfloat16mf2_t vs2, size_t vl) {
@@ -522,7 +522,7 @@ vfloat16mf2_t test_vfnmadd_vf_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, _Float
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfnmadd.mask.nxv4f16.nxv4f16.i64(<vscale x 4 x half> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfnmadd.mask.nxv4f16.nxv4f16.i64(<vscale x 4 x half> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x half> [[TMP0]]
 //
 vfloat16m1_t test_vfnmadd_vv_f16m1_m (vbool16_t mask, vfloat16m1_t vd, vfloat16m1_t vs1, vfloat16m1_t vs2, size_t vl) {
@@ -531,7 +531,7 @@ vfloat16m1_t test_vfnmadd_vv_f16m1_m (vbool16_t mask, vfloat16m1_t vd, vfloat16m
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfnmadd.mask.nxv4f16.f16.i64(<vscale x 4 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfnmadd.mask.nxv4f16.f16.i64(<vscale x 4 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x half> [[TMP0]]
 //
 vfloat16m1_t test_vfnmadd_vf_f16m1_m (vbool16_t mask, vfloat16m1_t vd, _Float16 rs1, vfloat16m1_t vs2, size_t vl) {
@@ -540,7 +540,7 @@ vfloat16m1_t test_vfnmadd_vf_f16m1_m (vbool16_t mask, vfloat16m1_t vd, _Float16
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfnmadd.mask.nxv8f16.nxv8f16.i64(<vscale x 8 x half> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfnmadd.mask.nxv8f16.nxv8f16.i64(<vscale x 8 x half> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x half> [[TMP0]]
 //
 vfloat16m2_t test_vfnmadd_vv_f16m2_m (vbool8_t mask, vfloat16m2_t vd, vfloat16m2_t vs1, vfloat16m2_t vs2, size_t vl) {
@@ -549,7 +549,7 @@ vfloat16m2_t test_vfnmadd_vv_f16m2_m (vbool8_t mask, vfloat16m2_t vd, vfloat16m2
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfnmadd.mask.nxv8f16.f16.i64(<vscale x 8 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfnmadd.mask.nxv8f16.f16.i64(<vscale x 8 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x half> [[TMP0]]
 //
 vfloat16m2_t test_vfnmadd_vf_f16m2_m (vbool8_t mask, vfloat16m2_t vd, _Float16 rs1, vfloat16m2_t vs2, size_t vl) {
@@ -558,7 +558,7 @@ vfloat16m2_t test_vfnmadd_vf_f16m2_m (vbool8_t mask, vfloat16m2_t vd, _Float16 r
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfnmadd.mask.nxv16f16.nxv16f16.i64(<vscale x 16 x half> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfnmadd.mask.nxv16f16.nxv16f16.i64(<vscale x 16 x half> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x half> [[TMP0]]
 //
 vfloat16m4_t test_vfnmadd_vv_f16m4_m (vbool4_t mask, vfloat16m4_t vd, vfloat16m4_t vs1, vfloat16m4_t vs2, size_t vl) {
@@ -567,7 +567,7 @@ vfloat16m4_t test_vfnmadd_vv_f16m4_m (vbool4_t mask, vfloat16m4_t vd, vfloat16m4
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfnmadd.mask.nxv16f16.f16.i64(<vscale x 16 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfnmadd.mask.nxv16f16.f16.i64(<vscale x 16 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x half> [[TMP0]]
 //
 vfloat16m4_t test_vfnmadd_vf_f16m4_m (vbool4_t mask, vfloat16m4_t vd, _Float16 rs1, vfloat16m4_t vs2, size_t vl) {
@@ -576,7 +576,7 @@ vfloat16m4_t test_vfnmadd_vf_f16m4_m (vbool4_t mask, vfloat16m4_t vd, _Float16 r
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vv_f16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfnmadd.mask.nxv32f16.nxv32f16.i64(<vscale x 32 x half> [[VD:%.*]], <vscale x 32 x half> [[VS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfnmadd.mask.nxv32f16.nxv32f16.i64(<vscale x 32 x half> [[VD:%.*]], <vscale x 32 x half> [[VS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x half> [[TMP0]]
 //
 vfloat16m8_t test_vfnmadd_vv_f16m8_m (vbool2_t mask, vfloat16m8_t vd, vfloat16m8_t vs1, vfloat16m8_t vs2, size_t vl) {
@@ -585,7 +585,7 @@ vfloat16m8_t test_vfnmadd_vv_f16m8_m (vbool2_t mask, vfloat16m8_t vd, vfloat16m8
 
 // CHECK-RV64-LABEL: @test_vfnmadd_vf_f16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfnmadd.mask.nxv32f16.f16.i64(<vscale x 32 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfnmadd.mask.nxv32f16.f16.i64(<vscale x 32 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x half> [[TMP0]]
 //
 vfloat16m8_t test_vfnmadd_vf_f16m8_m (vbool2_t mask, vfloat16m8_t vd, _Float16 rs1, vfloat16m8_t vs2, size_t vl) {
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfnmsac.c b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfnmsac.c
index 11d3b9816df4d7..c9f805328a1077 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfnmsac.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfnmsac.c
@@ -188,7 +188,7 @@ vfloat64m8_t test_vfnmsac_vf_f64m8(vfloat64m8_t acc, double op1,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmsac.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmsac.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfnmsac_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -199,7 +199,7 @@ vfloat32mf2_t test_vfnmsac_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmsac.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmsac.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfnmsac_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -210,7 +210,7 @@ vfloat32mf2_t test_vfnmsac_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmsac.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmsac.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfnmsac_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -221,7 +221,7 @@ vfloat32m1_t test_vfnmsac_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmsac.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmsac.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfnmsac_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -231,7 +231,7 @@ vfloat32m1_t test_vfnmsac_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmsac.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmsac.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfnmsac_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -242,7 +242,7 @@ vfloat32m2_t test_vfnmsac_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmsac.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmsac.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfnmsac_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -252,7 +252,7 @@ vfloat32m2_t test_vfnmsac_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmsac.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmsac.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfnmsac_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
@@ -263,7 +263,7 @@ vfloat32m4_t test_vfnmsac_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmsac.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmsac.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfnmsac_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
@@ -273,7 +273,7 @@ vfloat32m4_t test_vfnmsac_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmsac.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmsac.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfnmsac_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
@@ -284,7 +284,7 @@ vfloat32m8_t test_vfnmsac_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmsac.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmsac.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfnmsac_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
@@ -294,7 +294,7 @@ vfloat32m8_t test_vfnmsac_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmsac.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmsac.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfnmsac_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -305,7 +305,7 @@ vfloat64m1_t test_vfnmsac_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmsac.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmsac.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfnmsac_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -315,7 +315,7 @@ vfloat64m1_t test_vfnmsac_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmsac.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmsac.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfnmsac_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -326,7 +326,7 @@ vfloat64m2_t test_vfnmsac_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmsac.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmsac.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfnmsac_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -336,7 +336,7 @@ vfloat64m2_t test_vfnmsac_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmsac.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmsac.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfnmsac_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -347,7 +347,7 @@ vfloat64m4_t test_vfnmsac_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmsac.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmsac.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfnmsac_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -357,7 +357,7 @@ vfloat64m4_t test_vfnmsac_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmsac.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmsac.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfnmsac_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -368,7 +368,7 @@ vfloat64m8_t test_vfnmsac_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmsac.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmsac.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfnmsac_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -486,7 +486,7 @@ vfloat16m8_t test_vfnmsac_vf_f16m8 (vfloat16m8_t vd, _Float16 rs1, vfloat16m8_t
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfnmsac.mask.nxv1f16.nxv1f16.i64(<vscale x 1 x half> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfnmsac.mask.nxv1f16.nxv1f16.i64(<vscale x 1 x half> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x half> [[TMP0]]
 //
 vfloat16mf4_t test_vfnmsac_vv_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, vfloat16mf4_t vs1, vfloat16mf4_t vs2, size_t vl) {
@@ -495,7 +495,7 @@ vfloat16mf4_t test_vfnmsac_vv_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, vfloat
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfnmsac.mask.nxv1f16.f16.i64(<vscale x 1 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfnmsac.mask.nxv1f16.f16.i64(<vscale x 1 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x half> [[TMP0]]
 //
 vfloat16mf4_t test_vfnmsac_vf_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, _Float16 rs1, vfloat16mf4_t vs2, size_t vl) {
@@ -504,7 +504,7 @@ vfloat16mf4_t test_vfnmsac_vf_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, _Float
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfnmsac.mask.nxv2f16.nxv2f16.i64(<vscale x 2 x half> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfnmsac.mask.nxv2f16.nxv2f16.i64(<vscale x 2 x half> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x half> [[TMP0]]
 //
 vfloat16mf2_t test_vfnmsac_vv_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, vfloat16mf2_t vs1, vfloat16mf2_t vs2, size_t vl) {
@@ -513,7 +513,7 @@ vfloat16mf2_t test_vfnmsac_vv_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, vfloat
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfnmsac.mask.nxv2f16.f16.i64(<vscale x 2 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfnmsac.mask.nxv2f16.f16.i64(<vscale x 2 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x half> [[TMP0]]
 //
 vfloat16mf2_t test_vfnmsac_vf_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, _Float16 rs1, vfloat16mf2_t vs2, size_t vl) {
@@ -522,7 +522,7 @@ vfloat16mf2_t test_vfnmsac_vf_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, _Float
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfnmsac.mask.nxv4f16.nxv4f16.i64(<vscale x 4 x half> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfnmsac.mask.nxv4f16.nxv4f16.i64(<vscale x 4 x half> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x half> [[TMP0]]
 //
 vfloat16m1_t test_vfnmsac_vv_f16m1_m (vbool16_t mask, vfloat16m1_t vd, vfloat16m1_t vs1, vfloat16m1_t vs2, size_t vl) {
@@ -531,7 +531,7 @@ vfloat16m1_t test_vfnmsac_vv_f16m1_m (vbool16_t mask, vfloat16m1_t vd, vfloat16m
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfnmsac.mask.nxv4f16.f16.i64(<vscale x 4 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfnmsac.mask.nxv4f16.f16.i64(<vscale x 4 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x half> [[TMP0]]
 //
 vfloat16m1_t test_vfnmsac_vf_f16m1_m (vbool16_t mask, vfloat16m1_t vd, _Float16 rs1, vfloat16m1_t vs2, size_t vl) {
@@ -540,7 +540,7 @@ vfloat16m1_t test_vfnmsac_vf_f16m1_m (vbool16_t mask, vfloat16m1_t vd, _Float16
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfnmsac.mask.nxv8f16.nxv8f16.i64(<vscale x 8 x half> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfnmsac.mask.nxv8f16.nxv8f16.i64(<vscale x 8 x half> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x half> [[TMP0]]
 //
 vfloat16m2_t test_vfnmsac_vv_f16m2_m (vbool8_t mask, vfloat16m2_t vd, vfloat16m2_t vs1, vfloat16m2_t vs2, size_t vl) {
@@ -549,7 +549,7 @@ vfloat16m2_t test_vfnmsac_vv_f16m2_m (vbool8_t mask, vfloat16m2_t vd, vfloat16m2
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfnmsac.mask.nxv8f16.f16.i64(<vscale x 8 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfnmsac.mask.nxv8f16.f16.i64(<vscale x 8 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x half> [[TMP0]]
 //
 vfloat16m2_t test_vfnmsac_vf_f16m2_m (vbool8_t mask, vfloat16m2_t vd, _Float16 rs1, vfloat16m2_t vs2, size_t vl) {
@@ -558,7 +558,7 @@ vfloat16m2_t test_vfnmsac_vf_f16m2_m (vbool8_t mask, vfloat16m2_t vd, _Float16 r
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfnmsac.mask.nxv16f16.nxv16f16.i64(<vscale x 16 x half> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfnmsac.mask.nxv16f16.nxv16f16.i64(<vscale x 16 x half> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x half> [[TMP0]]
 //
 vfloat16m4_t test_vfnmsac_vv_f16m4_m (vbool4_t mask, vfloat16m4_t vd, vfloat16m4_t vs1, vfloat16m4_t vs2, size_t vl) {
@@ -567,7 +567,7 @@ vfloat16m4_t test_vfnmsac_vv_f16m4_m (vbool4_t mask, vfloat16m4_t vd, vfloat16m4
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfnmsac.mask.nxv16f16.f16.i64(<vscale x 16 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfnmsac.mask.nxv16f16.f16.i64(<vscale x 16 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x half> [[TMP0]]
 //
 vfloat16m4_t test_vfnmsac_vf_f16m4_m (vbool4_t mask, vfloat16m4_t vd, _Float16 rs1, vfloat16m4_t vs2, size_t vl) {
@@ -576,7 +576,7 @@ vfloat16m4_t test_vfnmsac_vf_f16m4_m (vbool4_t mask, vfloat16m4_t vd, _Float16 r
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vv_f16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfnmsac.mask.nxv32f16.nxv32f16.i64(<vscale x 32 x half> [[VD:%.*]], <vscale x 32 x half> [[VS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfnmsac.mask.nxv32f16.nxv32f16.i64(<vscale x 32 x half> [[VD:%.*]], <vscale x 32 x half> [[VS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x half> [[TMP0]]
 //
 vfloat16m8_t test_vfnmsac_vv_f16m8_m (vbool2_t mask, vfloat16m8_t vd, vfloat16m8_t vs1, vfloat16m8_t vs2, size_t vl) {
@@ -585,7 +585,7 @@ vfloat16m8_t test_vfnmsac_vv_f16m8_m (vbool2_t mask, vfloat16m8_t vd, vfloat16m8
 
 // CHECK-RV64-LABEL: @test_vfnmsac_vf_f16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfnmsac.mask.nxv32f16.f16.i64(<vscale x 32 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfnmsac.mask.nxv32f16.f16.i64(<vscale x 32 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x half> [[TMP0]]
 //
 vfloat16m8_t test_vfnmsac_vf_f16m8_m (vbool2_t mask, vfloat16m8_t vd, _Float16 rs1, vfloat16m8_t vs2, size_t vl) {
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfnmsub.c b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfnmsub.c
index bdb90cb40e609d..c2254419baa544 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfnmsub.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfnmsub.c
@@ -188,7 +188,7 @@ vfloat64m8_t test_vfnmsub_vf_f64m8(vfloat64m8_t acc, double op1,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmsub.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmsub.mask.nxv1f32.nxv1f32.i64(<vscale x 1 x float> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfnmsub_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -199,7 +199,7 @@ vfloat32mf2_t test_vfnmsub_vv_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmsub.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfnmsub.mask.nxv1f32.f32.i64(<vscale x 1 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfnmsub_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
@@ -210,7 +210,7 @@ vfloat32mf2_t test_vfnmsub_vf_f32mf2_m(vbool64_t mask, vfloat32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmsub.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmsub.mask.nxv2f32.nxv2f32.i64(<vscale x 2 x float> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfnmsub_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -221,7 +221,7 @@ vfloat32m1_t test_vfnmsub_vv_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmsub.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfnmsub.mask.nxv2f32.f32.i64(<vscale x 2 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfnmsub_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
@@ -231,7 +231,7 @@ vfloat32m1_t test_vfnmsub_vf_f32m1_m(vbool32_t mask, vfloat32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmsub.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmsub.mask.nxv4f32.nxv4f32.i64(<vscale x 4 x float> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfnmsub_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -242,7 +242,7 @@ vfloat32m2_t test_vfnmsub_vv_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmsub.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfnmsub.mask.nxv4f32.f32.i64(<vscale x 4 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfnmsub_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
@@ -252,7 +252,7 @@ vfloat32m2_t test_vfnmsub_vf_f32m2_m(vbool16_t mask, vfloat32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmsub.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmsub.mask.nxv8f32.nxv8f32.i64(<vscale x 8 x float> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfnmsub_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
@@ -263,7 +263,7 @@ vfloat32m4_t test_vfnmsub_vv_f32m4_m(vbool8_t mask, vfloat32m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmsub.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfnmsub.mask.nxv8f32.f32.i64(<vscale x 8 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfnmsub_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
@@ -273,7 +273,7 @@ vfloat32m4_t test_vfnmsub_vf_f32m4_m(vbool8_t mask, vfloat32m4_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmsub.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmsub.mask.nxv16f32.nxv16f32.i64(<vscale x 16 x float> [[ACC:%.*]], <vscale x 16 x float> [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfnmsub_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
@@ -284,7 +284,7 @@ vfloat32m8_t test_vfnmsub_vv_f32m8_m(vbool4_t mask, vfloat32m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmsub.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfnmsub.mask.nxv16f32.f32.i64(<vscale x 16 x float> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 16 x float> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfnmsub_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
@@ -294,7 +294,7 @@ vfloat32m8_t test_vfnmsub_vf_f32m8_m(vbool4_t mask, vfloat32m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmsub.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmsub.mask.nxv1f64.nxv1f64.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x double> [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfnmsub_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -305,7 +305,7 @@ vfloat64m1_t test_vfnmsub_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmsub.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfnmsub.mask.nxv1f64.f64.i64(<vscale x 1 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 1 x double> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfnmsub_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -315,7 +315,7 @@ vfloat64m1_t test_vfnmsub_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmsub.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmsub.mask.nxv2f64.nxv2f64.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x double> [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfnmsub_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -326,7 +326,7 @@ vfloat64m2_t test_vfnmsub_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmsub.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfnmsub.mask.nxv2f64.f64.i64(<vscale x 2 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 2 x double> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfnmsub_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -336,7 +336,7 @@ vfloat64m2_t test_vfnmsub_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmsub.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmsub.mask.nxv4f64.nxv4f64.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x double> [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfnmsub_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -347,7 +347,7 @@ vfloat64m4_t test_vfnmsub_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmsub.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfnmsub.mask.nxv4f64.f64.i64(<vscale x 4 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 4 x double> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfnmsub_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -357,7 +357,7 @@ vfloat64m4_t test_vfnmsub_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmsub.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmsub.mask.nxv8f64.nxv8f64.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x double> [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfnmsub_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -368,7 +368,7 @@ vfloat64m8_t test_vfnmsub_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmsub.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfnmsub.mask.nxv8f64.f64.i64(<vscale x 8 x double> [[ACC:%.*]], double [[OP1:%.*]], <vscale x 8 x double> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfnmsub_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -492,7 +492,7 @@ vfloat16m8_t test_vfnmsub_vf_f16m8 (vfloat16m8_t vd, _Float16 rs1, vfloat16m8_t
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfnmsub.mask.nxv1f16.nxv1f16.i64(<vscale x 1 x half> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfnmsub.mask.nxv1f16.nxv1f16.i64(<vscale x 1 x half> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x half> [[TMP0]]
 //
 vfloat16mf4_t test_vfnmsub_vv_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, vfloat16mf4_t vs1, vfloat16mf4_t vs2, size_t vl) {
@@ -501,7 +501,7 @@ vfloat16mf4_t test_vfnmsub_vv_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, vfloat
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfnmsub.mask.nxv1f16.f16.i64(<vscale x 1 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vfnmsub.mask.nxv1f16.f16.i64(<vscale x 1 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x half> [[TMP0]]
 //
 vfloat16mf4_t test_vfnmsub_vf_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, _Float16 rs1, vfloat16mf4_t vs2, size_t vl) {
@@ -510,7 +510,7 @@ vfloat16mf4_t test_vfnmsub_vf_f16mf4_m (vbool64_t mask, vfloat16mf4_t vd, _Float
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfnmsub.mask.nxv2f16.nxv2f16.i64(<vscale x 2 x half> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfnmsub.mask.nxv2f16.nxv2f16.i64(<vscale x 2 x half> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x half> [[TMP0]]
 //
 vfloat16mf2_t test_vfnmsub_vv_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, vfloat16mf2_t vs1, vfloat16mf2_t vs2, size_t vl) {
@@ -519,7 +519,7 @@ vfloat16mf2_t test_vfnmsub_vv_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, vfloat
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfnmsub.mask.nxv2f16.f16.i64(<vscale x 2 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vfnmsub.mask.nxv2f16.f16.i64(<vscale x 2 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x half> [[TMP0]]
 //
 vfloat16mf2_t test_vfnmsub_vf_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, _Float16 rs1, vfloat16mf2_t vs2, size_t vl) {
@@ -528,7 +528,7 @@ vfloat16mf2_t test_vfnmsub_vf_f16mf2_m (vbool32_t mask, vfloat16mf2_t vd, _Float
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfnmsub.mask.nxv4f16.nxv4f16.i64(<vscale x 4 x half> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfnmsub.mask.nxv4f16.nxv4f16.i64(<vscale x 4 x half> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x half> [[TMP0]]
 //
 vfloat16m1_t test_vfnmsub_vv_f16m1_m (vbool16_t mask, vfloat16m1_t vd, vfloat16m1_t vs1, vfloat16m1_t vs2, size_t vl) {
@@ -537,7 +537,7 @@ vfloat16m1_t test_vfnmsub_vv_f16m1_m (vbool16_t mask, vfloat16m1_t vd, vfloat16m
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfnmsub.mask.nxv4f16.f16.i64(<vscale x 4 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vfnmsub.mask.nxv4f16.f16.i64(<vscale x 4 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x half> [[TMP0]]
 //
 vfloat16m1_t test_vfnmsub_vf_f16m1_m (vbool16_t mask, vfloat16m1_t vd, _Float16 rs1, vfloat16m1_t vs2, size_t vl) {
@@ -546,7 +546,7 @@ vfloat16m1_t test_vfnmsub_vf_f16m1_m (vbool16_t mask, vfloat16m1_t vd, _Float16
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfnmsub.mask.nxv8f16.nxv8f16.i64(<vscale x 8 x half> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfnmsub.mask.nxv8f16.nxv8f16.i64(<vscale x 8 x half> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x half> [[TMP0]]
 //
 vfloat16m2_t test_vfnmsub_vv_f16m2_m (vbool8_t mask, vfloat16m2_t vd, vfloat16m2_t vs1, vfloat16m2_t vs2, size_t vl) {
@@ -555,7 +555,7 @@ vfloat16m2_t test_vfnmsub_vv_f16m2_m (vbool8_t mask, vfloat16m2_t vd, vfloat16m2
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfnmsub.mask.nxv8f16.f16.i64(<vscale x 8 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vfnmsub.mask.nxv8f16.f16.i64(<vscale x 8 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x half> [[TMP0]]
 //
 vfloat16m2_t test_vfnmsub_vf_f16m2_m (vbool8_t mask, vfloat16m2_t vd, _Float16 rs1, vfloat16m2_t vs2, size_t vl) {
@@ -564,7 +564,7 @@ vfloat16m2_t test_vfnmsub_vf_f16m2_m (vbool8_t mask, vfloat16m2_t vd, _Float16 r
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfnmsub.mask.nxv16f16.nxv16f16.i64(<vscale x 16 x half> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfnmsub.mask.nxv16f16.nxv16f16.i64(<vscale x 16 x half> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x half> [[TMP0]]
 //
 vfloat16m4_t test_vfnmsub_vv_f16m4_m (vbool4_t mask, vfloat16m4_t vd, vfloat16m4_t vs1, vfloat16m4_t vs2, size_t vl) {
@@ -573,7 +573,7 @@ vfloat16m4_t test_vfnmsub_vv_f16m4_m (vbool4_t mask, vfloat16m4_t vd, vfloat16m4
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfnmsub.mask.nxv16f16.f16.i64(<vscale x 16 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vfnmsub.mask.nxv16f16.f16.i64(<vscale x 16 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x half> [[TMP0]]
 //
 vfloat16m4_t test_vfnmsub_vf_f16m4_m (vbool4_t mask, vfloat16m4_t vd, _Float16 rs1, vfloat16m4_t vs2, size_t vl) {
@@ -582,7 +582,7 @@ vfloat16m4_t test_vfnmsub_vf_f16m4_m (vbool4_t mask, vfloat16m4_t vd, _Float16 r
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vv_f16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfnmsub.mask.nxv32f16.nxv32f16.i64(<vscale x 32 x half> [[VD:%.*]], <vscale x 32 x half> [[VS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfnmsub.mask.nxv32f16.nxv32f16.i64(<vscale x 32 x half> [[VD:%.*]], <vscale x 32 x half> [[VS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x half> [[TMP0]]
 //
 vfloat16m8_t test_vfnmsub_vv_f16m8_m (vbool2_t mask, vfloat16m8_t vd, vfloat16m8_t vs1, vfloat16m8_t vs2, size_t vl) {
@@ -591,7 +591,7 @@ vfloat16m8_t test_vfnmsub_vv_f16m8_m (vbool2_t mask, vfloat16m8_t vd, vfloat16m8
 
 // CHECK-RV64-LABEL: @test_vfnmsub_vf_f16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfnmsub.mask.nxv32f16.f16.i64(<vscale x 32 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vfnmsub.mask.nxv32f16.f16.i64(<vscale x 32 x half> [[VD:%.*]], half [[RS1:%.*]], <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x half> [[TMP0]]
 //
 vfloat16m8_t test_vfnmsub_vf_f16m8_m (vbool2_t mask, vfloat16m8_t vd, _Float16 rs1, vfloat16m8_t vs2, size_t vl) {
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfwmacc.c b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfwmacc.c
index 51c3a6edaaa079..c43b9ad466f72e 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfwmacc.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfwmacc.c
@@ -88,7 +88,7 @@ vfloat64m8_t test_vfwmacc_vf_f64m8(vfloat64m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwmacc.mask.nxv1f64.nxv1f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwmacc.mask.nxv1f64.nxv1f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfwmacc_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -99,7 +99,7 @@ vfloat64m1_t test_vfwmacc_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwmacc.mask.nxv1f64.f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwmacc.mask.nxv1f64.f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfwmacc_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -109,7 +109,7 @@ vfloat64m1_t test_vfwmacc_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwmacc.mask.nxv2f64.nxv2f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwmacc.mask.nxv2f64.nxv2f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfwmacc_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -120,7 +120,7 @@ vfloat64m2_t test_vfwmacc_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwmacc.mask.nxv2f64.f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwmacc.mask.nxv2f64.f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfwmacc_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -130,7 +130,7 @@ vfloat64m2_t test_vfwmacc_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwmacc.mask.nxv4f64.nxv4f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwmacc.mask.nxv4f64.nxv4f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfwmacc_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -141,7 +141,7 @@ vfloat64m4_t test_vfwmacc_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwmacc.mask.nxv4f64.f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwmacc.mask.nxv4f64.f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfwmacc_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -151,7 +151,7 @@ vfloat64m4_t test_vfwmacc_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwmacc.mask.nxv8f64.nxv8f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwmacc.mask.nxv8f64.nxv8f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfwmacc_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -162,7 +162,7 @@ vfloat64m8_t test_vfwmacc_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwmacc.mask.nxv8f64.f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwmacc.mask.nxv8f64.f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfwmacc_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc, float op1,
@@ -262,7 +262,7 @@ vfloat32m8_t test_vfwmacc_vf_f32m8 (vfloat32m8_t vd, _Float16 vs1, vfloat16m4_t
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vv_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfwmacc.mask.nxv1f32.nxv1f16.nxv1f16.i64(<vscale x 1 x float> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfwmacc.mask.nxv1f32.nxv1f16.nxv1f16.i64(<vscale x 1 x float> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfwmacc_vv_f32mf2_m (vbool64_t mask, vfloat32mf2_t vd, vfloat16mf4_t vs1, vfloat16mf4_t vs2, size_t vl) {
@@ -271,7 +271,7 @@ vfloat32mf2_t test_vfwmacc_vv_f32mf2_m (vbool64_t mask, vfloat32mf2_t vd, vfloat
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vf_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfwmacc.mask.nxv1f32.f16.nxv1f16.i64(<vscale x 1 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfwmacc.mask.nxv1f32.f16.nxv1f16.i64(<vscale x 1 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfwmacc_vf_f32mf2_m (vbool64_t mask, vfloat32mf2_t vd, _Float16 vs1, vfloat16mf4_t vs2, size_t vl) {
@@ -280,7 +280,7 @@ vfloat32mf2_t test_vfwmacc_vf_f32mf2_m (vbool64_t mask, vfloat32mf2_t vd, _Float
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vv_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfwmacc.mask.nxv2f32.nxv2f16.nxv2f16.i64(<vscale x 2 x float> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfwmacc.mask.nxv2f32.nxv2f16.nxv2f16.i64(<vscale x 2 x float> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfwmacc_vv_f32m1_m (vbool32_t mask, vfloat32m1_t vd, vfloat16mf2_t vs1, vfloat16mf2_t vs2, size_t vl) {
@@ -289,7 +289,7 @@ vfloat32m1_t test_vfwmacc_vv_f32m1_m (vbool32_t mask, vfloat32m1_t vd, vfloat16m
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vf_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfwmacc.mask.nxv2f32.f16.nxv2f16.i64(<vscale x 2 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfwmacc.mask.nxv2f32.f16.nxv2f16.i64(<vscale x 2 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfwmacc_vf_f32m1_m (vbool32_t mask, vfloat32m1_t vd, _Float16 vs1, vfloat16mf2_t vs2, size_t vl) {
@@ -298,7 +298,7 @@ vfloat32m1_t test_vfwmacc_vf_f32m1_m (vbool32_t mask, vfloat32m1_t vd, _Float16
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vv_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfwmacc.mask.nxv4f32.nxv4f16.nxv4f16.i64(<vscale x 4 x float> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfwmacc.mask.nxv4f32.nxv4f16.nxv4f16.i64(<vscale x 4 x float> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfwmacc_vv_f32m2_m (vbool16_t mask, vfloat32m2_t vd, vfloat16m1_t vs1, vfloat16m1_t vs2, size_t vl) {
@@ -307,7 +307,7 @@ vfloat32m2_t test_vfwmacc_vv_f32m2_m (vbool16_t mask, vfloat32m2_t vd, vfloat16m
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vf_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfwmacc.mask.nxv4f32.f16.nxv4f16.i64(<vscale x 4 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfwmacc.mask.nxv4f32.f16.nxv4f16.i64(<vscale x 4 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfwmacc_vf_f32m2_m (vbool16_t mask, vfloat32m2_t vd, _Float16 vs1, vfloat16m1_t vs2, size_t vl) {
@@ -316,7 +316,7 @@ vfloat32m2_t test_vfwmacc_vf_f32m2_m (vbool16_t mask, vfloat32m2_t vd, _Float16
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vv_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfwmacc.mask.nxv8f32.nxv8f16.nxv8f16.i64(<vscale x 8 x float> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfwmacc.mask.nxv8f32.nxv8f16.nxv8f16.i64(<vscale x 8 x float> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfwmacc_vv_f32m4_m (vbool8_t mask, vfloat32m4_t vd, vfloat16m2_t vs1, vfloat16m2_t vs2, size_t vl) {
@@ -325,7 +325,7 @@ vfloat32m4_t test_vfwmacc_vv_f32m4_m (vbool8_t mask, vfloat32m4_t vd, vfloat16m2
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vf_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfwmacc.mask.nxv8f32.f16.nxv8f16.i64(<vscale x 8 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfwmacc.mask.nxv8f32.f16.nxv8f16.i64(<vscale x 8 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfwmacc_vf_f32m4_m (vbool8_t mask, vfloat32m4_t vd, _Float16 vs1, vfloat16m2_t vs2, size_t vl) {
@@ -334,7 +334,7 @@ vfloat32m4_t test_vfwmacc_vf_f32m4_m (vbool8_t mask, vfloat32m4_t vd, _Float16 v
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vv_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfwmacc.mask.nxv16f32.nxv16f16.nxv16f16.i64(<vscale x 16 x float> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfwmacc.mask.nxv16f32.nxv16f16.nxv16f16.i64(<vscale x 16 x float> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfwmacc_vv_f32m8_m (vbool4_t mask, vfloat32m8_t vd, vfloat16m4_t vs1, vfloat16m4_t vs2, size_t vl) {
@@ -343,7 +343,7 @@ vfloat32m8_t test_vfwmacc_vv_f32m8_m (vbool4_t mask, vfloat32m8_t vd, vfloat16m4
 
 // CHECK-RV64-LABEL: @test_vfwmacc_vf_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfwmacc.mask.nxv16f32.f16.nxv16f16.i64(<vscale x 16 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfwmacc.mask.nxv16f32.f16.nxv16f16.i64(<vscale x 16 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfwmacc_vf_f32m8_m (vbool4_t mask, vfloat32m8_t vd, _Float16 vs1, vfloat16m4_t vs2, size_t vl) {
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfwmsac.c b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfwmsac.c
index f247017eb3ebc9..e6cb4861a4abc6 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfwmsac.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfwmsac.c
@@ -88,7 +88,7 @@ vfloat64m8_t test_vfwmsac_vf_f64m8(vfloat64m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwmsac.mask.nxv1f64.nxv1f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwmsac.mask.nxv1f64.nxv1f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfwmsac_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -99,7 +99,7 @@ vfloat64m1_t test_vfwmsac_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwmsac.mask.nxv1f64.f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwmsac.mask.nxv1f64.f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfwmsac_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -109,7 +109,7 @@ vfloat64m1_t test_vfwmsac_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwmsac.mask.nxv2f64.nxv2f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwmsac.mask.nxv2f64.nxv2f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfwmsac_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -120,7 +120,7 @@ vfloat64m2_t test_vfwmsac_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwmsac.mask.nxv2f64.f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwmsac.mask.nxv2f64.f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfwmsac_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -130,7 +130,7 @@ vfloat64m2_t test_vfwmsac_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwmsac.mask.nxv4f64.nxv4f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwmsac.mask.nxv4f64.nxv4f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfwmsac_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -141,7 +141,7 @@ vfloat64m4_t test_vfwmsac_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwmsac.mask.nxv4f64.f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwmsac.mask.nxv4f64.f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfwmsac_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -151,7 +151,7 @@ vfloat64m4_t test_vfwmsac_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwmsac.mask.nxv8f64.nxv8f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwmsac.mask.nxv8f64.nxv8f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfwmsac_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -162,7 +162,7 @@ vfloat64m8_t test_vfwmsac_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwmsac.mask.nxv8f64.f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwmsac.mask.nxv8f64.f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfwmsac_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc, float op1,
@@ -262,7 +262,7 @@ vfloat32m8_t test_vfwmsac_vf_f32m8 (vfloat32m8_t vd, _Float16 vs1, vfloat16m4_t
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vv_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfwmsac.mask.nxv1f32.nxv1f16.nxv1f16.i64(<vscale x 1 x float> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfwmsac.mask.nxv1f32.nxv1f16.nxv1f16.i64(<vscale x 1 x float> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfwmsac_vv_f32mf2_m (vbool64_t mask, vfloat32mf2_t vd, vfloat16mf4_t vs1, vfloat16mf4_t vs2, size_t vl) {
@@ -271,7 +271,7 @@ vfloat32mf2_t test_vfwmsac_vv_f32mf2_m (vbool64_t mask, vfloat32mf2_t vd, vfloat
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vf_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfwmsac.mask.nxv1f32.f16.nxv1f16.i64(<vscale x 1 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfwmsac.mask.nxv1f32.f16.nxv1f16.i64(<vscale x 1 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfwmsac_vf_f32mf2_m (vbool64_t mask, vfloat32mf2_t vd, _Float16 vs1, vfloat16mf4_t vs2, size_t vl) {
@@ -280,7 +280,7 @@ vfloat32mf2_t test_vfwmsac_vf_f32mf2_m (vbool64_t mask, vfloat32mf2_t vd, _Float
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vv_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfwmsac.mask.nxv2f32.nxv2f16.nxv2f16.i64(<vscale x 2 x float> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfwmsac.mask.nxv2f32.nxv2f16.nxv2f16.i64(<vscale x 2 x float> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfwmsac_vv_f32m1_m (vbool32_t mask, vfloat32m1_t vd, vfloat16mf2_t vs1, vfloat16mf2_t vs2, size_t vl) {
@@ -289,7 +289,7 @@ vfloat32m1_t test_vfwmsac_vv_f32m1_m (vbool32_t mask, vfloat32m1_t vd, vfloat16m
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vf_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfwmsac.mask.nxv2f32.f16.nxv2f16.i64(<vscale x 2 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfwmsac.mask.nxv2f32.f16.nxv2f16.i64(<vscale x 2 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfwmsac_vf_f32m1_m (vbool32_t mask, vfloat32m1_t vd, _Float16 vs1, vfloat16mf2_t vs2, size_t vl) {
@@ -298,7 +298,7 @@ vfloat32m1_t test_vfwmsac_vf_f32m1_m (vbool32_t mask, vfloat32m1_t vd, _Float16
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vv_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfwmsac.mask.nxv4f32.nxv4f16.nxv4f16.i64(<vscale x 4 x float> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfwmsac.mask.nxv4f32.nxv4f16.nxv4f16.i64(<vscale x 4 x float> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfwmsac_vv_f32m2_m (vbool16_t mask, vfloat32m2_t vd, vfloat16m1_t vs1, vfloat16m1_t vs2, size_t vl) {
@@ -307,7 +307,7 @@ vfloat32m2_t test_vfwmsac_vv_f32m2_m (vbool16_t mask, vfloat32m2_t vd, vfloat16m
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vf_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfwmsac.mask.nxv4f32.f16.nxv4f16.i64(<vscale x 4 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfwmsac.mask.nxv4f32.f16.nxv4f16.i64(<vscale x 4 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfwmsac_vf_f32m2_m (vbool16_t mask, vfloat32m2_t vd, _Float16 vs1, vfloat16m1_t vs2, size_t vl) {
@@ -316,7 +316,7 @@ vfloat32m2_t test_vfwmsac_vf_f32m2_m (vbool16_t mask, vfloat32m2_t vd, _Float16
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vv_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfwmsac.mask.nxv8f32.nxv8f16.nxv8f16.i64(<vscale x 8 x float> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfwmsac.mask.nxv8f32.nxv8f16.nxv8f16.i64(<vscale x 8 x float> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfwmsac_vv_f32m4_m (vbool8_t mask, vfloat32m4_t vd, vfloat16m2_t vs1, vfloat16m2_t vs2, size_t vl) {
@@ -325,7 +325,7 @@ vfloat32m4_t test_vfwmsac_vv_f32m4_m (vbool8_t mask, vfloat32m4_t vd, vfloat16m2
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vf_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfwmsac.mask.nxv8f32.f16.nxv8f16.i64(<vscale x 8 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfwmsac.mask.nxv8f32.f16.nxv8f16.i64(<vscale x 8 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfwmsac_vf_f32m4_m (vbool8_t mask, vfloat32m4_t vd, _Float16 vs1, vfloat16m2_t vs2, size_t vl) {
@@ -334,7 +334,7 @@ vfloat32m4_t test_vfwmsac_vf_f32m4_m (vbool8_t mask, vfloat32m4_t vd, _Float16 v
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vv_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfwmsac.mask.nxv16f32.nxv16f16.nxv16f16.i64(<vscale x 16 x float> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfwmsac.mask.nxv16f32.nxv16f16.nxv16f16.i64(<vscale x 16 x float> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfwmsac_vv_f32m8_m (vbool4_t mask, vfloat32m8_t vd, vfloat16m4_t vs1, vfloat16m4_t vs2, size_t vl) {
@@ -343,7 +343,7 @@ vfloat32m8_t test_vfwmsac_vv_f32m8_m (vbool4_t mask, vfloat32m8_t vd, vfloat16m4
 
 // CHECK-RV64-LABEL: @test_vfwmsac_vf_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfwmsac.mask.nxv16f32.f16.nxv16f16.i64(<vscale x 16 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfwmsac.mask.nxv16f32.f16.nxv16f16.i64(<vscale x 16 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfwmsac_vf_f32m8_m (vbool4_t mask, vfloat32m8_t vd, _Float16 vs1, vfloat16m4_t vs2, size_t vl) {
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfwnmacc.c b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfwnmacc.c
index f9d4cb6b60b84e..cff04ff4bd30d8 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfwnmacc.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfwnmacc.c
@@ -88,7 +88,7 @@ vfloat64m8_t test_vfwnmacc_vf_f64m8(vfloat64m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwnmacc.mask.nxv1f64.nxv1f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwnmacc.mask.nxv1f64.nxv1f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfwnmacc_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -99,7 +99,7 @@ vfloat64m1_t test_vfwnmacc_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwnmacc.mask.nxv1f64.f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwnmacc.mask.nxv1f64.f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfwnmacc_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -109,7 +109,7 @@ vfloat64m1_t test_vfwnmacc_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwnmacc.mask.nxv2f64.nxv2f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwnmacc.mask.nxv2f64.nxv2f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfwnmacc_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -120,7 +120,7 @@ vfloat64m2_t test_vfwnmacc_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwnmacc.mask.nxv2f64.f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwnmacc.mask.nxv2f64.f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfwnmacc_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -130,7 +130,7 @@ vfloat64m2_t test_vfwnmacc_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwnmacc.mask.nxv4f64.nxv4f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwnmacc.mask.nxv4f64.nxv4f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfwnmacc_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -141,7 +141,7 @@ vfloat64m4_t test_vfwnmacc_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwnmacc.mask.nxv4f64.f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwnmacc.mask.nxv4f64.f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfwnmacc_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -151,7 +151,7 @@ vfloat64m4_t test_vfwnmacc_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwnmacc.mask.nxv8f64.nxv8f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwnmacc.mask.nxv8f64.nxv8f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfwnmacc_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -162,7 +162,7 @@ vfloat64m8_t test_vfwnmacc_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwnmacc.mask.nxv8f64.f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwnmacc.mask.nxv8f64.f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfwnmacc_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -262,7 +262,7 @@ vfloat32m8_t test_vfwnmacc_vf_f32m8 (vfloat32m8_t vd, _Float16 vs1, vfloat16m4_t
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vv_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfwnmacc.mask.nxv1f32.nxv1f16.nxv1f16.i64(<vscale x 1 x float> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfwnmacc.mask.nxv1f32.nxv1f16.nxv1f16.i64(<vscale x 1 x float> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfwnmacc_vv_f32mf2_m (vbool64_t mask, vfloat32mf2_t vd, vfloat16mf4_t vs1, vfloat16mf4_t vs2, size_t vl) {
@@ -271,7 +271,7 @@ vfloat32mf2_t test_vfwnmacc_vv_f32mf2_m (vbool64_t mask, vfloat32mf2_t vd, vfloa
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vf_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfwnmacc.mask.nxv1f32.f16.nxv1f16.i64(<vscale x 1 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfwnmacc.mask.nxv1f32.f16.nxv1f16.i64(<vscale x 1 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfwnmacc_vf_f32mf2_m (vbool64_t mask, vfloat32mf2_t vd, _Float16 vs1, vfloat16mf4_t vs2, size_t vl) {
@@ -280,7 +280,7 @@ vfloat32mf2_t test_vfwnmacc_vf_f32mf2_m (vbool64_t mask, vfloat32mf2_t vd, _Floa
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vv_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfwnmacc.mask.nxv2f32.nxv2f16.nxv2f16.i64(<vscale x 2 x float> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfwnmacc.mask.nxv2f32.nxv2f16.nxv2f16.i64(<vscale x 2 x float> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfwnmacc_vv_f32m1_m (vbool32_t mask, vfloat32m1_t vd, vfloat16mf2_t vs1, vfloat16mf2_t vs2, size_t vl) {
@@ -289,7 +289,7 @@ vfloat32m1_t test_vfwnmacc_vv_f32m1_m (vbool32_t mask, vfloat32m1_t vd, vfloat16
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vf_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfwnmacc.mask.nxv2f32.f16.nxv2f16.i64(<vscale x 2 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfwnmacc.mask.nxv2f32.f16.nxv2f16.i64(<vscale x 2 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfwnmacc_vf_f32m1_m (vbool32_t mask, vfloat32m1_t vd, _Float16 vs1, vfloat16mf2_t vs2, size_t vl) {
@@ -298,7 +298,7 @@ vfloat32m1_t test_vfwnmacc_vf_f32m1_m (vbool32_t mask, vfloat32m1_t vd, _Float16
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vv_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfwnmacc.mask.nxv4f32.nxv4f16.nxv4f16.i64(<vscale x 4 x float> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfwnmacc.mask.nxv4f32.nxv4f16.nxv4f16.i64(<vscale x 4 x float> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfwnmacc_vv_f32m2_m (vbool16_t mask, vfloat32m2_t vd, vfloat16m1_t vs1, vfloat16m1_t vs2, size_t vl) {
@@ -307,7 +307,7 @@ vfloat32m2_t test_vfwnmacc_vv_f32m2_m (vbool16_t mask, vfloat32m2_t vd, vfloat16
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vf_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfwnmacc.mask.nxv4f32.f16.nxv4f16.i64(<vscale x 4 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfwnmacc.mask.nxv4f32.f16.nxv4f16.i64(<vscale x 4 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfwnmacc_vf_f32m2_m (vbool16_t mask, vfloat32m2_t vd, _Float16 vs1, vfloat16m1_t vs2, size_t vl) {
@@ -316,7 +316,7 @@ vfloat32m2_t test_vfwnmacc_vf_f32m2_m (vbool16_t mask, vfloat32m2_t vd, _Float16
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vv_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfwnmacc.mask.nxv8f32.nxv8f16.nxv8f16.i64(<vscale x 8 x float> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfwnmacc.mask.nxv8f32.nxv8f16.nxv8f16.i64(<vscale x 8 x float> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfwnmacc_vv_f32m4_m (vbool8_t mask, vfloat32m4_t vd, vfloat16m2_t vs1, vfloat16m2_t vs2, size_t vl) {
@@ -325,7 +325,7 @@ vfloat32m4_t test_vfwnmacc_vv_f32m4_m (vbool8_t mask, vfloat32m4_t vd, vfloat16m
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vf_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfwnmacc.mask.nxv8f32.f16.nxv8f16.i64(<vscale x 8 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfwnmacc.mask.nxv8f32.f16.nxv8f16.i64(<vscale x 8 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfwnmacc_vf_f32m4_m (vbool8_t mask, vfloat32m4_t vd, _Float16 vs1, vfloat16m2_t vs2, size_t vl) {
@@ -334,7 +334,7 @@ vfloat32m4_t test_vfwnmacc_vf_f32m4_m (vbool8_t mask, vfloat32m4_t vd, _Float16
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vv_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfwnmacc.mask.nxv16f32.nxv16f16.nxv16f16.i64(<vscale x 16 x float> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfwnmacc.mask.nxv16f32.nxv16f16.nxv16f16.i64(<vscale x 16 x float> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfwnmacc_vv_f32m8_m (vbool4_t mask, vfloat32m8_t vd, vfloat16m4_t vs1, vfloat16m4_t vs2, size_t vl) {
@@ -343,7 +343,7 @@ vfloat32m8_t test_vfwnmacc_vv_f32m8_m (vbool4_t mask, vfloat32m8_t vd, vfloat16m
 
 // CHECK-RV64-LABEL: @test_vfwnmacc_vf_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfwnmacc.mask.nxv16f32.f16.nxv16f16.i64(<vscale x 16 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfwnmacc.mask.nxv16f32.f16.nxv16f16.i64(<vscale x 16 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfwnmacc_vf_f32m8_m (vbool4_t mask, vfloat32m8_t vd, _Float16 vs1, vfloat16m4_t vs2, size_t vl) {
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfwnmsac.c b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfwnmsac.c
index afe3e32691febb..d4a8239312bbd5 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics/vfwnmsac.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics/vfwnmsac.c
@@ -88,7 +88,7 @@ vfloat64m8_t test_vfwnmsac_vf_f64m8(vfloat64m8_t acc, float op1,
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vv_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwnmsac.mask.nxv1f64.nxv1f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwnmsac.mask.nxv1f64.nxv1f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], <vscale x 1 x float> [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfwnmsac_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -99,7 +99,7 @@ vfloat64m1_t test_vfwnmsac_vv_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vf_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwnmsac.mask.nxv1f64.f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vfwnmsac.mask.nxv1f64.f32.nxv1f32.i64(<vscale x 1 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 1 x float> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vfwnmsac_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
@@ -109,7 +109,7 @@ vfloat64m1_t test_vfwnmsac_vf_f64m1_m(vbool64_t mask, vfloat64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vv_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwnmsac.mask.nxv2f64.nxv2f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwnmsac.mask.nxv2f64.nxv2f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], <vscale x 2 x float> [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfwnmsac_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -120,7 +120,7 @@ vfloat64m2_t test_vfwnmsac_vv_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vf_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwnmsac.mask.nxv2f64.f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vfwnmsac.mask.nxv2f64.f32.nxv2f32.i64(<vscale x 2 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 2 x float> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vfwnmsac_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
@@ -130,7 +130,7 @@ vfloat64m2_t test_vfwnmsac_vf_f64m2_m(vbool32_t mask, vfloat64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vv_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwnmsac.mask.nxv4f64.nxv4f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwnmsac.mask.nxv4f64.nxv4f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], <vscale x 4 x float> [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfwnmsac_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -141,7 +141,7 @@ vfloat64m4_t test_vfwnmsac_vv_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vf_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwnmsac.mask.nxv4f64.f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vfwnmsac.mask.nxv4f64.f32.nxv4f32.i64(<vscale x 4 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 4 x float> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vfwnmsac_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
@@ -151,7 +151,7 @@ vfloat64m4_t test_vfwnmsac_vf_f64m4_m(vbool16_t mask, vfloat64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vv_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwnmsac.mask.nxv8f64.nxv8f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwnmsac.mask.nxv8f64.nxv8f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], <vscale x 8 x float> [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfwnmsac_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -162,7 +162,7 @@ vfloat64m8_t test_vfwnmsac_vv_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vf_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwnmsac.mask.nxv8f64.f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vfwnmsac.mask.nxv8f64.f32.nxv8f32.i64(<vscale x 8 x double> [[ACC:%.*]], float [[OP1:%.*]], <vscale x 8 x float> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vfwnmsac_vf_f64m8_m(vbool8_t mask, vfloat64m8_t acc,
@@ -262,7 +262,7 @@ vfloat32m8_t test_vfwnmsac_vf_f32m8 (vfloat32m8_t vd, _Float16 vs1, vfloat16m4_t
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vv_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfwnmsac.mask.nxv1f32.nxv1f16.nxv1f16.i64(<vscale x 1 x float> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfwnmsac.mask.nxv1f32.nxv1f16.nxv1f16.i64(<vscale x 1 x float> [[VD:%.*]], <vscale x 1 x half> [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfwnmsac_vv_f32mf2_m (vbool64_t mask, vfloat32mf2_t vd, vfloat16mf4_t vs1, vfloat16mf4_t vs2, size_t vl) {
@@ -271,7 +271,7 @@ vfloat32mf2_t test_vfwnmsac_vv_f32mf2_m (vbool64_t mask, vfloat32mf2_t vd, vfloa
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vf_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfwnmsac.mask.nxv1f32.f16.nxv1f16.i64(<vscale x 1 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vfwnmsac.mask.nxv1f32.f16.nxv1f16.i64(<vscale x 1 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 1 x half> [[VS2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vfwnmsac_vf_f32mf2_m (vbool64_t mask, vfloat32mf2_t vd, _Float16 vs1, vfloat16mf4_t vs2, size_t vl) {
@@ -280,7 +280,7 @@ vfloat32mf2_t test_vfwnmsac_vf_f32mf2_m (vbool64_t mask, vfloat32mf2_t vd, _Floa
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vv_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfwnmsac.mask.nxv2f32.nxv2f16.nxv2f16.i64(<vscale x 2 x float> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfwnmsac.mask.nxv2f32.nxv2f16.nxv2f16.i64(<vscale x 2 x float> [[VD:%.*]], <vscale x 2 x half> [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfwnmsac_vv_f32m1_m (vbool32_t mask, vfloat32m1_t vd, vfloat16mf2_t vs1, vfloat16mf2_t vs2, size_t vl) {
@@ -289,7 +289,7 @@ vfloat32m1_t test_vfwnmsac_vv_f32m1_m (vbool32_t mask, vfloat32m1_t vd, vfloat16
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vf_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfwnmsac.mask.nxv2f32.f16.nxv2f16.i64(<vscale x 2 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vfwnmsac.mask.nxv2f32.f16.nxv2f16.i64(<vscale x 2 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 2 x half> [[VS2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vfwnmsac_vf_f32m1_m (vbool32_t mask, vfloat32m1_t vd, _Float16 vs1, vfloat16mf2_t vs2, size_t vl) {
@@ -298,7 +298,7 @@ vfloat32m1_t test_vfwnmsac_vf_f32m1_m (vbool32_t mask, vfloat32m1_t vd, _Float16
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vv_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfwnmsac.mask.nxv4f32.nxv4f16.nxv4f16.i64(<vscale x 4 x float> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfwnmsac.mask.nxv4f32.nxv4f16.nxv4f16.i64(<vscale x 4 x float> [[VD:%.*]], <vscale x 4 x half> [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfwnmsac_vv_f32m2_m (vbool16_t mask, vfloat32m2_t vd, vfloat16m1_t vs1, vfloat16m1_t vs2, size_t vl) {
@@ -307,7 +307,7 @@ vfloat32m2_t test_vfwnmsac_vv_f32m2_m (vbool16_t mask, vfloat32m2_t vd, vfloat16
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vf_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfwnmsac.mask.nxv4f32.f16.nxv4f16.i64(<vscale x 4 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vfwnmsac.mask.nxv4f32.f16.nxv4f16.i64(<vscale x 4 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 4 x half> [[VS2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vfwnmsac_vf_f32m2_m (vbool16_t mask, vfloat32m2_t vd, _Float16 vs1, vfloat16m1_t vs2, size_t vl) {
@@ -316,7 +316,7 @@ vfloat32m2_t test_vfwnmsac_vf_f32m2_m (vbool16_t mask, vfloat32m2_t vd, _Float16
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vv_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfwnmsac.mask.nxv8f32.nxv8f16.nxv8f16.i64(<vscale x 8 x float> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfwnmsac.mask.nxv8f32.nxv8f16.nxv8f16.i64(<vscale x 8 x float> [[VD:%.*]], <vscale x 8 x half> [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfwnmsac_vv_f32m4_m (vbool8_t mask, vfloat32m4_t vd, vfloat16m2_t vs1, vfloat16m2_t vs2, size_t vl) {
@@ -325,7 +325,7 @@ vfloat32m4_t test_vfwnmsac_vv_f32m4_m (vbool8_t mask, vfloat32m4_t vd, vfloat16m
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vf_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfwnmsac.mask.nxv8f32.f16.nxv8f16.i64(<vscale x 8 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vfwnmsac.mask.nxv8f32.f16.nxv8f16.i64(<vscale x 8 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 8 x half> [[VS2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vfwnmsac_vf_f32m4_m (vbool8_t mask, vfloat32m4_t vd, _Float16 vs1, vfloat16m2_t vs2, size_t vl) {
@@ -334,7 +334,7 @@ vfloat32m4_t test_vfwnmsac_vf_f32m4_m (vbool8_t mask, vfloat32m4_t vd, _Float16
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vv_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfwnmsac.mask.nxv16f32.nxv16f16.nxv16f16.i64(<vscale x 16 x float> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfwnmsac.mask.nxv16f32.nxv16f16.nxv16f16.i64(<vscale x 16 x float> [[VD:%.*]], <vscale x 16 x half> [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfwnmsac_vv_f32m8_m (vbool4_t mask, vfloat32m8_t vd, vfloat16m4_t vs1, vfloat16m4_t vs2, size_t vl) {
@@ -343,7 +343,7 @@ vfloat32m8_t test_vfwnmsac_vv_f32m8_m (vbool4_t mask, vfloat32m8_t vd, vfloat16m
 
 // CHECK-RV64-LABEL: @test_vfwnmsac_vf_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfwnmsac.mask.nxv16f32.f16.nxv16f16.i64(<vscale x 16 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vfwnmsac.mask.nxv16f32.f16.nxv16f16.i64(<vscale x 16 x float> [[VD:%.*]], half [[VS1:%.*]], <vscale x 16 x half> [[VS2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vfwnmsac_vf_f32m8_m (vbool4_t mask, vfloat32m8_t vd, _Float16 vs1, vfloat16m4_t vs2, size_t vl) {
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics/vmacc.c b/clang/test/CodeGen/RISCV/rvv-intrinsics/vmacc.c
index 478a67a16a5a0f..243567cfab13fe 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics/vmacc.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics/vmacc.c
@@ -798,7 +798,7 @@ vuint64m8_t test_vmacc_vx_u64m8(vuint64m8_t acc, uint64_t op1, vuint64m8_t op2,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vint8mf8_t test_vmacc_vv_i8mf8_m(vbool64_t mask, vint8mf8_t acc, vint8mf8_t op1, vint8mf8_t op2, size_t vl) {
@@ -807,7 +807,7 @@ vint8mf8_t test_vmacc_vv_i8mf8_m(vbool64_t mask, vint8mf8_t acc, vint8mf8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vint8mf8_t test_vmacc_vx_i8mf8_m(vbool64_t mask, vint8mf8_t acc, int8_t op1, vint8mf8_t op2, size_t vl) {
@@ -816,7 +816,7 @@ vint8mf8_t test_vmacc_vx_i8mf8_m(vbool64_t mask, vint8mf8_t acc, int8_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmacc.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmacc.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vint8mf4_t test_vmacc_vv_i8mf4_m(vbool32_t mask, vint8mf4_t acc, vint8mf4_t op1, vint8mf4_t op2, size_t vl) {
@@ -825,7 +825,7 @@ vint8mf4_t test_vmacc_vv_i8mf4_m(vbool32_t mask, vint8mf4_t acc, vint8mf4_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmacc.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmacc.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vint8mf4_t test_vmacc_vx_i8mf4_m(vbool32_t mask, vint8mf4_t acc, int8_t op1, vint8mf4_t op2, size_t vl) {
@@ -834,7 +834,7 @@ vint8mf4_t test_vmacc_vx_i8mf4_m(vbool32_t mask, vint8mf4_t acc, int8_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmacc.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmacc.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vint8mf2_t test_vmacc_vv_i8mf2_m(vbool16_t mask, vint8mf2_t acc, vint8mf2_t op1, vint8mf2_t op2, size_t vl) {
@@ -843,7 +843,7 @@ vint8mf2_t test_vmacc_vv_i8mf2_m(vbool16_t mask, vint8mf2_t acc, vint8mf2_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmacc.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmacc.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vint8mf2_t test_vmacc_vx_i8mf2_m(vbool16_t mask, vint8mf2_t acc, int8_t op1, vint8mf2_t op2, size_t vl) {
@@ -852,7 +852,7 @@ vint8mf2_t test_vmacc_vx_i8mf2_m(vbool16_t mask, vint8mf2_t acc, int8_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmacc.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmacc.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vint8m1_t test_vmacc_vv_i8m1_m(vbool8_t mask, vint8m1_t acc, vint8m1_t op1, vint8m1_t op2, size_t vl) {
@@ -861,7 +861,7 @@ vint8m1_t test_vmacc_vv_i8m1_m(vbool8_t mask, vint8m1_t acc, vint8m1_t op1, vint
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmacc.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmacc.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vint8m1_t test_vmacc_vx_i8m1_m(vbool8_t mask, vint8m1_t acc, int8_t op1, vint8m1_t op2, size_t vl) {
@@ -870,7 +870,7 @@ vint8m1_t test_vmacc_vx_i8m1_m(vbool8_t mask, vint8m1_t acc, int8_t op1, vint8m1
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmacc.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmacc.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vint8m2_t test_vmacc_vv_i8m2_m(vbool4_t mask, vint8m2_t acc, vint8m2_t op1, vint8m2_t op2, size_t vl) {
@@ -879,7 +879,7 @@ vint8m2_t test_vmacc_vv_i8m2_m(vbool4_t mask, vint8m2_t acc, vint8m2_t op1, vint
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmacc.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmacc.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vint8m2_t test_vmacc_vx_i8m2_m(vbool4_t mask, vint8m2_t acc, int8_t op1, vint8m2_t op2, size_t vl) {
@@ -888,7 +888,7 @@ vint8m2_t test_vmacc_vx_i8m2_m(vbool4_t mask, vint8m2_t acc, int8_t op1, vint8m2
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmacc.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmacc.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vint8m4_t test_vmacc_vv_i8m4_m(vbool2_t mask, vint8m4_t acc, vint8m4_t op1, vint8m4_t op2, size_t vl) {
@@ -897,7 +897,7 @@ vint8m4_t test_vmacc_vv_i8m4_m(vbool2_t mask, vint8m4_t acc, vint8m4_t op1, vint
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmacc.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmacc.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vint8m4_t test_vmacc_vx_i8m4_m(vbool2_t mask, vint8m4_t acc, int8_t op1, vint8m4_t op2, size_t vl) {
@@ -906,7 +906,7 @@ vint8m4_t test_vmacc_vx_i8m4_m(vbool2_t mask, vint8m4_t acc, int8_t op1, vint8m4
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmacc.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmacc.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vint8m8_t test_vmacc_vv_i8m8_m(vbool1_t mask, vint8m8_t acc, vint8m8_t op1, vint8m8_t op2, size_t vl) {
@@ -915,7 +915,7 @@ vint8m8_t test_vmacc_vv_i8m8_m(vbool1_t mask, vint8m8_t acc, vint8m8_t op1, vint
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmacc.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmacc.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vint8m8_t test_vmacc_vx_i8m8_m(vbool1_t mask, vint8m8_t acc, int8_t op1, vint8m8_t op2, size_t vl) {
@@ -924,7 +924,7 @@ vint8m8_t test_vmacc_vx_i8m8_m(vbool1_t mask, vint8m8_t acc, int8_t op1, vint8m8
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmacc.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmacc.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vmacc_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc, vint16mf4_t op1, vint16mf4_t op2, size_t vl) {
@@ -933,7 +933,7 @@ vint16mf4_t test_vmacc_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc, vint16mf4_t
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmacc.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmacc.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vmacc_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc, int16_t op1, vint16mf4_t op2, size_t vl) {
@@ -942,7 +942,7 @@ vint16mf4_t test_vmacc_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmacc.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmacc.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vmacc_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc, vint16mf2_t op1, vint16mf2_t op2, size_t vl) {
@@ -951,7 +951,7 @@ vint16mf2_t test_vmacc_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc, vint16mf2_t
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmacc.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmacc.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vmacc_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc, int16_t op1, vint16mf2_t op2, size_t vl) {
@@ -960,7 +960,7 @@ vint16mf2_t test_vmacc_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmacc.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmacc.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vmacc_vv_i16m1_m(vbool16_t mask, vint16m1_t acc, vint16m1_t op1, vint16m1_t op2, size_t vl) {
@@ -969,7 +969,7 @@ vint16m1_t test_vmacc_vv_i16m1_m(vbool16_t mask, vint16m1_t acc, vint16m1_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmacc.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmacc.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vmacc_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int16_t op1, vint16m1_t op2, size_t vl) {
@@ -978,7 +978,7 @@ vint16m1_t test_vmacc_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int16_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmacc.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmacc.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vmacc_vv_i16m2_m(vbool8_t mask, vint16m2_t acc, vint16m2_t op1, vint16m2_t op2, size_t vl) {
@@ -987,7 +987,7 @@ vint16m2_t test_vmacc_vv_i16m2_m(vbool8_t mask, vint16m2_t acc, vint16m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmacc.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmacc.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vmacc_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int16_t op1, vint16m2_t op2, size_t vl) {
@@ -996,7 +996,7 @@ vint16m2_t test_vmacc_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int16_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmacc.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmacc.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vmacc_vv_i16m4_m(vbool4_t mask, vint16m4_t acc, vint16m4_t op1, vint16m4_t op2, size_t vl) {
@@ -1005,7 +1005,7 @@ vint16m4_t test_vmacc_vv_i16m4_m(vbool4_t mask, vint16m4_t acc, vint16m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmacc.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmacc.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vmacc_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int16_t op1, vint16m4_t op2, size_t vl) {
@@ -1014,7 +1014,7 @@ vint16m4_t test_vmacc_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int16_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmacc.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmacc.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vmacc_vv_i16m8_m(vbool2_t mask, vint16m8_t acc, vint16m8_t op1, vint16m8_t op2, size_t vl) {
@@ -1023,7 +1023,7 @@ vint16m8_t test_vmacc_vv_i16m8_m(vbool2_t mask, vint16m8_t acc, vint16m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmacc.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmacc.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vmacc_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int16_t op1, vint16m8_t op2, size_t vl) {
@@ -1032,7 +1032,7 @@ vint16m8_t test_vmacc_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int16_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmacc.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmacc.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vmacc_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc, vint32mf2_t op1, vint32mf2_t op2, size_t vl) {
@@ -1041,7 +1041,7 @@ vint32mf2_t test_vmacc_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc, vint32mf2_t
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmacc.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmacc.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vmacc_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc, int32_t op1, vint32mf2_t op2, size_t vl) {
@@ -1050,7 +1050,7 @@ vint32mf2_t test_vmacc_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc, int32_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmacc.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmacc.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vmacc_vv_i32m1_m(vbool32_t mask, vint32m1_t acc, vint32m1_t op1, vint32m1_t op2, size_t vl) {
@@ -1059,7 +1059,7 @@ vint32m1_t test_vmacc_vv_i32m1_m(vbool32_t mask, vint32m1_t acc, vint32m1_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmacc.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmacc.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vmacc_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int32_t op1, vint32m1_t op2, size_t vl) {
@@ -1068,7 +1068,7 @@ vint32m1_t test_vmacc_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int32_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmacc.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmacc.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vmacc_vv_i32m2_m(vbool16_t mask, vint32m2_t acc, vint32m2_t op1, vint32m2_t op2, size_t vl) {
@@ -1077,7 +1077,7 @@ vint32m2_t test_vmacc_vv_i32m2_m(vbool16_t mask, vint32m2_t acc, vint32m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmacc.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmacc.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vmacc_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int32_t op1, vint32m2_t op2, size_t vl) {
@@ -1086,7 +1086,7 @@ vint32m2_t test_vmacc_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int32_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmacc.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmacc.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vmacc_vv_i32m4_m(vbool8_t mask, vint32m4_t acc, vint32m4_t op1, vint32m4_t op2, size_t vl) {
@@ -1095,7 +1095,7 @@ vint32m4_t test_vmacc_vv_i32m4_m(vbool8_t mask, vint32m4_t acc, vint32m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmacc.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmacc.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vmacc_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int32_t op1, vint32m4_t op2, size_t vl) {
@@ -1104,7 +1104,7 @@ vint32m4_t test_vmacc_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int32_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmacc.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmacc.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vmacc_vv_i32m8_m(vbool4_t mask, vint32m8_t acc, vint32m8_t op1, vint32m8_t op2, size_t vl) {
@@ -1113,7 +1113,7 @@ vint32m8_t test_vmacc_vv_i32m8_m(vbool4_t mask, vint32m8_t acc, vint32m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmacc.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmacc.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vmacc_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int32_t op1, vint32m8_t op2, size_t vl) {
@@ -1122,7 +1122,7 @@ vint32m8_t test_vmacc_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int32_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmacc.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmacc.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vmacc_vv_i64m1_m(vbool64_t mask, vint64m1_t acc, vint64m1_t op1, vint64m1_t op2, size_t vl) {
@@ -1131,7 +1131,7 @@ vint64m1_t test_vmacc_vv_i64m1_m(vbool64_t mask, vint64m1_t acc, vint64m1_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmacc.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmacc.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vmacc_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int64_t op1, vint64m1_t op2, size_t vl) {
@@ -1140,7 +1140,7 @@ vint64m1_t test_vmacc_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int64_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmacc.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmacc.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vmacc_vv_i64m2_m(vbool32_t mask, vint64m2_t acc, vint64m2_t op1, vint64m2_t op2, size_t vl) {
@@ -1149,7 +1149,7 @@ vint64m2_t test_vmacc_vv_i64m2_m(vbool32_t mask, vint64m2_t acc, vint64m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmacc.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmacc.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vmacc_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int64_t op1, vint64m2_t op2, size_t vl) {
@@ -1158,7 +1158,7 @@ vint64m2_t test_vmacc_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int64_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmacc.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmacc.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vmacc_vv_i64m4_m(vbool16_t mask, vint64m4_t acc, vint64m4_t op1, vint64m4_t op2, size_t vl) {
@@ -1167,7 +1167,7 @@ vint64m4_t test_vmacc_vv_i64m4_m(vbool16_t mask, vint64m4_t acc, vint64m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmacc.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmacc.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vmacc_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int64_t op1, vint64m4_t op2, size_t vl) {
@@ -1176,7 +1176,7 @@ vint64m4_t test_vmacc_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int64_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmacc.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmacc.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vmacc_vv_i64m8_m(vbool8_t mask, vint64m8_t acc, vint64m8_t op1, vint64m8_t op2, size_t vl) {
@@ -1185,7 +1185,7 @@ vint64m8_t test_vmacc_vv_i64m8_m(vbool8_t mask, vint64m8_t acc, vint64m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmacc.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmacc.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vmacc_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int64_t op1, vint64m8_t op2, size_t vl) {
@@ -1194,7 +1194,7 @@ vint64m8_t test_vmacc_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int64_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vuint8mf8_t test_vmacc_vv_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, vuint8mf8_t op1, vuint8mf8_t op2, size_t vl) {
@@ -1203,7 +1203,7 @@ vuint8mf8_t test_vmacc_vv_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, vuint8mf8_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vuint8mf8_t test_vmacc_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, uint8_t op1, vuint8mf8_t op2, size_t vl) {
@@ -1212,7 +1212,7 @@ vuint8mf8_t test_vmacc_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmacc.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmacc.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vuint8mf4_t test_vmacc_vv_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, vuint8mf4_t op1, vuint8mf4_t op2, size_t vl) {
@@ -1221,7 +1221,7 @@ vuint8mf4_t test_vmacc_vv_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, vuint8mf4_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmacc.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmacc.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vuint8mf4_t test_vmacc_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, uint8_t op1, vuint8mf4_t op2, size_t vl) {
@@ -1230,7 +1230,7 @@ vuint8mf4_t test_vmacc_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmacc.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmacc.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vuint8mf2_t test_vmacc_vv_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, vuint8mf2_t op1, vuint8mf2_t op2, size_t vl) {
@@ -1239,7 +1239,7 @@ vuint8mf2_t test_vmacc_vv_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, vuint8mf2_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmacc.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmacc.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vuint8mf2_t test_vmacc_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, uint8_t op1, vuint8mf2_t op2, size_t vl) {
@@ -1248,7 +1248,7 @@ vuint8mf2_t test_vmacc_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmacc.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmacc.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vuint8m1_t test_vmacc_vv_u8m1_m(vbool8_t mask, vuint8m1_t acc, vuint8m1_t op1, vuint8m1_t op2, size_t vl) {
@@ -1257,7 +1257,7 @@ vuint8m1_t test_vmacc_vv_u8m1_m(vbool8_t mask, vuint8m1_t acc, vuint8m1_t op1, v
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmacc.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmacc.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vuint8m1_t test_vmacc_vx_u8m1_m(vbool8_t mask, vuint8m1_t acc, uint8_t op1, vuint8m1_t op2, size_t vl) {
@@ -1266,7 +1266,7 @@ vuint8m1_t test_vmacc_vx_u8m1_m(vbool8_t mask, vuint8m1_t acc, uint8_t op1, vuin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmacc.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmacc.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vuint8m2_t test_vmacc_vv_u8m2_m(vbool4_t mask, vuint8m2_t acc, vuint8m2_t op1, vuint8m2_t op2, size_t vl) {
@@ -1275,7 +1275,7 @@ vuint8m2_t test_vmacc_vv_u8m2_m(vbool4_t mask, vuint8m2_t acc, vuint8m2_t op1, v
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmacc.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmacc.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vuint8m2_t test_vmacc_vx_u8m2_m(vbool4_t mask, vuint8m2_t acc, uint8_t op1, vuint8m2_t op2, size_t vl) {
@@ -1284,7 +1284,7 @@ vuint8m2_t test_vmacc_vx_u8m2_m(vbool4_t mask, vuint8m2_t acc, uint8_t op1, vuin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmacc.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmacc.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vuint8m4_t test_vmacc_vv_u8m4_m(vbool2_t mask, vuint8m4_t acc, vuint8m4_t op1, vuint8m4_t op2, size_t vl) {
@@ -1293,7 +1293,7 @@ vuint8m4_t test_vmacc_vv_u8m4_m(vbool2_t mask, vuint8m4_t acc, vuint8m4_t op1, v
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmacc.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmacc.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vuint8m4_t test_vmacc_vx_u8m4_m(vbool2_t mask, vuint8m4_t acc, uint8_t op1, vuint8m4_t op2, size_t vl) {
@@ -1302,7 +1302,7 @@ vuint8m4_t test_vmacc_vx_u8m4_m(vbool2_t mask, vuint8m4_t acc, uint8_t op1, vuin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmacc.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmacc.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vuint8m8_t test_vmacc_vv_u8m8_m(vbool1_t mask, vuint8m8_t acc, vuint8m8_t op1, vuint8m8_t op2, size_t vl) {
@@ -1311,7 +1311,7 @@ vuint8m8_t test_vmacc_vv_u8m8_m(vbool1_t mask, vuint8m8_t acc, vuint8m8_t op1, v
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmacc.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmacc.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vuint8m8_t test_vmacc_vx_u8m8_m(vbool1_t mask, vuint8m8_t acc, uint8_t op1, vuint8m8_t op2, size_t vl) {
@@ -1320,7 +1320,7 @@ vuint8m8_t test_vmacc_vx_u8m8_m(vbool1_t mask, vuint8m8_t acc, uint8_t op1, vuin
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmacc.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmacc.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vmacc_vv_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, vuint16mf4_t op1, vuint16mf4_t op2, size_t vl) {
@@ -1329,7 +1329,7 @@ vuint16mf4_t test_vmacc_vv_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, vuint16mf4
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmacc.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmacc.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vmacc_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, uint16_t op1, vuint16mf4_t op2, size_t vl) {
@@ -1338,7 +1338,7 @@ vuint16mf4_t test_vmacc_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, uint16_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmacc.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmacc.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vmacc_vv_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, vuint16mf2_t op1, vuint16mf2_t op2, size_t vl) {
@@ -1347,7 +1347,7 @@ vuint16mf2_t test_vmacc_vv_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, vuint16mf2
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmacc.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmacc.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vmacc_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, uint16_t op1, vuint16mf2_t op2, size_t vl) {
@@ -1356,7 +1356,7 @@ vuint16mf2_t test_vmacc_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, uint16_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmacc.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmacc.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vmacc_vv_u16m1_m(vbool16_t mask, vuint16m1_t acc, vuint16m1_t op1, vuint16m1_t op2, size_t vl) {
@@ -1365,7 +1365,7 @@ vuint16m1_t test_vmacc_vv_u16m1_m(vbool16_t mask, vuint16m1_t acc, vuint16m1_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmacc.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmacc.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vmacc_vx_u16m1_m(vbool16_t mask, vuint16m1_t acc, uint16_t op1, vuint16m1_t op2, size_t vl) {
@@ -1374,7 +1374,7 @@ vuint16m1_t test_vmacc_vx_u16m1_m(vbool16_t mask, vuint16m1_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmacc.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmacc.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vmacc_vv_u16m2_m(vbool8_t mask, vuint16m2_t acc, vuint16m2_t op1, vuint16m2_t op2, size_t vl) {
@@ -1383,7 +1383,7 @@ vuint16m2_t test_vmacc_vv_u16m2_m(vbool8_t mask, vuint16m2_t acc, vuint16m2_t op
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmacc.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmacc.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vmacc_vx_u16m2_m(vbool8_t mask, vuint16m2_t acc, uint16_t op1, vuint16m2_t op2, size_t vl) {
@@ -1392,7 +1392,7 @@ vuint16m2_t test_vmacc_vx_u16m2_m(vbool8_t mask, vuint16m2_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmacc.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmacc.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vmacc_vv_u16m4_m(vbool4_t mask, vuint16m4_t acc, vuint16m4_t op1, vuint16m4_t op2, size_t vl) {
@@ -1401,7 +1401,7 @@ vuint16m4_t test_vmacc_vv_u16m4_m(vbool4_t mask, vuint16m4_t acc, vuint16m4_t op
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmacc.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmacc.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vmacc_vx_u16m4_m(vbool4_t mask, vuint16m4_t acc, uint16_t op1, vuint16m4_t op2, size_t vl) {
@@ -1410,7 +1410,7 @@ vuint16m4_t test_vmacc_vx_u16m4_m(vbool4_t mask, vuint16m4_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmacc.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmacc.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vmacc_vv_u16m8_m(vbool2_t mask, vuint16m8_t acc, vuint16m8_t op1, vuint16m8_t op2, size_t vl) {
@@ -1419,7 +1419,7 @@ vuint16m8_t test_vmacc_vv_u16m8_m(vbool2_t mask, vuint16m8_t acc, vuint16m8_t op
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmacc.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmacc.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vmacc_vx_u16m8_m(vbool2_t mask, vuint16m8_t acc, uint16_t op1, vuint16m8_t op2, size_t vl) {
@@ -1428,7 +1428,7 @@ vuint16m8_t test_vmacc_vx_u16m8_m(vbool2_t mask, vuint16m8_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmacc.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmacc.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vmacc_vv_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, vuint32mf2_t op1, vuint32mf2_t op2, size_t vl) {
@@ -1437,7 +1437,7 @@ vuint32mf2_t test_vmacc_vv_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, vuint32mf2
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmacc.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmacc.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vmacc_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, uint32_t op1, vuint32mf2_t op2, size_t vl) {
@@ -1446,7 +1446,7 @@ vuint32mf2_t test_vmacc_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, uint32_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmacc.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmacc.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vmacc_vv_u32m1_m(vbool32_t mask, vuint32m1_t acc, vuint32m1_t op1, vuint32m1_t op2, size_t vl) {
@@ -1455,7 +1455,7 @@ vuint32m1_t test_vmacc_vv_u32m1_m(vbool32_t mask, vuint32m1_t acc, vuint32m1_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmacc.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmacc.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vmacc_vx_u32m1_m(vbool32_t mask, vuint32m1_t acc, uint32_t op1, vuint32m1_t op2, size_t vl) {
@@ -1464,7 +1464,7 @@ vuint32m1_t test_vmacc_vx_u32m1_m(vbool32_t mask, vuint32m1_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmacc.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmacc.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vmacc_vv_u32m2_m(vbool16_t mask, vuint32m2_t acc, vuint32m2_t op1, vuint32m2_t op2, size_t vl) {
@@ -1473,7 +1473,7 @@ vuint32m2_t test_vmacc_vv_u32m2_m(vbool16_t mask, vuint32m2_t acc, vuint32m2_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmacc.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmacc.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vmacc_vx_u32m2_m(vbool16_t mask, vuint32m2_t acc, uint32_t op1, vuint32m2_t op2, size_t vl) {
@@ -1482,7 +1482,7 @@ vuint32m2_t test_vmacc_vx_u32m2_m(vbool16_t mask, vuint32m2_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmacc.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmacc.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vmacc_vv_u32m4_m(vbool8_t mask, vuint32m4_t acc, vuint32m4_t op1, vuint32m4_t op2, size_t vl) {
@@ -1491,7 +1491,7 @@ vuint32m4_t test_vmacc_vv_u32m4_m(vbool8_t mask, vuint32m4_t acc, vuint32m4_t op
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmacc.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmacc.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vmacc_vx_u32m4_m(vbool8_t mask, vuint32m4_t acc, uint32_t op1, vuint32m4_t op2, size_t vl) {
@@ -1500,7 +1500,7 @@ vuint32m4_t test_vmacc_vx_u32m4_m(vbool8_t mask, vuint32m4_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmacc.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmacc.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vmacc_vv_u32m8_m(vbool4_t mask, vuint32m8_t acc, vuint32m8_t op1, vuint32m8_t op2, size_t vl) {
@@ -1509,7 +1509,7 @@ vuint32m8_t test_vmacc_vv_u32m8_m(vbool4_t mask, vuint32m8_t acc, vuint32m8_t op
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmacc.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmacc.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vmacc_vx_u32m8_m(vbool4_t mask, vuint32m8_t acc, uint32_t op1, vuint32m8_t op2, size_t vl) {
@@ -1518,7 +1518,7 @@ vuint32m8_t test_vmacc_vx_u32m8_m(vbool4_t mask, vuint32m8_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmacc.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmacc.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vmacc_vv_u64m1_m(vbool64_t mask, vuint64m1_t acc, vuint64m1_t op1, vuint64m1_t op2, size_t vl) {
@@ -1527,7 +1527,7 @@ vuint64m1_t test_vmacc_vv_u64m1_m(vbool64_t mask, vuint64m1_t acc, vuint64m1_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmacc.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmacc.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vmacc_vx_u64m1_m(vbool64_t mask, vuint64m1_t acc, uint64_t op1, vuint64m1_t op2, size_t vl) {
@@ -1536,7 +1536,7 @@ vuint64m1_t test_vmacc_vx_u64m1_m(vbool64_t mask, vuint64m1_t acc, uint64_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmacc.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmacc.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vmacc_vv_u64m2_m(vbool32_t mask, vuint64m2_t acc, vuint64m2_t op1, vuint64m2_t op2, size_t vl) {
@@ -1545,7 +1545,7 @@ vuint64m2_t test_vmacc_vv_u64m2_m(vbool32_t mask, vuint64m2_t acc, vuint64m2_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmacc.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmacc.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vmacc_vx_u64m2_m(vbool32_t mask, vuint64m2_t acc, uint64_t op1, vuint64m2_t op2, size_t vl) {
@@ -1554,7 +1554,7 @@ vuint64m2_t test_vmacc_vx_u64m2_m(vbool32_t mask, vuint64m2_t acc, uint64_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmacc.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmacc.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vmacc_vv_u64m4_m(vbool16_t mask, vuint64m4_t acc, vuint64m4_t op1, vuint64m4_t op2, size_t vl) {
@@ -1563,7 +1563,7 @@ vuint64m4_t test_vmacc_vv_u64m4_m(vbool16_t mask, vuint64m4_t acc, vuint64m4_t o
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmacc.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmacc.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vmacc_vx_u64m4_m(vbool16_t mask, vuint64m4_t acc, uint64_t op1, vuint64m4_t op2, size_t vl) {
@@ -1572,7 +1572,7 @@ vuint64m4_t test_vmacc_vx_u64m4_m(vbool16_t mask, vuint64m4_t acc, uint64_t op1,
 
 // CHECK-RV64-LABEL: @test_vmacc_vv_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmacc.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmacc.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vmacc_vv_u64m8_m(vbool8_t mask, vuint64m8_t acc, vuint64m8_t op1, vuint64m8_t op2, size_t vl) {
@@ -1581,7 +1581,7 @@ vuint64m8_t test_vmacc_vv_u64m8_m(vbool8_t mask, vuint64m8_t acc, vuint64m8_t op
 
 // CHECK-RV64-LABEL: @test_vmacc_vx_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmacc.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmacc.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vmacc_vx_u64m8_m(vbool8_t mask, vuint64m8_t acc, uint64_t op1, vuint64m8_t op2, size_t vl) {
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics/vmadd.c b/clang/test/CodeGen/RISCV/rvv-intrinsics/vmadd.c
index de7241136561be..9835a68f21651c 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics/vmadd.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics/vmadd.c
@@ -798,7 +798,7 @@ vuint64m8_t test_vmadd_vx_u64m8(vuint64m8_t acc, uint64_t op1, vuint64m8_t op2,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmadd.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmadd.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vint8mf8_t test_vmadd_vv_i8mf8_m(vbool64_t mask, vint8mf8_t acc, vint8mf8_t op1, vint8mf8_t op2, size_t vl) {
@@ -807,7 +807,7 @@ vint8mf8_t test_vmadd_vv_i8mf8_m(vbool64_t mask, vint8mf8_t acc, vint8mf8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmadd.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmadd.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vint8mf8_t test_vmadd_vx_i8mf8_m(vbool64_t mask, vint8mf8_t acc, int8_t op1, vint8mf8_t op2, size_t vl) {
@@ -816,7 +816,7 @@ vint8mf8_t test_vmadd_vx_i8mf8_m(vbool64_t mask, vint8mf8_t acc, int8_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmadd.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmadd.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vint8mf4_t test_vmadd_vv_i8mf4_m(vbool32_t mask, vint8mf4_t acc, vint8mf4_t op1, vint8mf4_t op2, size_t vl) {
@@ -825,7 +825,7 @@ vint8mf4_t test_vmadd_vv_i8mf4_m(vbool32_t mask, vint8mf4_t acc, vint8mf4_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmadd.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmadd.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vint8mf4_t test_vmadd_vx_i8mf4_m(vbool32_t mask, vint8mf4_t acc, int8_t op1, vint8mf4_t op2, size_t vl) {
@@ -834,7 +834,7 @@ vint8mf4_t test_vmadd_vx_i8mf4_m(vbool32_t mask, vint8mf4_t acc, int8_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmadd.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmadd.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vint8mf2_t test_vmadd_vv_i8mf2_m(vbool16_t mask, vint8mf2_t acc, vint8mf2_t op1, vint8mf2_t op2, size_t vl) {
@@ -843,7 +843,7 @@ vint8mf2_t test_vmadd_vv_i8mf2_m(vbool16_t mask, vint8mf2_t acc, vint8mf2_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmadd.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmadd.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vint8mf2_t test_vmadd_vx_i8mf2_m(vbool16_t mask, vint8mf2_t acc, int8_t op1, vint8mf2_t op2, size_t vl) {
@@ -852,7 +852,7 @@ vint8mf2_t test_vmadd_vx_i8mf2_m(vbool16_t mask, vint8mf2_t acc, int8_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmadd.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmadd.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vint8m1_t test_vmadd_vv_i8m1_m(vbool8_t mask, vint8m1_t acc, vint8m1_t op1, vint8m1_t op2, size_t vl) {
@@ -861,7 +861,7 @@ vint8m1_t test_vmadd_vv_i8m1_m(vbool8_t mask, vint8m1_t acc, vint8m1_t op1, vint
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmadd.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmadd.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vint8m1_t test_vmadd_vx_i8m1_m(vbool8_t mask, vint8m1_t acc, int8_t op1, vint8m1_t op2, size_t vl) {
@@ -870,7 +870,7 @@ vint8m1_t test_vmadd_vx_i8m1_m(vbool8_t mask, vint8m1_t acc, int8_t op1, vint8m1
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmadd.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmadd.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vint8m2_t test_vmadd_vv_i8m2_m(vbool4_t mask, vint8m2_t acc, vint8m2_t op1, vint8m2_t op2, size_t vl) {
@@ -879,7 +879,7 @@ vint8m2_t test_vmadd_vv_i8m2_m(vbool4_t mask, vint8m2_t acc, vint8m2_t op1, vint
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmadd.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmadd.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vint8m2_t test_vmadd_vx_i8m2_m(vbool4_t mask, vint8m2_t acc, int8_t op1, vint8m2_t op2, size_t vl) {
@@ -888,7 +888,7 @@ vint8m2_t test_vmadd_vx_i8m2_m(vbool4_t mask, vint8m2_t acc, int8_t op1, vint8m2
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmadd.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmadd.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vint8m4_t test_vmadd_vv_i8m4_m(vbool2_t mask, vint8m4_t acc, vint8m4_t op1, vint8m4_t op2, size_t vl) {
@@ -897,7 +897,7 @@ vint8m4_t test_vmadd_vv_i8m4_m(vbool2_t mask, vint8m4_t acc, vint8m4_t op1, vint
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmadd.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmadd.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vint8m4_t test_vmadd_vx_i8m4_m(vbool2_t mask, vint8m4_t acc, int8_t op1, vint8m4_t op2, size_t vl) {
@@ -906,7 +906,7 @@ vint8m4_t test_vmadd_vx_i8m4_m(vbool2_t mask, vint8m4_t acc, int8_t op1, vint8m4
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmadd.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmadd.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vint8m8_t test_vmadd_vv_i8m8_m(vbool1_t mask, vint8m8_t acc, vint8m8_t op1, vint8m8_t op2, size_t vl) {
@@ -915,7 +915,7 @@ vint8m8_t test_vmadd_vv_i8m8_m(vbool1_t mask, vint8m8_t acc, vint8m8_t op1, vint
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmadd.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmadd.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vint8m8_t test_vmadd_vx_i8m8_m(vbool1_t mask, vint8m8_t acc, int8_t op1, vint8m8_t op2, size_t vl) {
@@ -924,7 +924,7 @@ vint8m8_t test_vmadd_vx_i8m8_m(vbool1_t mask, vint8m8_t acc, int8_t op1, vint8m8
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmadd.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmadd.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vmadd_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc, vint16mf4_t op1, vint16mf4_t op2, size_t vl) {
@@ -933,7 +933,7 @@ vint16mf4_t test_vmadd_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc, vint16mf4_t
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmadd.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmadd.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vmadd_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc, int16_t op1, vint16mf4_t op2, size_t vl) {
@@ -942,7 +942,7 @@ vint16mf4_t test_vmadd_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmadd.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmadd.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vmadd_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc, vint16mf2_t op1, vint16mf2_t op2, size_t vl) {
@@ -951,7 +951,7 @@ vint16mf2_t test_vmadd_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc, vint16mf2_t
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmadd.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmadd.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vmadd_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc, int16_t op1, vint16mf2_t op2, size_t vl) {
@@ -960,7 +960,7 @@ vint16mf2_t test_vmadd_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmadd.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmadd.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vmadd_vv_i16m1_m(vbool16_t mask, vint16m1_t acc, vint16m1_t op1, vint16m1_t op2, size_t vl) {
@@ -969,7 +969,7 @@ vint16m1_t test_vmadd_vv_i16m1_m(vbool16_t mask, vint16m1_t acc, vint16m1_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmadd.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmadd.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vmadd_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int16_t op1, vint16m1_t op2, size_t vl) {
@@ -978,7 +978,7 @@ vint16m1_t test_vmadd_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int16_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmadd.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmadd.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vmadd_vv_i16m2_m(vbool8_t mask, vint16m2_t acc, vint16m2_t op1, vint16m2_t op2, size_t vl) {
@@ -987,7 +987,7 @@ vint16m2_t test_vmadd_vv_i16m2_m(vbool8_t mask, vint16m2_t acc, vint16m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmadd.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmadd.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vmadd_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int16_t op1, vint16m2_t op2, size_t vl) {
@@ -996,7 +996,7 @@ vint16m2_t test_vmadd_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int16_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmadd.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmadd.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vmadd_vv_i16m4_m(vbool4_t mask, vint16m4_t acc, vint16m4_t op1, vint16m4_t op2, size_t vl) {
@@ -1005,7 +1005,7 @@ vint16m4_t test_vmadd_vv_i16m4_m(vbool4_t mask, vint16m4_t acc, vint16m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmadd.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmadd.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vmadd_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int16_t op1, vint16m4_t op2, size_t vl) {
@@ -1014,7 +1014,7 @@ vint16m4_t test_vmadd_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int16_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmadd.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmadd.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vmadd_vv_i16m8_m(vbool2_t mask, vint16m8_t acc, vint16m8_t op1, vint16m8_t op2, size_t vl) {
@@ -1023,7 +1023,7 @@ vint16m8_t test_vmadd_vv_i16m8_m(vbool2_t mask, vint16m8_t acc, vint16m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmadd.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmadd.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vmadd_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int16_t op1, vint16m8_t op2, size_t vl) {
@@ -1032,7 +1032,7 @@ vint16m8_t test_vmadd_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int16_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmadd.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmadd.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vmadd_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc, vint32mf2_t op1, vint32mf2_t op2, size_t vl) {
@@ -1041,7 +1041,7 @@ vint32mf2_t test_vmadd_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc, vint32mf2_t
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmadd.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmadd.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vmadd_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc, int32_t op1, vint32mf2_t op2, size_t vl) {
@@ -1050,7 +1050,7 @@ vint32mf2_t test_vmadd_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc, int32_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmadd.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmadd.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vmadd_vv_i32m1_m(vbool32_t mask, vint32m1_t acc, vint32m1_t op1, vint32m1_t op2, size_t vl) {
@@ -1059,7 +1059,7 @@ vint32m1_t test_vmadd_vv_i32m1_m(vbool32_t mask, vint32m1_t acc, vint32m1_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmadd.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmadd.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vmadd_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int32_t op1, vint32m1_t op2, size_t vl) {
@@ -1068,7 +1068,7 @@ vint32m1_t test_vmadd_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int32_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmadd.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmadd.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vmadd_vv_i32m2_m(vbool16_t mask, vint32m2_t acc, vint32m2_t op1, vint32m2_t op2, size_t vl) {
@@ -1077,7 +1077,7 @@ vint32m2_t test_vmadd_vv_i32m2_m(vbool16_t mask, vint32m2_t acc, vint32m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmadd.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmadd.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vmadd_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int32_t op1, vint32m2_t op2, size_t vl) {
@@ -1086,7 +1086,7 @@ vint32m2_t test_vmadd_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int32_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmadd.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmadd.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vmadd_vv_i32m4_m(vbool8_t mask, vint32m4_t acc, vint32m4_t op1, vint32m4_t op2, size_t vl) {
@@ -1095,7 +1095,7 @@ vint32m4_t test_vmadd_vv_i32m4_m(vbool8_t mask, vint32m4_t acc, vint32m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmadd.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmadd.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vmadd_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int32_t op1, vint32m4_t op2, size_t vl) {
@@ -1104,7 +1104,7 @@ vint32m4_t test_vmadd_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int32_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmadd.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmadd.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vmadd_vv_i32m8_m(vbool4_t mask, vint32m8_t acc, vint32m8_t op1, vint32m8_t op2, size_t vl) {
@@ -1113,7 +1113,7 @@ vint32m8_t test_vmadd_vv_i32m8_m(vbool4_t mask, vint32m8_t acc, vint32m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmadd.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmadd.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vmadd_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int32_t op1, vint32m8_t op2, size_t vl) {
@@ -1122,7 +1122,7 @@ vint32m8_t test_vmadd_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int32_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmadd.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmadd.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vmadd_vv_i64m1_m(vbool64_t mask, vint64m1_t acc, vint64m1_t op1, vint64m1_t op2, size_t vl) {
@@ -1131,7 +1131,7 @@ vint64m1_t test_vmadd_vv_i64m1_m(vbool64_t mask, vint64m1_t acc, vint64m1_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmadd.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmadd.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vmadd_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int64_t op1, vint64m1_t op2, size_t vl) {
@@ -1140,7 +1140,7 @@ vint64m1_t test_vmadd_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int64_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmadd.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmadd.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vmadd_vv_i64m2_m(vbool32_t mask, vint64m2_t acc, vint64m2_t op1, vint64m2_t op2, size_t vl) {
@@ -1149,7 +1149,7 @@ vint64m2_t test_vmadd_vv_i64m2_m(vbool32_t mask, vint64m2_t acc, vint64m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmadd.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmadd.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vmadd_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int64_t op1, vint64m2_t op2, size_t vl) {
@@ -1158,7 +1158,7 @@ vint64m2_t test_vmadd_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int64_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmadd.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmadd.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vmadd_vv_i64m4_m(vbool16_t mask, vint64m4_t acc, vint64m4_t op1, vint64m4_t op2, size_t vl) {
@@ -1167,7 +1167,7 @@ vint64m4_t test_vmadd_vv_i64m4_m(vbool16_t mask, vint64m4_t acc, vint64m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmadd.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmadd.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vmadd_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int64_t op1, vint64m4_t op2, size_t vl) {
@@ -1176,7 +1176,7 @@ vint64m4_t test_vmadd_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int64_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmadd.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmadd.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vmadd_vv_i64m8_m(vbool8_t mask, vint64m8_t acc, vint64m8_t op1, vint64m8_t op2, size_t vl) {
@@ -1185,7 +1185,7 @@ vint64m8_t test_vmadd_vv_i64m8_m(vbool8_t mask, vint64m8_t acc, vint64m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmadd.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmadd.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vmadd_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int64_t op1, vint64m8_t op2, size_t vl) {
@@ -1194,7 +1194,7 @@ vint64m8_t test_vmadd_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int64_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmadd.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmadd.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vuint8mf8_t test_vmadd_vv_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, vuint8mf8_t op1, vuint8mf8_t op2, size_t vl) {
@@ -1203,7 +1203,7 @@ vuint8mf8_t test_vmadd_vv_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, vuint8mf8_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmadd.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vmadd.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vuint8mf8_t test_vmadd_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, uint8_t op1, vuint8mf8_t op2, size_t vl) {
@@ -1212,7 +1212,7 @@ vuint8mf8_t test_vmadd_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmadd.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmadd.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vuint8mf4_t test_vmadd_vv_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, vuint8mf4_t op1, vuint8mf4_t op2, size_t vl) {
@@ -1221,7 +1221,7 @@ vuint8mf4_t test_vmadd_vv_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, vuint8mf4_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmadd.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vmadd.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vuint8mf4_t test_vmadd_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, uint8_t op1, vuint8mf4_t op2, size_t vl) {
@@ -1230,7 +1230,7 @@ vuint8mf4_t test_vmadd_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmadd.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmadd.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vuint8mf2_t test_vmadd_vv_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, vuint8mf2_t op1, vuint8mf2_t op2, size_t vl) {
@@ -1239,7 +1239,7 @@ vuint8mf2_t test_vmadd_vv_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, vuint8mf2_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmadd.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vmadd.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vuint8mf2_t test_vmadd_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, uint8_t op1, vuint8mf2_t op2, size_t vl) {
@@ -1248,7 +1248,7 @@ vuint8mf2_t test_vmadd_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmadd.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmadd.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vuint8m1_t test_vmadd_vv_u8m1_m(vbool8_t mask, vuint8m1_t acc, vuint8m1_t op1, vuint8m1_t op2, size_t vl) {
@@ -1257,7 +1257,7 @@ vuint8m1_t test_vmadd_vv_u8m1_m(vbool8_t mask, vuint8m1_t acc, vuint8m1_t op1, v
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmadd.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vmadd.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vuint8m1_t test_vmadd_vx_u8m1_m(vbool8_t mask, vuint8m1_t acc, uint8_t op1, vuint8m1_t op2, size_t vl) {
@@ -1266,7 +1266,7 @@ vuint8m1_t test_vmadd_vx_u8m1_m(vbool8_t mask, vuint8m1_t acc, uint8_t op1, vuin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmadd.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmadd.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vuint8m2_t test_vmadd_vv_u8m2_m(vbool4_t mask, vuint8m2_t acc, vuint8m2_t op1, vuint8m2_t op2, size_t vl) {
@@ -1275,7 +1275,7 @@ vuint8m2_t test_vmadd_vv_u8m2_m(vbool4_t mask, vuint8m2_t acc, vuint8m2_t op1, v
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmadd.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vmadd.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vuint8m2_t test_vmadd_vx_u8m2_m(vbool4_t mask, vuint8m2_t acc, uint8_t op1, vuint8m2_t op2, size_t vl) {
@@ -1284,7 +1284,7 @@ vuint8m2_t test_vmadd_vx_u8m2_m(vbool4_t mask, vuint8m2_t acc, uint8_t op1, vuin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmadd.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmadd.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vuint8m4_t test_vmadd_vv_u8m4_m(vbool2_t mask, vuint8m4_t acc, vuint8m4_t op1, vuint8m4_t op2, size_t vl) {
@@ -1293,7 +1293,7 @@ vuint8m4_t test_vmadd_vv_u8m4_m(vbool2_t mask, vuint8m4_t acc, vuint8m4_t op1, v
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmadd.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vmadd.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vuint8m4_t test_vmadd_vx_u8m4_m(vbool2_t mask, vuint8m4_t acc, uint8_t op1, vuint8m4_t op2, size_t vl) {
@@ -1302,7 +1302,7 @@ vuint8m4_t test_vmadd_vx_u8m4_m(vbool2_t mask, vuint8m4_t acc, uint8_t op1, vuin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmadd.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmadd.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vuint8m8_t test_vmadd_vv_u8m8_m(vbool1_t mask, vuint8m8_t acc, vuint8m8_t op1, vuint8m8_t op2, size_t vl) {
@@ -1311,7 +1311,7 @@ vuint8m8_t test_vmadd_vv_u8m8_m(vbool1_t mask, vuint8m8_t acc, vuint8m8_t op1, v
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmadd.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vmadd.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vuint8m8_t test_vmadd_vx_u8m8_m(vbool1_t mask, vuint8m8_t acc, uint8_t op1, vuint8m8_t op2, size_t vl) {
@@ -1320,7 +1320,7 @@ vuint8m8_t test_vmadd_vx_u8m8_m(vbool1_t mask, vuint8m8_t acc, uint8_t op1, vuin
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmadd.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmadd.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vmadd_vv_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, vuint16mf4_t op1, vuint16mf4_t op2, size_t vl) {
@@ -1329,7 +1329,7 @@ vuint16mf4_t test_vmadd_vv_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, vuint16mf4
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmadd.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vmadd.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vmadd_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, uint16_t op1, vuint16mf4_t op2, size_t vl) {
@@ -1338,7 +1338,7 @@ vuint16mf4_t test_vmadd_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, uint16_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmadd.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmadd.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vmadd_vv_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, vuint16mf2_t op1, vuint16mf2_t op2, size_t vl) {
@@ -1347,7 +1347,7 @@ vuint16mf2_t test_vmadd_vv_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, vuint16mf2
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmadd.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vmadd.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vmadd_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, uint16_t op1, vuint16mf2_t op2, size_t vl) {
@@ -1356,7 +1356,7 @@ vuint16mf2_t test_vmadd_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, uint16_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmadd.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmadd.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vmadd_vv_u16m1_m(vbool16_t mask, vuint16m1_t acc, vuint16m1_t op1, vuint16m1_t op2, size_t vl) {
@@ -1365,7 +1365,7 @@ vuint16m1_t test_vmadd_vv_u16m1_m(vbool16_t mask, vuint16m1_t acc, vuint16m1_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmadd.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vmadd.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vmadd_vx_u16m1_m(vbool16_t mask, vuint16m1_t acc, uint16_t op1, vuint16m1_t op2, size_t vl) {
@@ -1374,7 +1374,7 @@ vuint16m1_t test_vmadd_vx_u16m1_m(vbool16_t mask, vuint16m1_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmadd.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmadd.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vmadd_vv_u16m2_m(vbool8_t mask, vuint16m2_t acc, vuint16m2_t op1, vuint16m2_t op2, size_t vl) {
@@ -1383,7 +1383,7 @@ vuint16m2_t test_vmadd_vv_u16m2_m(vbool8_t mask, vuint16m2_t acc, vuint16m2_t op
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmadd.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vmadd.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vmadd_vx_u16m2_m(vbool8_t mask, vuint16m2_t acc, uint16_t op1, vuint16m2_t op2, size_t vl) {
@@ -1392,7 +1392,7 @@ vuint16m2_t test_vmadd_vx_u16m2_m(vbool8_t mask, vuint16m2_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmadd.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmadd.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vmadd_vv_u16m4_m(vbool4_t mask, vuint16m4_t acc, vuint16m4_t op1, vuint16m4_t op2, size_t vl) {
@@ -1401,7 +1401,7 @@ vuint16m4_t test_vmadd_vv_u16m4_m(vbool4_t mask, vuint16m4_t acc, vuint16m4_t op
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmadd.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vmadd.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vmadd_vx_u16m4_m(vbool4_t mask, vuint16m4_t acc, uint16_t op1, vuint16m4_t op2, size_t vl) {
@@ -1410,7 +1410,7 @@ vuint16m4_t test_vmadd_vx_u16m4_m(vbool4_t mask, vuint16m4_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmadd.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmadd.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vmadd_vv_u16m8_m(vbool2_t mask, vuint16m8_t acc, vuint16m8_t op1, vuint16m8_t op2, size_t vl) {
@@ -1419,7 +1419,7 @@ vuint16m8_t test_vmadd_vv_u16m8_m(vbool2_t mask, vuint16m8_t acc, vuint16m8_t op
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmadd.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vmadd.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vmadd_vx_u16m8_m(vbool2_t mask, vuint16m8_t acc, uint16_t op1, vuint16m8_t op2, size_t vl) {
@@ -1428,7 +1428,7 @@ vuint16m8_t test_vmadd_vx_u16m8_m(vbool2_t mask, vuint16m8_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmadd.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmadd.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vmadd_vv_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, vuint32mf2_t op1, vuint32mf2_t op2, size_t vl) {
@@ -1437,7 +1437,7 @@ vuint32mf2_t test_vmadd_vv_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, vuint32mf2
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmadd.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vmadd.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vmadd_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, uint32_t op1, vuint32mf2_t op2, size_t vl) {
@@ -1446,7 +1446,7 @@ vuint32mf2_t test_vmadd_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, uint32_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmadd.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmadd.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vmadd_vv_u32m1_m(vbool32_t mask, vuint32m1_t acc, vuint32m1_t op1, vuint32m1_t op2, size_t vl) {
@@ -1455,7 +1455,7 @@ vuint32m1_t test_vmadd_vv_u32m1_m(vbool32_t mask, vuint32m1_t acc, vuint32m1_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmadd.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vmadd.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vmadd_vx_u32m1_m(vbool32_t mask, vuint32m1_t acc, uint32_t op1, vuint32m1_t op2, size_t vl) {
@@ -1464,7 +1464,7 @@ vuint32m1_t test_vmadd_vx_u32m1_m(vbool32_t mask, vuint32m1_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmadd.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmadd.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vmadd_vv_u32m2_m(vbool16_t mask, vuint32m2_t acc, vuint32m2_t op1, vuint32m2_t op2, size_t vl) {
@@ -1473,7 +1473,7 @@ vuint32m2_t test_vmadd_vv_u32m2_m(vbool16_t mask, vuint32m2_t acc, vuint32m2_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmadd.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vmadd.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vmadd_vx_u32m2_m(vbool16_t mask, vuint32m2_t acc, uint32_t op1, vuint32m2_t op2, size_t vl) {
@@ -1482,7 +1482,7 @@ vuint32m2_t test_vmadd_vx_u32m2_m(vbool16_t mask, vuint32m2_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmadd.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmadd.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vmadd_vv_u32m4_m(vbool8_t mask, vuint32m4_t acc, vuint32m4_t op1, vuint32m4_t op2, size_t vl) {
@@ -1491,7 +1491,7 @@ vuint32m4_t test_vmadd_vv_u32m4_m(vbool8_t mask, vuint32m4_t acc, vuint32m4_t op
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmadd.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vmadd.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vmadd_vx_u32m4_m(vbool8_t mask, vuint32m4_t acc, uint32_t op1, vuint32m4_t op2, size_t vl) {
@@ -1500,7 +1500,7 @@ vuint32m4_t test_vmadd_vx_u32m4_m(vbool8_t mask, vuint32m4_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmadd.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmadd.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vmadd_vv_u32m8_m(vbool4_t mask, vuint32m8_t acc, vuint32m8_t op1, vuint32m8_t op2, size_t vl) {
@@ -1509,7 +1509,7 @@ vuint32m8_t test_vmadd_vv_u32m8_m(vbool4_t mask, vuint32m8_t acc, vuint32m8_t op
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmadd.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vmadd.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vmadd_vx_u32m8_m(vbool4_t mask, vuint32m8_t acc, uint32_t op1, vuint32m8_t op2, size_t vl) {
@@ -1518,7 +1518,7 @@ vuint32m8_t test_vmadd_vx_u32m8_m(vbool4_t mask, vuint32m8_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmadd.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmadd.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vmadd_vv_u64m1_m(vbool64_t mask, vuint64m1_t acc, vuint64m1_t op1, vuint64m1_t op2, size_t vl) {
@@ -1527,7 +1527,7 @@ vuint64m1_t test_vmadd_vv_u64m1_m(vbool64_t mask, vuint64m1_t acc, vuint64m1_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmadd.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vmadd.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vmadd_vx_u64m1_m(vbool64_t mask, vuint64m1_t acc, uint64_t op1, vuint64m1_t op2, size_t vl) {
@@ -1536,7 +1536,7 @@ vuint64m1_t test_vmadd_vx_u64m1_m(vbool64_t mask, vuint64m1_t acc, uint64_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmadd.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmadd.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vmadd_vv_u64m2_m(vbool32_t mask, vuint64m2_t acc, vuint64m2_t op1, vuint64m2_t op2, size_t vl) {
@@ -1545,7 +1545,7 @@ vuint64m2_t test_vmadd_vv_u64m2_m(vbool32_t mask, vuint64m2_t acc, vuint64m2_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmadd.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vmadd.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vmadd_vx_u64m2_m(vbool32_t mask, vuint64m2_t acc, uint64_t op1, vuint64m2_t op2, size_t vl) {
@@ -1554,7 +1554,7 @@ vuint64m2_t test_vmadd_vx_u64m2_m(vbool32_t mask, vuint64m2_t acc, uint64_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmadd.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmadd.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vmadd_vv_u64m4_m(vbool16_t mask, vuint64m4_t acc, vuint64m4_t op1, vuint64m4_t op2, size_t vl) {
@@ -1563,7 +1563,7 @@ vuint64m4_t test_vmadd_vv_u64m4_m(vbool16_t mask, vuint64m4_t acc, vuint64m4_t o
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmadd.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vmadd.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vmadd_vx_u64m4_m(vbool16_t mask, vuint64m4_t acc, uint64_t op1, vuint64m4_t op2, size_t vl) {
@@ -1572,7 +1572,7 @@ vuint64m4_t test_vmadd_vx_u64m4_m(vbool16_t mask, vuint64m4_t acc, uint64_t op1,
 
 // CHECK-RV64-LABEL: @test_vmadd_vv_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmadd.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmadd.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vmadd_vv_u64m8_m(vbool8_t mask, vuint64m8_t acc, vuint64m8_t op1, vuint64m8_t op2, size_t vl) {
@@ -1581,7 +1581,7 @@ vuint64m8_t test_vmadd_vv_u64m8_m(vbool8_t mask, vuint64m8_t acc, vuint64m8_t op
 
 // CHECK-RV64-LABEL: @test_vmadd_vx_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmadd.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vmadd.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vmadd_vx_u64m8_m(vbool8_t mask, vuint64m8_t acc, uint64_t op1, vuint64m8_t op2, size_t vl) {
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics/vnmsac.c b/clang/test/CodeGen/RISCV/rvv-intrinsics/vnmsac.c
index 65b97bdeccd40a..e0aa06746cc558 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics/vnmsac.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics/vnmsac.c
@@ -798,7 +798,7 @@ vuint64m8_t test_vnmsac_vx_u64m8(vuint64m8_t acc, uint64_t op1, vuint64m8_t op2,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsac.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsac.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vint8mf8_t test_vnmsac_vv_i8mf8_m(vbool64_t mask, vint8mf8_t acc, vint8mf8_t op1, vint8mf8_t op2, size_t vl) {
@@ -807,7 +807,7 @@ vint8mf8_t test_vnmsac_vv_i8mf8_m(vbool64_t mask, vint8mf8_t acc, vint8mf8_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsac.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsac.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vint8mf8_t test_vnmsac_vx_i8mf8_m(vbool64_t mask, vint8mf8_t acc, int8_t op1, vint8mf8_t op2, size_t vl) {
@@ -816,7 +816,7 @@ vint8mf8_t test_vnmsac_vx_i8mf8_m(vbool64_t mask, vint8mf8_t acc, int8_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsac.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsac.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vint8mf4_t test_vnmsac_vv_i8mf4_m(vbool32_t mask, vint8mf4_t acc, vint8mf4_t op1, vint8mf4_t op2, size_t vl) {
@@ -825,7 +825,7 @@ vint8mf4_t test_vnmsac_vv_i8mf4_m(vbool32_t mask, vint8mf4_t acc, vint8mf4_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsac.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsac.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vint8mf4_t test_vnmsac_vx_i8mf4_m(vbool32_t mask, vint8mf4_t acc, int8_t op1, vint8mf4_t op2, size_t vl) {
@@ -834,7 +834,7 @@ vint8mf4_t test_vnmsac_vx_i8mf4_m(vbool32_t mask, vint8mf4_t acc, int8_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsac.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsac.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vint8mf2_t test_vnmsac_vv_i8mf2_m(vbool16_t mask, vint8mf2_t acc, vint8mf2_t op1, vint8mf2_t op2, size_t vl) {
@@ -843,7 +843,7 @@ vint8mf2_t test_vnmsac_vv_i8mf2_m(vbool16_t mask, vint8mf2_t acc, vint8mf2_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsac.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsac.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vint8mf2_t test_vnmsac_vx_i8mf2_m(vbool16_t mask, vint8mf2_t acc, int8_t op1, vint8mf2_t op2, size_t vl) {
@@ -852,7 +852,7 @@ vint8mf2_t test_vnmsac_vx_i8mf2_m(vbool16_t mask, vint8mf2_t acc, int8_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsac.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsac.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vint8m1_t test_vnmsac_vv_i8m1_m(vbool8_t mask, vint8m1_t acc, vint8m1_t op1, vint8m1_t op2, size_t vl) {
@@ -861,7 +861,7 @@ vint8m1_t test_vnmsac_vv_i8m1_m(vbool8_t mask, vint8m1_t acc, vint8m1_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsac.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsac.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vint8m1_t test_vnmsac_vx_i8m1_m(vbool8_t mask, vint8m1_t acc, int8_t op1, vint8m1_t op2, size_t vl) {
@@ -870,7 +870,7 @@ vint8m1_t test_vnmsac_vx_i8m1_m(vbool8_t mask, vint8m1_t acc, int8_t op1, vint8m
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsac.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsac.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vint8m2_t test_vnmsac_vv_i8m2_m(vbool4_t mask, vint8m2_t acc, vint8m2_t op1, vint8m2_t op2, size_t vl) {
@@ -879,7 +879,7 @@ vint8m2_t test_vnmsac_vv_i8m2_m(vbool4_t mask, vint8m2_t acc, vint8m2_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsac.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsac.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vint8m2_t test_vnmsac_vx_i8m2_m(vbool4_t mask, vint8m2_t acc, int8_t op1, vint8m2_t op2, size_t vl) {
@@ -888,7 +888,7 @@ vint8m2_t test_vnmsac_vx_i8m2_m(vbool4_t mask, vint8m2_t acc, int8_t op1, vint8m
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsac.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsac.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vint8m4_t test_vnmsac_vv_i8m4_m(vbool2_t mask, vint8m4_t acc, vint8m4_t op1, vint8m4_t op2, size_t vl) {
@@ -897,7 +897,7 @@ vint8m4_t test_vnmsac_vv_i8m4_m(vbool2_t mask, vint8m4_t acc, vint8m4_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsac.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsac.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vint8m4_t test_vnmsac_vx_i8m4_m(vbool2_t mask, vint8m4_t acc, int8_t op1, vint8m4_t op2, size_t vl) {
@@ -906,7 +906,7 @@ vint8m4_t test_vnmsac_vx_i8m4_m(vbool2_t mask, vint8m4_t acc, int8_t op1, vint8m
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsac.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsac.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vint8m8_t test_vnmsac_vv_i8m8_m(vbool1_t mask, vint8m8_t acc, vint8m8_t op1, vint8m8_t op2, size_t vl) {
@@ -915,7 +915,7 @@ vint8m8_t test_vnmsac_vv_i8m8_m(vbool1_t mask, vint8m8_t acc, vint8m8_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsac.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsac.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vint8m8_t test_vnmsac_vx_i8m8_m(vbool1_t mask, vint8m8_t acc, int8_t op1, vint8m8_t op2, size_t vl) {
@@ -924,7 +924,7 @@ vint8m8_t test_vnmsac_vx_i8m8_m(vbool1_t mask, vint8m8_t acc, int8_t op1, vint8m
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsac.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsac.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vnmsac_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc, vint16mf4_t op1, vint16mf4_t op2, size_t vl) {
@@ -933,7 +933,7 @@ vint16mf4_t test_vnmsac_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc, vint16mf4_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsac.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsac.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vnmsac_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc, int16_t op1, vint16mf4_t op2, size_t vl) {
@@ -942,7 +942,7 @@ vint16mf4_t test_vnmsac_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc, int16_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsac.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsac.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vnmsac_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc, vint16mf2_t op1, vint16mf2_t op2, size_t vl) {
@@ -951,7 +951,7 @@ vint16mf2_t test_vnmsac_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc, vint16mf2_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsac.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsac.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vnmsac_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc, int16_t op1, vint16mf2_t op2, size_t vl) {
@@ -960,7 +960,7 @@ vint16mf2_t test_vnmsac_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc, int16_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsac.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsac.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vnmsac_vv_i16m1_m(vbool16_t mask, vint16m1_t acc, vint16m1_t op1, vint16m1_t op2, size_t vl) {
@@ -969,7 +969,7 @@ vint16m1_t test_vnmsac_vv_i16m1_m(vbool16_t mask, vint16m1_t acc, vint16m1_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsac.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsac.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vnmsac_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int16_t op1, vint16m1_t op2, size_t vl) {
@@ -978,7 +978,7 @@ vint16m1_t test_vnmsac_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int16_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsac.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsac.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vnmsac_vv_i16m2_m(vbool8_t mask, vint16m2_t acc, vint16m2_t op1, vint16m2_t op2, size_t vl) {
@@ -987,7 +987,7 @@ vint16m2_t test_vnmsac_vv_i16m2_m(vbool8_t mask, vint16m2_t acc, vint16m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsac.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsac.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vnmsac_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int16_t op1, vint16m2_t op2, size_t vl) {
@@ -996,7 +996,7 @@ vint16m2_t test_vnmsac_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int16_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsac.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsac.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vnmsac_vv_i16m4_m(vbool4_t mask, vint16m4_t acc, vint16m4_t op1, vint16m4_t op2, size_t vl) {
@@ -1005,7 +1005,7 @@ vint16m4_t test_vnmsac_vv_i16m4_m(vbool4_t mask, vint16m4_t acc, vint16m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsac.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsac.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vnmsac_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int16_t op1, vint16m4_t op2, size_t vl) {
@@ -1014,7 +1014,7 @@ vint16m4_t test_vnmsac_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int16_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsac.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsac.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vnmsac_vv_i16m8_m(vbool2_t mask, vint16m8_t acc, vint16m8_t op1, vint16m8_t op2, size_t vl) {
@@ -1023,7 +1023,7 @@ vint16m8_t test_vnmsac_vv_i16m8_m(vbool2_t mask, vint16m8_t acc, vint16m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsac.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsac.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vnmsac_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int16_t op1, vint16m8_t op2, size_t vl) {
@@ -1032,7 +1032,7 @@ vint16m8_t test_vnmsac_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int16_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsac.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsac.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vnmsac_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc, vint32mf2_t op1, vint32mf2_t op2, size_t vl) {
@@ -1041,7 +1041,7 @@ vint32mf2_t test_vnmsac_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc, vint32mf2_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsac.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsac.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vnmsac_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc, int32_t op1, vint32mf2_t op2, size_t vl) {
@@ -1050,7 +1050,7 @@ vint32mf2_t test_vnmsac_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc, int32_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsac.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsac.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vnmsac_vv_i32m1_m(vbool32_t mask, vint32m1_t acc, vint32m1_t op1, vint32m1_t op2, size_t vl) {
@@ -1059,7 +1059,7 @@ vint32m1_t test_vnmsac_vv_i32m1_m(vbool32_t mask, vint32m1_t acc, vint32m1_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsac.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsac.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vnmsac_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int32_t op1, vint32m1_t op2, size_t vl) {
@@ -1068,7 +1068,7 @@ vint32m1_t test_vnmsac_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int32_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsac.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsac.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vnmsac_vv_i32m2_m(vbool16_t mask, vint32m2_t acc, vint32m2_t op1, vint32m2_t op2, size_t vl) {
@@ -1077,7 +1077,7 @@ vint32m2_t test_vnmsac_vv_i32m2_m(vbool16_t mask, vint32m2_t acc, vint32m2_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsac.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsac.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vnmsac_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int32_t op1, vint32m2_t op2, size_t vl) {
@@ -1086,7 +1086,7 @@ vint32m2_t test_vnmsac_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int32_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsac.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsac.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vnmsac_vv_i32m4_m(vbool8_t mask, vint32m4_t acc, vint32m4_t op1, vint32m4_t op2, size_t vl) {
@@ -1095,7 +1095,7 @@ vint32m4_t test_vnmsac_vv_i32m4_m(vbool8_t mask, vint32m4_t acc, vint32m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsac.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsac.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vnmsac_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int32_t op1, vint32m4_t op2, size_t vl) {
@@ -1104,7 +1104,7 @@ vint32m4_t test_vnmsac_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int32_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsac.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsac.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vnmsac_vv_i32m8_m(vbool4_t mask, vint32m8_t acc, vint32m8_t op1, vint32m8_t op2, size_t vl) {
@@ -1113,7 +1113,7 @@ vint32m8_t test_vnmsac_vv_i32m8_m(vbool4_t mask, vint32m8_t acc, vint32m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsac.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsac.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vnmsac_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int32_t op1, vint32m8_t op2, size_t vl) {
@@ -1122,7 +1122,7 @@ vint32m8_t test_vnmsac_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int32_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsac.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsac.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vnmsac_vv_i64m1_m(vbool64_t mask, vint64m1_t acc, vint64m1_t op1, vint64m1_t op2, size_t vl) {
@@ -1131,7 +1131,7 @@ vint64m1_t test_vnmsac_vv_i64m1_m(vbool64_t mask, vint64m1_t acc, vint64m1_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsac.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsac.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vnmsac_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int64_t op1, vint64m1_t op2, size_t vl) {
@@ -1140,7 +1140,7 @@ vint64m1_t test_vnmsac_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int64_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsac.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsac.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vnmsac_vv_i64m2_m(vbool32_t mask, vint64m2_t acc, vint64m2_t op1, vint64m2_t op2, size_t vl) {
@@ -1149,7 +1149,7 @@ vint64m2_t test_vnmsac_vv_i64m2_m(vbool32_t mask, vint64m2_t acc, vint64m2_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsac.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsac.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vnmsac_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int64_t op1, vint64m2_t op2, size_t vl) {
@@ -1158,7 +1158,7 @@ vint64m2_t test_vnmsac_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int64_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsac.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsac.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vnmsac_vv_i64m4_m(vbool16_t mask, vint64m4_t acc, vint64m4_t op1, vint64m4_t op2, size_t vl) {
@@ -1167,7 +1167,7 @@ vint64m4_t test_vnmsac_vv_i64m4_m(vbool16_t mask, vint64m4_t acc, vint64m4_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsac.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsac.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vnmsac_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int64_t op1, vint64m4_t op2, size_t vl) {
@@ -1176,7 +1176,7 @@ vint64m4_t test_vnmsac_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int64_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsac.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsac.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vnmsac_vv_i64m8_m(vbool8_t mask, vint64m8_t acc, vint64m8_t op1, vint64m8_t op2, size_t vl) {
@@ -1185,7 +1185,7 @@ vint64m8_t test_vnmsac_vv_i64m8_m(vbool8_t mask, vint64m8_t acc, vint64m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsac.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsac.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vnmsac_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int64_t op1, vint64m8_t op2, size_t vl) {
@@ -1194,7 +1194,7 @@ vint64m8_t test_vnmsac_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int64_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsac.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsac.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vuint8mf8_t test_vnmsac_vv_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, vuint8mf8_t op1, vuint8mf8_t op2, size_t vl) {
@@ -1203,7 +1203,7 @@ vuint8mf8_t test_vnmsac_vv_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, vuint8mf8_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsac.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsac.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vuint8mf8_t test_vnmsac_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, uint8_t op1, vuint8mf8_t op2, size_t vl) {
@@ -1212,7 +1212,7 @@ vuint8mf8_t test_vnmsac_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsac.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsac.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vuint8mf4_t test_vnmsac_vv_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, vuint8mf4_t op1, vuint8mf4_t op2, size_t vl) {
@@ -1221,7 +1221,7 @@ vuint8mf4_t test_vnmsac_vv_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, vuint8mf4_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsac.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsac.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vuint8mf4_t test_vnmsac_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, uint8_t op1, vuint8mf4_t op2, size_t vl) {
@@ -1230,7 +1230,7 @@ vuint8mf4_t test_vnmsac_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsac.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsac.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vuint8mf2_t test_vnmsac_vv_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, vuint8mf2_t op1, vuint8mf2_t op2, size_t vl) {
@@ -1239,7 +1239,7 @@ vuint8mf2_t test_vnmsac_vv_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, vuint8mf2_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsac.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsac.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vuint8mf2_t test_vnmsac_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, uint8_t op1, vuint8mf2_t op2, size_t vl) {
@@ -1248,7 +1248,7 @@ vuint8mf2_t test_vnmsac_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsac.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsac.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vuint8m1_t test_vnmsac_vv_u8m1_m(vbool8_t mask, vuint8m1_t acc, vuint8m1_t op1, vuint8m1_t op2, size_t vl) {
@@ -1257,7 +1257,7 @@ vuint8m1_t test_vnmsac_vv_u8m1_m(vbool8_t mask, vuint8m1_t acc, vuint8m1_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsac.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsac.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vuint8m1_t test_vnmsac_vx_u8m1_m(vbool8_t mask, vuint8m1_t acc, uint8_t op1, vuint8m1_t op2, size_t vl) {
@@ -1266,7 +1266,7 @@ vuint8m1_t test_vnmsac_vx_u8m1_m(vbool8_t mask, vuint8m1_t acc, uint8_t op1, vui
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsac.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsac.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vuint8m2_t test_vnmsac_vv_u8m2_m(vbool4_t mask, vuint8m2_t acc, vuint8m2_t op1, vuint8m2_t op2, size_t vl) {
@@ -1275,7 +1275,7 @@ vuint8m2_t test_vnmsac_vv_u8m2_m(vbool4_t mask, vuint8m2_t acc, vuint8m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsac.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsac.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vuint8m2_t test_vnmsac_vx_u8m2_m(vbool4_t mask, vuint8m2_t acc, uint8_t op1, vuint8m2_t op2, size_t vl) {
@@ -1284,7 +1284,7 @@ vuint8m2_t test_vnmsac_vx_u8m2_m(vbool4_t mask, vuint8m2_t acc, uint8_t op1, vui
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsac.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsac.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vuint8m4_t test_vnmsac_vv_u8m4_m(vbool2_t mask, vuint8m4_t acc, vuint8m4_t op1, vuint8m4_t op2, size_t vl) {
@@ -1293,7 +1293,7 @@ vuint8m4_t test_vnmsac_vv_u8m4_m(vbool2_t mask, vuint8m4_t acc, vuint8m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsac.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsac.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vuint8m4_t test_vnmsac_vx_u8m4_m(vbool2_t mask, vuint8m4_t acc, uint8_t op1, vuint8m4_t op2, size_t vl) {
@@ -1302,7 +1302,7 @@ vuint8m4_t test_vnmsac_vx_u8m4_m(vbool2_t mask, vuint8m4_t acc, uint8_t op1, vui
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsac.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsac.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vuint8m8_t test_vnmsac_vv_u8m8_m(vbool1_t mask, vuint8m8_t acc, vuint8m8_t op1, vuint8m8_t op2, size_t vl) {
@@ -1311,7 +1311,7 @@ vuint8m8_t test_vnmsac_vv_u8m8_m(vbool1_t mask, vuint8m8_t acc, vuint8m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsac.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsac.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vuint8m8_t test_vnmsac_vx_u8m8_m(vbool1_t mask, vuint8m8_t acc, uint8_t op1, vuint8m8_t op2, size_t vl) {
@@ -1320,7 +1320,7 @@ vuint8m8_t test_vnmsac_vx_u8m8_m(vbool1_t mask, vuint8m8_t acc, uint8_t op1, vui
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsac.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsac.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vnmsac_vv_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, vuint16mf4_t op1, vuint16mf4_t op2, size_t vl) {
@@ -1329,7 +1329,7 @@ vuint16mf4_t test_vnmsac_vv_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, vuint16mf
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsac.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsac.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vnmsac_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, uint16_t op1, vuint16mf4_t op2, size_t vl) {
@@ -1338,7 +1338,7 @@ vuint16mf4_t test_vnmsac_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, uint16_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsac.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsac.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vnmsac_vv_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, vuint16mf2_t op1, vuint16mf2_t op2, size_t vl) {
@@ -1347,7 +1347,7 @@ vuint16mf2_t test_vnmsac_vv_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, vuint16mf
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsac.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsac.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vnmsac_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, uint16_t op1, vuint16mf2_t op2, size_t vl) {
@@ -1356,7 +1356,7 @@ vuint16mf2_t test_vnmsac_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, uint16_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsac.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsac.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vnmsac_vv_u16m1_m(vbool16_t mask, vuint16m1_t acc, vuint16m1_t op1, vuint16m1_t op2, size_t vl) {
@@ -1365,7 +1365,7 @@ vuint16m1_t test_vnmsac_vv_u16m1_m(vbool16_t mask, vuint16m1_t acc, vuint16m1_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsac.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsac.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vnmsac_vx_u16m1_m(vbool16_t mask, vuint16m1_t acc, uint16_t op1, vuint16m1_t op2, size_t vl) {
@@ -1374,7 +1374,7 @@ vuint16m1_t test_vnmsac_vx_u16m1_m(vbool16_t mask, vuint16m1_t acc, uint16_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsac.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsac.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vnmsac_vv_u16m2_m(vbool8_t mask, vuint16m2_t acc, vuint16m2_t op1, vuint16m2_t op2, size_t vl) {
@@ -1383,7 +1383,7 @@ vuint16m2_t test_vnmsac_vv_u16m2_m(vbool8_t mask, vuint16m2_t acc, vuint16m2_t o
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsac.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsac.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vnmsac_vx_u16m2_m(vbool8_t mask, vuint16m2_t acc, uint16_t op1, vuint16m2_t op2, size_t vl) {
@@ -1392,7 +1392,7 @@ vuint16m2_t test_vnmsac_vx_u16m2_m(vbool8_t mask, vuint16m2_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsac.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsac.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vnmsac_vv_u16m4_m(vbool4_t mask, vuint16m4_t acc, vuint16m4_t op1, vuint16m4_t op2, size_t vl) {
@@ -1401,7 +1401,7 @@ vuint16m4_t test_vnmsac_vv_u16m4_m(vbool4_t mask, vuint16m4_t acc, vuint16m4_t o
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsac.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsac.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vnmsac_vx_u16m4_m(vbool4_t mask, vuint16m4_t acc, uint16_t op1, vuint16m4_t op2, size_t vl) {
@@ -1410,7 +1410,7 @@ vuint16m4_t test_vnmsac_vx_u16m4_m(vbool4_t mask, vuint16m4_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsac.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsac.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vnmsac_vv_u16m8_m(vbool2_t mask, vuint16m8_t acc, vuint16m8_t op1, vuint16m8_t op2, size_t vl) {
@@ -1419,7 +1419,7 @@ vuint16m8_t test_vnmsac_vv_u16m8_m(vbool2_t mask, vuint16m8_t acc, vuint16m8_t o
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsac.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsac.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vnmsac_vx_u16m8_m(vbool2_t mask, vuint16m8_t acc, uint16_t op1, vuint16m8_t op2, size_t vl) {
@@ -1428,7 +1428,7 @@ vuint16m8_t test_vnmsac_vx_u16m8_m(vbool2_t mask, vuint16m8_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsac.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsac.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vnmsac_vv_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, vuint32mf2_t op1, vuint32mf2_t op2, size_t vl) {
@@ -1437,7 +1437,7 @@ vuint32mf2_t test_vnmsac_vv_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, vuint32mf
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsac.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsac.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vnmsac_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, uint32_t op1, vuint32mf2_t op2, size_t vl) {
@@ -1446,7 +1446,7 @@ vuint32mf2_t test_vnmsac_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, uint32_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsac.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsac.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vnmsac_vv_u32m1_m(vbool32_t mask, vuint32m1_t acc, vuint32m1_t op1, vuint32m1_t op2, size_t vl) {
@@ -1455,7 +1455,7 @@ vuint32m1_t test_vnmsac_vv_u32m1_m(vbool32_t mask, vuint32m1_t acc, vuint32m1_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsac.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsac.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vnmsac_vx_u32m1_m(vbool32_t mask, vuint32m1_t acc, uint32_t op1, vuint32m1_t op2, size_t vl) {
@@ -1464,7 +1464,7 @@ vuint32m1_t test_vnmsac_vx_u32m1_m(vbool32_t mask, vuint32m1_t acc, uint32_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsac.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsac.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vnmsac_vv_u32m2_m(vbool16_t mask, vuint32m2_t acc, vuint32m2_t op1, vuint32m2_t op2, size_t vl) {
@@ -1473,7 +1473,7 @@ vuint32m2_t test_vnmsac_vv_u32m2_m(vbool16_t mask, vuint32m2_t acc, vuint32m2_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsac.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsac.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vnmsac_vx_u32m2_m(vbool16_t mask, vuint32m2_t acc, uint32_t op1, vuint32m2_t op2, size_t vl) {
@@ -1482,7 +1482,7 @@ vuint32m2_t test_vnmsac_vx_u32m2_m(vbool16_t mask, vuint32m2_t acc, uint32_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsac.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsac.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vnmsac_vv_u32m4_m(vbool8_t mask, vuint32m4_t acc, vuint32m4_t op1, vuint32m4_t op2, size_t vl) {
@@ -1491,7 +1491,7 @@ vuint32m4_t test_vnmsac_vv_u32m4_m(vbool8_t mask, vuint32m4_t acc, vuint32m4_t o
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsac.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsac.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vnmsac_vx_u32m4_m(vbool8_t mask, vuint32m4_t acc, uint32_t op1, vuint32m4_t op2, size_t vl) {
@@ -1500,7 +1500,7 @@ vuint32m4_t test_vnmsac_vx_u32m4_m(vbool8_t mask, vuint32m4_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsac.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsac.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vnmsac_vv_u32m8_m(vbool4_t mask, vuint32m8_t acc, vuint32m8_t op1, vuint32m8_t op2, size_t vl) {
@@ -1509,7 +1509,7 @@ vuint32m8_t test_vnmsac_vv_u32m8_m(vbool4_t mask, vuint32m8_t acc, vuint32m8_t o
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsac.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsac.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vnmsac_vx_u32m8_m(vbool4_t mask, vuint32m8_t acc, uint32_t op1, vuint32m8_t op2, size_t vl) {
@@ -1518,7 +1518,7 @@ vuint32m8_t test_vnmsac_vx_u32m8_m(vbool4_t mask, vuint32m8_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsac.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsac.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vnmsac_vv_u64m1_m(vbool64_t mask, vuint64m1_t acc, vuint64m1_t op1, vuint64m1_t op2, size_t vl) {
@@ -1527,7 +1527,7 @@ vuint64m1_t test_vnmsac_vv_u64m1_m(vbool64_t mask, vuint64m1_t acc, vuint64m1_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsac.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsac.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vnmsac_vx_u64m1_m(vbool64_t mask, vuint64m1_t acc, uint64_t op1, vuint64m1_t op2, size_t vl) {
@@ -1536,7 +1536,7 @@ vuint64m1_t test_vnmsac_vx_u64m1_m(vbool64_t mask, vuint64m1_t acc, uint64_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsac.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsac.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vnmsac_vv_u64m2_m(vbool32_t mask, vuint64m2_t acc, vuint64m2_t op1, vuint64m2_t op2, size_t vl) {
@@ -1545,7 +1545,7 @@ vuint64m2_t test_vnmsac_vv_u64m2_m(vbool32_t mask, vuint64m2_t acc, vuint64m2_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsac.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsac.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vnmsac_vx_u64m2_m(vbool32_t mask, vuint64m2_t acc, uint64_t op1, vuint64m2_t op2, size_t vl) {
@@ -1554,7 +1554,7 @@ vuint64m2_t test_vnmsac_vx_u64m2_m(vbool32_t mask, vuint64m2_t acc, uint64_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsac.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsac.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vnmsac_vv_u64m4_m(vbool16_t mask, vuint64m4_t acc, vuint64m4_t op1, vuint64m4_t op2, size_t vl) {
@@ -1563,7 +1563,7 @@ vuint64m4_t test_vnmsac_vv_u64m4_m(vbool16_t mask, vuint64m4_t acc, vuint64m4_t
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsac.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsac.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vnmsac_vx_u64m4_m(vbool16_t mask, vuint64m4_t acc, uint64_t op1, vuint64m4_t op2, size_t vl) {
@@ -1572,7 +1572,7 @@ vuint64m4_t test_vnmsac_vx_u64m4_m(vbool16_t mask, vuint64m4_t acc, uint64_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsac_vv_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsac.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsac.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vnmsac_vv_u64m8_m(vbool8_t mask, vuint64m8_t acc, vuint64m8_t op1, vuint64m8_t op2, size_t vl) {
@@ -1581,7 +1581,7 @@ vuint64m8_t test_vnmsac_vv_u64m8_m(vbool8_t mask, vuint64m8_t acc, vuint64m8_t o
 
 // CHECK-RV64-LABEL: @test_vnmsac_vx_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsac.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsac.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vnmsac_vx_u64m8_m(vbool8_t mask, vuint64m8_t acc, uint64_t op1, vuint64m8_t op2, size_t vl) {
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics/vnmsub.c b/clang/test/CodeGen/RISCV/rvv-intrinsics/vnmsub.c
index 6bc003f150cda7..9a409c7abb662d 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics/vnmsub.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics/vnmsub.c
@@ -798,7 +798,7 @@ vuint64m8_t test_vnmsub_vx_u64m8(vuint64m8_t acc, uint64_t op1, vuint64m8_t op2,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsub.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsub.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vint8mf8_t test_vnmsub_vv_i8mf8_m(vbool64_t mask, vint8mf8_t acc, vint8mf8_t op1, vint8mf8_t op2, size_t vl) {
@@ -807,7 +807,7 @@ vint8mf8_t test_vnmsub_vv_i8mf8_m(vbool64_t mask, vint8mf8_t acc, vint8mf8_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsub.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsub.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vint8mf8_t test_vnmsub_vx_i8mf8_m(vbool64_t mask, vint8mf8_t acc, int8_t op1, vint8mf8_t op2, size_t vl) {
@@ -816,7 +816,7 @@ vint8mf8_t test_vnmsub_vx_i8mf8_m(vbool64_t mask, vint8mf8_t acc, int8_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsub.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsub.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vint8mf4_t test_vnmsub_vv_i8mf4_m(vbool32_t mask, vint8mf4_t acc, vint8mf4_t op1, vint8mf4_t op2, size_t vl) {
@@ -825,7 +825,7 @@ vint8mf4_t test_vnmsub_vv_i8mf4_m(vbool32_t mask, vint8mf4_t acc, vint8mf4_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsub.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsub.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vint8mf4_t test_vnmsub_vx_i8mf4_m(vbool32_t mask, vint8mf4_t acc, int8_t op1, vint8mf4_t op2, size_t vl) {
@@ -834,7 +834,7 @@ vint8mf4_t test_vnmsub_vx_i8mf4_m(vbool32_t mask, vint8mf4_t acc, int8_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsub.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsub.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vint8mf2_t test_vnmsub_vv_i8mf2_m(vbool16_t mask, vint8mf2_t acc, vint8mf2_t op1, vint8mf2_t op2, size_t vl) {
@@ -843,7 +843,7 @@ vint8mf2_t test_vnmsub_vv_i8mf2_m(vbool16_t mask, vint8mf2_t acc, vint8mf2_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsub.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsub.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vint8mf2_t test_vnmsub_vx_i8mf2_m(vbool16_t mask, vint8mf2_t acc, int8_t op1, vint8mf2_t op2, size_t vl) {
@@ -852,7 +852,7 @@ vint8mf2_t test_vnmsub_vx_i8mf2_m(vbool16_t mask, vint8mf2_t acc, int8_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsub.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsub.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vint8m1_t test_vnmsub_vv_i8m1_m(vbool8_t mask, vint8m1_t acc, vint8m1_t op1, vint8m1_t op2, size_t vl) {
@@ -861,7 +861,7 @@ vint8m1_t test_vnmsub_vv_i8m1_m(vbool8_t mask, vint8m1_t acc, vint8m1_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsub.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsub.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vint8m1_t test_vnmsub_vx_i8m1_m(vbool8_t mask, vint8m1_t acc, int8_t op1, vint8m1_t op2, size_t vl) {
@@ -870,7 +870,7 @@ vint8m1_t test_vnmsub_vx_i8m1_m(vbool8_t mask, vint8m1_t acc, int8_t op1, vint8m
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsub.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsub.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vint8m2_t test_vnmsub_vv_i8m2_m(vbool4_t mask, vint8m2_t acc, vint8m2_t op1, vint8m2_t op2, size_t vl) {
@@ -879,7 +879,7 @@ vint8m2_t test_vnmsub_vv_i8m2_m(vbool4_t mask, vint8m2_t acc, vint8m2_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsub.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsub.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vint8m2_t test_vnmsub_vx_i8m2_m(vbool4_t mask, vint8m2_t acc, int8_t op1, vint8m2_t op2, size_t vl) {
@@ -888,7 +888,7 @@ vint8m2_t test_vnmsub_vx_i8m2_m(vbool4_t mask, vint8m2_t acc, int8_t op1, vint8m
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsub.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsub.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vint8m4_t test_vnmsub_vv_i8m4_m(vbool2_t mask, vint8m4_t acc, vint8m4_t op1, vint8m4_t op2, size_t vl) {
@@ -897,7 +897,7 @@ vint8m4_t test_vnmsub_vv_i8m4_m(vbool2_t mask, vint8m4_t acc, vint8m4_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsub.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsub.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vint8m4_t test_vnmsub_vx_i8m4_m(vbool2_t mask, vint8m4_t acc, int8_t op1, vint8m4_t op2, size_t vl) {
@@ -906,7 +906,7 @@ vint8m4_t test_vnmsub_vx_i8m4_m(vbool2_t mask, vint8m4_t acc, int8_t op1, vint8m
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsub.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsub.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vint8m8_t test_vnmsub_vv_i8m8_m(vbool1_t mask, vint8m8_t acc, vint8m8_t op1, vint8m8_t op2, size_t vl) {
@@ -915,7 +915,7 @@ vint8m8_t test_vnmsub_vv_i8m8_m(vbool1_t mask, vint8m8_t acc, vint8m8_t op1, vin
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsub.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsub.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vint8m8_t test_vnmsub_vx_i8m8_m(vbool1_t mask, vint8m8_t acc, int8_t op1, vint8m8_t op2, size_t vl) {
@@ -924,7 +924,7 @@ vint8m8_t test_vnmsub_vx_i8m8_m(vbool1_t mask, vint8m8_t acc, int8_t op1, vint8m
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsub.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsub.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vnmsub_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc, vint16mf4_t op1, vint16mf4_t op2, size_t vl) {
@@ -933,7 +933,7 @@ vint16mf4_t test_vnmsub_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc, vint16mf4_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsub.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsub.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vnmsub_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc, int16_t op1, vint16mf4_t op2, size_t vl) {
@@ -942,7 +942,7 @@ vint16mf4_t test_vnmsub_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc, int16_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsub.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsub.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vnmsub_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc, vint16mf2_t op1, vint16mf2_t op2, size_t vl) {
@@ -951,7 +951,7 @@ vint16mf2_t test_vnmsub_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc, vint16mf2_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsub.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsub.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vnmsub_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc, int16_t op1, vint16mf2_t op2, size_t vl) {
@@ -960,7 +960,7 @@ vint16mf2_t test_vnmsub_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc, int16_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsub.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsub.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vnmsub_vv_i16m1_m(vbool16_t mask, vint16m1_t acc, vint16m1_t op1, vint16m1_t op2, size_t vl) {
@@ -969,7 +969,7 @@ vint16m1_t test_vnmsub_vv_i16m1_m(vbool16_t mask, vint16m1_t acc, vint16m1_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsub.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsub.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vnmsub_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int16_t op1, vint16m1_t op2, size_t vl) {
@@ -978,7 +978,7 @@ vint16m1_t test_vnmsub_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int16_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsub.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsub.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vnmsub_vv_i16m2_m(vbool8_t mask, vint16m2_t acc, vint16m2_t op1, vint16m2_t op2, size_t vl) {
@@ -987,7 +987,7 @@ vint16m2_t test_vnmsub_vv_i16m2_m(vbool8_t mask, vint16m2_t acc, vint16m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsub.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsub.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vnmsub_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int16_t op1, vint16m2_t op2, size_t vl) {
@@ -996,7 +996,7 @@ vint16m2_t test_vnmsub_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int16_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsub.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsub.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vnmsub_vv_i16m4_m(vbool4_t mask, vint16m4_t acc, vint16m4_t op1, vint16m4_t op2, size_t vl) {
@@ -1005,7 +1005,7 @@ vint16m4_t test_vnmsub_vv_i16m4_m(vbool4_t mask, vint16m4_t acc, vint16m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsub.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsub.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vnmsub_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int16_t op1, vint16m4_t op2, size_t vl) {
@@ -1014,7 +1014,7 @@ vint16m4_t test_vnmsub_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int16_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsub.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsub.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vnmsub_vv_i16m8_m(vbool2_t mask, vint16m8_t acc, vint16m8_t op1, vint16m8_t op2, size_t vl) {
@@ -1023,7 +1023,7 @@ vint16m8_t test_vnmsub_vv_i16m8_m(vbool2_t mask, vint16m8_t acc, vint16m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsub.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsub.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vnmsub_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int16_t op1, vint16m8_t op2, size_t vl) {
@@ -1032,7 +1032,7 @@ vint16m8_t test_vnmsub_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int16_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsub.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsub.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vnmsub_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc, vint32mf2_t op1, vint32mf2_t op2, size_t vl) {
@@ -1041,7 +1041,7 @@ vint32mf2_t test_vnmsub_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc, vint32mf2_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsub.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsub.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vnmsub_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc, int32_t op1, vint32mf2_t op2, size_t vl) {
@@ -1050,7 +1050,7 @@ vint32mf2_t test_vnmsub_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc, int32_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsub.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsub.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vnmsub_vv_i32m1_m(vbool32_t mask, vint32m1_t acc, vint32m1_t op1, vint32m1_t op2, size_t vl) {
@@ -1059,7 +1059,7 @@ vint32m1_t test_vnmsub_vv_i32m1_m(vbool32_t mask, vint32m1_t acc, vint32m1_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsub.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsub.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vnmsub_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int32_t op1, vint32m1_t op2, size_t vl) {
@@ -1068,7 +1068,7 @@ vint32m1_t test_vnmsub_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int32_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsub.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsub.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vnmsub_vv_i32m2_m(vbool16_t mask, vint32m2_t acc, vint32m2_t op1, vint32m2_t op2, size_t vl) {
@@ -1077,7 +1077,7 @@ vint32m2_t test_vnmsub_vv_i32m2_m(vbool16_t mask, vint32m2_t acc, vint32m2_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsub.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsub.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vnmsub_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int32_t op1, vint32m2_t op2, size_t vl) {
@@ -1086,7 +1086,7 @@ vint32m2_t test_vnmsub_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int32_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsub.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsub.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vnmsub_vv_i32m4_m(vbool8_t mask, vint32m4_t acc, vint32m4_t op1, vint32m4_t op2, size_t vl) {
@@ -1095,7 +1095,7 @@ vint32m4_t test_vnmsub_vv_i32m4_m(vbool8_t mask, vint32m4_t acc, vint32m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsub.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsub.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vnmsub_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int32_t op1, vint32m4_t op2, size_t vl) {
@@ -1104,7 +1104,7 @@ vint32m4_t test_vnmsub_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int32_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsub.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsub.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vnmsub_vv_i32m8_m(vbool4_t mask, vint32m8_t acc, vint32m8_t op1, vint32m8_t op2, size_t vl) {
@@ -1113,7 +1113,7 @@ vint32m8_t test_vnmsub_vv_i32m8_m(vbool4_t mask, vint32m8_t acc, vint32m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsub.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsub.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vnmsub_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int32_t op1, vint32m8_t op2, size_t vl) {
@@ -1122,7 +1122,7 @@ vint32m8_t test_vnmsub_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int32_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsub.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsub.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vnmsub_vv_i64m1_m(vbool64_t mask, vint64m1_t acc, vint64m1_t op1, vint64m1_t op2, size_t vl) {
@@ -1131,7 +1131,7 @@ vint64m1_t test_vnmsub_vv_i64m1_m(vbool64_t mask, vint64m1_t acc, vint64m1_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsub.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsub.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vnmsub_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int64_t op1, vint64m1_t op2, size_t vl) {
@@ -1140,7 +1140,7 @@ vint64m1_t test_vnmsub_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int64_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsub.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsub.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vnmsub_vv_i64m2_m(vbool32_t mask, vint64m2_t acc, vint64m2_t op1, vint64m2_t op2, size_t vl) {
@@ -1149,7 +1149,7 @@ vint64m2_t test_vnmsub_vv_i64m2_m(vbool32_t mask, vint64m2_t acc, vint64m2_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsub.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsub.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vnmsub_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int64_t op1, vint64m2_t op2, size_t vl) {
@@ -1158,7 +1158,7 @@ vint64m2_t test_vnmsub_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int64_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsub.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsub.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vnmsub_vv_i64m4_m(vbool16_t mask, vint64m4_t acc, vint64m4_t op1, vint64m4_t op2, size_t vl) {
@@ -1167,7 +1167,7 @@ vint64m4_t test_vnmsub_vv_i64m4_m(vbool16_t mask, vint64m4_t acc, vint64m4_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsub.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsub.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vnmsub_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int64_t op1, vint64m4_t op2, size_t vl) {
@@ -1176,7 +1176,7 @@ vint64m4_t test_vnmsub_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int64_t op1, v
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsub.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsub.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vnmsub_vv_i64m8_m(vbool8_t mask, vint64m8_t acc, vint64m8_t op1, vint64m8_t op2, size_t vl) {
@@ -1185,7 +1185,7 @@ vint64m8_t test_vnmsub_vv_i64m8_m(vbool8_t mask, vint64m8_t acc, vint64m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsub.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsub.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vnmsub_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int64_t op1, vint64m8_t op2, size_t vl) {
@@ -1194,7 +1194,7 @@ vint64m8_t test_vnmsub_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int64_t op1, vi
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsub.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsub.mask.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vuint8mf8_t test_vnmsub_vv_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, vuint8mf8_t op1, vuint8mf8_t op2, size_t vl) {
@@ -1203,7 +1203,7 @@ vuint8mf8_t test_vnmsub_vv_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, vuint8mf8_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsub.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnmsub.mask.nxv1i8.i8.i64(<vscale x 1 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vuint8mf8_t test_vnmsub_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, uint8_t op1, vuint8mf8_t op2, size_t vl) {
@@ -1212,7 +1212,7 @@ vuint8mf8_t test_vnmsub_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsub.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsub.mask.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vuint8mf4_t test_vnmsub_vv_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, vuint8mf4_t op1, vuint8mf4_t op2, size_t vl) {
@@ -1221,7 +1221,7 @@ vuint8mf4_t test_vnmsub_vv_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, vuint8mf4_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsub.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnmsub.mask.nxv2i8.i8.i64(<vscale x 2 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vuint8mf4_t test_vnmsub_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, uint8_t op1, vuint8mf4_t op2, size_t vl) {
@@ -1230,7 +1230,7 @@ vuint8mf4_t test_vnmsub_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsub.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsub.mask.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vuint8mf2_t test_vnmsub_vv_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, vuint8mf2_t op1, vuint8mf2_t op2, size_t vl) {
@@ -1239,7 +1239,7 @@ vuint8mf2_t test_vnmsub_vv_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, vuint8mf2_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsub.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnmsub.mask.nxv4i8.i8.i64(<vscale x 4 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vuint8mf2_t test_vnmsub_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, uint8_t op1, vuint8mf2_t op2, size_t vl) {
@@ -1248,7 +1248,7 @@ vuint8mf2_t test_vnmsub_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsub.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsub.mask.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vuint8m1_t test_vnmsub_vv_u8m1_m(vbool8_t mask, vuint8m1_t acc, vuint8m1_t op1, vuint8m1_t op2, size_t vl) {
@@ -1257,7 +1257,7 @@ vuint8m1_t test_vnmsub_vv_u8m1_m(vbool8_t mask, vuint8m1_t acc, vuint8m1_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsub.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnmsub.mask.nxv8i8.i8.i64(<vscale x 8 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vuint8m1_t test_vnmsub_vx_u8m1_m(vbool8_t mask, vuint8m1_t acc, uint8_t op1, vuint8m1_t op2, size_t vl) {
@@ -1266,7 +1266,7 @@ vuint8m1_t test_vnmsub_vx_u8m1_m(vbool8_t mask, vuint8m1_t acc, uint8_t op1, vui
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsub.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsub.mask.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vuint8m2_t test_vnmsub_vv_u8m2_m(vbool4_t mask, vuint8m2_t acc, vuint8m2_t op1, vuint8m2_t op2, size_t vl) {
@@ -1275,7 +1275,7 @@ vuint8m2_t test_vnmsub_vv_u8m2_m(vbool4_t mask, vuint8m2_t acc, vuint8m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsub.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnmsub.mask.nxv16i8.i8.i64(<vscale x 16 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vuint8m2_t test_vnmsub_vx_u8m2_m(vbool4_t mask, vuint8m2_t acc, uint8_t op1, vuint8m2_t op2, size_t vl) {
@@ -1284,7 +1284,7 @@ vuint8m2_t test_vnmsub_vx_u8m2_m(vbool4_t mask, vuint8m2_t acc, uint8_t op1, vui
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsub.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsub.mask.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vuint8m4_t test_vnmsub_vv_u8m4_m(vbool2_t mask, vuint8m4_t acc, vuint8m4_t op1, vuint8m4_t op2, size_t vl) {
@@ -1293,7 +1293,7 @@ vuint8m4_t test_vnmsub_vv_u8m4_m(vbool2_t mask, vuint8m4_t acc, vuint8m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsub.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnmsub.mask.nxv32i8.i8.i64(<vscale x 32 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vuint8m4_t test_vnmsub_vx_u8m4_m(vbool2_t mask, vuint8m4_t acc, uint8_t op1, vuint8m4_t op2, size_t vl) {
@@ -1302,7 +1302,7 @@ vuint8m4_t test_vnmsub_vx_u8m4_m(vbool2_t mask, vuint8m4_t acc, uint8_t op1, vui
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsub.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsub.mask.nxv64i8.nxv64i8.i64(<vscale x 64 x i8> [[ACC:%.*]], <vscale x 64 x i8> [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vuint8m8_t test_vnmsub_vv_u8m8_m(vbool1_t mask, vuint8m8_t acc, vuint8m8_t op1, vuint8m8_t op2, size_t vl) {
@@ -1311,7 +1311,7 @@ vuint8m8_t test_vnmsub_vv_u8m8_m(vbool1_t mask, vuint8m8_t acc, vuint8m8_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsub.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vnmsub.mask.nxv64i8.i8.i64(<vscale x 64 x i8> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 64 x i8> [[OP2:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vuint8m8_t test_vnmsub_vx_u8m8_m(vbool1_t mask, vuint8m8_t acc, uint8_t op1, vuint8m8_t op2, size_t vl) {
@@ -1320,7 +1320,7 @@ vuint8m8_t test_vnmsub_vx_u8m8_m(vbool1_t mask, vuint8m8_t acc, uint8_t op1, vui
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsub.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsub.mask.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vnmsub_vv_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, vuint16mf4_t op1, vuint16mf4_t op2, size_t vl) {
@@ -1329,7 +1329,7 @@ vuint16mf4_t test_vnmsub_vv_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, vuint16mf
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsub.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnmsub.mask.nxv1i16.i16.i64(<vscale x 1 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vnmsub_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, uint16_t op1, vuint16mf4_t op2, size_t vl) {
@@ -1338,7 +1338,7 @@ vuint16mf4_t test_vnmsub_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t acc, uint16_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsub.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsub.mask.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vnmsub_vv_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, vuint16mf2_t op1, vuint16mf2_t op2, size_t vl) {
@@ -1347,7 +1347,7 @@ vuint16mf2_t test_vnmsub_vv_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, vuint16mf
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsub.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnmsub.mask.nxv2i16.i16.i64(<vscale x 2 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vnmsub_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, uint16_t op1, vuint16mf2_t op2, size_t vl) {
@@ -1356,7 +1356,7 @@ vuint16mf2_t test_vnmsub_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t acc, uint16_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsub.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsub.mask.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vnmsub_vv_u16m1_m(vbool16_t mask, vuint16m1_t acc, vuint16m1_t op1, vuint16m1_t op2, size_t vl) {
@@ -1365,7 +1365,7 @@ vuint16m1_t test_vnmsub_vv_u16m1_m(vbool16_t mask, vuint16m1_t acc, vuint16m1_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsub.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnmsub.mask.nxv4i16.i16.i64(<vscale x 4 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vnmsub_vx_u16m1_m(vbool16_t mask, vuint16m1_t acc, uint16_t op1, vuint16m1_t op2, size_t vl) {
@@ -1374,7 +1374,7 @@ vuint16m1_t test_vnmsub_vx_u16m1_m(vbool16_t mask, vuint16m1_t acc, uint16_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsub.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsub.mask.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vnmsub_vv_u16m2_m(vbool8_t mask, vuint16m2_t acc, vuint16m2_t op1, vuint16m2_t op2, size_t vl) {
@@ -1383,7 +1383,7 @@ vuint16m2_t test_vnmsub_vv_u16m2_m(vbool8_t mask, vuint16m2_t acc, vuint16m2_t o
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsub.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnmsub.mask.nxv8i16.i16.i64(<vscale x 8 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vnmsub_vx_u16m2_m(vbool8_t mask, vuint16m2_t acc, uint16_t op1, vuint16m2_t op2, size_t vl) {
@@ -1392,7 +1392,7 @@ vuint16m2_t test_vnmsub_vx_u16m2_m(vbool8_t mask, vuint16m2_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsub.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsub.mask.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vnmsub_vv_u16m4_m(vbool4_t mask, vuint16m4_t acc, vuint16m4_t op1, vuint16m4_t op2, size_t vl) {
@@ -1401,7 +1401,7 @@ vuint16m4_t test_vnmsub_vv_u16m4_m(vbool4_t mask, vuint16m4_t acc, vuint16m4_t o
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsub.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnmsub.mask.nxv16i16.i16.i64(<vscale x 16 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vnmsub_vx_u16m4_m(vbool4_t mask, vuint16m4_t acc, uint16_t op1, vuint16m4_t op2, size_t vl) {
@@ -1410,7 +1410,7 @@ vuint16m4_t test_vnmsub_vx_u16m4_m(vbool4_t mask, vuint16m4_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsub.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsub.mask.nxv32i16.nxv32i16.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vnmsub_vv_u16m8_m(vbool2_t mask, vuint16m8_t acc, vuint16m8_t op1, vuint16m8_t op2, size_t vl) {
@@ -1419,7 +1419,7 @@ vuint16m8_t test_vnmsub_vv_u16m8_m(vbool2_t mask, vuint16m8_t acc, vuint16m8_t o
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsub.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vnmsub.mask.nxv32i16.i16.i64(<vscale x 32 x i16> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 32 x i16> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vnmsub_vx_u16m8_m(vbool2_t mask, vuint16m8_t acc, uint16_t op1, vuint16m8_t op2, size_t vl) {
@@ -1428,7 +1428,7 @@ vuint16m8_t test_vnmsub_vx_u16m8_m(vbool2_t mask, vuint16m8_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsub.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsub.mask.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vnmsub_vv_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, vuint32mf2_t op1, vuint32mf2_t op2, size_t vl) {
@@ -1437,7 +1437,7 @@ vuint32mf2_t test_vnmsub_vv_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, vuint32mf
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsub.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnmsub.mask.nxv1i32.i32.i64(<vscale x 1 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vnmsub_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, uint32_t op1, vuint32mf2_t op2, size_t vl) {
@@ -1446,7 +1446,7 @@ vuint32mf2_t test_vnmsub_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t acc, uint32_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsub.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsub.mask.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vnmsub_vv_u32m1_m(vbool32_t mask, vuint32m1_t acc, vuint32m1_t op1, vuint32m1_t op2, size_t vl) {
@@ -1455,7 +1455,7 @@ vuint32m1_t test_vnmsub_vv_u32m1_m(vbool32_t mask, vuint32m1_t acc, vuint32m1_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsub.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnmsub.mask.nxv2i32.i32.i64(<vscale x 2 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vnmsub_vx_u32m1_m(vbool32_t mask, vuint32m1_t acc, uint32_t op1, vuint32m1_t op2, size_t vl) {
@@ -1464,7 +1464,7 @@ vuint32m1_t test_vnmsub_vx_u32m1_m(vbool32_t mask, vuint32m1_t acc, uint32_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsub.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsub.mask.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vnmsub_vv_u32m2_m(vbool16_t mask, vuint32m2_t acc, vuint32m2_t op1, vuint32m2_t op2, size_t vl) {
@@ -1473,7 +1473,7 @@ vuint32m2_t test_vnmsub_vv_u32m2_m(vbool16_t mask, vuint32m2_t acc, vuint32m2_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsub.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnmsub.mask.nxv4i32.i32.i64(<vscale x 4 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vnmsub_vx_u32m2_m(vbool16_t mask, vuint32m2_t acc, uint32_t op1, vuint32m2_t op2, size_t vl) {
@@ -1482,7 +1482,7 @@ vuint32m2_t test_vnmsub_vx_u32m2_m(vbool16_t mask, vuint32m2_t acc, uint32_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsub.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsub.mask.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vnmsub_vv_u32m4_m(vbool8_t mask, vuint32m4_t acc, vuint32m4_t op1, vuint32m4_t op2, size_t vl) {
@@ -1491,7 +1491,7 @@ vuint32m4_t test_vnmsub_vv_u32m4_m(vbool8_t mask, vuint32m4_t acc, vuint32m4_t o
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsub.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnmsub.mask.nxv8i32.i32.i64(<vscale x 8 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vnmsub_vx_u32m4_m(vbool8_t mask, vuint32m4_t acc, uint32_t op1, vuint32m4_t op2, size_t vl) {
@@ -1500,7 +1500,7 @@ vuint32m4_t test_vnmsub_vx_u32m4_m(vbool8_t mask, vuint32m4_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsub.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsub.mask.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vnmsub_vv_u32m8_m(vbool4_t mask, vuint32m8_t acc, vuint32m8_t op1, vuint32m8_t op2, size_t vl) {
@@ -1509,7 +1509,7 @@ vuint32m8_t test_vnmsub_vv_u32m8_m(vbool4_t mask, vuint32m8_t acc, vuint32m8_t o
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsub.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vnmsub.mask.nxv16i32.i32.i64(<vscale x 16 x i32> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 16 x i32> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vnmsub_vx_u32m8_m(vbool4_t mask, vuint32m8_t acc, uint32_t op1, vuint32m8_t op2, size_t vl) {
@@ -1518,7 +1518,7 @@ vuint32m8_t test_vnmsub_vx_u32m8_m(vbool4_t mask, vuint32m8_t acc, uint32_t op1,
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsub.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsub.mask.nxv1i64.nxv1i64.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vnmsub_vv_u64m1_m(vbool64_t mask, vuint64m1_t acc, vuint64m1_t op1, vuint64m1_t op2, size_t vl) {
@@ -1527,7 +1527,7 @@ vuint64m1_t test_vnmsub_vv_u64m1_m(vbool64_t mask, vuint64m1_t acc, vuint64m1_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsub.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vnmsub.mask.nxv1i64.i64.i64(<vscale x 1 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 1 x i64> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vnmsub_vx_u64m1_m(vbool64_t mask, vuint64m1_t acc, uint64_t op1, vuint64m1_t op2, size_t vl) {
@@ -1536,7 +1536,7 @@ vuint64m1_t test_vnmsub_vx_u64m1_m(vbool64_t mask, vuint64m1_t acc, uint64_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsub.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsub.mask.nxv2i64.nxv2i64.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vnmsub_vv_u64m2_m(vbool32_t mask, vuint64m2_t acc, vuint64m2_t op1, vuint64m2_t op2, size_t vl) {
@@ -1545,7 +1545,7 @@ vuint64m2_t test_vnmsub_vv_u64m2_m(vbool32_t mask, vuint64m2_t acc, vuint64m2_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsub.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vnmsub.mask.nxv2i64.i64.i64(<vscale x 2 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vnmsub_vx_u64m2_m(vbool32_t mask, vuint64m2_t acc, uint64_t op1, vuint64m2_t op2, size_t vl) {
@@ -1554,7 +1554,7 @@ vuint64m2_t test_vnmsub_vx_u64m2_m(vbool32_t mask, vuint64m2_t acc, uint64_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsub.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsub.mask.nxv4i64.nxv4i64.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vnmsub_vv_u64m4_m(vbool16_t mask, vuint64m4_t acc, vuint64m4_t op1, vuint64m4_t op2, size_t vl) {
@@ -1563,7 +1563,7 @@ vuint64m4_t test_vnmsub_vv_u64m4_m(vbool16_t mask, vuint64m4_t acc, vuint64m4_t
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsub.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vnmsub.mask.nxv4i64.i64.i64(<vscale x 4 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 4 x i64> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vnmsub_vx_u64m4_m(vbool16_t mask, vuint64m4_t acc, uint64_t op1, vuint64m4_t op2, size_t vl) {
@@ -1572,7 +1572,7 @@ vuint64m4_t test_vnmsub_vx_u64m4_m(vbool16_t mask, vuint64m4_t acc, uint64_t op1
 
 // CHECK-RV64-LABEL: @test_vnmsub_vv_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsub.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsub.mask.nxv8i64.nxv8i64.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vnmsub_vv_u64m8_m(vbool8_t mask, vuint64m8_t acc, vuint64m8_t op1, vuint64m8_t op2, size_t vl) {
@@ -1581,7 +1581,7 @@ vuint64m8_t test_vnmsub_vv_u64m8_m(vbool8_t mask, vuint64m8_t acc, vuint64m8_t o
 
 // CHECK-RV64-LABEL: @test_vnmsub_vx_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsub.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vnmsub.mask.nxv8i64.i64.i64(<vscale x 8 x i64> [[ACC:%.*]], i64 [[OP1:%.*]], <vscale x 8 x i64> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vnmsub_vx_u64m8_m(vbool8_t mask, vuint64m8_t acc, uint64_t op1, vuint64m8_t op2, size_t vl) {
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics/vslidedown.c b/clang/test/CodeGen/RISCV/rvv-intrinsics/vslidedown.c
index 86b7f9f0ff4a3d..96b6233920f940 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics/vslidedown.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics/vslidedown.c
@@ -538,7 +538,7 @@ vfloat64m8_t test_vslidedown_vx_f64m8(vfloat64m8_t dst, vfloat64m8_t src,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vslidedown.mask.nxv1i8.i64(<vscale x 1 x i8> [[DST:%.*]], <vscale x 1 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vslidedown.mask.nxv1i8.i64(<vscale x 1 x i8> [[DST:%.*]], <vscale x 1 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vint8mf8_t test_vslidedown_vx_i8mf8_m(vbool64_t mask, vint8mf8_t dst,
@@ -549,7 +549,7 @@ vint8mf8_t test_vslidedown_vx_i8mf8_m(vbool64_t mask, vint8mf8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vslidedown.mask.nxv2i8.i64(<vscale x 2 x i8> [[DST:%.*]], <vscale x 2 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vslidedown.mask.nxv2i8.i64(<vscale x 2 x i8> [[DST:%.*]], <vscale x 2 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vint8mf4_t test_vslidedown_vx_i8mf4_m(vbool32_t mask, vint8mf4_t dst,
@@ -560,7 +560,7 @@ vint8mf4_t test_vslidedown_vx_i8mf4_m(vbool32_t mask, vint8mf4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vslidedown.mask.nxv4i8.i64(<vscale x 4 x i8> [[DST:%.*]], <vscale x 4 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vslidedown.mask.nxv4i8.i64(<vscale x 4 x i8> [[DST:%.*]], <vscale x 4 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vint8mf2_t test_vslidedown_vx_i8mf2_m(vbool16_t mask, vint8mf2_t dst,
@@ -571,7 +571,7 @@ vint8mf2_t test_vslidedown_vx_i8mf2_m(vbool16_t mask, vint8mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vslidedown.mask.nxv8i8.i64(<vscale x 8 x i8> [[DST:%.*]], <vscale x 8 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vslidedown.mask.nxv8i8.i64(<vscale x 8 x i8> [[DST:%.*]], <vscale x 8 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vint8m1_t test_vslidedown_vx_i8m1_m(vbool8_t mask, vint8m1_t dst, vint8m1_t src,
@@ -581,7 +581,7 @@ vint8m1_t test_vslidedown_vx_i8m1_m(vbool8_t mask, vint8m1_t dst, vint8m1_t src,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vslidedown.mask.nxv16i8.i64(<vscale x 16 x i8> [[DST:%.*]], <vscale x 16 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vslidedown.mask.nxv16i8.i64(<vscale x 16 x i8> [[DST:%.*]], <vscale x 16 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vint8m2_t test_vslidedown_vx_i8m2_m(vbool4_t mask, vint8m2_t dst, vint8m2_t src,
@@ -591,7 +591,7 @@ vint8m2_t test_vslidedown_vx_i8m2_m(vbool4_t mask, vint8m2_t dst, vint8m2_t src,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vslidedown.mask.nxv32i8.i64(<vscale x 32 x i8> [[DST:%.*]], <vscale x 32 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vslidedown.mask.nxv32i8.i64(<vscale x 32 x i8> [[DST:%.*]], <vscale x 32 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vint8m4_t test_vslidedown_vx_i8m4_m(vbool2_t mask, vint8m4_t dst, vint8m4_t src,
@@ -601,7 +601,7 @@ vint8m4_t test_vslidedown_vx_i8m4_m(vbool2_t mask, vint8m4_t dst, vint8m4_t src,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vslidedown.mask.nxv64i8.i64(<vscale x 64 x i8> [[DST:%.*]], <vscale x 64 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vslidedown.mask.nxv64i8.i64(<vscale x 64 x i8> [[DST:%.*]], <vscale x 64 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vint8m8_t test_vslidedown_vx_i8m8_m(vbool1_t mask, vint8m8_t dst, vint8m8_t src,
@@ -611,7 +611,7 @@ vint8m8_t test_vslidedown_vx_i8m8_m(vbool1_t mask, vint8m8_t dst, vint8m8_t src,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vslidedown.mask.nxv1i16.i64(<vscale x 1 x i16> [[DST:%.*]], <vscale x 1 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vslidedown.mask.nxv1i16.i64(<vscale x 1 x i16> [[DST:%.*]], <vscale x 1 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vslidedown_vx_i16mf4_m(vbool64_t mask, vint16mf4_t dst,
@@ -622,7 +622,7 @@ vint16mf4_t test_vslidedown_vx_i16mf4_m(vbool64_t mask, vint16mf4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vslidedown.mask.nxv2i16.i64(<vscale x 2 x i16> [[DST:%.*]], <vscale x 2 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vslidedown.mask.nxv2i16.i64(<vscale x 2 x i16> [[DST:%.*]], <vscale x 2 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vslidedown_vx_i16mf2_m(vbool32_t mask, vint16mf2_t dst,
@@ -633,7 +633,7 @@ vint16mf2_t test_vslidedown_vx_i16mf2_m(vbool32_t mask, vint16mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vslidedown.mask.nxv4i16.i64(<vscale x 4 x i16> [[DST:%.*]], <vscale x 4 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vslidedown.mask.nxv4i16.i64(<vscale x 4 x i16> [[DST:%.*]], <vscale x 4 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vslidedown_vx_i16m1_m(vbool16_t mask, vint16m1_t dst,
@@ -644,7 +644,7 @@ vint16m1_t test_vslidedown_vx_i16m1_m(vbool16_t mask, vint16m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vslidedown.mask.nxv8i16.i64(<vscale x 8 x i16> [[DST:%.*]], <vscale x 8 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vslidedown.mask.nxv8i16.i64(<vscale x 8 x i16> [[DST:%.*]], <vscale x 8 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vslidedown_vx_i16m2_m(vbool8_t mask, vint16m2_t dst,
@@ -655,7 +655,7 @@ vint16m2_t test_vslidedown_vx_i16m2_m(vbool8_t mask, vint16m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vslidedown.mask.nxv16i16.i64(<vscale x 16 x i16> [[DST:%.*]], <vscale x 16 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vslidedown.mask.nxv16i16.i64(<vscale x 16 x i16> [[DST:%.*]], <vscale x 16 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vslidedown_vx_i16m4_m(vbool4_t mask, vint16m4_t dst,
@@ -666,7 +666,7 @@ vint16m4_t test_vslidedown_vx_i16m4_m(vbool4_t mask, vint16m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vslidedown.mask.nxv32i16.i64(<vscale x 32 x i16> [[DST:%.*]], <vscale x 32 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vslidedown.mask.nxv32i16.i64(<vscale x 32 x i16> [[DST:%.*]], <vscale x 32 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vslidedown_vx_i16m8_m(vbool2_t mask, vint16m8_t dst,
@@ -677,7 +677,7 @@ vint16m8_t test_vslidedown_vx_i16m8_m(vbool2_t mask, vint16m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vslidedown.mask.nxv1i32.i64(<vscale x 1 x i32> [[DST:%.*]], <vscale x 1 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vslidedown.mask.nxv1i32.i64(<vscale x 1 x i32> [[DST:%.*]], <vscale x 1 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vslidedown_vx_i32mf2_m(vbool64_t mask, vint32mf2_t dst,
@@ -688,7 +688,7 @@ vint32mf2_t test_vslidedown_vx_i32mf2_m(vbool64_t mask, vint32mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vslidedown.mask.nxv2i32.i64(<vscale x 2 x i32> [[DST:%.*]], <vscale x 2 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vslidedown.mask.nxv2i32.i64(<vscale x 2 x i32> [[DST:%.*]], <vscale x 2 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vslidedown_vx_i32m1_m(vbool32_t mask, vint32m1_t dst,
@@ -699,7 +699,7 @@ vint32m1_t test_vslidedown_vx_i32m1_m(vbool32_t mask, vint32m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vslidedown.mask.nxv4i32.i64(<vscale x 4 x i32> [[DST:%.*]], <vscale x 4 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vslidedown.mask.nxv4i32.i64(<vscale x 4 x i32> [[DST:%.*]], <vscale x 4 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vslidedown_vx_i32m2_m(vbool16_t mask, vint32m2_t dst,
@@ -710,7 +710,7 @@ vint32m2_t test_vslidedown_vx_i32m2_m(vbool16_t mask, vint32m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vslidedown.mask.nxv8i32.i64(<vscale x 8 x i32> [[DST:%.*]], <vscale x 8 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vslidedown.mask.nxv8i32.i64(<vscale x 8 x i32> [[DST:%.*]], <vscale x 8 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vslidedown_vx_i32m4_m(vbool8_t mask, vint32m4_t dst,
@@ -721,7 +721,7 @@ vint32m4_t test_vslidedown_vx_i32m4_m(vbool8_t mask, vint32m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vslidedown.mask.nxv16i32.i64(<vscale x 16 x i32> [[DST:%.*]], <vscale x 16 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vslidedown.mask.nxv16i32.i64(<vscale x 16 x i32> [[DST:%.*]], <vscale x 16 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vslidedown_vx_i32m8_m(vbool4_t mask, vint32m8_t dst,
@@ -732,7 +732,7 @@ vint32m8_t test_vslidedown_vx_i32m8_m(vbool4_t mask, vint32m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vslidedown.mask.nxv1i64.i64(<vscale x 1 x i64> [[DST:%.*]], <vscale x 1 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vslidedown.mask.nxv1i64.i64(<vscale x 1 x i64> [[DST:%.*]], <vscale x 1 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vslidedown_vx_i64m1_m(vbool64_t mask, vint64m1_t dst,
@@ -743,7 +743,7 @@ vint64m1_t test_vslidedown_vx_i64m1_m(vbool64_t mask, vint64m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vslidedown.mask.nxv2i64.i64(<vscale x 2 x i64> [[DST:%.*]], <vscale x 2 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vslidedown.mask.nxv2i64.i64(<vscale x 2 x i64> [[DST:%.*]], <vscale x 2 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vslidedown_vx_i64m2_m(vbool32_t mask, vint64m2_t dst,
@@ -754,7 +754,7 @@ vint64m2_t test_vslidedown_vx_i64m2_m(vbool32_t mask, vint64m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vslidedown.mask.nxv4i64.i64(<vscale x 4 x i64> [[DST:%.*]], <vscale x 4 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vslidedown.mask.nxv4i64.i64(<vscale x 4 x i64> [[DST:%.*]], <vscale x 4 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vslidedown_vx_i64m4_m(vbool16_t mask, vint64m4_t dst,
@@ -765,7 +765,7 @@ vint64m4_t test_vslidedown_vx_i64m4_m(vbool16_t mask, vint64m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vslidedown.mask.nxv8i64.i64(<vscale x 8 x i64> [[DST:%.*]], <vscale x 8 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vslidedown.mask.nxv8i64.i64(<vscale x 8 x i64> [[DST:%.*]], <vscale x 8 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vslidedown_vx_i64m8_m(vbool8_t mask, vint64m8_t dst,
@@ -776,7 +776,7 @@ vint64m8_t test_vslidedown_vx_i64m8_m(vbool8_t mask, vint64m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vslidedown.mask.nxv1i8.i64(<vscale x 1 x i8> [[DST:%.*]], <vscale x 1 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vslidedown.mask.nxv1i8.i64(<vscale x 1 x i8> [[DST:%.*]], <vscale x 1 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vuint8mf8_t test_vslidedown_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t dst,
@@ -787,7 +787,7 @@ vuint8mf8_t test_vslidedown_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vslidedown.mask.nxv2i8.i64(<vscale x 2 x i8> [[DST:%.*]], <vscale x 2 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vslidedown.mask.nxv2i8.i64(<vscale x 2 x i8> [[DST:%.*]], <vscale x 2 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vuint8mf4_t test_vslidedown_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t dst,
@@ -798,7 +798,7 @@ vuint8mf4_t test_vslidedown_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vslidedown.mask.nxv4i8.i64(<vscale x 4 x i8> [[DST:%.*]], <vscale x 4 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vslidedown.mask.nxv4i8.i64(<vscale x 4 x i8> [[DST:%.*]], <vscale x 4 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vuint8mf2_t test_vslidedown_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t dst,
@@ -809,7 +809,7 @@ vuint8mf2_t test_vslidedown_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vslidedown.mask.nxv8i8.i64(<vscale x 8 x i8> [[DST:%.*]], <vscale x 8 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vslidedown.mask.nxv8i8.i64(<vscale x 8 x i8> [[DST:%.*]], <vscale x 8 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vuint8m1_t test_vslidedown_vx_u8m1_m(vbool8_t mask, vuint8m1_t dst,
@@ -819,7 +819,7 @@ vuint8m1_t test_vslidedown_vx_u8m1_m(vbool8_t mask, vuint8m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vslidedown.mask.nxv16i8.i64(<vscale x 16 x i8> [[DST:%.*]], <vscale x 16 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vslidedown.mask.nxv16i8.i64(<vscale x 16 x i8> [[DST:%.*]], <vscale x 16 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vuint8m2_t test_vslidedown_vx_u8m2_m(vbool4_t mask, vuint8m2_t dst,
@@ -829,7 +829,7 @@ vuint8m2_t test_vslidedown_vx_u8m2_m(vbool4_t mask, vuint8m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vslidedown.mask.nxv32i8.i64(<vscale x 32 x i8> [[DST:%.*]], <vscale x 32 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vslidedown.mask.nxv32i8.i64(<vscale x 32 x i8> [[DST:%.*]], <vscale x 32 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vuint8m4_t test_vslidedown_vx_u8m4_m(vbool2_t mask, vuint8m4_t dst,
@@ -839,7 +839,7 @@ vuint8m4_t test_vslidedown_vx_u8m4_m(vbool2_t mask, vuint8m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vslidedown.mask.nxv64i8.i64(<vscale x 64 x i8> [[DST:%.*]], <vscale x 64 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vslidedown.mask.nxv64i8.i64(<vscale x 64 x i8> [[DST:%.*]], <vscale x 64 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vuint8m8_t test_vslidedown_vx_u8m8_m(vbool1_t mask, vuint8m8_t dst,
@@ -849,7 +849,7 @@ vuint8m8_t test_vslidedown_vx_u8m8_m(vbool1_t mask, vuint8m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vslidedown.mask.nxv1i16.i64(<vscale x 1 x i16> [[DST:%.*]], <vscale x 1 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vslidedown.mask.nxv1i16.i64(<vscale x 1 x i16> [[DST:%.*]], <vscale x 1 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vslidedown_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t dst,
@@ -860,7 +860,7 @@ vuint16mf4_t test_vslidedown_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vslidedown.mask.nxv2i16.i64(<vscale x 2 x i16> [[DST:%.*]], <vscale x 2 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vslidedown.mask.nxv2i16.i64(<vscale x 2 x i16> [[DST:%.*]], <vscale x 2 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vslidedown_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t dst,
@@ -871,7 +871,7 @@ vuint16mf2_t test_vslidedown_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vslidedown.mask.nxv4i16.i64(<vscale x 4 x i16> [[DST:%.*]], <vscale x 4 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vslidedown.mask.nxv4i16.i64(<vscale x 4 x i16> [[DST:%.*]], <vscale x 4 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vslidedown_vx_u16m1_m(vbool16_t mask, vuint16m1_t dst,
@@ -882,7 +882,7 @@ vuint16m1_t test_vslidedown_vx_u16m1_m(vbool16_t mask, vuint16m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vslidedown.mask.nxv8i16.i64(<vscale x 8 x i16> [[DST:%.*]], <vscale x 8 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vslidedown.mask.nxv8i16.i64(<vscale x 8 x i16> [[DST:%.*]], <vscale x 8 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vslidedown_vx_u16m2_m(vbool8_t mask, vuint16m2_t dst,
@@ -893,7 +893,7 @@ vuint16m2_t test_vslidedown_vx_u16m2_m(vbool8_t mask, vuint16m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vslidedown.mask.nxv16i16.i64(<vscale x 16 x i16> [[DST:%.*]], <vscale x 16 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vslidedown.mask.nxv16i16.i64(<vscale x 16 x i16> [[DST:%.*]], <vscale x 16 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vslidedown_vx_u16m4_m(vbool4_t mask, vuint16m4_t dst,
@@ -904,7 +904,7 @@ vuint16m4_t test_vslidedown_vx_u16m4_m(vbool4_t mask, vuint16m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vslidedown.mask.nxv32i16.i64(<vscale x 32 x i16> [[DST:%.*]], <vscale x 32 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vslidedown.mask.nxv32i16.i64(<vscale x 32 x i16> [[DST:%.*]], <vscale x 32 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vslidedown_vx_u16m8_m(vbool2_t mask, vuint16m8_t dst,
@@ -915,7 +915,7 @@ vuint16m8_t test_vslidedown_vx_u16m8_m(vbool2_t mask, vuint16m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vslidedown.mask.nxv1i32.i64(<vscale x 1 x i32> [[DST:%.*]], <vscale x 1 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vslidedown.mask.nxv1i32.i64(<vscale x 1 x i32> [[DST:%.*]], <vscale x 1 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vslidedown_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t dst,
@@ -926,7 +926,7 @@ vuint32mf2_t test_vslidedown_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vslidedown.mask.nxv2i32.i64(<vscale x 2 x i32> [[DST:%.*]], <vscale x 2 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vslidedown.mask.nxv2i32.i64(<vscale x 2 x i32> [[DST:%.*]], <vscale x 2 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vslidedown_vx_u32m1_m(vbool32_t mask, vuint32m1_t dst,
@@ -937,7 +937,7 @@ vuint32m1_t test_vslidedown_vx_u32m1_m(vbool32_t mask, vuint32m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vslidedown.mask.nxv4i32.i64(<vscale x 4 x i32> [[DST:%.*]], <vscale x 4 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vslidedown.mask.nxv4i32.i64(<vscale x 4 x i32> [[DST:%.*]], <vscale x 4 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vslidedown_vx_u32m2_m(vbool16_t mask, vuint32m2_t dst,
@@ -948,7 +948,7 @@ vuint32m2_t test_vslidedown_vx_u32m2_m(vbool16_t mask, vuint32m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vslidedown.mask.nxv8i32.i64(<vscale x 8 x i32> [[DST:%.*]], <vscale x 8 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vslidedown.mask.nxv8i32.i64(<vscale x 8 x i32> [[DST:%.*]], <vscale x 8 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vslidedown_vx_u32m4_m(vbool8_t mask, vuint32m4_t dst,
@@ -959,7 +959,7 @@ vuint32m4_t test_vslidedown_vx_u32m4_m(vbool8_t mask, vuint32m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vslidedown.mask.nxv16i32.i64(<vscale x 16 x i32> [[DST:%.*]], <vscale x 16 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vslidedown.mask.nxv16i32.i64(<vscale x 16 x i32> [[DST:%.*]], <vscale x 16 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vslidedown_vx_u32m8_m(vbool4_t mask, vuint32m8_t dst,
@@ -970,7 +970,7 @@ vuint32m8_t test_vslidedown_vx_u32m8_m(vbool4_t mask, vuint32m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vslidedown.mask.nxv1i64.i64(<vscale x 1 x i64> [[DST:%.*]], <vscale x 1 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vslidedown.mask.nxv1i64.i64(<vscale x 1 x i64> [[DST:%.*]], <vscale x 1 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vslidedown_vx_u64m1_m(vbool64_t mask, vuint64m1_t dst,
@@ -981,7 +981,7 @@ vuint64m1_t test_vslidedown_vx_u64m1_m(vbool64_t mask, vuint64m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vslidedown.mask.nxv2i64.i64(<vscale x 2 x i64> [[DST:%.*]], <vscale x 2 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vslidedown.mask.nxv2i64.i64(<vscale x 2 x i64> [[DST:%.*]], <vscale x 2 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vslidedown_vx_u64m2_m(vbool32_t mask, vuint64m2_t dst,
@@ -992,7 +992,7 @@ vuint64m2_t test_vslidedown_vx_u64m2_m(vbool32_t mask, vuint64m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vslidedown.mask.nxv4i64.i64(<vscale x 4 x i64> [[DST:%.*]], <vscale x 4 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vslidedown.mask.nxv4i64.i64(<vscale x 4 x i64> [[DST:%.*]], <vscale x 4 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vslidedown_vx_u64m4_m(vbool16_t mask, vuint64m4_t dst,
@@ -1003,7 +1003,7 @@ vuint64m4_t test_vslidedown_vx_u64m4_m(vbool16_t mask, vuint64m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vslidedown.mask.nxv8i64.i64(<vscale x 8 x i64> [[DST:%.*]], <vscale x 8 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vslidedown.mask.nxv8i64.i64(<vscale x 8 x i64> [[DST:%.*]], <vscale x 8 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vslidedown_vx_u64m8_m(vbool8_t mask, vuint64m8_t dst,
@@ -1014,7 +1014,7 @@ vuint64m8_t test_vslidedown_vx_u64m8_m(vbool8_t mask, vuint64m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vslidedown.mask.nxv1f32.i64(<vscale x 1 x float> [[DST:%.*]], <vscale x 1 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vslidedown.mask.nxv1f32.i64(<vscale x 1 x float> [[DST:%.*]], <vscale x 1 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vslidedown_vx_f32mf2_m(vbool64_t mask, vfloat32mf2_t dst,
@@ -1025,7 +1025,7 @@ vfloat32mf2_t test_vslidedown_vx_f32mf2_m(vbool64_t mask, vfloat32mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vslidedown.mask.nxv2f32.i64(<vscale x 2 x float> [[DST:%.*]], <vscale x 2 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vslidedown.mask.nxv2f32.i64(<vscale x 2 x float> [[DST:%.*]], <vscale x 2 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vslidedown_vx_f32m1_m(vbool32_t mask, vfloat32m1_t dst,
@@ -1036,7 +1036,7 @@ vfloat32m1_t test_vslidedown_vx_f32m1_m(vbool32_t mask, vfloat32m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vslidedown.mask.nxv4f32.i64(<vscale x 4 x float> [[DST:%.*]], <vscale x 4 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vslidedown.mask.nxv4f32.i64(<vscale x 4 x float> [[DST:%.*]], <vscale x 4 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vslidedown_vx_f32m2_m(vbool16_t mask, vfloat32m2_t dst,
@@ -1047,7 +1047,7 @@ vfloat32m2_t test_vslidedown_vx_f32m2_m(vbool16_t mask, vfloat32m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vslidedown.mask.nxv8f32.i64(<vscale x 8 x float> [[DST:%.*]], <vscale x 8 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vslidedown.mask.nxv8f32.i64(<vscale x 8 x float> [[DST:%.*]], <vscale x 8 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vslidedown_vx_f32m4_m(vbool8_t mask, vfloat32m4_t dst,
@@ -1058,7 +1058,7 @@ vfloat32m4_t test_vslidedown_vx_f32m4_m(vbool8_t mask, vfloat32m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vslidedown.mask.nxv16f32.i64(<vscale x 16 x float> [[DST:%.*]], <vscale x 16 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vslidedown.mask.nxv16f32.i64(<vscale x 16 x float> [[DST:%.*]], <vscale x 16 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vslidedown_vx_f32m8_m(vbool4_t mask, vfloat32m8_t dst,
@@ -1069,7 +1069,7 @@ vfloat32m8_t test_vslidedown_vx_f32m8_m(vbool4_t mask, vfloat32m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vslidedown.mask.nxv1f64.i64(<vscale x 1 x double> [[DST:%.*]], <vscale x 1 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vslidedown.mask.nxv1f64.i64(<vscale x 1 x double> [[DST:%.*]], <vscale x 1 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vslidedown_vx_f64m1_m(vbool64_t mask, vfloat64m1_t dst,
@@ -1080,7 +1080,7 @@ vfloat64m1_t test_vslidedown_vx_f64m1_m(vbool64_t mask, vfloat64m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vslidedown.mask.nxv2f64.i64(<vscale x 2 x double> [[DST:%.*]], <vscale x 2 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vslidedown.mask.nxv2f64.i64(<vscale x 2 x double> [[DST:%.*]], <vscale x 2 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vslidedown_vx_f64m2_m(vbool32_t mask, vfloat64m2_t dst,
@@ -1091,7 +1091,7 @@ vfloat64m2_t test_vslidedown_vx_f64m2_m(vbool32_t mask, vfloat64m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vslidedown.mask.nxv4f64.i64(<vscale x 4 x double> [[DST:%.*]], <vscale x 4 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vslidedown.mask.nxv4f64.i64(<vscale x 4 x double> [[DST:%.*]], <vscale x 4 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vslidedown_vx_f64m4_m(vbool16_t mask, vfloat64m4_t dst,
@@ -1102,7 +1102,7 @@ vfloat64m4_t test_vslidedown_vx_f64m4_m(vbool16_t mask, vfloat64m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vslidedown.mask.nxv8f64.i64(<vscale x 8 x double> [[DST:%.*]], <vscale x 8 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vslidedown.mask.nxv8f64.i64(<vscale x 8 x double> [[DST:%.*]], <vscale x 8 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vslidedown_vx_f64m8_m(vbool8_t mask, vfloat64m8_t dst,
@@ -1167,7 +1167,7 @@ vfloat16m8_t test_vslidedown_vx_f16m8 (vfloat16m8_t dest, vfloat16m8_t src, size
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vslidedown.mask.nxv1f16.i64(<vscale x 1 x half> [[DEST:%.*]], <vscale x 1 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vslidedown.mask.nxv1f16.i64(<vscale x 1 x half> [[DEST:%.*]], <vscale x 1 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x half> [[TMP0]]
 //
 vfloat16mf4_t test_vslidedown_vx_f16mf4_m (vbool64_t mask, vfloat16mf4_t dest, vfloat16mf4_t src, size_t offset, size_t vl) {
@@ -1176,7 +1176,7 @@ vfloat16mf4_t test_vslidedown_vx_f16mf4_m (vbool64_t mask, vfloat16mf4_t dest, v
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vslidedown.mask.nxv2f16.i64(<vscale x 2 x half> [[DEST:%.*]], <vscale x 2 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vslidedown.mask.nxv2f16.i64(<vscale x 2 x half> [[DEST:%.*]], <vscale x 2 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x half> [[TMP0]]
 //
 vfloat16mf2_t test_vslidedown_vx_f16mf2_m (vbool32_t mask, vfloat16mf2_t dest, vfloat16mf2_t src, size_t offset, size_t vl) {
@@ -1185,7 +1185,7 @@ vfloat16mf2_t test_vslidedown_vx_f16mf2_m (vbool32_t mask, vfloat16mf2_t dest, v
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vslidedown.mask.nxv4f16.i64(<vscale x 4 x half> [[DEST:%.*]], <vscale x 4 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vslidedown.mask.nxv4f16.i64(<vscale x 4 x half> [[DEST:%.*]], <vscale x 4 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x half> [[TMP0]]
 //
 vfloat16m1_t test_vslidedown_vx_f16m1_m (vbool16_t mask, vfloat16m1_t dest, vfloat16m1_t src, size_t offset, size_t vl) {
@@ -1194,7 +1194,7 @@ vfloat16m1_t test_vslidedown_vx_f16m1_m (vbool16_t mask, vfloat16m1_t dest, vflo
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vslidedown.mask.nxv8f16.i64(<vscale x 8 x half> [[DEST:%.*]], <vscale x 8 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vslidedown.mask.nxv8f16.i64(<vscale x 8 x half> [[DEST:%.*]], <vscale x 8 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x half> [[TMP0]]
 //
 vfloat16m2_t test_vslidedown_vx_f16m2_m (vbool8_t mask, vfloat16m2_t dest, vfloat16m2_t src, size_t offset, size_t vl) {
@@ -1203,7 +1203,7 @@ vfloat16m2_t test_vslidedown_vx_f16m2_m (vbool8_t mask, vfloat16m2_t dest, vfloa
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vslidedown.mask.nxv16f16.i64(<vscale x 16 x half> [[DEST:%.*]], <vscale x 16 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vslidedown.mask.nxv16f16.i64(<vscale x 16 x half> [[DEST:%.*]], <vscale x 16 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x half> [[TMP0]]
 //
 vfloat16m4_t test_vslidedown_vx_f16m4_m (vbool4_t mask, vfloat16m4_t dest, vfloat16m4_t src, size_t offset, size_t vl) {
@@ -1212,7 +1212,7 @@ vfloat16m4_t test_vslidedown_vx_f16m4_m (vbool4_t mask, vfloat16m4_t dest, vfloa
 
 // CHECK-RV64-LABEL: @test_vslidedown_vx_f16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vslidedown.mask.nxv32f16.i64(<vscale x 32 x half> [[DEST:%.*]], <vscale x 32 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vslidedown.mask.nxv32f16.i64(<vscale x 32 x half> [[DEST:%.*]], <vscale x 32 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x half> [[TMP0]]
 //
 vfloat16m8_t test_vslidedown_vx_f16m8_m (vbool2_t mask, vfloat16m8_t dest, vfloat16m8_t src, size_t offset, size_t vl) {
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics/vslideup.c b/clang/test/CodeGen/RISCV/rvv-intrinsics/vslideup.c
index ca7cada4e51b6d..54718f2c109598 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics/vslideup.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics/vslideup.c
@@ -538,7 +538,7 @@ vfloat64m8_t test_vslideup_vx_f64m8(vfloat64m8_t dst, vfloat64m8_t src,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vslideup.mask.nxv1i8.i64(<vscale x 1 x i8> [[DST:%.*]], <vscale x 1 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vslideup.mask.nxv1i8.i64(<vscale x 1 x i8> [[DST:%.*]], <vscale x 1 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vint8mf8_t test_vslideup_vx_i8mf8_m(vbool64_t mask, vint8mf8_t dst,
@@ -548,7 +548,7 @@ vint8mf8_t test_vslideup_vx_i8mf8_m(vbool64_t mask, vint8mf8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vslideup.mask.nxv2i8.i64(<vscale x 2 x i8> [[DST:%.*]], <vscale x 2 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vslideup.mask.nxv2i8.i64(<vscale x 2 x i8> [[DST:%.*]], <vscale x 2 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vint8mf4_t test_vslideup_vx_i8mf4_m(vbool32_t mask, vint8mf4_t dst,
@@ -558,7 +558,7 @@ vint8mf4_t test_vslideup_vx_i8mf4_m(vbool32_t mask, vint8mf4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vslideup.mask.nxv4i8.i64(<vscale x 4 x i8> [[DST:%.*]], <vscale x 4 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vslideup.mask.nxv4i8.i64(<vscale x 4 x i8> [[DST:%.*]], <vscale x 4 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vint8mf2_t test_vslideup_vx_i8mf2_m(vbool16_t mask, vint8mf2_t dst,
@@ -568,7 +568,7 @@ vint8mf2_t test_vslideup_vx_i8mf2_m(vbool16_t mask, vint8mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vslideup.mask.nxv8i8.i64(<vscale x 8 x i8> [[DST:%.*]], <vscale x 8 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vslideup.mask.nxv8i8.i64(<vscale x 8 x i8> [[DST:%.*]], <vscale x 8 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vint8m1_t test_vslideup_vx_i8m1_m(vbool8_t mask, vint8m1_t dst, vint8m1_t src,
@@ -578,7 +578,7 @@ vint8m1_t test_vslideup_vx_i8m1_m(vbool8_t mask, vint8m1_t dst, vint8m1_t src,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vslideup.mask.nxv16i8.i64(<vscale x 16 x i8> [[DST:%.*]], <vscale x 16 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vslideup.mask.nxv16i8.i64(<vscale x 16 x i8> [[DST:%.*]], <vscale x 16 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vint8m2_t test_vslideup_vx_i8m2_m(vbool4_t mask, vint8m2_t dst, vint8m2_t src,
@@ -588,7 +588,7 @@ vint8m2_t test_vslideup_vx_i8m2_m(vbool4_t mask, vint8m2_t dst, vint8m2_t src,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vslideup.mask.nxv32i8.i64(<vscale x 32 x i8> [[DST:%.*]], <vscale x 32 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vslideup.mask.nxv32i8.i64(<vscale x 32 x i8> [[DST:%.*]], <vscale x 32 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vint8m4_t test_vslideup_vx_i8m4_m(vbool2_t mask, vint8m4_t dst, vint8m4_t src,
@@ -598,7 +598,7 @@ vint8m4_t test_vslideup_vx_i8m4_m(vbool2_t mask, vint8m4_t dst, vint8m4_t src,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vslideup.mask.nxv64i8.i64(<vscale x 64 x i8> [[DST:%.*]], <vscale x 64 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vslideup.mask.nxv64i8.i64(<vscale x 64 x i8> [[DST:%.*]], <vscale x 64 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vint8m8_t test_vslideup_vx_i8m8_m(vbool1_t mask, vint8m8_t dst, vint8m8_t src,
@@ -608,7 +608,7 @@ vint8m8_t test_vslideup_vx_i8m8_m(vbool1_t mask, vint8m8_t dst, vint8m8_t src,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vslideup.mask.nxv1i16.i64(<vscale x 1 x i16> [[DST:%.*]], <vscale x 1 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vslideup.mask.nxv1i16.i64(<vscale x 1 x i16> [[DST:%.*]], <vscale x 1 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vslideup_vx_i16mf4_m(vbool64_t mask, vint16mf4_t dst,
@@ -619,7 +619,7 @@ vint16mf4_t test_vslideup_vx_i16mf4_m(vbool64_t mask, vint16mf4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vslideup.mask.nxv2i16.i64(<vscale x 2 x i16> [[DST:%.*]], <vscale x 2 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vslideup.mask.nxv2i16.i64(<vscale x 2 x i16> [[DST:%.*]], <vscale x 2 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vslideup_vx_i16mf2_m(vbool32_t mask, vint16mf2_t dst,
@@ -630,7 +630,7 @@ vint16mf2_t test_vslideup_vx_i16mf2_m(vbool32_t mask, vint16mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vslideup.mask.nxv4i16.i64(<vscale x 4 x i16> [[DST:%.*]], <vscale x 4 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vslideup.mask.nxv4i16.i64(<vscale x 4 x i16> [[DST:%.*]], <vscale x 4 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vslideup_vx_i16m1_m(vbool16_t mask, vint16m1_t dst,
@@ -640,7 +640,7 @@ vint16m1_t test_vslideup_vx_i16m1_m(vbool16_t mask, vint16m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vslideup.mask.nxv8i16.i64(<vscale x 8 x i16> [[DST:%.*]], <vscale x 8 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vslideup.mask.nxv8i16.i64(<vscale x 8 x i16> [[DST:%.*]], <vscale x 8 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vslideup_vx_i16m2_m(vbool8_t mask, vint16m2_t dst,
@@ -650,7 +650,7 @@ vint16m2_t test_vslideup_vx_i16m2_m(vbool8_t mask, vint16m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vslideup.mask.nxv16i16.i64(<vscale x 16 x i16> [[DST:%.*]], <vscale x 16 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vslideup.mask.nxv16i16.i64(<vscale x 16 x i16> [[DST:%.*]], <vscale x 16 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vslideup_vx_i16m4_m(vbool4_t mask, vint16m4_t dst,
@@ -660,7 +660,7 @@ vint16m4_t test_vslideup_vx_i16m4_m(vbool4_t mask, vint16m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vslideup.mask.nxv32i16.i64(<vscale x 32 x i16> [[DST:%.*]], <vscale x 32 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vslideup.mask.nxv32i16.i64(<vscale x 32 x i16> [[DST:%.*]], <vscale x 32 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vslideup_vx_i16m8_m(vbool2_t mask, vint16m8_t dst,
@@ -670,7 +670,7 @@ vint16m8_t test_vslideup_vx_i16m8_m(vbool2_t mask, vint16m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vslideup.mask.nxv1i32.i64(<vscale x 1 x i32> [[DST:%.*]], <vscale x 1 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vslideup.mask.nxv1i32.i64(<vscale x 1 x i32> [[DST:%.*]], <vscale x 1 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vslideup_vx_i32mf2_m(vbool64_t mask, vint32mf2_t dst,
@@ -681,7 +681,7 @@ vint32mf2_t test_vslideup_vx_i32mf2_m(vbool64_t mask, vint32mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vslideup.mask.nxv2i32.i64(<vscale x 2 x i32> [[DST:%.*]], <vscale x 2 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vslideup.mask.nxv2i32.i64(<vscale x 2 x i32> [[DST:%.*]], <vscale x 2 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vslideup_vx_i32m1_m(vbool32_t mask, vint32m1_t dst,
@@ -691,7 +691,7 @@ vint32m1_t test_vslideup_vx_i32m1_m(vbool32_t mask, vint32m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vslideup.mask.nxv4i32.i64(<vscale x 4 x i32> [[DST:%.*]], <vscale x 4 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vslideup.mask.nxv4i32.i64(<vscale x 4 x i32> [[DST:%.*]], <vscale x 4 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vslideup_vx_i32m2_m(vbool16_t mask, vint32m2_t dst,
@@ -701,7 +701,7 @@ vint32m2_t test_vslideup_vx_i32m2_m(vbool16_t mask, vint32m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vslideup.mask.nxv8i32.i64(<vscale x 8 x i32> [[DST:%.*]], <vscale x 8 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vslideup.mask.nxv8i32.i64(<vscale x 8 x i32> [[DST:%.*]], <vscale x 8 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vslideup_vx_i32m4_m(vbool8_t mask, vint32m4_t dst,
@@ -711,7 +711,7 @@ vint32m4_t test_vslideup_vx_i32m4_m(vbool8_t mask, vint32m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vslideup.mask.nxv16i32.i64(<vscale x 16 x i32> [[DST:%.*]], <vscale x 16 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vslideup.mask.nxv16i32.i64(<vscale x 16 x i32> [[DST:%.*]], <vscale x 16 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vslideup_vx_i32m8_m(vbool4_t mask, vint32m8_t dst,
@@ -721,7 +721,7 @@ vint32m8_t test_vslideup_vx_i32m8_m(vbool4_t mask, vint32m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vslideup.mask.nxv1i64.i64(<vscale x 1 x i64> [[DST:%.*]], <vscale x 1 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vslideup.mask.nxv1i64.i64(<vscale x 1 x i64> [[DST:%.*]], <vscale x 1 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vslideup_vx_i64m1_m(vbool64_t mask, vint64m1_t dst,
@@ -731,7 +731,7 @@ vint64m1_t test_vslideup_vx_i64m1_m(vbool64_t mask, vint64m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vslideup.mask.nxv2i64.i64(<vscale x 2 x i64> [[DST:%.*]], <vscale x 2 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vslideup.mask.nxv2i64.i64(<vscale x 2 x i64> [[DST:%.*]], <vscale x 2 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vslideup_vx_i64m2_m(vbool32_t mask, vint64m2_t dst,
@@ -741,7 +741,7 @@ vint64m2_t test_vslideup_vx_i64m2_m(vbool32_t mask, vint64m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vslideup.mask.nxv4i64.i64(<vscale x 4 x i64> [[DST:%.*]], <vscale x 4 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vslideup.mask.nxv4i64.i64(<vscale x 4 x i64> [[DST:%.*]], <vscale x 4 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vslideup_vx_i64m4_m(vbool16_t mask, vint64m4_t dst,
@@ -751,7 +751,7 @@ vint64m4_t test_vslideup_vx_i64m4_m(vbool16_t mask, vint64m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vslideup.mask.nxv8i64.i64(<vscale x 8 x i64> [[DST:%.*]], <vscale x 8 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vslideup.mask.nxv8i64.i64(<vscale x 8 x i64> [[DST:%.*]], <vscale x 8 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vslideup_vx_i64m8_m(vbool8_t mask, vint64m8_t dst,
@@ -761,7 +761,7 @@ vint64m8_t test_vslideup_vx_i64m8_m(vbool8_t mask, vint64m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u8mf8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vslideup.mask.nxv1i8.i64(<vscale x 1 x i8> [[DST:%.*]], <vscale x 1 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vslideup.mask.nxv1i8.i64(<vscale x 1 x i8> [[DST:%.*]], <vscale x 1 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
 //
 vuint8mf8_t test_vslideup_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t dst,
@@ -772,7 +772,7 @@ vuint8mf8_t test_vslideup_vx_u8mf8_m(vbool64_t mask, vuint8mf8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u8mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vslideup.mask.nxv2i8.i64(<vscale x 2 x i8> [[DST:%.*]], <vscale x 2 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vslideup.mask.nxv2i8.i64(<vscale x 2 x i8> [[DST:%.*]], <vscale x 2 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
 //
 vuint8mf4_t test_vslideup_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t dst,
@@ -783,7 +783,7 @@ vuint8mf4_t test_vslideup_vx_u8mf4_m(vbool32_t mask, vuint8mf4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u8mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vslideup.mask.nxv4i8.i64(<vscale x 4 x i8> [[DST:%.*]], <vscale x 4 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vslideup.mask.nxv4i8.i64(<vscale x 4 x i8> [[DST:%.*]], <vscale x 4 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
 //
 vuint8mf2_t test_vslideup_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t dst,
@@ -794,7 +794,7 @@ vuint8mf2_t test_vslideup_vx_u8mf2_m(vbool16_t mask, vuint8mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u8m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vslideup.mask.nxv8i8.i64(<vscale x 8 x i8> [[DST:%.*]], <vscale x 8 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vslideup.mask.nxv8i8.i64(<vscale x 8 x i8> [[DST:%.*]], <vscale x 8 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
 //
 vuint8m1_t test_vslideup_vx_u8m1_m(vbool8_t mask, vuint8m1_t dst,
@@ -804,7 +804,7 @@ vuint8m1_t test_vslideup_vx_u8m1_m(vbool8_t mask, vuint8m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u8m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vslideup.mask.nxv16i8.i64(<vscale x 16 x i8> [[DST:%.*]], <vscale x 16 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vslideup.mask.nxv16i8.i64(<vscale x 16 x i8> [[DST:%.*]], <vscale x 16 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
 //
 vuint8m2_t test_vslideup_vx_u8m2_m(vbool4_t mask, vuint8m2_t dst,
@@ -814,7 +814,7 @@ vuint8m2_t test_vslideup_vx_u8m2_m(vbool4_t mask, vuint8m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u8m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vslideup.mask.nxv32i8.i64(<vscale x 32 x i8> [[DST:%.*]], <vscale x 32 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vslideup.mask.nxv32i8.i64(<vscale x 32 x i8> [[DST:%.*]], <vscale x 32 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
 //
 vuint8m4_t test_vslideup_vx_u8m4_m(vbool2_t mask, vuint8m4_t dst,
@@ -824,7 +824,7 @@ vuint8m4_t test_vslideup_vx_u8m4_m(vbool2_t mask, vuint8m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u8m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vslideup.mask.nxv64i8.i64(<vscale x 64 x i8> [[DST:%.*]], <vscale x 64 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.vslideup.mask.nxv64i8.i64(<vscale x 64 x i8> [[DST:%.*]], <vscale x 64 x i8> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 64 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 64 x i8> [[TMP0]]
 //
 vuint8m8_t test_vslideup_vx_u8m8_m(vbool1_t mask, vuint8m8_t dst,
@@ -834,7 +834,7 @@ vuint8m8_t test_vslideup_vx_u8m8_m(vbool1_t mask, vuint8m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vslideup.mask.nxv1i16.i64(<vscale x 1 x i16> [[DST:%.*]], <vscale x 1 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vslideup.mask.nxv1i16.i64(<vscale x 1 x i16> [[DST:%.*]], <vscale x 1 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vslideup_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t dst,
@@ -845,7 +845,7 @@ vuint16mf4_t test_vslideup_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vslideup.mask.nxv2i16.i64(<vscale x 2 x i16> [[DST:%.*]], <vscale x 2 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vslideup.mask.nxv2i16.i64(<vscale x 2 x i16> [[DST:%.*]], <vscale x 2 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vslideup_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t dst,
@@ -856,7 +856,7 @@ vuint16mf2_t test_vslideup_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vslideup.mask.nxv4i16.i64(<vscale x 4 x i16> [[DST:%.*]], <vscale x 4 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vslideup.mask.nxv4i16.i64(<vscale x 4 x i16> [[DST:%.*]], <vscale x 4 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vslideup_vx_u16m1_m(vbool16_t mask, vuint16m1_t dst,
@@ -867,7 +867,7 @@ vuint16m1_t test_vslideup_vx_u16m1_m(vbool16_t mask, vuint16m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vslideup.mask.nxv8i16.i64(<vscale x 8 x i16> [[DST:%.*]], <vscale x 8 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vslideup.mask.nxv8i16.i64(<vscale x 8 x i16> [[DST:%.*]], <vscale x 8 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vslideup_vx_u16m2_m(vbool8_t mask, vuint16m2_t dst,
@@ -878,7 +878,7 @@ vuint16m2_t test_vslideup_vx_u16m2_m(vbool8_t mask, vuint16m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vslideup.mask.nxv16i16.i64(<vscale x 16 x i16> [[DST:%.*]], <vscale x 16 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vslideup.mask.nxv16i16.i64(<vscale x 16 x i16> [[DST:%.*]], <vscale x 16 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vslideup_vx_u16m4_m(vbool4_t mask, vuint16m4_t dst,
@@ -889,7 +889,7 @@ vuint16m4_t test_vslideup_vx_u16m4_m(vbool4_t mask, vuint16m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vslideup.mask.nxv32i16.i64(<vscale x 32 x i16> [[DST:%.*]], <vscale x 32 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vslideup.mask.nxv32i16.i64(<vscale x 32 x i16> [[DST:%.*]], <vscale x 32 x i16> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vslideup_vx_u16m8_m(vbool2_t mask, vuint16m8_t dst,
@@ -900,7 +900,7 @@ vuint16m8_t test_vslideup_vx_u16m8_m(vbool2_t mask, vuint16m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vslideup.mask.nxv1i32.i64(<vscale x 1 x i32> [[DST:%.*]], <vscale x 1 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vslideup.mask.nxv1i32.i64(<vscale x 1 x i32> [[DST:%.*]], <vscale x 1 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vslideup_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t dst,
@@ -911,7 +911,7 @@ vuint32mf2_t test_vslideup_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vslideup.mask.nxv2i32.i64(<vscale x 2 x i32> [[DST:%.*]], <vscale x 2 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vslideup.mask.nxv2i32.i64(<vscale x 2 x i32> [[DST:%.*]], <vscale x 2 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vslideup_vx_u32m1_m(vbool32_t mask, vuint32m1_t dst,
@@ -922,7 +922,7 @@ vuint32m1_t test_vslideup_vx_u32m1_m(vbool32_t mask, vuint32m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vslideup.mask.nxv4i32.i64(<vscale x 4 x i32> [[DST:%.*]], <vscale x 4 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vslideup.mask.nxv4i32.i64(<vscale x 4 x i32> [[DST:%.*]], <vscale x 4 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vslideup_vx_u32m2_m(vbool16_t mask, vuint32m2_t dst,
@@ -933,7 +933,7 @@ vuint32m2_t test_vslideup_vx_u32m2_m(vbool16_t mask, vuint32m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vslideup.mask.nxv8i32.i64(<vscale x 8 x i32> [[DST:%.*]], <vscale x 8 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vslideup.mask.nxv8i32.i64(<vscale x 8 x i32> [[DST:%.*]], <vscale x 8 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vslideup_vx_u32m4_m(vbool8_t mask, vuint32m4_t dst,
@@ -944,7 +944,7 @@ vuint32m4_t test_vslideup_vx_u32m4_m(vbool8_t mask, vuint32m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vslideup.mask.nxv16i32.i64(<vscale x 16 x i32> [[DST:%.*]], <vscale x 16 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vslideup.mask.nxv16i32.i64(<vscale x 16 x i32> [[DST:%.*]], <vscale x 16 x i32> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vslideup_vx_u32m8_m(vbool4_t mask, vuint32m8_t dst,
@@ -955,7 +955,7 @@ vuint32m8_t test_vslideup_vx_u32m8_m(vbool4_t mask, vuint32m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vslideup.mask.nxv1i64.i64(<vscale x 1 x i64> [[DST:%.*]], <vscale x 1 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vslideup.mask.nxv1i64.i64(<vscale x 1 x i64> [[DST:%.*]], <vscale x 1 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vslideup_vx_u64m1_m(vbool64_t mask, vuint64m1_t dst,
@@ -966,7 +966,7 @@ vuint64m1_t test_vslideup_vx_u64m1_m(vbool64_t mask, vuint64m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vslideup.mask.nxv2i64.i64(<vscale x 2 x i64> [[DST:%.*]], <vscale x 2 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vslideup.mask.nxv2i64.i64(<vscale x 2 x i64> [[DST:%.*]], <vscale x 2 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vslideup_vx_u64m2_m(vbool32_t mask, vuint64m2_t dst,
@@ -977,7 +977,7 @@ vuint64m2_t test_vslideup_vx_u64m2_m(vbool32_t mask, vuint64m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vslideup.mask.nxv4i64.i64(<vscale x 4 x i64> [[DST:%.*]], <vscale x 4 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vslideup.mask.nxv4i64.i64(<vscale x 4 x i64> [[DST:%.*]], <vscale x 4 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vslideup_vx_u64m4_m(vbool16_t mask, vuint64m4_t dst,
@@ -988,7 +988,7 @@ vuint64m4_t test_vslideup_vx_u64m4_m(vbool16_t mask, vuint64m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vslideup.mask.nxv8i64.i64(<vscale x 8 x i64> [[DST:%.*]], <vscale x 8 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vslideup.mask.nxv8i64.i64(<vscale x 8 x i64> [[DST:%.*]], <vscale x 8 x i64> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vslideup_vx_u64m8_m(vbool8_t mask, vuint64m8_t dst,
@@ -999,7 +999,7 @@ vuint64m8_t test_vslideup_vx_u64m8_m(vbool8_t mask, vuint64m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vslideup.mask.nxv1f32.i64(<vscale x 1 x float> [[DST:%.*]], <vscale x 1 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x float> @llvm.riscv.vslideup.mask.nxv1f32.i64(<vscale x 1 x float> [[DST:%.*]], <vscale x 1 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x float> [[TMP0]]
 //
 vfloat32mf2_t test_vslideup_vx_f32mf2_m(vbool64_t mask, vfloat32mf2_t dst,
@@ -1010,7 +1010,7 @@ vfloat32mf2_t test_vslideup_vx_f32mf2_m(vbool64_t mask, vfloat32mf2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vslideup.mask.nxv2f32.i64(<vscale x 2 x float> [[DST:%.*]], <vscale x 2 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x float> @llvm.riscv.vslideup.mask.nxv2f32.i64(<vscale x 2 x float> [[DST:%.*]], <vscale x 2 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x float> [[TMP0]]
 //
 vfloat32m1_t test_vslideup_vx_f32m1_m(vbool32_t mask, vfloat32m1_t dst,
@@ -1021,7 +1021,7 @@ vfloat32m1_t test_vslideup_vx_f32m1_m(vbool32_t mask, vfloat32m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vslideup.mask.nxv4f32.i64(<vscale x 4 x float> [[DST:%.*]], <vscale x 4 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x float> @llvm.riscv.vslideup.mask.nxv4f32.i64(<vscale x 4 x float> [[DST:%.*]], <vscale x 4 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x float> [[TMP0]]
 //
 vfloat32m2_t test_vslideup_vx_f32m2_m(vbool16_t mask, vfloat32m2_t dst,
@@ -1032,7 +1032,7 @@ vfloat32m2_t test_vslideup_vx_f32m2_m(vbool16_t mask, vfloat32m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vslideup.mask.nxv8f32.i64(<vscale x 8 x float> [[DST:%.*]], <vscale x 8 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x float> @llvm.riscv.vslideup.mask.nxv8f32.i64(<vscale x 8 x float> [[DST:%.*]], <vscale x 8 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x float> [[TMP0]]
 //
 vfloat32m4_t test_vslideup_vx_f32m4_m(vbool8_t mask, vfloat32m4_t dst,
@@ -1043,7 +1043,7 @@ vfloat32m4_t test_vslideup_vx_f32m4_m(vbool8_t mask, vfloat32m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vslideup.mask.nxv16f32.i64(<vscale x 16 x float> [[DST:%.*]], <vscale x 16 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.vslideup.mask.nxv16f32.i64(<vscale x 16 x float> [[DST:%.*]], <vscale x 16 x float> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x float> [[TMP0]]
 //
 vfloat32m8_t test_vslideup_vx_f32m8_m(vbool4_t mask, vfloat32m8_t dst,
@@ -1054,7 +1054,7 @@ vfloat32m8_t test_vslideup_vx_f32m8_m(vbool4_t mask, vfloat32m8_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vslideup.mask.nxv1f64.i64(<vscale x 1 x double> [[DST:%.*]], <vscale x 1 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x double> @llvm.riscv.vslideup.mask.nxv1f64.i64(<vscale x 1 x double> [[DST:%.*]], <vscale x 1 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x double> [[TMP0]]
 //
 vfloat64m1_t test_vslideup_vx_f64m1_m(vbool64_t mask, vfloat64m1_t dst,
@@ -1065,7 +1065,7 @@ vfloat64m1_t test_vslideup_vx_f64m1_m(vbool64_t mask, vfloat64m1_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vslideup.mask.nxv2f64.i64(<vscale x 2 x double> [[DST:%.*]], <vscale x 2 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x double> @llvm.riscv.vslideup.mask.nxv2f64.i64(<vscale x 2 x double> [[DST:%.*]], <vscale x 2 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x double> [[TMP0]]
 //
 vfloat64m2_t test_vslideup_vx_f64m2_m(vbool32_t mask, vfloat64m2_t dst,
@@ -1076,7 +1076,7 @@ vfloat64m2_t test_vslideup_vx_f64m2_m(vbool32_t mask, vfloat64m2_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vslideup.mask.nxv4f64.i64(<vscale x 4 x double> [[DST:%.*]], <vscale x 4 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x double> @llvm.riscv.vslideup.mask.nxv4f64.i64(<vscale x 4 x double> [[DST:%.*]], <vscale x 4 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x double> [[TMP0]]
 //
 vfloat64m4_t test_vslideup_vx_f64m4_m(vbool16_t mask, vfloat64m4_t dst,
@@ -1087,7 +1087,7 @@ vfloat64m4_t test_vslideup_vx_f64m4_m(vbool16_t mask, vfloat64m4_t dst,
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vslideup.mask.nxv8f64.i64(<vscale x 8 x double> [[DST:%.*]], <vscale x 8 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.vslideup.mask.nxv8f64.i64(<vscale x 8 x double> [[DST:%.*]], <vscale x 8 x double> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x double> [[TMP0]]
 //
 vfloat64m8_t test_vslideup_vx_f64m8_m(vbool8_t mask, vfloat64m8_t dst,
@@ -1152,7 +1152,7 @@ vfloat16m8_t test_vslideup_vx_f16m8 (vfloat16m8_t dest, vfloat16m8_t src, size_t
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vslideup.mask.nxv1f16.i64(<vscale x 1 x half> [[DEST:%.*]], <vscale x 1 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x half> @llvm.riscv.vslideup.mask.nxv1f16.i64(<vscale x 1 x half> [[DEST:%.*]], <vscale x 1 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x half> [[TMP0]]
 //
 vfloat16mf4_t test_vslideup_vx_f16mf4_m (vbool64_t mask, vfloat16mf4_t dest, vfloat16mf4_t src, size_t offset, size_t vl) {
@@ -1161,7 +1161,7 @@ vfloat16mf4_t test_vslideup_vx_f16mf4_m (vbool64_t mask, vfloat16mf4_t dest, vfl
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vslideup.mask.nxv2f16.i64(<vscale x 2 x half> [[DEST:%.*]], <vscale x 2 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x half> @llvm.riscv.vslideup.mask.nxv2f16.i64(<vscale x 2 x half> [[DEST:%.*]], <vscale x 2 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x half> [[TMP0]]
 //
 vfloat16mf2_t test_vslideup_vx_f16mf2_m (vbool32_t mask, vfloat16mf2_t dest, vfloat16mf2_t src, size_t offset, size_t vl) {
@@ -1170,7 +1170,7 @@ vfloat16mf2_t test_vslideup_vx_f16mf2_m (vbool32_t mask, vfloat16mf2_t dest, vfl
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vslideup.mask.nxv4f16.i64(<vscale x 4 x half> [[DEST:%.*]], <vscale x 4 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x half> @llvm.riscv.vslideup.mask.nxv4f16.i64(<vscale x 4 x half> [[DEST:%.*]], <vscale x 4 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x half> [[TMP0]]
 //
 vfloat16m1_t test_vslideup_vx_f16m1_m (vbool16_t mask, vfloat16m1_t dest, vfloat16m1_t src, size_t offset, size_t vl) {
@@ -1179,7 +1179,7 @@ vfloat16m1_t test_vslideup_vx_f16m1_m (vbool16_t mask, vfloat16m1_t dest, vfloat
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vslideup.mask.nxv8f16.i64(<vscale x 8 x half> [[DEST:%.*]], <vscale x 8 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x half> @llvm.riscv.vslideup.mask.nxv8f16.i64(<vscale x 8 x half> [[DEST:%.*]], <vscale x 8 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x half> [[TMP0]]
 //
 vfloat16m2_t test_vslideup_vx_f16m2_m (vbool8_t mask, vfloat16m2_t dest, vfloat16m2_t src, size_t offset, size_t vl) {
@@ -1188,7 +1188,7 @@ vfloat16m2_t test_vslideup_vx_f16m2_m (vbool8_t mask, vfloat16m2_t dest, vfloat1
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vslideup.mask.nxv16f16.i64(<vscale x 16 x half> [[DEST:%.*]], <vscale x 16 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x half> @llvm.riscv.vslideup.mask.nxv16f16.i64(<vscale x 16 x half> [[DEST:%.*]], <vscale x 16 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x half> [[TMP0]]
 //
 vfloat16m4_t test_vslideup_vx_f16m4_m (vbool4_t mask, vfloat16m4_t dest, vfloat16m4_t src, size_t offset, size_t vl) {
@@ -1197,7 +1197,7 @@ vfloat16m4_t test_vslideup_vx_f16m4_m (vbool4_t mask, vfloat16m4_t dest, vfloat1
 
 // CHECK-RV64-LABEL: @test_vslideup_vx_f16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vslideup.mask.nxv32f16.i64(<vscale x 32 x half> [[DEST:%.*]], <vscale x 32 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.vslideup.mask.nxv32f16.i64(<vscale x 32 x half> [[DEST:%.*]], <vscale x 32 x half> [[SRC:%.*]], i64 [[OFFSET:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x half> [[TMP0]]
 //
 vfloat16m8_t test_vslideup_vx_f16m8_m (vbool2_t mask, vfloat16m8_t dest, vfloat16m8_t src, size_t offset, size_t vl) {
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics/vwmacc.c b/clang/test/CodeGen/RISCV/rvv-intrinsics/vwmacc.c
index a0f993feb7c4de..f8eb6e6d111a16 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics/vwmacc.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics/vwmacc.c
@@ -1056,7 +1056,7 @@ vint64m8_t test_vwmaccus_vx_i64m8(vint64m8_t acc, uint32_t op1, vint32m4_t op2,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vwmacc_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc,
@@ -1066,7 +1066,7 @@ vint16mf4_t test_vwmacc_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vwmacc_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc, int8_t op1,
@@ -1076,7 +1076,7 @@ vint16mf4_t test_vwmacc_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc, int8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmacc.mask.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmacc.mask.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vwmacc_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc,
@@ -1086,7 +1086,7 @@ vint16mf2_t test_vwmacc_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmacc.mask.nxv2i16.i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmacc.mask.nxv2i16.i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vwmacc_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc, int8_t op1,
@@ -1096,7 +1096,7 @@ vint16mf2_t test_vwmacc_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc, int8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmacc.mask.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmacc.mask.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vwmacc_vv_i16m1_m(vbool16_t mask, vint16m1_t acc,
@@ -1106,7 +1106,7 @@ vint16m1_t test_vwmacc_vv_i16m1_m(vbool16_t mask, vint16m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmacc.mask.nxv4i16.i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmacc.mask.nxv4i16.i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vwmacc_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int8_t op1,
@@ -1116,7 +1116,7 @@ vint16m1_t test_vwmacc_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmacc.mask.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmacc.mask.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vwmacc_vv_i16m2_m(vbool8_t mask, vint16m2_t acc, vint8m1_t op1,
@@ -1126,7 +1126,7 @@ vint16m2_t test_vwmacc_vv_i16m2_m(vbool8_t mask, vint16m2_t acc, vint8m1_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmacc.mask.nxv8i16.i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmacc.mask.nxv8i16.i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vwmacc_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int8_t op1,
@@ -1136,7 +1136,7 @@ vint16m2_t test_vwmacc_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmacc.mask.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmacc.mask.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vwmacc_vv_i16m4_m(vbool4_t mask, vint16m4_t acc, vint8m2_t op1,
@@ -1146,7 +1146,7 @@ vint16m4_t test_vwmacc_vv_i16m4_m(vbool4_t mask, vint16m4_t acc, vint8m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmacc.mask.nxv16i16.i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmacc.mask.nxv16i16.i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vwmacc_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int8_t op1,
@@ -1156,7 +1156,7 @@ vint16m4_t test_vwmacc_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmacc.mask.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmacc.mask.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vwmacc_vv_i16m8_m(vbool2_t mask, vint16m8_t acc, vint8m4_t op1,
@@ -1166,7 +1166,7 @@ vint16m8_t test_vwmacc_vv_i16m8_m(vbool2_t mask, vint16m8_t acc, vint8m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmacc.mask.nxv32i16.i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmacc.mask.nxv32i16.i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vwmacc_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int8_t op1,
@@ -1176,7 +1176,7 @@ vint16m8_t test_vwmacc_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmacc.mask.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmacc.mask.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vwmacc_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc,
@@ -1187,7 +1187,7 @@ vint32mf2_t test_vwmacc_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmacc.mask.nxv1i32.i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmacc.mask.nxv1i32.i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vwmacc_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc,
@@ -1197,7 +1197,7 @@ vint32mf2_t test_vwmacc_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmacc.mask.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmacc.mask.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vwmacc_vv_i32m1_m(vbool32_t mask, vint32m1_t acc,
@@ -1207,7 +1207,7 @@ vint32m1_t test_vwmacc_vv_i32m1_m(vbool32_t mask, vint32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmacc.mask.nxv2i32.i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmacc.mask.nxv2i32.i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vwmacc_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int16_t op1,
@@ -1217,7 +1217,7 @@ vint32m1_t test_vwmacc_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmacc.mask.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmacc.mask.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vwmacc_vv_i32m2_m(vbool16_t mask, vint32m2_t acc,
@@ -1227,7 +1227,7 @@ vint32m2_t test_vwmacc_vv_i32m2_m(vbool16_t mask, vint32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmacc.mask.nxv4i32.i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmacc.mask.nxv4i32.i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vwmacc_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int16_t op1,
@@ -1237,7 +1237,7 @@ vint32m2_t test_vwmacc_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmacc.mask.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmacc.mask.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vwmacc_vv_i32m4_m(vbool8_t mask, vint32m4_t acc, vint16m2_t op1,
@@ -1247,7 +1247,7 @@ vint32m4_t test_vwmacc_vv_i32m4_m(vbool8_t mask, vint32m4_t acc, vint16m2_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmacc.mask.nxv8i32.i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmacc.mask.nxv8i32.i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vwmacc_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int16_t op1,
@@ -1257,7 +1257,7 @@ vint32m4_t test_vwmacc_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmacc.mask.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmacc.mask.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vwmacc_vv_i32m8_m(vbool4_t mask, vint32m8_t acc, vint16m4_t op1,
@@ -1267,7 +1267,7 @@ vint32m8_t test_vwmacc_vv_i32m8_m(vbool4_t mask, vint32m8_t acc, vint16m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmacc.mask.nxv16i32.i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmacc.mask.nxv16i32.i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vwmacc_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int16_t op1,
@@ -1277,7 +1277,7 @@ vint32m8_t test_vwmacc_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmacc.mask.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmacc.mask.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vwmacc_vv_i64m1_m(vbool64_t mask, vint64m1_t acc,
@@ -1287,7 +1287,7 @@ vint64m1_t test_vwmacc_vv_i64m1_m(vbool64_t mask, vint64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmacc.mask.nxv1i64.i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmacc.mask.nxv1i64.i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vwmacc_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int32_t op1,
@@ -1297,7 +1297,7 @@ vint64m1_t test_vwmacc_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int32_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmacc.mask.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmacc.mask.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vwmacc_vv_i64m2_m(vbool32_t mask, vint64m2_t acc,
@@ -1307,7 +1307,7 @@ vint64m2_t test_vwmacc_vv_i64m2_m(vbool32_t mask, vint64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmacc.mask.nxv2i64.i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmacc.mask.nxv2i64.i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vwmacc_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int32_t op1,
@@ -1317,7 +1317,7 @@ vint64m2_t test_vwmacc_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int32_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmacc.mask.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmacc.mask.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vwmacc_vv_i64m4_m(vbool16_t mask, vint64m4_t acc,
@@ -1327,7 +1327,7 @@ vint64m4_t test_vwmacc_vv_i64m4_m(vbool16_t mask, vint64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmacc.mask.nxv4i64.i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmacc.mask.nxv4i64.i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vwmacc_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int32_t op1,
@@ -1337,7 +1337,7 @@ vint64m4_t test_vwmacc_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int32_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vv_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmacc.mask.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmacc.mask.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vwmacc_vv_i64m8_m(vbool8_t mask, vint64m8_t acc, vint32m4_t op1,
@@ -1347,7 +1347,7 @@ vint64m8_t test_vwmacc_vv_i64m8_m(vbool8_t mask, vint64m8_t acc, vint32m4_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmacc_vx_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmacc.mask.nxv8i64.i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmacc.mask.nxv8i64.i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vwmacc_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int32_t op1,
@@ -1357,7 +1357,7 @@ vint64m8_t test_vwmacc_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int32_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmaccu.mask.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmaccu.mask.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vwmaccu_vv_u16mf4_m(vbool64_t mask, vuint16mf4_t acc,
@@ -1368,7 +1368,7 @@ vuint16mf4_t test_vwmaccu_vv_u16mf4_m(vbool64_t mask, vuint16mf4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmaccu.mask.nxv1i16.i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmaccu.mask.nxv1i16.i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vuint16mf4_t test_vwmaccu_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t acc,
@@ -1378,7 +1378,7 @@ vuint16mf4_t test_vwmaccu_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmaccu.mask.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmaccu.mask.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vwmaccu_vv_u16mf2_m(vbool32_t mask, vuint16mf2_t acc,
@@ -1389,7 +1389,7 @@ vuint16mf2_t test_vwmaccu_vv_u16mf2_m(vbool32_t mask, vuint16mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmaccu.mask.nxv2i16.i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmaccu.mask.nxv2i16.i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vuint16mf2_t test_vwmaccu_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t acc,
@@ -1399,7 +1399,7 @@ vuint16mf2_t test_vwmaccu_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmaccu.mask.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmaccu.mask.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vwmaccu_vv_u16m1_m(vbool16_t mask, vuint16m1_t acc,
@@ -1410,7 +1410,7 @@ vuint16m1_t test_vwmaccu_vv_u16m1_m(vbool16_t mask, vuint16m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmaccu.mask.nxv4i16.i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmaccu.mask.nxv4i16.i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vuint16m1_t test_vwmaccu_vx_u16m1_m(vbool16_t mask, vuint16m1_t acc,
@@ -1420,7 +1420,7 @@ vuint16m1_t test_vwmaccu_vx_u16m1_m(vbool16_t mask, vuint16m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmaccu.mask.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmaccu.mask.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vwmaccu_vv_u16m2_m(vbool8_t mask, vuint16m2_t acc,
@@ -1430,7 +1430,7 @@ vuint16m2_t test_vwmaccu_vv_u16m2_m(vbool8_t mask, vuint16m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmaccu.mask.nxv8i16.i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmaccu.mask.nxv8i16.i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vuint16m2_t test_vwmaccu_vx_u16m2_m(vbool8_t mask, vuint16m2_t acc, uint8_t op1,
@@ -1440,7 +1440,7 @@ vuint16m2_t test_vwmaccu_vx_u16m2_m(vbool8_t mask, vuint16m2_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmaccu.mask.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmaccu.mask.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vwmaccu_vv_u16m4_m(vbool4_t mask, vuint16m4_t acc,
@@ -1450,7 +1450,7 @@ vuint16m4_t test_vwmaccu_vv_u16m4_m(vbool4_t mask, vuint16m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmaccu.mask.nxv16i16.i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmaccu.mask.nxv16i16.i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vuint16m4_t test_vwmaccu_vx_u16m4_m(vbool4_t mask, vuint16m4_t acc, uint8_t op1,
@@ -1460,7 +1460,7 @@ vuint16m4_t test_vwmaccu_vx_u16m4_m(vbool4_t mask, vuint16m4_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmaccu.mask.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmaccu.mask.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vwmaccu_vv_u16m8_m(vbool2_t mask, vuint16m8_t acc,
@@ -1470,7 +1470,7 @@ vuint16m8_t test_vwmaccu_vv_u16m8_m(vbool2_t mask, vuint16m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmaccu.mask.nxv32i16.i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmaccu.mask.nxv32i16.i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vuint16m8_t test_vwmaccu_vx_u16m8_m(vbool2_t mask, vuint16m8_t acc, uint8_t op1,
@@ -1480,7 +1480,7 @@ vuint16m8_t test_vwmaccu_vx_u16m8_m(vbool2_t mask, vuint16m8_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmaccu.mask.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmaccu.mask.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vwmaccu_vv_u32mf2_m(vbool64_t mask, vuint32mf2_t acc,
@@ -1491,7 +1491,7 @@ vuint32mf2_t test_vwmaccu_vv_u32mf2_m(vbool64_t mask, vuint32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmaccu.mask.nxv1i32.i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmaccu.mask.nxv1i32.i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vuint32mf2_t test_vwmaccu_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t acc,
@@ -1502,7 +1502,7 @@ vuint32mf2_t test_vwmaccu_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmaccu.mask.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmaccu.mask.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vwmaccu_vv_u32m1_m(vbool32_t mask, vuint32m1_t acc,
@@ -1513,7 +1513,7 @@ vuint32m1_t test_vwmaccu_vv_u32m1_m(vbool32_t mask, vuint32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmaccu.mask.nxv2i32.i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmaccu.mask.nxv2i32.i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vuint32m1_t test_vwmaccu_vx_u32m1_m(vbool32_t mask, vuint32m1_t acc,
@@ -1523,7 +1523,7 @@ vuint32m1_t test_vwmaccu_vx_u32m1_m(vbool32_t mask, vuint32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmaccu.mask.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmaccu.mask.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vwmaccu_vv_u32m2_m(vbool16_t mask, vuint32m2_t acc,
@@ -1534,7 +1534,7 @@ vuint32m2_t test_vwmaccu_vv_u32m2_m(vbool16_t mask, vuint32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmaccu.mask.nxv4i32.i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmaccu.mask.nxv4i32.i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vuint32m2_t test_vwmaccu_vx_u32m2_m(vbool16_t mask, vuint32m2_t acc,
@@ -1544,7 +1544,7 @@ vuint32m2_t test_vwmaccu_vx_u32m2_m(vbool16_t mask, vuint32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmaccu.mask.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmaccu.mask.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vwmaccu_vv_u32m4_m(vbool8_t mask, vuint32m4_t acc,
@@ -1555,7 +1555,7 @@ vuint32m4_t test_vwmaccu_vv_u32m4_m(vbool8_t mask, vuint32m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmaccu.mask.nxv8i32.i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmaccu.mask.nxv8i32.i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vuint32m4_t test_vwmaccu_vx_u32m4_m(vbool8_t mask, vuint32m4_t acc,
@@ -1565,7 +1565,7 @@ vuint32m4_t test_vwmaccu_vx_u32m4_m(vbool8_t mask, vuint32m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmaccu.mask.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmaccu.mask.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vwmaccu_vv_u32m8_m(vbool4_t mask, vuint32m8_t acc,
@@ -1576,7 +1576,7 @@ vuint32m8_t test_vwmaccu_vv_u32m8_m(vbool4_t mask, vuint32m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmaccu.mask.nxv16i32.i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmaccu.mask.nxv16i32.i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vuint32m8_t test_vwmaccu_vx_u32m8_m(vbool4_t mask, vuint32m8_t acc,
@@ -1586,7 +1586,7 @@ vuint32m8_t test_vwmaccu_vx_u32m8_m(vbool4_t mask, vuint32m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmaccu.mask.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmaccu.mask.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vwmaccu_vv_u64m1_m(vbool64_t mask, vuint64m1_t acc,
@@ -1597,7 +1597,7 @@ vuint64m1_t test_vwmaccu_vv_u64m1_m(vbool64_t mask, vuint64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmaccu.mask.nxv1i64.i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmaccu.mask.nxv1i64.i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vuint64m1_t test_vwmaccu_vx_u64m1_m(vbool64_t mask, vuint64m1_t acc,
@@ -1607,7 +1607,7 @@ vuint64m1_t test_vwmaccu_vx_u64m1_m(vbool64_t mask, vuint64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmaccu.mask.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmaccu.mask.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vwmaccu_vv_u64m2_m(vbool32_t mask, vuint64m2_t acc,
@@ -1618,7 +1618,7 @@ vuint64m2_t test_vwmaccu_vv_u64m2_m(vbool32_t mask, vuint64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmaccu.mask.nxv2i64.i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmaccu.mask.nxv2i64.i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vuint64m2_t test_vwmaccu_vx_u64m2_m(vbool32_t mask, vuint64m2_t acc,
@@ -1628,7 +1628,7 @@ vuint64m2_t test_vwmaccu_vx_u64m2_m(vbool32_t mask, vuint64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmaccu.mask.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmaccu.mask.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vwmaccu_vv_u64m4_m(vbool16_t mask, vuint64m4_t acc,
@@ -1639,7 +1639,7 @@ vuint64m4_t test_vwmaccu_vv_u64m4_m(vbool16_t mask, vuint64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmaccu.mask.nxv4i64.i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmaccu.mask.nxv4i64.i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vuint64m4_t test_vwmaccu_vx_u64m4_m(vbool16_t mask, vuint64m4_t acc,
@@ -1649,7 +1649,7 @@ vuint64m4_t test_vwmaccu_vx_u64m4_m(vbool16_t mask, vuint64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vv_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmaccu.mask.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmaccu.mask.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vwmaccu_vv_u64m8_m(vbool8_t mask, vuint64m8_t acc,
@@ -1660,7 +1660,7 @@ vuint64m8_t test_vwmaccu_vv_u64m8_m(vbool8_t mask, vuint64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccu_vx_u64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmaccu.mask.nxv8i64.i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmaccu.mask.nxv8i64.i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vuint64m8_t test_vwmaccu_vx_u64m8_m(vbool8_t mask, vuint64m8_t acc,
@@ -1670,7 +1670,7 @@ vuint64m8_t test_vwmaccu_vx_u64m8_m(vbool8_t mask, vuint64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmaccsu.mask.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmaccsu.mask.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vwmaccsu_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc,
@@ -1681,7 +1681,7 @@ vint16mf4_t test_vwmaccsu_vv_i16mf4_m(vbool64_t mask, vint16mf4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmaccsu.mask.nxv1i16.i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmaccsu.mask.nxv1i16.i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vwmaccsu_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc,
@@ -1691,7 +1691,7 @@ vint16mf4_t test_vwmaccsu_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmaccsu.mask.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmaccsu.mask.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vwmaccsu_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc,
@@ -1702,7 +1702,7 @@ vint16mf2_t test_vwmaccsu_vv_i16mf2_m(vbool32_t mask, vint16mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmaccsu.mask.nxv2i16.i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmaccsu.mask.nxv2i16.i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vwmaccsu_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc,
@@ -1712,7 +1712,7 @@ vint16mf2_t test_vwmaccsu_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmaccsu.mask.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmaccsu.mask.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vwmaccsu_vv_i16m1_m(vbool16_t mask, vint16m1_t acc,
@@ -1723,7 +1723,7 @@ vint16m1_t test_vwmaccsu_vv_i16m1_m(vbool16_t mask, vint16m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmaccsu.mask.nxv4i16.i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmaccsu.mask.nxv4i16.i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vwmaccsu_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int8_t op1,
@@ -1733,7 +1733,7 @@ vint16m1_t test_vwmaccsu_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, int8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmaccsu.mask.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmaccsu.mask.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vwmaccsu_vv_i16m2_m(vbool8_t mask, vint16m2_t acc,
@@ -1743,7 +1743,7 @@ vint16m2_t test_vwmaccsu_vv_i16m2_m(vbool8_t mask, vint16m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmaccsu.mask.nxv8i16.i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmaccsu.mask.nxv8i16.i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vwmaccsu_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int8_t op1,
@@ -1753,7 +1753,7 @@ vint16m2_t test_vwmaccsu_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, int8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmaccsu.mask.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmaccsu.mask.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vwmaccsu_vv_i16m4_m(vbool4_t mask, vint16m4_t acc,
@@ -1763,7 +1763,7 @@ vint16m4_t test_vwmaccsu_vv_i16m4_m(vbool4_t mask, vint16m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmaccsu.mask.nxv16i16.i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmaccsu.mask.nxv16i16.i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vwmaccsu_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int8_t op1,
@@ -1773,7 +1773,7 @@ vint16m4_t test_vwmaccsu_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, int8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmaccsu.mask.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmaccsu.mask.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vwmaccsu_vv_i16m8_m(vbool2_t mask, vint16m8_t acc,
@@ -1783,7 +1783,7 @@ vint16m8_t test_vwmaccsu_vv_i16m8_m(vbool2_t mask, vint16m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmaccsu.mask.nxv32i16.i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmaccsu.mask.nxv32i16.i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vwmaccsu_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int8_t op1,
@@ -1793,7 +1793,7 @@ vint16m8_t test_vwmaccsu_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, int8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmaccsu.mask.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmaccsu.mask.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vwmaccsu_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc,
@@ -1804,7 +1804,7 @@ vint32mf2_t test_vwmaccsu_vv_i32mf2_m(vbool64_t mask, vint32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmaccsu.mask.nxv1i32.i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmaccsu.mask.nxv1i32.i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vwmaccsu_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc,
@@ -1815,7 +1815,7 @@ vint32mf2_t test_vwmaccsu_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmaccsu.mask.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmaccsu.mask.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vwmaccsu_vv_i32m1_m(vbool32_t mask, vint32m1_t acc,
@@ -1826,7 +1826,7 @@ vint32m1_t test_vwmaccsu_vv_i32m1_m(vbool32_t mask, vint32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmaccsu.mask.nxv2i32.i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmaccsu.mask.nxv2i32.i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vwmaccsu_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int16_t op1,
@@ -1836,7 +1836,7 @@ vint32m1_t test_vwmaccsu_vx_i32m1_m(vbool32_t mask, vint32m1_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmaccsu.mask.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmaccsu.mask.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vwmaccsu_vv_i32m2_m(vbool16_t mask, vint32m2_t acc,
@@ -1847,7 +1847,7 @@ vint32m2_t test_vwmaccsu_vv_i32m2_m(vbool16_t mask, vint32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmaccsu.mask.nxv4i32.i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmaccsu.mask.nxv4i32.i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vwmaccsu_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int16_t op1,
@@ -1857,7 +1857,7 @@ vint32m2_t test_vwmaccsu_vx_i32m2_m(vbool16_t mask, vint32m2_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmaccsu.mask.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmaccsu.mask.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vwmaccsu_vv_i32m4_m(vbool8_t mask, vint32m4_t acc,
@@ -1868,7 +1868,7 @@ vint32m4_t test_vwmaccsu_vv_i32m4_m(vbool8_t mask, vint32m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmaccsu.mask.nxv8i32.i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmaccsu.mask.nxv8i32.i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vwmaccsu_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int16_t op1,
@@ -1878,7 +1878,7 @@ vint32m4_t test_vwmaccsu_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmaccsu.mask.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmaccsu.mask.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vwmaccsu_vv_i32m8_m(vbool4_t mask, vint32m8_t acc,
@@ -1889,7 +1889,7 @@ vint32m8_t test_vwmaccsu_vv_i32m8_m(vbool4_t mask, vint32m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmaccsu.mask.nxv16i32.i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmaccsu.mask.nxv16i32.i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vwmaccsu_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int16_t op1,
@@ -1899,7 +1899,7 @@ vint32m8_t test_vwmaccsu_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, int16_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmaccsu.mask.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmaccsu.mask.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vwmaccsu_vv_i64m1_m(vbool64_t mask, vint64m1_t acc,
@@ -1910,7 +1910,7 @@ vint64m1_t test_vwmaccsu_vv_i64m1_m(vbool64_t mask, vint64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmaccsu.mask.nxv1i64.i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmaccsu.mask.nxv1i64.i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vwmaccsu_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int32_t op1,
@@ -1920,7 +1920,7 @@ vint64m1_t test_vwmaccsu_vx_i64m1_m(vbool64_t mask, vint64m1_t acc, int32_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmaccsu.mask.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmaccsu.mask.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vwmaccsu_vv_i64m2_m(vbool32_t mask, vint64m2_t acc,
@@ -1931,7 +1931,7 @@ vint64m2_t test_vwmaccsu_vv_i64m2_m(vbool32_t mask, vint64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmaccsu.mask.nxv2i64.i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmaccsu.mask.nxv2i64.i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vwmaccsu_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int32_t op1,
@@ -1941,7 +1941,7 @@ vint64m2_t test_vwmaccsu_vx_i64m2_m(vbool32_t mask, vint64m2_t acc, int32_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmaccsu.mask.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmaccsu.mask.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vwmaccsu_vv_i64m4_m(vbool16_t mask, vint64m4_t acc,
@@ -1952,7 +1952,7 @@ vint64m4_t test_vwmaccsu_vv_i64m4_m(vbool16_t mask, vint64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmaccsu.mask.nxv4i64.i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmaccsu.mask.nxv4i64.i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vwmaccsu_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int32_t op1,
@@ -1962,7 +1962,7 @@ vint64m4_t test_vwmaccsu_vx_i64m4_m(vbool16_t mask, vint64m4_t acc, int32_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vv_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmaccsu.mask.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmaccsu.mask.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vwmaccsu_vv_i64m8_m(vbool8_t mask, vint64m8_t acc,
@@ -1973,7 +1973,7 @@ vint64m8_t test_vwmaccsu_vv_i64m8_m(vbool8_t mask, vint64m8_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccsu_vx_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmaccsu.mask.nxv8i64.i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmaccsu.mask.nxv8i64.i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vwmaccsu_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int32_t op1,
@@ -1983,7 +1983,7 @@ vint64m8_t test_vwmaccsu_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, int32_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i16mf4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmaccus.mask.nxv1i16.i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmaccus.mask.nxv1i16.i8.nxv1i8.i64(<vscale x 1 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
 //
 vint16mf4_t test_vwmaccus_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc,
@@ -1993,7 +1993,7 @@ vint16mf4_t test_vwmaccus_vx_i16mf4_m(vbool64_t mask, vint16mf4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i16mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmaccus.mask.nxv2i16.i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmaccus.mask.nxv2i16.i8.nxv2i8.i64(<vscale x 2 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
 //
 vint16mf2_t test_vwmaccus_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc,
@@ -2003,7 +2003,7 @@ vint16mf2_t test_vwmaccus_vx_i16mf2_m(vbool32_t mask, vint16mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i16m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmaccus.mask.nxv4i16.i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmaccus.mask.nxv4i16.i8.nxv4i8.i64(<vscale x 4 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
 //
 vint16m1_t test_vwmaccus_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, uint8_t op1,
@@ -2013,7 +2013,7 @@ vint16m1_t test_vwmaccus_vx_i16m1_m(vbool16_t mask, vint16m1_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i16m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmaccus.mask.nxv8i16.i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmaccus.mask.nxv8i16.i8.nxv8i8.i64(<vscale x 8 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
 //
 vint16m2_t test_vwmaccus_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, uint8_t op1,
@@ -2023,7 +2023,7 @@ vint16m2_t test_vwmaccus_vx_i16m2_m(vbool8_t mask, vint16m2_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i16m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmaccus.mask.nxv16i16.i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmaccus.mask.nxv16i16.i8.nxv16i8.i64(<vscale x 16 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
 //
 vint16m4_t test_vwmaccus_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, uint8_t op1,
@@ -2033,7 +2033,7 @@ vint16m4_t test_vwmaccus_vx_i16m4_m(vbool4_t mask, vint16m4_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i16m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmaccus.mask.nxv32i16.i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmaccus.mask.nxv32i16.i8.nxv32i8.i64(<vscale x 32 x i16> [[ACC:%.*]], i8 [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
 //
 vint16m8_t test_vwmaccus_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, uint8_t op1,
@@ -2043,7 +2043,7 @@ vint16m8_t test_vwmaccus_vx_i16m8_m(vbool2_t mask, vint16m8_t acc, uint8_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i32mf2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmaccus.mask.nxv1i32.i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmaccus.mask.nxv1i32.i16.nxv1i16.i64(<vscale x 1 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
 //
 vint32mf2_t test_vwmaccus_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc,
@@ -2054,7 +2054,7 @@ vint32mf2_t test_vwmaccus_vx_i32mf2_m(vbool64_t mask, vint32mf2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i32m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmaccus.mask.nxv2i32.i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmaccus.mask.nxv2i32.i16.nxv2i16.i64(<vscale x 2 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
 //
 vint32m1_t test_vwmaccus_vx_i32m1_m(vbool32_t mask, vint32m1_t acc,
@@ -2064,7 +2064,7 @@ vint32m1_t test_vwmaccus_vx_i32m1_m(vbool32_t mask, vint32m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i32m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmaccus.mask.nxv4i32.i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmaccus.mask.nxv4i32.i16.nxv4i16.i64(<vscale x 4 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
 //
 vint32m2_t test_vwmaccus_vx_i32m2_m(vbool16_t mask, vint32m2_t acc,
@@ -2074,7 +2074,7 @@ vint32m2_t test_vwmaccus_vx_i32m2_m(vbool16_t mask, vint32m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i32m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmaccus.mask.nxv8i32.i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmaccus.mask.nxv8i32.i16.nxv8i16.i64(<vscale x 8 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
 //
 vint32m4_t test_vwmaccus_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, uint16_t op1,
@@ -2084,7 +2084,7 @@ vint32m4_t test_vwmaccus_vx_i32m4_m(vbool8_t mask, vint32m4_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i32m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmaccus.mask.nxv16i32.i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmaccus.mask.nxv16i32.i16.nxv16i16.i64(<vscale x 16 x i32> [[ACC:%.*]], i16 [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
 //
 vint32m8_t test_vwmaccus_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, uint16_t op1,
@@ -2094,7 +2094,7 @@ vint32m8_t test_vwmaccus_vx_i32m8_m(vbool4_t mask, vint32m8_t acc, uint16_t op1,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i64m1_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmaccus.mask.nxv1i64.i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmaccus.mask.nxv1i64.i32.nxv1i32.i64(<vscale x 1 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
 //
 vint64m1_t test_vwmaccus_vx_i64m1_m(vbool64_t mask, vint64m1_t acc,
@@ -2104,7 +2104,7 @@ vint64m1_t test_vwmaccus_vx_i64m1_m(vbool64_t mask, vint64m1_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i64m2_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmaccus.mask.nxv2i64.i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmaccus.mask.nxv2i64.i32.nxv2i32.i64(<vscale x 2 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
 //
 vint64m2_t test_vwmaccus_vx_i64m2_m(vbool32_t mask, vint64m2_t acc,
@@ -2114,7 +2114,7 @@ vint64m2_t test_vwmaccus_vx_i64m2_m(vbool32_t mask, vint64m2_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i64m4_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmaccus.mask.nxv4i64.i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmaccus.mask.nxv4i64.i32.nxv4i32.i64(<vscale x 4 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
 //
 vint64m4_t test_vwmaccus_vx_i64m4_m(vbool16_t mask, vint64m4_t acc,
@@ -2124,7 +2124,7 @@ vint64m4_t test_vwmaccus_vx_i64m4_m(vbool16_t mask, vint64m4_t acc,
 
 // CHECK-RV64-LABEL: @test_vwmaccus_vx_i64m8_m(
 // CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmaccus.mask.nxv8i64.i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmaccus.mask.nxv8i64.i32.nxv8i32.i64(<vscale x 8 x i64> [[ACC:%.*]], i32 [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]], i64 0)
 // CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
 //
 vint64m8_t test_vwmaccus_vx_i64m8_m(vbool8_t mask, vint64m8_t acc, uint32_t op1,
diff --git a/llvm/include/llvm/IR/IntrinsicsRISCV.td b/llvm/include/llvm/IR/IntrinsicsRISCV.td
index 6780436bd701a7..5e44fb900c0ab4 100644
--- a/llvm/include/llvm/IR/IntrinsicsRISCV.td
+++ b/llvm/include/llvm/IR/IntrinsicsRISCV.td
@@ -669,8 +669,9 @@ let TargetPrefix = "riscv" in {
   class RISCVTernaryAAAXMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, LLVMMatchType<0>, llvm_anyint_ty,
-                     LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, LLVMMatchType<1>],
-                    [IntrNoMem]>, RISCVVIntrinsic {
+                     LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
+                     LLVMMatchType<1>, LLVMMatchType<1>],
+                    [ImmArg<ArgIndex<5>>, IntrNoMem]>, RISCVVIntrinsic {
     let VLOperand = 4;
   }
   class RISCVTernaryAAXANoMask
@@ -681,11 +682,13 @@ let TargetPrefix = "riscv" in {
     let SplatOperand = 1;
     let VLOperand = 3;
   }
+  // Input: (vector_in, vector_in/scalar, vector_in, mask, vl, policy
   class RISCVTernaryAAXAMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, llvm_any_ty, LLVMMatchType<0>,
-                     LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty],
-                    [IntrNoMem]>, RISCVVIntrinsic {
+                     LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
+                     llvm_anyint_ty, LLVMMatchType<2>],
+                    [ImmArg<ArgIndex<5>>, IntrNoMem]>, RISCVVIntrinsic {
     let SplatOperand = 1;
     let VLOperand = 4;
   }
@@ -697,11 +700,13 @@ let TargetPrefix = "riscv" in {
     let SplatOperand = 1;
     let VLOperand = 3;
   }
+  // Input: (vector_in, vector_in/scalar, vector_in, mask, vl, policy
   class RISCVTernaryWideMask
         : Intrinsic< [llvm_anyvector_ty],
                      [LLVMMatchType<0>, llvm_any_ty, llvm_anyvector_ty,
-                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty],
-                     [IntrNoMem]>, RISCVVIntrinsic {
+                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
+                      llvm_anyint_ty, LLVMMatchType<3>],
+                     [ImmArg<ArgIndex<5>>, IntrNoMem]>, RISCVVIntrinsic {
     let SplatOperand = 1;
     let VLOperand = 4;
   }
diff --git a/llvm/lib/Target/RISCV/RISCVInstrInfoVPseudos.td b/llvm/lib/Target/RISCV/RISCVInstrInfoVPseudos.td
index 41fa5087e93033..25b02824c14501 100644
--- a/llvm/lib/Target/RISCV/RISCVInstrInfoVPseudos.td
+++ b/llvm/lib/Target/RISCV/RISCVInstrInfoVPseudos.td
@@ -1065,10 +1065,10 @@ class VPseudoBinaryMask<VReg RetClass,
   let BaseInstr = !cast<Instruction>(PseudoToVInst<NAME>.VInst);
 }
 
-class VPseudoBinaryMaskTA<VReg RetClass,
-                          RegisterClass Op1Class,
-                          DAGOperand Op2Class,
-                          string Constraint> :
+class VPseudoBinaryMaskPolicy<VReg RetClass,
+                              RegisterClass Op1Class,
+                              DAGOperand Op2Class,
+                              string Constraint> :
         Pseudo<(outs GetVRegNoV0<RetClass>.R:$rd),
                 (ins GetVRegNoV0<RetClass>.R:$merge,
                      Op1Class:$rs2, Op2Class:$rs1,
@@ -1652,8 +1652,8 @@ multiclass VPseudoBinary<VReg RetClass,
   let VLMul = MInfo.value in {
     def "_" # MInfo.MX : VPseudoBinaryNoMask<RetClass, Op1Class, Op2Class,
                                              Constraint>;
-    def "_" # MInfo.MX # "_MASK" : VPseudoBinaryMaskTA<RetClass, Op1Class, Op2Class,
-                                                       Constraint>,
+    def "_" # MInfo.MX # "_MASK" : VPseudoBinaryMaskPolicy<RetClass, Op1Class, Op2Class,
+                                                           Constraint>,
                                    RISCVMaskedPseudo</*MaskOpIdx*/ 3>;
   }
 }
@@ -1681,8 +1681,8 @@ multiclass VPseudoBinaryEmul<VReg RetClass,
   let VLMul = lmul.value in {
     def "_" # lmul.MX # "_" # emul.MX : VPseudoBinaryNoMask<RetClass, Op1Class, Op2Class,
                                                             Constraint>;
-    def "_" # lmul.MX # "_" # emul.MX # "_MASK" : VPseudoBinaryMaskTA<RetClass, Op1Class, Op2Class,
-                                                                      Constraint>;
+    def "_" # lmul.MX # "_" # emul.MX # "_MASK" : VPseudoBinaryMaskPolicy<RetClass, Op1Class, Op2Class,
+                                                                          Constraint>;
   }
 }
 
@@ -2333,6 +2333,19 @@ multiclass VPseudoTernary<VReg RetClass,
   }
 }
 
+multiclass VPseudoTernaryNoMaskNoPolicy<VReg RetClass,
+                                        RegisterClass Op1Class,
+                                        DAGOperand Op2Class,
+                                        LMULInfo MInfo,
+                                        string Constraint = ""> {
+  let VLMul = MInfo.value in {
+    def "_" # MInfo.MX : VPseudoTernaryNoMask<RetClass, Op1Class, Op2Class, Constraint>;
+    def "_" # MInfo.MX # "_MASK" : VPseudoBinaryMaskPolicy<RetClass, Op1Class, Op2Class,
+                                                           Constraint>;
+                                   
+  }
+}
+
 multiclass VPseudoTernaryWithPolicy<VReg RetClass,
                                     RegisterClass Op1Class,
                                     DAGOperand Op2Class,
@@ -2342,7 +2355,7 @@ multiclass VPseudoTernaryWithPolicy<VReg RetClass,
   let VLMul = MInfo.value in {
     let isCommutable = Commutable in
     def "_" # MInfo.MX : VPseudoTernaryNoMaskWithPolicy<RetClass, Op1Class, Op2Class, Constraint>;
-    def "_" # MInfo.MX # "_MASK" : VPseudoBinaryMask<RetClass, Op1Class, Op2Class, Constraint>;
+    def "_" # MInfo.MX # "_MASK" : VPseudoBinaryMaskPolicy<RetClass, Op1Class, Op2Class, Constraint>;
   }
 }
 
@@ -2356,7 +2369,7 @@ multiclass VPseudoTernaryV_VV_AAXA<string Constraint = "",
 
 multiclass VPseudoTernaryV_VX<string Constraint = ""> {
   foreach m = MxList in
-    defm _VX : VPseudoTernary<m.vrclass, m.vrclass, GPR, m, Constraint>;
+    defm _VX : VPseudoTernaryNoMaskNoPolicy<m.vrclass, m.vrclass, GPR, m, Constraint>;
 }
 
 multiclass VPseudoTernaryV_VX_AAXA<string Constraint = ""> {
@@ -2397,7 +2410,7 @@ multiclass VPseudoTernaryW_VF {
 
 multiclass VPseudoTernaryV_VI<Operand ImmType = simm5, string Constraint = ""> {
   foreach m = MxList in
-    defm _VI : VPseudoTernary<m.vrclass, m.vrclass, ImmType, m, Constraint>;
+    defm _VI : VPseudoTernaryNoMaskNoPolicy<m.vrclass, m.vrclass, ImmType, m, Constraint>;
 }
 
 multiclass VPseudoVMAC_VV_VX_AAXA<string Constraint = ""> {
@@ -3013,6 +3026,31 @@ class VPatTernaryMask<string intrinsic,
                     (mask_type V0),
                     GPR:$vl, sew)>;
 
+class VPatTernaryMaskPolicy<string intrinsic,
+                            string inst,
+                            string kind,
+                            ValueType result_type,
+                            ValueType op1_type,
+                            ValueType op2_type,
+                            ValueType mask_type,
+                            int sew,
+                            LMULInfo vlmul,
+                            VReg result_reg_class,
+                            RegisterClass op1_reg_class,
+                            DAGOperand op2_kind> :
+  Pat<(result_type (!cast<Intrinsic>(intrinsic#"_mask")
+                    (result_type result_reg_class:$rs3),
+                    (op1_type op1_reg_class:$rs1),
+                    (op2_type op2_kind:$rs2),
+                    (mask_type V0),
+                    VLOpFrag, (XLenVT timm:$policy))),
+                   (!cast<Instruction>(inst#"_"#kind#"_"#vlmul.MX # "_MASK")
+                    result_reg_class:$rs3,
+                    (op1_type op1_reg_class:$rs1),
+                    op2_kind:$rs2,
+                    (mask_type V0),
+                    GPR:$vl, sew, (XLenVT timm:$policy))>;
+
 multiclass VPatUnaryS_M<string intrinsic_name,
                              string inst>
 {
@@ -3584,6 +3622,26 @@ multiclass VPatTernary<string intrinsic,
                         op2_kind>;
 }
 
+multiclass VPatTernaryNoMaskNoPolicy<string intrinsic,
+                                     string inst,
+                                     string kind,
+                                     ValueType result_type,
+                                     ValueType op1_type,
+                                     ValueType op2_type,
+                                     ValueType mask_type,
+                                     int sew,
+                                     LMULInfo vlmul,
+                                     VReg result_reg_class,
+                                     RegisterClass op1_reg_class,
+                                     DAGOperand op2_kind> {
+  def : VPatTernaryNoMask<intrinsic, inst, kind, result_type, op1_type, op2_type,
+                          sew, vlmul, result_reg_class, op1_reg_class,
+                          op2_kind>;
+  def : VPatTernaryMaskPolicy<intrinsic, inst, kind, result_type, op1_type, op2_type,
+                              mask_type, sew, vlmul, result_reg_class, op1_reg_class,
+                              op2_kind>;
+}
+
 multiclass VPatTernaryWithPolicy<string intrinsic,
                                  string inst,
                                  string kind,
@@ -3599,9 +3657,9 @@ multiclass VPatTernaryWithPolicy<string intrinsic,
   def : VPatTernaryNoMaskWithPolicy<intrinsic, inst, kind, result_type, op1_type,
                                     op2_type, sew, vlmul, result_reg_class,
                                     op1_reg_class, op2_kind>;
-  def : VPatTernaryMask<intrinsic, inst, kind, result_type, op1_type, op2_type,
-                        mask_type, sew, vlmul, result_reg_class, op1_reg_class,
-                        op2_kind>;
+  def : VPatTernaryMaskPolicy<intrinsic, inst, kind, result_type, op1_type, op2_type,
+                              mask_type, sew, vlmul, result_reg_class, op1_reg_class,
+                              op2_kind>;
 }
 
 multiclass VPatTernaryV_VV_AAXA<string intrinsic, string instruction,
@@ -3616,10 +3674,10 @@ multiclass VPatTernaryV_VV_AAXA<string intrinsic, string instruction,
 multiclass VPatTernaryV_VX<string intrinsic, string instruction,
                            list<VTypeInfo> vtilist> {
   foreach vti = vtilist in
-    defm : VPatTernary<intrinsic, instruction, "VX",
-                       vti.Vector, vti.Vector, XLenVT, vti.Mask,
-                       vti.Log2SEW, vti.LMul, vti.RegClass,
-                       vti.RegClass, GPR>;
+    defm : VPatTernaryNoMaskNoPolicy<intrinsic, instruction, "VX",
+                                     vti.Vector, vti.Vector, XLenVT, vti.Mask,
+                                     vti.Log2SEW, vti.LMul, vti.RegClass,
+                                     vti.RegClass, GPR>;
 }
 
 multiclass VPatTernaryV_VX_AAXA<string intrinsic, string instruction,
@@ -3635,10 +3693,10 @@ multiclass VPatTernaryV_VX_AAXA<string intrinsic, string instruction,
 multiclass VPatTernaryV_VI<string intrinsic, string instruction,
                            list<VTypeInfo> vtilist, Operand Imm_type> {
   foreach vti = vtilist in
-    defm : VPatTernary<intrinsic, instruction, "VI",
-                      vti.Vector, vti.Vector, XLenVT, vti.Mask,
-                      vti.Log2SEW, vti.LMul, vti.RegClass,
-                      vti.RegClass, Imm_type>;
+    defm : VPatTernaryNoMaskNoPolicy<intrinsic, instruction, "VI",
+                                     vti.Vector, vti.Vector, XLenVT, vti.Mask,
+                                     vti.Log2SEW, vti.LMul, vti.RegClass,
+                                     vti.RegClass, Imm_type>;
 }
 
 multiclass VPatTernaryW_VV<string intrinsic, string instruction,
diff --git a/llvm/lib/Target/RISCV/RISCVInstrInfoVVLPatterns.td b/llvm/lib/Target/RISCV/RISCVInstrInfoVVLPatterns.td
index 41bda9e241232d..7f5630456afece 100644
--- a/llvm/lib/Target/RISCV/RISCVInstrInfoVVLPatterns.td
+++ b/llvm/lib/Target/RISCV/RISCVInstrInfoVVLPatterns.td
@@ -1111,7 +1111,7 @@ foreach vti = AllFloatVectors in {
                                       VLOpFrag)),
             (!cast<Instruction>("PseudoVFMADD_VV_"# suffix #"_MASK")
                  vti.RegClass:$rd, vti.RegClass:$rs1, vti.RegClass:$rs2,
-                 (vti.Mask V0), GPR:$vl, vti.Log2SEW)>;
+                 (vti.Mask V0), GPR:$vl, vti.Log2SEW, TAIL_UNDISTURBED)>;
 
   def : Pat<(vti.Vector (riscv_fma_vl vti.RegClass:$rs1, vti.RegClass:$rd,
                                       (riscv_fneg_vl vti.RegClass:$rs2,
@@ -1161,7 +1161,7 @@ foreach vti = AllFloatVectors in {
                                        VLOpFrag)),
             (!cast<Instruction>("PseudoVFMADD_V" # vti.ScalarSuffix # "_" # suffix # "_MASK")
                  vti.RegClass:$rd, vti.ScalarRegClass:$rs1, vti.RegClass:$rs2,
-                 (vti.Mask V0), GPR:$vl, vti.Log2SEW)>;
+                 (vti.Mask V0), GPR:$vl, vti.Log2SEW, TAIL_UNDISTURBED)>;
 
   def : Pat<(vti.Vector (riscv_fma_vl (SplatFPOp vti.ScalarRegClass:$rs1),
                                        vti.RegClass:$rd,
diff --git a/llvm/test/CodeGen/RISCV/rvv/vfmacc.ll b/llvm/test/CodeGen/RISCV/rvv/vfmacc.ll
index 5115a7548e2ceb..02fa45976203b7 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vfmacc.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vfmacc.ll
@@ -30,7 +30,7 @@ declare <vscale x 1 x half> @llvm.riscv.vfmacc.mask.nxv1f16.nxv1f16(
   <vscale x 1 x half>,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x half>  @intrinsic_vfmacc_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vv_nxv1f16_nxv1f16_nxv1f16:
@@ -44,7 +44,7 @@ entry:
     <vscale x 1 x half> %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 1 x half> %a
 }
@@ -76,7 +76,7 @@ declare <vscale x 2 x half> @llvm.riscv.vfmacc.mask.nxv2f16.nxv2f16(
   <vscale x 2 x half>,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x half>  @intrinsic_vfmacc_mask_vv_nxv2f16_nxv2f16_nxv2f16(<vscale x 2 x half> %0, <vscale x 2 x half> %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vv_nxv2f16_nxv2f16_nxv2f16:
@@ -90,7 +90,7 @@ entry:
     <vscale x 2 x half> %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 2 x half> %a
 }
@@ -122,7 +122,7 @@ declare <vscale x 4 x half> @llvm.riscv.vfmacc.mask.nxv4f16.nxv4f16(
   <vscale x 4 x half>,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x half>  @intrinsic_vfmacc_mask_vv_nxv4f16_nxv4f16_nxv4f16(<vscale x 4 x half> %0, <vscale x 4 x half> %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vv_nxv4f16_nxv4f16_nxv4f16:
@@ -136,7 +136,7 @@ entry:
     <vscale x 4 x half> %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 4 x half> %a
 }
@@ -168,7 +168,7 @@ declare <vscale x 8 x half> @llvm.riscv.vfmacc.mask.nxv8f16.nxv8f16(
   <vscale x 8 x half>,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x half>  @intrinsic_vfmacc_mask_vv_nxv8f16_nxv8f16_nxv8f16(<vscale x 8 x half> %0, <vscale x 8 x half> %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vv_nxv8f16_nxv8f16_nxv8f16:
@@ -182,7 +182,7 @@ entry:
     <vscale x 8 x half> %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 8 x half> %a
 }
@@ -214,7 +214,7 @@ declare <vscale x 16 x half> @llvm.riscv.vfmacc.mask.nxv16f16.nxv16f16(
   <vscale x 16 x half>,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x half>  @intrinsic_vfmacc_mask_vv_nxv16f16_nxv16f16_nxv16f16(<vscale x 16 x half> %0, <vscale x 16 x half> %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vv_nxv16f16_nxv16f16_nxv16f16:
@@ -228,7 +228,7 @@ entry:
     <vscale x 16 x half> %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 16 x half> %a
 }
@@ -260,7 +260,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfmacc.mask.nxv1f32.nxv1f32(
   <vscale x 1 x float>,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float>  @intrinsic_vfmacc_mask_vv_nxv1f32_nxv1f32_nxv1f32(<vscale x 1 x float> %0, <vscale x 1 x float> %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vv_nxv1f32_nxv1f32_nxv1f32:
@@ -274,7 +274,7 @@ entry:
     <vscale x 1 x float> %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 1 x float> %a
 }
@@ -306,7 +306,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfmacc.mask.nxv2f32.nxv2f32(
   <vscale x 2 x float>,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float>  @intrinsic_vfmacc_mask_vv_nxv2f32_nxv2f32_nxv2f32(<vscale x 2 x float> %0, <vscale x 2 x float> %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vv_nxv2f32_nxv2f32_nxv2f32:
@@ -320,7 +320,7 @@ entry:
     <vscale x 2 x float> %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 2 x float> %a
 }
@@ -352,7 +352,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfmacc.mask.nxv4f32.nxv4f32(
   <vscale x 4 x float>,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float>  @intrinsic_vfmacc_mask_vv_nxv4f32_nxv4f32_nxv4f32(<vscale x 4 x float> %0, <vscale x 4 x float> %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vv_nxv4f32_nxv4f32_nxv4f32:
@@ -366,7 +366,7 @@ entry:
     <vscale x 4 x float> %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 4 x float> %a
 }
@@ -398,7 +398,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfmacc.mask.nxv8f32.nxv8f32(
   <vscale x 8 x float>,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float>  @intrinsic_vfmacc_mask_vv_nxv8f32_nxv8f32_nxv8f32(<vscale x 8 x float> %0, <vscale x 8 x float> %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vv_nxv8f32_nxv8f32_nxv8f32:
@@ -412,7 +412,7 @@ entry:
     <vscale x 8 x float> %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 8 x float> %a
 }
@@ -444,7 +444,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfmacc.mask.nxv1f64.nxv1f64(
   <vscale x 1 x double>,
   <vscale x 1 x double>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double>  @intrinsic_vfmacc_mask_vv_nxv1f64_nxv1f64_nxv1f64(<vscale x 1 x double> %0, <vscale x 1 x double> %1, <vscale x 1 x double> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vv_nxv1f64_nxv1f64_nxv1f64:
@@ -458,7 +458,7 @@ entry:
     <vscale x 1 x double> %1,
     <vscale x 1 x double> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 1 x double> %a
 }
@@ -490,7 +490,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfmacc.mask.nxv2f64.nxv2f64(
   <vscale x 2 x double>,
   <vscale x 2 x double>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double>  @intrinsic_vfmacc_mask_vv_nxv2f64_nxv2f64_nxv2f64(<vscale x 2 x double> %0, <vscale x 2 x double> %1, <vscale x 2 x double> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vv_nxv2f64_nxv2f64_nxv2f64:
@@ -504,7 +504,7 @@ entry:
     <vscale x 2 x double> %1,
     <vscale x 2 x double> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 2 x double> %a
 }
@@ -536,7 +536,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfmacc.mask.nxv4f64.nxv4f64(
   <vscale x 4 x double>,
   <vscale x 4 x double>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double>  @intrinsic_vfmacc_mask_vv_nxv4f64_nxv4f64_nxv4f64(<vscale x 4 x double> %0, <vscale x 4 x double> %1, <vscale x 4 x double> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vv_nxv4f64_nxv4f64_nxv4f64:
@@ -550,7 +550,7 @@ entry:
     <vscale x 4 x double> %1,
     <vscale x 4 x double> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 4 x double> %a
 }
@@ -582,7 +582,7 @@ declare <vscale x 1 x half> @llvm.riscv.vfmacc.mask.nxv1f16.f16(
   half,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x half> @intrinsic_vfmacc_mask_vf_nxv1f16_f16_nxv1f16(<vscale x 1 x half> %0, half %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vf_nxv1f16_f16_nxv1f16:
@@ -596,7 +596,7 @@ entry:
     half %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 1 x half> %a
 }
@@ -628,7 +628,7 @@ declare <vscale x 2 x half> @llvm.riscv.vfmacc.mask.nxv2f16.f16(
   half,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x half> @intrinsic_vfmacc_mask_vf_nxv2f16_f16_nxv2f16(<vscale x 2 x half> %0, half %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vf_nxv2f16_f16_nxv2f16:
@@ -642,7 +642,7 @@ entry:
     half %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 2 x half> %a
 }
@@ -674,7 +674,7 @@ declare <vscale x 4 x half> @llvm.riscv.vfmacc.mask.nxv4f16.f16(
   half,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x half> @intrinsic_vfmacc_mask_vf_nxv4f16_f16_nxv4f16(<vscale x 4 x half> %0, half %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vf_nxv4f16_f16_nxv4f16:
@@ -688,7 +688,7 @@ entry:
     half %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 4 x half> %a
 }
@@ -720,7 +720,7 @@ declare <vscale x 8 x half> @llvm.riscv.vfmacc.mask.nxv8f16.f16(
   half,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x half> @intrinsic_vfmacc_mask_vf_nxv8f16_f16_nxv8f16(<vscale x 8 x half> %0, half %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vf_nxv8f16_f16_nxv8f16:
@@ -734,7 +734,7 @@ entry:
     half %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 8 x half> %a
 }
@@ -766,7 +766,7 @@ declare <vscale x 16 x half> @llvm.riscv.vfmacc.mask.nxv16f16.f16(
   half,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x half> @intrinsic_vfmacc_mask_vf_nxv16f16_f16_nxv16f16(<vscale x 16 x half> %0, half %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vf_nxv16f16_f16_nxv16f16:
@@ -780,7 +780,7 @@ entry:
     half %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 16 x half> %a
 }
@@ -812,7 +812,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfmacc.mask.nxv1f32.f32(
   float,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float> @intrinsic_vfmacc_mask_vf_nxv1f32_f32_nxv1f32(<vscale x 1 x float> %0, float %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vf_nxv1f32_f32_nxv1f32:
@@ -826,7 +826,7 @@ entry:
     float %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 1 x float> %a
 }
@@ -858,7 +858,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfmacc.mask.nxv2f32.f32(
   float,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float> @intrinsic_vfmacc_mask_vf_nxv2f32_f32_nxv2f32(<vscale x 2 x float> %0, float %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vf_nxv2f32_f32_nxv2f32:
@@ -872,7 +872,7 @@ entry:
     float %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 2 x float> %a
 }
@@ -904,7 +904,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfmacc.mask.nxv4f32.f32(
   float,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float> @intrinsic_vfmacc_mask_vf_nxv4f32_f32_nxv4f32(<vscale x 4 x float> %0, float %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vf_nxv4f32_f32_nxv4f32:
@@ -918,7 +918,7 @@ entry:
     float %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 4 x float> %a
 }
@@ -950,7 +950,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfmacc.mask.nxv8f32.f32(
   float,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float> @intrinsic_vfmacc_mask_vf_nxv8f32_f32_nxv8f32(<vscale x 8 x float> %0, float %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vf_nxv8f32_f32_nxv8f32:
@@ -964,7 +964,7 @@ entry:
     float %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 8 x float> %a
 }
@@ -996,7 +996,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfmacc.mask.nxv1f64.f64(
   double,
   <vscale x 1 x double>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double> @intrinsic_vfmacc_mask_vf_nxv1f64_f64_nxv1f64(<vscale x 1 x double> %0, double %1, <vscale x 1 x double> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vf_nxv1f64_f64_nxv1f64:
@@ -1010,7 +1010,7 @@ entry:
     double %1,
     <vscale x 1 x double> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 1 x double> %a
 }
@@ -1042,7 +1042,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfmacc.mask.nxv2f64.f64(
   double,
   <vscale x 2 x double>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double> @intrinsic_vfmacc_mask_vf_nxv2f64_f64_nxv2f64(<vscale x 2 x double> %0, double %1, <vscale x 2 x double> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vf_nxv2f64_f64_nxv2f64:
@@ -1056,7 +1056,7 @@ entry:
     double %1,
     <vscale x 2 x double> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 2 x double> %a
 }
@@ -1088,7 +1088,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfmacc.mask.nxv4f64.f64(
   double,
   <vscale x 4 x double>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double> @intrinsic_vfmacc_mask_vf_nxv4f64_f64_nxv4f64(<vscale x 4 x double> %0, double %1, <vscale x 4 x double> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmacc_mask_vf_nxv4f64_f64_nxv4f64:
@@ -1102,7 +1102,7 @@ entry:
     double %1,
     <vscale x 4 x double> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0)
 
   ret <vscale x 4 x double> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vfmadd.ll b/llvm/test/CodeGen/RISCV/rvv/vfmadd.ll
index 9313e440e500f9..4b2ffbdc863e03 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vfmadd.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vfmadd.ll
@@ -30,7 +30,7 @@ declare <vscale x 1 x half> @llvm.riscv.vfmadd.mask.nxv1f16.nxv1f16(
   <vscale x 1 x half>,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x half>  @intrinsic_vfmadd_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vv_nxv1f16_nxv1f16_nxv1f16:
@@ -44,7 +44,7 @@ entry:
     <vscale x 1 x half> %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x half> %a
 }
@@ -76,7 +76,7 @@ declare <vscale x 2 x half> @llvm.riscv.vfmadd.mask.nxv2f16.nxv2f16(
   <vscale x 2 x half>,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x half>  @intrinsic_vfmadd_mask_vv_nxv2f16_nxv2f16_nxv2f16(<vscale x 2 x half> %0, <vscale x 2 x half> %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vv_nxv2f16_nxv2f16_nxv2f16:
@@ -90,7 +90,7 @@ entry:
     <vscale x 2 x half> %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x half> %a
 }
@@ -122,7 +122,7 @@ declare <vscale x 4 x half> @llvm.riscv.vfmadd.mask.nxv4f16.nxv4f16(
   <vscale x 4 x half>,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x half>  @intrinsic_vfmadd_mask_vv_nxv4f16_nxv4f16_nxv4f16(<vscale x 4 x half> %0, <vscale x 4 x half> %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vv_nxv4f16_nxv4f16_nxv4f16:
@@ -136,7 +136,7 @@ entry:
     <vscale x 4 x half> %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x half> %a
 }
@@ -168,7 +168,7 @@ declare <vscale x 8 x half> @llvm.riscv.vfmadd.mask.nxv8f16.nxv8f16(
   <vscale x 8 x half>,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x half>  @intrinsic_vfmadd_mask_vv_nxv8f16_nxv8f16_nxv8f16(<vscale x 8 x half> %0, <vscale x 8 x half> %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vv_nxv8f16_nxv8f16_nxv8f16:
@@ -182,7 +182,7 @@ entry:
     <vscale x 8 x half> %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x half> %a
 }
@@ -214,7 +214,7 @@ declare <vscale x 16 x half> @llvm.riscv.vfmadd.mask.nxv16f16.nxv16f16(
   <vscale x 16 x half>,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x half>  @intrinsic_vfmadd_mask_vv_nxv16f16_nxv16f16_nxv16f16(<vscale x 16 x half> %0, <vscale x 16 x half> %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vv_nxv16f16_nxv16f16_nxv16f16:
@@ -228,7 +228,7 @@ entry:
     <vscale x 16 x half> %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 16 x half> %a
 }
@@ -260,7 +260,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfmadd.mask.nxv1f32.nxv1f32(
   <vscale x 1 x float>,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float>  @intrinsic_vfmadd_mask_vv_nxv1f32_nxv1f32_nxv1f32(<vscale x 1 x float> %0, <vscale x 1 x float> %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vv_nxv1f32_nxv1f32_nxv1f32:
@@ -274,7 +274,7 @@ entry:
     <vscale x 1 x float> %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x float> %a
 }
@@ -306,7 +306,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfmadd.mask.nxv2f32.nxv2f32(
   <vscale x 2 x float>,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float>  @intrinsic_vfmadd_mask_vv_nxv2f32_nxv2f32_nxv2f32(<vscale x 2 x float> %0, <vscale x 2 x float> %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vv_nxv2f32_nxv2f32_nxv2f32:
@@ -320,7 +320,7 @@ entry:
     <vscale x 2 x float> %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x float> %a
 }
@@ -352,7 +352,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfmadd.mask.nxv4f32.nxv4f32(
   <vscale x 4 x float>,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float>  @intrinsic_vfmadd_mask_vv_nxv4f32_nxv4f32_nxv4f32(<vscale x 4 x float> %0, <vscale x 4 x float> %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vv_nxv4f32_nxv4f32_nxv4f32:
@@ -366,7 +366,7 @@ entry:
     <vscale x 4 x float> %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x float> %a
 }
@@ -398,7 +398,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfmadd.mask.nxv8f32.nxv8f32(
   <vscale x 8 x float>,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float>  @intrinsic_vfmadd_mask_vv_nxv8f32_nxv8f32_nxv8f32(<vscale x 8 x float> %0, <vscale x 8 x float> %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vv_nxv8f32_nxv8f32_nxv8f32:
@@ -412,7 +412,7 @@ entry:
     <vscale x 8 x float> %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x float> %a
 }
@@ -444,7 +444,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfmadd.mask.nxv1f64.nxv1f64(
   <vscale x 1 x double>,
   <vscale x 1 x double>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double>  @intrinsic_vfmadd_mask_vv_nxv1f64_nxv1f64_nxv1f64(<vscale x 1 x double> %0, <vscale x 1 x double> %1, <vscale x 1 x double> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vv_nxv1f64_nxv1f64_nxv1f64:
@@ -458,7 +458,7 @@ entry:
     <vscale x 1 x double> %1,
     <vscale x 1 x double> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x double> %a
 }
@@ -490,7 +490,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfmadd.mask.nxv2f64.nxv2f64(
   <vscale x 2 x double>,
   <vscale x 2 x double>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double>  @intrinsic_vfmadd_mask_vv_nxv2f64_nxv2f64_nxv2f64(<vscale x 2 x double> %0, <vscale x 2 x double> %1, <vscale x 2 x double> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vv_nxv2f64_nxv2f64_nxv2f64:
@@ -504,7 +504,7 @@ entry:
     <vscale x 2 x double> %1,
     <vscale x 2 x double> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x double> %a
 }
@@ -536,7 +536,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfmadd.mask.nxv4f64.nxv4f64(
   <vscale x 4 x double>,
   <vscale x 4 x double>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double>  @intrinsic_vfmadd_mask_vv_nxv4f64_nxv4f64_nxv4f64(<vscale x 4 x double> %0, <vscale x 4 x double> %1, <vscale x 4 x double> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vv_nxv4f64_nxv4f64_nxv4f64:
@@ -550,7 +550,7 @@ entry:
     <vscale x 4 x double> %1,
     <vscale x 4 x double> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x double> %a
 }
@@ -582,7 +582,7 @@ declare <vscale x 1 x half> @llvm.riscv.vfmadd.mask.nxv1f16.f16(
   half,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x half> @intrinsic_vfmadd_mask_vf_nxv1f16_f16_nxv1f16(<vscale x 1 x half> %0, half %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vf_nxv1f16_f16_nxv1f16:
@@ -596,7 +596,7 @@ entry:
     half %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x half> %a
 }
@@ -628,7 +628,7 @@ declare <vscale x 2 x half> @llvm.riscv.vfmadd.mask.nxv2f16.f16(
   half,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x half> @intrinsic_vfmadd_mask_vf_nxv2f16_f16_nxv2f16(<vscale x 2 x half> %0, half %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vf_nxv2f16_f16_nxv2f16:
@@ -642,7 +642,7 @@ entry:
     half %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x half> %a
 }
@@ -674,7 +674,7 @@ declare <vscale x 4 x half> @llvm.riscv.vfmadd.mask.nxv4f16.f16(
   half,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x half> @intrinsic_vfmadd_mask_vf_nxv4f16_f16_nxv4f16(<vscale x 4 x half> %0, half %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vf_nxv4f16_f16_nxv4f16:
@@ -688,7 +688,7 @@ entry:
     half %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x half> %a
 }
@@ -720,7 +720,7 @@ declare <vscale x 8 x half> @llvm.riscv.vfmadd.mask.nxv8f16.f16(
   half,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x half> @intrinsic_vfmadd_mask_vf_nxv8f16_f16_nxv8f16(<vscale x 8 x half> %0, half %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vf_nxv8f16_f16_nxv8f16:
@@ -734,7 +734,7 @@ entry:
     half %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x half> %a
 }
@@ -766,7 +766,7 @@ declare <vscale x 16 x half> @llvm.riscv.vfmadd.mask.nxv16f16.f16(
   half,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x half> @intrinsic_vfmadd_mask_vf_nxv16f16_f16_nxv16f16(<vscale x 16 x half> %0, half %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vf_nxv16f16_f16_nxv16f16:
@@ -780,7 +780,7 @@ entry:
     half %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 16 x half> %a
 }
@@ -812,7 +812,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfmadd.mask.nxv1f32.f32(
   float,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float> @intrinsic_vfmadd_mask_vf_nxv1f32_f32_nxv1f32(<vscale x 1 x float> %0, float %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vf_nxv1f32_f32_nxv1f32:
@@ -826,7 +826,7 @@ entry:
     float %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x float> %a
 }
@@ -858,7 +858,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfmadd.mask.nxv2f32.f32(
   float,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float> @intrinsic_vfmadd_mask_vf_nxv2f32_f32_nxv2f32(<vscale x 2 x float> %0, float %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vf_nxv2f32_f32_nxv2f32:
@@ -872,7 +872,7 @@ entry:
     float %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x float> %a
 }
@@ -904,7 +904,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfmadd.mask.nxv4f32.f32(
   float,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float> @intrinsic_vfmadd_mask_vf_nxv4f32_f32_nxv4f32(<vscale x 4 x float> %0, float %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vf_nxv4f32_f32_nxv4f32:
@@ -918,7 +918,7 @@ entry:
     float %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x float> %a
 }
@@ -950,7 +950,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfmadd.mask.nxv8f32.f32(
   float,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float> @intrinsic_vfmadd_mask_vf_nxv8f32_f32_nxv8f32(<vscale x 8 x float> %0, float %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vf_nxv8f32_f32_nxv8f32:
@@ -964,7 +964,7 @@ entry:
     float %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x float> %a
 }
@@ -996,7 +996,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfmadd.mask.nxv1f64.f64(
   double,
   <vscale x 1 x double>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double> @intrinsic_vfmadd_mask_vf_nxv1f64_f64_nxv1f64(<vscale x 1 x double> %0, double %1, <vscale x 1 x double> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vf_nxv1f64_f64_nxv1f64:
@@ -1010,7 +1010,7 @@ entry:
     double %1,
     <vscale x 1 x double> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x double> %a
 }
@@ -1042,7 +1042,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfmadd.mask.nxv2f64.f64(
   double,
   <vscale x 2 x double>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double> @intrinsic_vfmadd_mask_vf_nxv2f64_f64_nxv2f64(<vscale x 2 x double> %0, double %1, <vscale x 2 x double> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vf_nxv2f64_f64_nxv2f64:
@@ -1056,7 +1056,7 @@ entry:
     double %1,
     <vscale x 2 x double> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x double> %a
 }
@@ -1088,7 +1088,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfmadd.mask.nxv4f64.f64(
   double,
   <vscale x 4 x double>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double> @intrinsic_vfmadd_mask_vf_nxv4f64_f64_nxv4f64(<vscale x 4 x double> %0, double %1, <vscale x 4 x double> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmadd_mask_vf_nxv4f64_f64_nxv4f64:
@@ -1102,7 +1102,7 @@ entry:
     double %1,
     <vscale x 4 x double> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x double> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vfmsac.ll b/llvm/test/CodeGen/RISCV/rvv/vfmsac.ll
index cf9df7550fcc7e..6e7e3acadea40a 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vfmsac.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vfmsac.ll
@@ -30,7 +30,7 @@ declare <vscale x 1 x half> @llvm.riscv.vfmsac.mask.nxv1f16.nxv1f16(
   <vscale x 1 x half>,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x half>  @intrinsic_vfmsac_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vv_nxv1f16_nxv1f16_nxv1f16:
@@ -44,7 +44,7 @@ entry:
     <vscale x 1 x half> %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x half> %a
 }
@@ -76,7 +76,7 @@ declare <vscale x 2 x half> @llvm.riscv.vfmsac.mask.nxv2f16.nxv2f16(
   <vscale x 2 x half>,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x half>  @intrinsic_vfmsac_mask_vv_nxv2f16_nxv2f16_nxv2f16(<vscale x 2 x half> %0, <vscale x 2 x half> %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vv_nxv2f16_nxv2f16_nxv2f16:
@@ -90,7 +90,7 @@ entry:
     <vscale x 2 x half> %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x half> %a
 }
@@ -122,7 +122,7 @@ declare <vscale x 4 x half> @llvm.riscv.vfmsac.mask.nxv4f16.nxv4f16(
   <vscale x 4 x half>,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x half>  @intrinsic_vfmsac_mask_vv_nxv4f16_nxv4f16_nxv4f16(<vscale x 4 x half> %0, <vscale x 4 x half> %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vv_nxv4f16_nxv4f16_nxv4f16:
@@ -136,7 +136,7 @@ entry:
     <vscale x 4 x half> %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x half> %a
 }
@@ -168,7 +168,7 @@ declare <vscale x 8 x half> @llvm.riscv.vfmsac.mask.nxv8f16.nxv8f16(
   <vscale x 8 x half>,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x half>  @intrinsic_vfmsac_mask_vv_nxv8f16_nxv8f16_nxv8f16(<vscale x 8 x half> %0, <vscale x 8 x half> %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vv_nxv8f16_nxv8f16_nxv8f16:
@@ -182,7 +182,7 @@ entry:
     <vscale x 8 x half> %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x half> %a
 }
@@ -214,7 +214,7 @@ declare <vscale x 16 x half> @llvm.riscv.vfmsac.mask.nxv16f16.nxv16f16(
   <vscale x 16 x half>,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x half>  @intrinsic_vfmsac_mask_vv_nxv16f16_nxv16f16_nxv16f16(<vscale x 16 x half> %0, <vscale x 16 x half> %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vv_nxv16f16_nxv16f16_nxv16f16:
@@ -228,7 +228,7 @@ entry:
     <vscale x 16 x half> %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 16 x half> %a
 }
@@ -260,7 +260,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfmsac.mask.nxv1f32.nxv1f32(
   <vscale x 1 x float>,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float>  @intrinsic_vfmsac_mask_vv_nxv1f32_nxv1f32_nxv1f32(<vscale x 1 x float> %0, <vscale x 1 x float> %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vv_nxv1f32_nxv1f32_nxv1f32:
@@ -274,7 +274,7 @@ entry:
     <vscale x 1 x float> %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x float> %a
 }
@@ -306,7 +306,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfmsac.mask.nxv2f32.nxv2f32(
   <vscale x 2 x float>,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float>  @intrinsic_vfmsac_mask_vv_nxv2f32_nxv2f32_nxv2f32(<vscale x 2 x float> %0, <vscale x 2 x float> %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vv_nxv2f32_nxv2f32_nxv2f32:
@@ -320,7 +320,7 @@ entry:
     <vscale x 2 x float> %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x float> %a
 }
@@ -352,7 +352,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfmsac.mask.nxv4f32.nxv4f32(
   <vscale x 4 x float>,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float>  @intrinsic_vfmsac_mask_vv_nxv4f32_nxv4f32_nxv4f32(<vscale x 4 x float> %0, <vscale x 4 x float> %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vv_nxv4f32_nxv4f32_nxv4f32:
@@ -366,7 +366,7 @@ entry:
     <vscale x 4 x float> %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x float> %a
 }
@@ -398,7 +398,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfmsac.mask.nxv8f32.nxv8f32(
   <vscale x 8 x float>,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float>  @intrinsic_vfmsac_mask_vv_nxv8f32_nxv8f32_nxv8f32(<vscale x 8 x float> %0, <vscale x 8 x float> %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vv_nxv8f32_nxv8f32_nxv8f32:
@@ -412,7 +412,7 @@ entry:
     <vscale x 8 x float> %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x float> %a
 }
@@ -444,7 +444,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfmsac.mask.nxv1f64.nxv1f64(
   <vscale x 1 x double>,
   <vscale x 1 x double>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double>  @intrinsic_vfmsac_mask_vv_nxv1f64_nxv1f64_nxv1f64(<vscale x 1 x double> %0, <vscale x 1 x double> %1, <vscale x 1 x double> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vv_nxv1f64_nxv1f64_nxv1f64:
@@ -458,7 +458,7 @@ entry:
     <vscale x 1 x double> %1,
     <vscale x 1 x double> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x double> %a
 }
@@ -490,7 +490,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfmsac.mask.nxv2f64.nxv2f64(
   <vscale x 2 x double>,
   <vscale x 2 x double>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double>  @intrinsic_vfmsac_mask_vv_nxv2f64_nxv2f64_nxv2f64(<vscale x 2 x double> %0, <vscale x 2 x double> %1, <vscale x 2 x double> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vv_nxv2f64_nxv2f64_nxv2f64:
@@ -504,7 +504,7 @@ entry:
     <vscale x 2 x double> %1,
     <vscale x 2 x double> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x double> %a
 }
@@ -536,7 +536,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfmsac.mask.nxv4f64.nxv4f64(
   <vscale x 4 x double>,
   <vscale x 4 x double>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double>  @intrinsic_vfmsac_mask_vv_nxv4f64_nxv4f64_nxv4f64(<vscale x 4 x double> %0, <vscale x 4 x double> %1, <vscale x 4 x double> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vv_nxv4f64_nxv4f64_nxv4f64:
@@ -550,7 +550,7 @@ entry:
     <vscale x 4 x double> %1,
     <vscale x 4 x double> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x double> %a
 }
@@ -582,7 +582,7 @@ declare <vscale x 1 x half> @llvm.riscv.vfmsac.mask.nxv1f16.f16(
   half,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x half> @intrinsic_vfmsac_mask_vf_nxv1f16_f16_nxv1f16(<vscale x 1 x half> %0, half %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vf_nxv1f16_f16_nxv1f16:
@@ -596,7 +596,7 @@ entry:
     half %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x half> %a
 }
@@ -628,7 +628,7 @@ declare <vscale x 2 x half> @llvm.riscv.vfmsac.mask.nxv2f16.f16(
   half,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x half> @intrinsic_vfmsac_mask_vf_nxv2f16_f16_nxv2f16(<vscale x 2 x half> %0, half %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vf_nxv2f16_f16_nxv2f16:
@@ -642,7 +642,7 @@ entry:
     half %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x half> %a
 }
@@ -674,7 +674,7 @@ declare <vscale x 4 x half> @llvm.riscv.vfmsac.mask.nxv4f16.f16(
   half,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x half> @intrinsic_vfmsac_mask_vf_nxv4f16_f16_nxv4f16(<vscale x 4 x half> %0, half %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vf_nxv4f16_f16_nxv4f16:
@@ -688,7 +688,7 @@ entry:
     half %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x half> %a
 }
@@ -720,7 +720,7 @@ declare <vscale x 8 x half> @llvm.riscv.vfmsac.mask.nxv8f16.f16(
   half,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x half> @intrinsic_vfmsac_mask_vf_nxv8f16_f16_nxv8f16(<vscale x 8 x half> %0, half %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vf_nxv8f16_f16_nxv8f16:
@@ -734,7 +734,7 @@ entry:
     half %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x half> %a
 }
@@ -766,7 +766,7 @@ declare <vscale x 16 x half> @llvm.riscv.vfmsac.mask.nxv16f16.f16(
   half,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x half> @intrinsic_vfmsac_mask_vf_nxv16f16_f16_nxv16f16(<vscale x 16 x half> %0, half %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vf_nxv16f16_f16_nxv16f16:
@@ -780,7 +780,7 @@ entry:
     half %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 16 x half> %a
 }
@@ -812,7 +812,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfmsac.mask.nxv1f32.f32(
   float,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float> @intrinsic_vfmsac_mask_vf_nxv1f32_f32_nxv1f32(<vscale x 1 x float> %0, float %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vf_nxv1f32_f32_nxv1f32:
@@ -826,7 +826,7 @@ entry:
     float %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x float> %a
 }
@@ -858,7 +858,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfmsac.mask.nxv2f32.f32(
   float,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float> @intrinsic_vfmsac_mask_vf_nxv2f32_f32_nxv2f32(<vscale x 2 x float> %0, float %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vf_nxv2f32_f32_nxv2f32:
@@ -872,7 +872,7 @@ entry:
     float %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x float> %a
 }
@@ -904,7 +904,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfmsac.mask.nxv4f32.f32(
   float,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float> @intrinsic_vfmsac_mask_vf_nxv4f32_f32_nxv4f32(<vscale x 4 x float> %0, float %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vf_nxv4f32_f32_nxv4f32:
@@ -918,7 +918,7 @@ entry:
     float %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x float> %a
 }
@@ -950,7 +950,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfmsac.mask.nxv8f32.f32(
   float,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float> @intrinsic_vfmsac_mask_vf_nxv8f32_f32_nxv8f32(<vscale x 8 x float> %0, float %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vf_nxv8f32_f32_nxv8f32:
@@ -964,7 +964,7 @@ entry:
     float %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x float> %a
 }
@@ -996,7 +996,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfmsac.mask.nxv1f64.f64(
   double,
   <vscale x 1 x double>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double> @intrinsic_vfmsac_mask_vf_nxv1f64_f64_nxv1f64(<vscale x 1 x double> %0, double %1, <vscale x 1 x double> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vf_nxv1f64_f64_nxv1f64:
@@ -1010,7 +1010,7 @@ entry:
     double %1,
     <vscale x 1 x double> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x double> %a
 }
@@ -1042,7 +1042,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfmsac.mask.nxv2f64.f64(
   double,
   <vscale x 2 x double>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double> @intrinsic_vfmsac_mask_vf_nxv2f64_f64_nxv2f64(<vscale x 2 x double> %0, double %1, <vscale x 2 x double> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vf_nxv2f64_f64_nxv2f64:
@@ -1056,7 +1056,7 @@ entry:
     double %1,
     <vscale x 2 x double> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x double> %a
 }
@@ -1088,7 +1088,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfmsac.mask.nxv4f64.f64(
   double,
   <vscale x 4 x double>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double> @intrinsic_vfmsac_mask_vf_nxv4f64_f64_nxv4f64(<vscale x 4 x double> %0, double %1, <vscale x 4 x double> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsac_mask_vf_nxv4f64_f64_nxv4f64:
@@ -1102,7 +1102,7 @@ entry:
     double %1,
     <vscale x 4 x double> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x double> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vfmsub.ll b/llvm/test/CodeGen/RISCV/rvv/vfmsub.ll
index d071893ceb0850..0f7a58a9480acf 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vfmsub.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vfmsub.ll
@@ -30,7 +30,7 @@ declare <vscale x 1 x half> @llvm.riscv.vfmsub.mask.nxv1f16.nxv1f16(
   <vscale x 1 x half>,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x half>  @intrinsic_vfmsub_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vv_nxv1f16_nxv1f16_nxv1f16:
@@ -44,7 +44,7 @@ entry:
     <vscale x 1 x half> %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x half> %a
 }
@@ -76,7 +76,7 @@ declare <vscale x 2 x half> @llvm.riscv.vfmsub.mask.nxv2f16.nxv2f16(
   <vscale x 2 x half>,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x half>  @intrinsic_vfmsub_mask_vv_nxv2f16_nxv2f16_nxv2f16(<vscale x 2 x half> %0, <vscale x 2 x half> %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vv_nxv2f16_nxv2f16_nxv2f16:
@@ -90,7 +90,7 @@ entry:
     <vscale x 2 x half> %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x half> %a
 }
@@ -122,7 +122,7 @@ declare <vscale x 4 x half> @llvm.riscv.vfmsub.mask.nxv4f16.nxv4f16(
   <vscale x 4 x half>,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x half>  @intrinsic_vfmsub_mask_vv_nxv4f16_nxv4f16_nxv4f16(<vscale x 4 x half> %0, <vscale x 4 x half> %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vv_nxv4f16_nxv4f16_nxv4f16:
@@ -136,7 +136,7 @@ entry:
     <vscale x 4 x half> %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x half> %a
 }
@@ -168,7 +168,7 @@ declare <vscale x 8 x half> @llvm.riscv.vfmsub.mask.nxv8f16.nxv8f16(
   <vscale x 8 x half>,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x half>  @intrinsic_vfmsub_mask_vv_nxv8f16_nxv8f16_nxv8f16(<vscale x 8 x half> %0, <vscale x 8 x half> %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vv_nxv8f16_nxv8f16_nxv8f16:
@@ -182,7 +182,7 @@ entry:
     <vscale x 8 x half> %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x half> %a
 }
@@ -214,7 +214,7 @@ declare <vscale x 16 x half> @llvm.riscv.vfmsub.mask.nxv16f16.nxv16f16(
   <vscale x 16 x half>,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x half>  @intrinsic_vfmsub_mask_vv_nxv16f16_nxv16f16_nxv16f16(<vscale x 16 x half> %0, <vscale x 16 x half> %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vv_nxv16f16_nxv16f16_nxv16f16:
@@ -228,7 +228,7 @@ entry:
     <vscale x 16 x half> %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 16 x half> %a
 }
@@ -260,7 +260,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfmsub.mask.nxv1f32.nxv1f32(
   <vscale x 1 x float>,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float>  @intrinsic_vfmsub_mask_vv_nxv1f32_nxv1f32_nxv1f32(<vscale x 1 x float> %0, <vscale x 1 x float> %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vv_nxv1f32_nxv1f32_nxv1f32:
@@ -274,7 +274,7 @@ entry:
     <vscale x 1 x float> %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x float> %a
 }
@@ -306,7 +306,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfmsub.mask.nxv2f32.nxv2f32(
   <vscale x 2 x float>,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float>  @intrinsic_vfmsub_mask_vv_nxv2f32_nxv2f32_nxv2f32(<vscale x 2 x float> %0, <vscale x 2 x float> %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vv_nxv2f32_nxv2f32_nxv2f32:
@@ -320,7 +320,7 @@ entry:
     <vscale x 2 x float> %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x float> %a
 }
@@ -352,7 +352,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfmsub.mask.nxv4f32.nxv4f32(
   <vscale x 4 x float>,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float>  @intrinsic_vfmsub_mask_vv_nxv4f32_nxv4f32_nxv4f32(<vscale x 4 x float> %0, <vscale x 4 x float> %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vv_nxv4f32_nxv4f32_nxv4f32:
@@ -366,7 +366,7 @@ entry:
     <vscale x 4 x float> %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x float> %a
 }
@@ -398,7 +398,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfmsub.mask.nxv8f32.nxv8f32(
   <vscale x 8 x float>,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float>  @intrinsic_vfmsub_mask_vv_nxv8f32_nxv8f32_nxv8f32(<vscale x 8 x float> %0, <vscale x 8 x float> %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vv_nxv8f32_nxv8f32_nxv8f32:
@@ -412,7 +412,7 @@ entry:
     <vscale x 8 x float> %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x float> %a
 }
@@ -444,7 +444,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfmsub.mask.nxv1f64.nxv1f64(
   <vscale x 1 x double>,
   <vscale x 1 x double>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double>  @intrinsic_vfmsub_mask_vv_nxv1f64_nxv1f64_nxv1f64(<vscale x 1 x double> %0, <vscale x 1 x double> %1, <vscale x 1 x double> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vv_nxv1f64_nxv1f64_nxv1f64:
@@ -458,7 +458,7 @@ entry:
     <vscale x 1 x double> %1,
     <vscale x 1 x double> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x double> %a
 }
@@ -490,7 +490,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfmsub.mask.nxv2f64.nxv2f64(
   <vscale x 2 x double>,
   <vscale x 2 x double>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double>  @intrinsic_vfmsub_mask_vv_nxv2f64_nxv2f64_nxv2f64(<vscale x 2 x double> %0, <vscale x 2 x double> %1, <vscale x 2 x double> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vv_nxv2f64_nxv2f64_nxv2f64:
@@ -504,7 +504,7 @@ entry:
     <vscale x 2 x double> %1,
     <vscale x 2 x double> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x double> %a
 }
@@ -536,7 +536,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfmsub.mask.nxv4f64.nxv4f64(
   <vscale x 4 x double>,
   <vscale x 4 x double>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double>  @intrinsic_vfmsub_mask_vv_nxv4f64_nxv4f64_nxv4f64(<vscale x 4 x double> %0, <vscale x 4 x double> %1, <vscale x 4 x double> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vv_nxv4f64_nxv4f64_nxv4f64:
@@ -550,7 +550,7 @@ entry:
     <vscale x 4 x double> %1,
     <vscale x 4 x double> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x double> %a
 }
@@ -582,7 +582,7 @@ declare <vscale x 1 x half> @llvm.riscv.vfmsub.mask.nxv1f16.f16(
   half,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x half> @intrinsic_vfmsub_mask_vf_nxv1f16_f16_nxv1f16(<vscale x 1 x half> %0, half %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vf_nxv1f16_f16_nxv1f16:
@@ -596,7 +596,7 @@ entry:
     half %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x half> %a
 }
@@ -628,7 +628,7 @@ declare <vscale x 2 x half> @llvm.riscv.vfmsub.mask.nxv2f16.f16(
   half,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x half> @intrinsic_vfmsub_mask_vf_nxv2f16_f16_nxv2f16(<vscale x 2 x half> %0, half %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vf_nxv2f16_f16_nxv2f16:
@@ -642,7 +642,7 @@ entry:
     half %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x half> %a
 }
@@ -674,7 +674,7 @@ declare <vscale x 4 x half> @llvm.riscv.vfmsub.mask.nxv4f16.f16(
   half,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x half> @intrinsic_vfmsub_mask_vf_nxv4f16_f16_nxv4f16(<vscale x 4 x half> %0, half %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vf_nxv4f16_f16_nxv4f16:
@@ -688,7 +688,7 @@ entry:
     half %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x half> %a
 }
@@ -720,7 +720,7 @@ declare <vscale x 8 x half> @llvm.riscv.vfmsub.mask.nxv8f16.f16(
   half,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x half> @intrinsic_vfmsub_mask_vf_nxv8f16_f16_nxv8f16(<vscale x 8 x half> %0, half %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vf_nxv8f16_f16_nxv8f16:
@@ -734,7 +734,7 @@ entry:
     half %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x half> %a
 }
@@ -766,7 +766,7 @@ declare <vscale x 16 x half> @llvm.riscv.vfmsub.mask.nxv16f16.f16(
   half,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x half> @intrinsic_vfmsub_mask_vf_nxv16f16_f16_nxv16f16(<vscale x 16 x half> %0, half %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vf_nxv16f16_f16_nxv16f16:
@@ -780,7 +780,7 @@ entry:
     half %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 16 x half> %a
 }
@@ -812,7 +812,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfmsub.mask.nxv1f32.f32(
   float,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float> @intrinsic_vfmsub_mask_vf_nxv1f32_f32_nxv1f32(<vscale x 1 x float> %0, float %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vf_nxv1f32_f32_nxv1f32:
@@ -826,7 +826,7 @@ entry:
     float %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x float> %a
 }
@@ -858,7 +858,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfmsub.mask.nxv2f32.f32(
   float,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float> @intrinsic_vfmsub_mask_vf_nxv2f32_f32_nxv2f32(<vscale x 2 x float> %0, float %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vf_nxv2f32_f32_nxv2f32:
@@ -872,7 +872,7 @@ entry:
     float %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x float> %a
 }
@@ -904,7 +904,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfmsub.mask.nxv4f32.f32(
   float,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float> @intrinsic_vfmsub_mask_vf_nxv4f32_f32_nxv4f32(<vscale x 4 x float> %0, float %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vf_nxv4f32_f32_nxv4f32:
@@ -918,7 +918,7 @@ entry:
     float %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x float> %a
 }
@@ -950,7 +950,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfmsub.mask.nxv8f32.f32(
   float,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float> @intrinsic_vfmsub_mask_vf_nxv8f32_f32_nxv8f32(<vscale x 8 x float> %0, float %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vf_nxv8f32_f32_nxv8f32:
@@ -964,7 +964,7 @@ entry:
     float %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x float> %a
 }
@@ -996,7 +996,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfmsub.mask.nxv1f64.f64(
   double,
   <vscale x 1 x double>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double> @intrinsic_vfmsub_mask_vf_nxv1f64_f64_nxv1f64(<vscale x 1 x double> %0, double %1, <vscale x 1 x double> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vf_nxv1f64_f64_nxv1f64:
@@ -1010,7 +1010,7 @@ entry:
     double %1,
     <vscale x 1 x double> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x double> %a
 }
@@ -1042,7 +1042,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfmsub.mask.nxv2f64.f64(
   double,
   <vscale x 2 x double>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double> @intrinsic_vfmsub_mask_vf_nxv2f64_f64_nxv2f64(<vscale x 2 x double> %0, double %1, <vscale x 2 x double> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vf_nxv2f64_f64_nxv2f64:
@@ -1056,7 +1056,7 @@ entry:
     double %1,
     <vscale x 2 x double> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x double> %a
 }
@@ -1088,7 +1088,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfmsub.mask.nxv4f64.f64(
   double,
   <vscale x 4 x double>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double> @intrinsic_vfmsub_mask_vf_nxv4f64_f64_nxv4f64(<vscale x 4 x double> %0, double %1, <vscale x 4 x double> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfmsub_mask_vf_nxv4f64_f64_nxv4f64:
@@ -1102,7 +1102,7 @@ entry:
     double %1,
     <vscale x 4 x double> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x double> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vfnmacc.ll b/llvm/test/CodeGen/RISCV/rvv/vfnmacc.ll
index d46c29f3be78b2..2d22dc269643c4 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vfnmacc.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vfnmacc.ll
@@ -30,7 +30,7 @@ declare <vscale x 1 x half> @llvm.riscv.vfnmacc.mask.nxv1f16.nxv1f16(
   <vscale x 1 x half>,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x half>  @intrinsic_vfnmacc_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vv_nxv1f16_nxv1f16_nxv1f16:
@@ -44,7 +44,7 @@ entry:
     <vscale x 1 x half> %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x half> %a
 }
@@ -76,7 +76,7 @@ declare <vscale x 2 x half> @llvm.riscv.vfnmacc.mask.nxv2f16.nxv2f16(
   <vscale x 2 x half>,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x half>  @intrinsic_vfnmacc_mask_vv_nxv2f16_nxv2f16_nxv2f16(<vscale x 2 x half> %0, <vscale x 2 x half> %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vv_nxv2f16_nxv2f16_nxv2f16:
@@ -90,7 +90,7 @@ entry:
     <vscale x 2 x half> %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x half> %a
 }
@@ -122,7 +122,7 @@ declare <vscale x 4 x half> @llvm.riscv.vfnmacc.mask.nxv4f16.nxv4f16(
   <vscale x 4 x half>,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x half>  @intrinsic_vfnmacc_mask_vv_nxv4f16_nxv4f16_nxv4f16(<vscale x 4 x half> %0, <vscale x 4 x half> %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vv_nxv4f16_nxv4f16_nxv4f16:
@@ -136,7 +136,7 @@ entry:
     <vscale x 4 x half> %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x half> %a
 }
@@ -168,7 +168,7 @@ declare <vscale x 8 x half> @llvm.riscv.vfnmacc.mask.nxv8f16.nxv8f16(
   <vscale x 8 x half>,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x half>  @intrinsic_vfnmacc_mask_vv_nxv8f16_nxv8f16_nxv8f16(<vscale x 8 x half> %0, <vscale x 8 x half> %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vv_nxv8f16_nxv8f16_nxv8f16:
@@ -182,7 +182,7 @@ entry:
     <vscale x 8 x half> %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x half> %a
 }
@@ -214,7 +214,7 @@ declare <vscale x 16 x half> @llvm.riscv.vfnmacc.mask.nxv16f16.nxv16f16(
   <vscale x 16 x half>,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x half>  @intrinsic_vfnmacc_mask_vv_nxv16f16_nxv16f16_nxv16f16(<vscale x 16 x half> %0, <vscale x 16 x half> %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vv_nxv16f16_nxv16f16_nxv16f16:
@@ -228,7 +228,7 @@ entry:
     <vscale x 16 x half> %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 16 x half> %a
 }
@@ -260,7 +260,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfnmacc.mask.nxv1f32.nxv1f32(
   <vscale x 1 x float>,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float>  @intrinsic_vfnmacc_mask_vv_nxv1f32_nxv1f32_nxv1f32(<vscale x 1 x float> %0, <vscale x 1 x float> %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vv_nxv1f32_nxv1f32_nxv1f32:
@@ -274,7 +274,7 @@ entry:
     <vscale x 1 x float> %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x float> %a
 }
@@ -306,7 +306,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfnmacc.mask.nxv2f32.nxv2f32(
   <vscale x 2 x float>,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float>  @intrinsic_vfnmacc_mask_vv_nxv2f32_nxv2f32_nxv2f32(<vscale x 2 x float> %0, <vscale x 2 x float> %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vv_nxv2f32_nxv2f32_nxv2f32:
@@ -320,7 +320,7 @@ entry:
     <vscale x 2 x float> %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x float> %a
 }
@@ -352,7 +352,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfnmacc.mask.nxv4f32.nxv4f32(
   <vscale x 4 x float>,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float>  @intrinsic_vfnmacc_mask_vv_nxv4f32_nxv4f32_nxv4f32(<vscale x 4 x float> %0, <vscale x 4 x float> %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vv_nxv4f32_nxv4f32_nxv4f32:
@@ -366,7 +366,7 @@ entry:
     <vscale x 4 x float> %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x float> %a
 }
@@ -398,7 +398,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfnmacc.mask.nxv8f32.nxv8f32(
   <vscale x 8 x float>,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float>  @intrinsic_vfnmacc_mask_vv_nxv8f32_nxv8f32_nxv8f32(<vscale x 8 x float> %0, <vscale x 8 x float> %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vv_nxv8f32_nxv8f32_nxv8f32:
@@ -412,7 +412,7 @@ entry:
     <vscale x 8 x float> %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x float> %a
 }
@@ -444,7 +444,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfnmacc.mask.nxv1f64.nxv1f64(
   <vscale x 1 x double>,
   <vscale x 1 x double>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double>  @intrinsic_vfnmacc_mask_vv_nxv1f64_nxv1f64_nxv1f64(<vscale x 1 x double> %0, <vscale x 1 x double> %1, <vscale x 1 x double> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vv_nxv1f64_nxv1f64_nxv1f64:
@@ -458,7 +458,7 @@ entry:
     <vscale x 1 x double> %1,
     <vscale x 1 x double> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x double> %a
 }
@@ -490,7 +490,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfnmacc.mask.nxv2f64.nxv2f64(
   <vscale x 2 x double>,
   <vscale x 2 x double>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double>  @intrinsic_vfnmacc_mask_vv_nxv2f64_nxv2f64_nxv2f64(<vscale x 2 x double> %0, <vscale x 2 x double> %1, <vscale x 2 x double> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vv_nxv2f64_nxv2f64_nxv2f64:
@@ -504,7 +504,7 @@ entry:
     <vscale x 2 x double> %1,
     <vscale x 2 x double> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x double> %a
 }
@@ -536,7 +536,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfnmacc.mask.nxv4f64.nxv4f64(
   <vscale x 4 x double>,
   <vscale x 4 x double>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double>  @intrinsic_vfnmacc_mask_vv_nxv4f64_nxv4f64_nxv4f64(<vscale x 4 x double> %0, <vscale x 4 x double> %1, <vscale x 4 x double> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vv_nxv4f64_nxv4f64_nxv4f64:
@@ -550,7 +550,7 @@ entry:
     <vscale x 4 x double> %1,
     <vscale x 4 x double> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x double> %a
 }
@@ -582,7 +582,7 @@ declare <vscale x 1 x half> @llvm.riscv.vfnmacc.mask.nxv1f16.f16(
   half,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x half> @intrinsic_vfnmacc_mask_vf_nxv1f16_f16_nxv1f16(<vscale x 1 x half> %0, half %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vf_nxv1f16_f16_nxv1f16:
@@ -596,7 +596,7 @@ entry:
     half %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x half> %a
 }
@@ -628,7 +628,7 @@ declare <vscale x 2 x half> @llvm.riscv.vfnmacc.mask.nxv2f16.f16(
   half,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x half> @intrinsic_vfnmacc_mask_vf_nxv2f16_f16_nxv2f16(<vscale x 2 x half> %0, half %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vf_nxv2f16_f16_nxv2f16:
@@ -642,7 +642,7 @@ entry:
     half %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x half> %a
 }
@@ -674,7 +674,7 @@ declare <vscale x 4 x half> @llvm.riscv.vfnmacc.mask.nxv4f16.f16(
   half,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x half> @intrinsic_vfnmacc_mask_vf_nxv4f16_f16_nxv4f16(<vscale x 4 x half> %0, half %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vf_nxv4f16_f16_nxv4f16:
@@ -688,7 +688,7 @@ entry:
     half %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x half> %a
 }
@@ -720,7 +720,7 @@ declare <vscale x 8 x half> @llvm.riscv.vfnmacc.mask.nxv8f16.f16(
   half,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x half> @intrinsic_vfnmacc_mask_vf_nxv8f16_f16_nxv8f16(<vscale x 8 x half> %0, half %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vf_nxv8f16_f16_nxv8f16:
@@ -734,7 +734,7 @@ entry:
     half %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x half> %a
 }
@@ -766,7 +766,7 @@ declare <vscale x 16 x half> @llvm.riscv.vfnmacc.mask.nxv16f16.f16(
   half,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x half> @intrinsic_vfnmacc_mask_vf_nxv16f16_f16_nxv16f16(<vscale x 16 x half> %0, half %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vf_nxv16f16_f16_nxv16f16:
@@ -780,7 +780,7 @@ entry:
     half %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 16 x half> %a
 }
@@ -812,7 +812,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfnmacc.mask.nxv1f32.f32(
   float,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float> @intrinsic_vfnmacc_mask_vf_nxv1f32_f32_nxv1f32(<vscale x 1 x float> %0, float %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vf_nxv1f32_f32_nxv1f32:
@@ -826,7 +826,7 @@ entry:
     float %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x float> %a
 }
@@ -858,7 +858,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfnmacc.mask.nxv2f32.f32(
   float,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float> @intrinsic_vfnmacc_mask_vf_nxv2f32_f32_nxv2f32(<vscale x 2 x float> %0, float %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vf_nxv2f32_f32_nxv2f32:
@@ -872,7 +872,7 @@ entry:
     float %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x float> %a
 }
@@ -904,7 +904,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfnmacc.mask.nxv4f32.f32(
   float,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float> @intrinsic_vfnmacc_mask_vf_nxv4f32_f32_nxv4f32(<vscale x 4 x float> %0, float %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vf_nxv4f32_f32_nxv4f32:
@@ -918,7 +918,7 @@ entry:
     float %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x float> %a
 }
@@ -950,7 +950,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfnmacc.mask.nxv8f32.f32(
   float,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float> @intrinsic_vfnmacc_mask_vf_nxv8f32_f32_nxv8f32(<vscale x 8 x float> %0, float %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vf_nxv8f32_f32_nxv8f32:
@@ -964,7 +964,7 @@ entry:
     float %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x float> %a
 }
@@ -996,7 +996,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfnmacc.mask.nxv1f64.f64(
   double,
   <vscale x 1 x double>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double> @intrinsic_vfnmacc_mask_vf_nxv1f64_f64_nxv1f64(<vscale x 1 x double> %0, double %1, <vscale x 1 x double> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vf_nxv1f64_f64_nxv1f64:
@@ -1010,7 +1010,7 @@ entry:
     double %1,
     <vscale x 1 x double> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x double> %a
 }
@@ -1042,7 +1042,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfnmacc.mask.nxv2f64.f64(
   double,
   <vscale x 2 x double>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double> @intrinsic_vfnmacc_mask_vf_nxv2f64_f64_nxv2f64(<vscale x 2 x double> %0, double %1, <vscale x 2 x double> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vf_nxv2f64_f64_nxv2f64:
@@ -1056,7 +1056,7 @@ entry:
     double %1,
     <vscale x 2 x double> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x double> %a
 }
@@ -1088,7 +1088,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfnmacc.mask.nxv4f64.f64(
   double,
   <vscale x 4 x double>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double> @intrinsic_vfnmacc_mask_vf_nxv4f64_f64_nxv4f64(<vscale x 4 x double> %0, double %1, <vscale x 4 x double> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmacc_mask_vf_nxv4f64_f64_nxv4f64:
@@ -1102,7 +1102,7 @@ entry:
     double %1,
     <vscale x 4 x double> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x double> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vfnmadd.ll b/llvm/test/CodeGen/RISCV/rvv/vfnmadd.ll
index 44810af5ab31b6..4a7cf4d010256f 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vfnmadd.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vfnmadd.ll
@@ -30,7 +30,7 @@ declare <vscale x 1 x half> @llvm.riscv.vfnmadd.mask.nxv1f16.nxv1f16(
   <vscale x 1 x half>,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x half>  @intrinsic_vfnmadd_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vv_nxv1f16_nxv1f16_nxv1f16:
@@ -44,7 +44,7 @@ entry:
     <vscale x 1 x half> %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x half> %a
 }
@@ -76,7 +76,7 @@ declare <vscale x 2 x half> @llvm.riscv.vfnmadd.mask.nxv2f16.nxv2f16(
   <vscale x 2 x half>,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x half>  @intrinsic_vfnmadd_mask_vv_nxv2f16_nxv2f16_nxv2f16(<vscale x 2 x half> %0, <vscale x 2 x half> %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vv_nxv2f16_nxv2f16_nxv2f16:
@@ -90,7 +90,7 @@ entry:
     <vscale x 2 x half> %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x half> %a
 }
@@ -122,7 +122,7 @@ declare <vscale x 4 x half> @llvm.riscv.vfnmadd.mask.nxv4f16.nxv4f16(
   <vscale x 4 x half>,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x half>  @intrinsic_vfnmadd_mask_vv_nxv4f16_nxv4f16_nxv4f16(<vscale x 4 x half> %0, <vscale x 4 x half> %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vv_nxv4f16_nxv4f16_nxv4f16:
@@ -136,7 +136,7 @@ entry:
     <vscale x 4 x half> %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x half> %a
 }
@@ -168,7 +168,7 @@ declare <vscale x 8 x half> @llvm.riscv.vfnmadd.mask.nxv8f16.nxv8f16(
   <vscale x 8 x half>,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x half>  @intrinsic_vfnmadd_mask_vv_nxv8f16_nxv8f16_nxv8f16(<vscale x 8 x half> %0, <vscale x 8 x half> %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vv_nxv8f16_nxv8f16_nxv8f16:
@@ -182,7 +182,7 @@ entry:
     <vscale x 8 x half> %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x half> %a
 }
@@ -214,7 +214,7 @@ declare <vscale x 16 x half> @llvm.riscv.vfnmadd.mask.nxv16f16.nxv16f16(
   <vscale x 16 x half>,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x half>  @intrinsic_vfnmadd_mask_vv_nxv16f16_nxv16f16_nxv16f16(<vscale x 16 x half> %0, <vscale x 16 x half> %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vv_nxv16f16_nxv16f16_nxv16f16:
@@ -228,7 +228,7 @@ entry:
     <vscale x 16 x half> %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 16 x half> %a
 }
@@ -260,7 +260,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfnmadd.mask.nxv1f32.nxv1f32(
   <vscale x 1 x float>,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float>  @intrinsic_vfnmadd_mask_vv_nxv1f32_nxv1f32_nxv1f32(<vscale x 1 x float> %0, <vscale x 1 x float> %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vv_nxv1f32_nxv1f32_nxv1f32:
@@ -274,7 +274,7 @@ entry:
     <vscale x 1 x float> %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x float> %a
 }
@@ -306,7 +306,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfnmadd.mask.nxv2f32.nxv2f32(
   <vscale x 2 x float>,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float>  @intrinsic_vfnmadd_mask_vv_nxv2f32_nxv2f32_nxv2f32(<vscale x 2 x float> %0, <vscale x 2 x float> %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vv_nxv2f32_nxv2f32_nxv2f32:
@@ -320,7 +320,7 @@ entry:
     <vscale x 2 x float> %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x float> %a
 }
@@ -352,7 +352,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfnmadd.mask.nxv4f32.nxv4f32(
   <vscale x 4 x float>,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float>  @intrinsic_vfnmadd_mask_vv_nxv4f32_nxv4f32_nxv4f32(<vscale x 4 x float> %0, <vscale x 4 x float> %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vv_nxv4f32_nxv4f32_nxv4f32:
@@ -366,7 +366,7 @@ entry:
     <vscale x 4 x float> %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x float> %a
 }
@@ -398,7 +398,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfnmadd.mask.nxv8f32.nxv8f32(
   <vscale x 8 x float>,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float>  @intrinsic_vfnmadd_mask_vv_nxv8f32_nxv8f32_nxv8f32(<vscale x 8 x float> %0, <vscale x 8 x float> %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vv_nxv8f32_nxv8f32_nxv8f32:
@@ -412,7 +412,7 @@ entry:
     <vscale x 8 x float> %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x float> %a
 }
@@ -444,7 +444,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfnmadd.mask.nxv1f64.nxv1f64(
   <vscale x 1 x double>,
   <vscale x 1 x double>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double>  @intrinsic_vfnmadd_mask_vv_nxv1f64_nxv1f64_nxv1f64(<vscale x 1 x double> %0, <vscale x 1 x double> %1, <vscale x 1 x double> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vv_nxv1f64_nxv1f64_nxv1f64:
@@ -458,7 +458,7 @@ entry:
     <vscale x 1 x double> %1,
     <vscale x 1 x double> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x double> %a
 }
@@ -490,7 +490,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfnmadd.mask.nxv2f64.nxv2f64(
   <vscale x 2 x double>,
   <vscale x 2 x double>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double>  @intrinsic_vfnmadd_mask_vv_nxv2f64_nxv2f64_nxv2f64(<vscale x 2 x double> %0, <vscale x 2 x double> %1, <vscale x 2 x double> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vv_nxv2f64_nxv2f64_nxv2f64:
@@ -504,7 +504,7 @@ entry:
     <vscale x 2 x double> %1,
     <vscale x 2 x double> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x double> %a
 }
@@ -536,7 +536,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfnmadd.mask.nxv4f64.nxv4f64(
   <vscale x 4 x double>,
   <vscale x 4 x double>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double>  @intrinsic_vfnmadd_mask_vv_nxv4f64_nxv4f64_nxv4f64(<vscale x 4 x double> %0, <vscale x 4 x double> %1, <vscale x 4 x double> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vv_nxv4f64_nxv4f64_nxv4f64:
@@ -550,7 +550,7 @@ entry:
     <vscale x 4 x double> %1,
     <vscale x 4 x double> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x double> %a
 }
@@ -582,7 +582,7 @@ declare <vscale x 1 x half> @llvm.riscv.vfnmadd.mask.nxv1f16.f16(
   half,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x half> @intrinsic_vfnmadd_mask_vf_nxv1f16_f16_nxv1f16(<vscale x 1 x half> %0, half %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vf_nxv1f16_f16_nxv1f16:
@@ -596,7 +596,7 @@ entry:
     half %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x half> %a
 }
@@ -628,7 +628,7 @@ declare <vscale x 2 x half> @llvm.riscv.vfnmadd.mask.nxv2f16.f16(
   half,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x half> @intrinsic_vfnmadd_mask_vf_nxv2f16_f16_nxv2f16(<vscale x 2 x half> %0, half %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vf_nxv2f16_f16_nxv2f16:
@@ -642,7 +642,7 @@ entry:
     half %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x half> %a
 }
@@ -674,7 +674,7 @@ declare <vscale x 4 x half> @llvm.riscv.vfnmadd.mask.nxv4f16.f16(
   half,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x half> @intrinsic_vfnmadd_mask_vf_nxv4f16_f16_nxv4f16(<vscale x 4 x half> %0, half %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vf_nxv4f16_f16_nxv4f16:
@@ -688,7 +688,7 @@ entry:
     half %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x half> %a
 }
@@ -720,7 +720,7 @@ declare <vscale x 8 x half> @llvm.riscv.vfnmadd.mask.nxv8f16.f16(
   half,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x half> @intrinsic_vfnmadd_mask_vf_nxv8f16_f16_nxv8f16(<vscale x 8 x half> %0, half %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vf_nxv8f16_f16_nxv8f16:
@@ -734,7 +734,7 @@ entry:
     half %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x half> %a
 }
@@ -766,7 +766,7 @@ declare <vscale x 16 x half> @llvm.riscv.vfnmadd.mask.nxv16f16.f16(
   half,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x half> @intrinsic_vfnmadd_mask_vf_nxv16f16_f16_nxv16f16(<vscale x 16 x half> %0, half %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vf_nxv16f16_f16_nxv16f16:
@@ -780,7 +780,7 @@ entry:
     half %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 16 x half> %a
 }
@@ -812,7 +812,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfnmadd.mask.nxv1f32.f32(
   float,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float> @intrinsic_vfnmadd_mask_vf_nxv1f32_f32_nxv1f32(<vscale x 1 x float> %0, float %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vf_nxv1f32_f32_nxv1f32:
@@ -826,7 +826,7 @@ entry:
     float %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x float> %a
 }
@@ -858,7 +858,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfnmadd.mask.nxv2f32.f32(
   float,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float> @intrinsic_vfnmadd_mask_vf_nxv2f32_f32_nxv2f32(<vscale x 2 x float> %0, float %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vf_nxv2f32_f32_nxv2f32:
@@ -872,7 +872,7 @@ entry:
     float %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x float> %a
 }
@@ -904,7 +904,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfnmadd.mask.nxv4f32.f32(
   float,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float> @intrinsic_vfnmadd_mask_vf_nxv4f32_f32_nxv4f32(<vscale x 4 x float> %0, float %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vf_nxv4f32_f32_nxv4f32:
@@ -918,7 +918,7 @@ entry:
     float %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x float> %a
 }
@@ -950,7 +950,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfnmadd.mask.nxv8f32.f32(
   float,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float> @intrinsic_vfnmadd_mask_vf_nxv8f32_f32_nxv8f32(<vscale x 8 x float> %0, float %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vf_nxv8f32_f32_nxv8f32:
@@ -964,7 +964,7 @@ entry:
     float %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x float> %a
 }
@@ -996,7 +996,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfnmadd.mask.nxv1f64.f64(
   double,
   <vscale x 1 x double>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double> @intrinsic_vfnmadd_mask_vf_nxv1f64_f64_nxv1f64(<vscale x 1 x double> %0, double %1, <vscale x 1 x double> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vf_nxv1f64_f64_nxv1f64:
@@ -1010,7 +1010,7 @@ entry:
     double %1,
     <vscale x 1 x double> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x double> %a
 }
@@ -1042,7 +1042,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfnmadd.mask.nxv2f64.f64(
   double,
   <vscale x 2 x double>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double> @intrinsic_vfnmadd_mask_vf_nxv2f64_f64_nxv2f64(<vscale x 2 x double> %0, double %1, <vscale x 2 x double> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vf_nxv2f64_f64_nxv2f64:
@@ -1056,7 +1056,7 @@ entry:
     double %1,
     <vscale x 2 x double> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x double> %a
 }
@@ -1088,7 +1088,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfnmadd.mask.nxv4f64.f64(
   double,
   <vscale x 4 x double>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double> @intrinsic_vfnmadd_mask_vf_nxv4f64_f64_nxv4f64(<vscale x 4 x double> %0, double %1, <vscale x 4 x double> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmadd_mask_vf_nxv4f64_f64_nxv4f64:
@@ -1102,7 +1102,7 @@ entry:
     double %1,
     <vscale x 4 x double> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x double> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vfnmsac.ll b/llvm/test/CodeGen/RISCV/rvv/vfnmsac.ll
index ff1bcfa86d3a80..dc77c03041dbd0 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vfnmsac.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vfnmsac.ll
@@ -30,7 +30,7 @@ declare <vscale x 1 x half> @llvm.riscv.vfnmsac.mask.nxv1f16.nxv1f16(
   <vscale x 1 x half>,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x half>  @intrinsic_vfnmsac_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vv_nxv1f16_nxv1f16_nxv1f16:
@@ -44,7 +44,7 @@ entry:
     <vscale x 1 x half> %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x half> %a
 }
@@ -76,7 +76,7 @@ declare <vscale x 2 x half> @llvm.riscv.vfnmsac.mask.nxv2f16.nxv2f16(
   <vscale x 2 x half>,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x half>  @intrinsic_vfnmsac_mask_vv_nxv2f16_nxv2f16_nxv2f16(<vscale x 2 x half> %0, <vscale x 2 x half> %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vv_nxv2f16_nxv2f16_nxv2f16:
@@ -90,7 +90,7 @@ entry:
     <vscale x 2 x half> %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x half> %a
 }
@@ -122,7 +122,7 @@ declare <vscale x 4 x half> @llvm.riscv.vfnmsac.mask.nxv4f16.nxv4f16(
   <vscale x 4 x half>,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x half>  @intrinsic_vfnmsac_mask_vv_nxv4f16_nxv4f16_nxv4f16(<vscale x 4 x half> %0, <vscale x 4 x half> %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vv_nxv4f16_nxv4f16_nxv4f16:
@@ -136,7 +136,7 @@ entry:
     <vscale x 4 x half> %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x half> %a
 }
@@ -168,7 +168,7 @@ declare <vscale x 8 x half> @llvm.riscv.vfnmsac.mask.nxv8f16.nxv8f16(
   <vscale x 8 x half>,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x half>  @intrinsic_vfnmsac_mask_vv_nxv8f16_nxv8f16_nxv8f16(<vscale x 8 x half> %0, <vscale x 8 x half> %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vv_nxv8f16_nxv8f16_nxv8f16:
@@ -182,7 +182,7 @@ entry:
     <vscale x 8 x half> %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x half> %a
 }
@@ -214,7 +214,7 @@ declare <vscale x 16 x half> @llvm.riscv.vfnmsac.mask.nxv16f16.nxv16f16(
   <vscale x 16 x half>,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x half>  @intrinsic_vfnmsac_mask_vv_nxv16f16_nxv16f16_nxv16f16(<vscale x 16 x half> %0, <vscale x 16 x half> %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vv_nxv16f16_nxv16f16_nxv16f16:
@@ -228,7 +228,7 @@ entry:
     <vscale x 16 x half> %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 16 x half> %a
 }
@@ -260,7 +260,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfnmsac.mask.nxv1f32.nxv1f32(
   <vscale x 1 x float>,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float>  @intrinsic_vfnmsac_mask_vv_nxv1f32_nxv1f32_nxv1f32(<vscale x 1 x float> %0, <vscale x 1 x float> %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vv_nxv1f32_nxv1f32_nxv1f32:
@@ -274,7 +274,7 @@ entry:
     <vscale x 1 x float> %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x float> %a
 }
@@ -306,7 +306,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfnmsac.mask.nxv2f32.nxv2f32(
   <vscale x 2 x float>,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float>  @intrinsic_vfnmsac_mask_vv_nxv2f32_nxv2f32_nxv2f32(<vscale x 2 x float> %0, <vscale x 2 x float> %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vv_nxv2f32_nxv2f32_nxv2f32:
@@ -320,7 +320,7 @@ entry:
     <vscale x 2 x float> %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x float> %a
 }
@@ -352,7 +352,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfnmsac.mask.nxv4f32.nxv4f32(
   <vscale x 4 x float>,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float>  @intrinsic_vfnmsac_mask_vv_nxv4f32_nxv4f32_nxv4f32(<vscale x 4 x float> %0, <vscale x 4 x float> %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vv_nxv4f32_nxv4f32_nxv4f32:
@@ -366,7 +366,7 @@ entry:
     <vscale x 4 x float> %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x float> %a
 }
@@ -398,7 +398,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfnmsac.mask.nxv8f32.nxv8f32(
   <vscale x 8 x float>,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float>  @intrinsic_vfnmsac_mask_vv_nxv8f32_nxv8f32_nxv8f32(<vscale x 8 x float> %0, <vscale x 8 x float> %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vv_nxv8f32_nxv8f32_nxv8f32:
@@ -412,7 +412,7 @@ entry:
     <vscale x 8 x float> %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x float> %a
 }
@@ -444,7 +444,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfnmsac.mask.nxv1f64.nxv1f64(
   <vscale x 1 x double>,
   <vscale x 1 x double>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double>  @intrinsic_vfnmsac_mask_vv_nxv1f64_nxv1f64_nxv1f64(<vscale x 1 x double> %0, <vscale x 1 x double> %1, <vscale x 1 x double> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vv_nxv1f64_nxv1f64_nxv1f64:
@@ -458,7 +458,7 @@ entry:
     <vscale x 1 x double> %1,
     <vscale x 1 x double> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x double> %a
 }
@@ -490,7 +490,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfnmsac.mask.nxv2f64.nxv2f64(
   <vscale x 2 x double>,
   <vscale x 2 x double>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double>  @intrinsic_vfnmsac_mask_vv_nxv2f64_nxv2f64_nxv2f64(<vscale x 2 x double> %0, <vscale x 2 x double> %1, <vscale x 2 x double> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vv_nxv2f64_nxv2f64_nxv2f64:
@@ -504,7 +504,7 @@ entry:
     <vscale x 2 x double> %1,
     <vscale x 2 x double> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x double> %a
 }
@@ -536,7 +536,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfnmsac.mask.nxv4f64.nxv4f64(
   <vscale x 4 x double>,
   <vscale x 4 x double>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double>  @intrinsic_vfnmsac_mask_vv_nxv4f64_nxv4f64_nxv4f64(<vscale x 4 x double> %0, <vscale x 4 x double> %1, <vscale x 4 x double> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vv_nxv4f64_nxv4f64_nxv4f64:
@@ -550,7 +550,7 @@ entry:
     <vscale x 4 x double> %1,
     <vscale x 4 x double> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x double> %a
 }
@@ -582,7 +582,7 @@ declare <vscale x 1 x half> @llvm.riscv.vfnmsac.mask.nxv1f16.f16(
   half,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x half> @intrinsic_vfnmsac_mask_vf_nxv1f16_f16_nxv1f16(<vscale x 1 x half> %0, half %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vf_nxv1f16_f16_nxv1f16:
@@ -596,7 +596,7 @@ entry:
     half %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x half> %a
 }
@@ -628,7 +628,7 @@ declare <vscale x 2 x half> @llvm.riscv.vfnmsac.mask.nxv2f16.f16(
   half,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x half> @intrinsic_vfnmsac_mask_vf_nxv2f16_f16_nxv2f16(<vscale x 2 x half> %0, half %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vf_nxv2f16_f16_nxv2f16:
@@ -642,7 +642,7 @@ entry:
     half %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x half> %a
 }
@@ -674,7 +674,7 @@ declare <vscale x 4 x half> @llvm.riscv.vfnmsac.mask.nxv4f16.f16(
   half,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x half> @intrinsic_vfnmsac_mask_vf_nxv4f16_f16_nxv4f16(<vscale x 4 x half> %0, half %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vf_nxv4f16_f16_nxv4f16:
@@ -688,7 +688,7 @@ entry:
     half %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x half> %a
 }
@@ -720,7 +720,7 @@ declare <vscale x 8 x half> @llvm.riscv.vfnmsac.mask.nxv8f16.f16(
   half,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x half> @intrinsic_vfnmsac_mask_vf_nxv8f16_f16_nxv8f16(<vscale x 8 x half> %0, half %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vf_nxv8f16_f16_nxv8f16:
@@ -734,7 +734,7 @@ entry:
     half %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x half> %a
 }
@@ -766,7 +766,7 @@ declare <vscale x 16 x half> @llvm.riscv.vfnmsac.mask.nxv16f16.f16(
   half,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x half> @intrinsic_vfnmsac_mask_vf_nxv16f16_f16_nxv16f16(<vscale x 16 x half> %0, half %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vf_nxv16f16_f16_nxv16f16:
@@ -780,7 +780,7 @@ entry:
     half %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 16 x half> %a
 }
@@ -812,7 +812,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfnmsac.mask.nxv1f32.f32(
   float,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float> @intrinsic_vfnmsac_mask_vf_nxv1f32_f32_nxv1f32(<vscale x 1 x float> %0, float %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vf_nxv1f32_f32_nxv1f32:
@@ -826,7 +826,7 @@ entry:
     float %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x float> %a
 }
@@ -858,7 +858,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfnmsac.mask.nxv2f32.f32(
   float,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float> @intrinsic_vfnmsac_mask_vf_nxv2f32_f32_nxv2f32(<vscale x 2 x float> %0, float %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vf_nxv2f32_f32_nxv2f32:
@@ -872,7 +872,7 @@ entry:
     float %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x float> %a
 }
@@ -904,7 +904,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfnmsac.mask.nxv4f32.f32(
   float,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float> @intrinsic_vfnmsac_mask_vf_nxv4f32_f32_nxv4f32(<vscale x 4 x float> %0, float %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vf_nxv4f32_f32_nxv4f32:
@@ -918,7 +918,7 @@ entry:
     float %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x float> %a
 }
@@ -950,7 +950,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfnmsac.mask.nxv8f32.f32(
   float,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float> @intrinsic_vfnmsac_mask_vf_nxv8f32_f32_nxv8f32(<vscale x 8 x float> %0, float %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vf_nxv8f32_f32_nxv8f32:
@@ -964,7 +964,7 @@ entry:
     float %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x float> %a
 }
@@ -996,7 +996,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfnmsac.mask.nxv1f64.f64(
   double,
   <vscale x 1 x double>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double> @intrinsic_vfnmsac_mask_vf_nxv1f64_f64_nxv1f64(<vscale x 1 x double> %0, double %1, <vscale x 1 x double> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vf_nxv1f64_f64_nxv1f64:
@@ -1010,7 +1010,7 @@ entry:
     double %1,
     <vscale x 1 x double> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x double> %a
 }
@@ -1042,7 +1042,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfnmsac.mask.nxv2f64.f64(
   double,
   <vscale x 2 x double>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double> @intrinsic_vfnmsac_mask_vf_nxv2f64_f64_nxv2f64(<vscale x 2 x double> %0, double %1, <vscale x 2 x double> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vf_nxv2f64_f64_nxv2f64:
@@ -1056,7 +1056,7 @@ entry:
     double %1,
     <vscale x 2 x double> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x double> %a
 }
@@ -1088,7 +1088,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfnmsac.mask.nxv4f64.f64(
   double,
   <vscale x 4 x double>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double> @intrinsic_vfnmsac_mask_vf_nxv4f64_f64_nxv4f64(<vscale x 4 x double> %0, double %1, <vscale x 4 x double> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsac_mask_vf_nxv4f64_f64_nxv4f64:
@@ -1102,7 +1102,7 @@ entry:
     double %1,
     <vscale x 4 x double> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x double> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vfnmsub.ll b/llvm/test/CodeGen/RISCV/rvv/vfnmsub.ll
index e6ca32f34752c1..6a3a7bbf376077 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vfnmsub.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vfnmsub.ll
@@ -30,7 +30,7 @@ declare <vscale x 1 x half> @llvm.riscv.vfnmsub.mask.nxv1f16.nxv1f16(
   <vscale x 1 x half>,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x half>  @intrinsic_vfnmsub_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vv_nxv1f16_nxv1f16_nxv1f16:
@@ -44,7 +44,7 @@ entry:
     <vscale x 1 x half> %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x half> %a
 }
@@ -76,7 +76,7 @@ declare <vscale x 2 x half> @llvm.riscv.vfnmsub.mask.nxv2f16.nxv2f16(
   <vscale x 2 x half>,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x half>  @intrinsic_vfnmsub_mask_vv_nxv2f16_nxv2f16_nxv2f16(<vscale x 2 x half> %0, <vscale x 2 x half> %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vv_nxv2f16_nxv2f16_nxv2f16:
@@ -90,7 +90,7 @@ entry:
     <vscale x 2 x half> %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x half> %a
 }
@@ -122,7 +122,7 @@ declare <vscale x 4 x half> @llvm.riscv.vfnmsub.mask.nxv4f16.nxv4f16(
   <vscale x 4 x half>,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x half>  @intrinsic_vfnmsub_mask_vv_nxv4f16_nxv4f16_nxv4f16(<vscale x 4 x half> %0, <vscale x 4 x half> %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vv_nxv4f16_nxv4f16_nxv4f16:
@@ -136,7 +136,7 @@ entry:
     <vscale x 4 x half> %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x half> %a
 }
@@ -168,7 +168,7 @@ declare <vscale x 8 x half> @llvm.riscv.vfnmsub.mask.nxv8f16.nxv8f16(
   <vscale x 8 x half>,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x half>  @intrinsic_vfnmsub_mask_vv_nxv8f16_nxv8f16_nxv8f16(<vscale x 8 x half> %0, <vscale x 8 x half> %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vv_nxv8f16_nxv8f16_nxv8f16:
@@ -182,7 +182,7 @@ entry:
     <vscale x 8 x half> %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x half> %a
 }
@@ -214,7 +214,7 @@ declare <vscale x 16 x half> @llvm.riscv.vfnmsub.mask.nxv16f16.nxv16f16(
   <vscale x 16 x half>,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x half>  @intrinsic_vfnmsub_mask_vv_nxv16f16_nxv16f16_nxv16f16(<vscale x 16 x half> %0, <vscale x 16 x half> %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vv_nxv16f16_nxv16f16_nxv16f16:
@@ -228,7 +228,7 @@ entry:
     <vscale x 16 x half> %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 16 x half> %a
 }
@@ -260,7 +260,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfnmsub.mask.nxv1f32.nxv1f32(
   <vscale x 1 x float>,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float>  @intrinsic_vfnmsub_mask_vv_nxv1f32_nxv1f32_nxv1f32(<vscale x 1 x float> %0, <vscale x 1 x float> %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vv_nxv1f32_nxv1f32_nxv1f32:
@@ -274,7 +274,7 @@ entry:
     <vscale x 1 x float> %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x float> %a
 }
@@ -306,7 +306,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfnmsub.mask.nxv2f32.nxv2f32(
   <vscale x 2 x float>,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float>  @intrinsic_vfnmsub_mask_vv_nxv2f32_nxv2f32_nxv2f32(<vscale x 2 x float> %0, <vscale x 2 x float> %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vv_nxv2f32_nxv2f32_nxv2f32:
@@ -320,7 +320,7 @@ entry:
     <vscale x 2 x float> %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x float> %a
 }
@@ -352,7 +352,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfnmsub.mask.nxv4f32.nxv4f32(
   <vscale x 4 x float>,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float>  @intrinsic_vfnmsub_mask_vv_nxv4f32_nxv4f32_nxv4f32(<vscale x 4 x float> %0, <vscale x 4 x float> %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vv_nxv4f32_nxv4f32_nxv4f32:
@@ -366,7 +366,7 @@ entry:
     <vscale x 4 x float> %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x float> %a
 }
@@ -398,7 +398,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfnmsub.mask.nxv8f32.nxv8f32(
   <vscale x 8 x float>,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float>  @intrinsic_vfnmsub_mask_vv_nxv8f32_nxv8f32_nxv8f32(<vscale x 8 x float> %0, <vscale x 8 x float> %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vv_nxv8f32_nxv8f32_nxv8f32:
@@ -412,7 +412,7 @@ entry:
     <vscale x 8 x float> %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x float> %a
 }
@@ -444,7 +444,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfnmsub.mask.nxv1f64.nxv1f64(
   <vscale x 1 x double>,
   <vscale x 1 x double>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double>  @intrinsic_vfnmsub_mask_vv_nxv1f64_nxv1f64_nxv1f64(<vscale x 1 x double> %0, <vscale x 1 x double> %1, <vscale x 1 x double> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vv_nxv1f64_nxv1f64_nxv1f64:
@@ -458,7 +458,7 @@ entry:
     <vscale x 1 x double> %1,
     <vscale x 1 x double> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x double> %a
 }
@@ -490,7 +490,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfnmsub.mask.nxv2f64.nxv2f64(
   <vscale x 2 x double>,
   <vscale x 2 x double>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double>  @intrinsic_vfnmsub_mask_vv_nxv2f64_nxv2f64_nxv2f64(<vscale x 2 x double> %0, <vscale x 2 x double> %1, <vscale x 2 x double> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vv_nxv2f64_nxv2f64_nxv2f64:
@@ -504,7 +504,7 @@ entry:
     <vscale x 2 x double> %1,
     <vscale x 2 x double> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x double> %a
 }
@@ -536,7 +536,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfnmsub.mask.nxv4f64.nxv4f64(
   <vscale x 4 x double>,
   <vscale x 4 x double>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double>  @intrinsic_vfnmsub_mask_vv_nxv4f64_nxv4f64_nxv4f64(<vscale x 4 x double> %0, <vscale x 4 x double> %1, <vscale x 4 x double> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vv_nxv4f64_nxv4f64_nxv4f64:
@@ -550,7 +550,7 @@ entry:
     <vscale x 4 x double> %1,
     <vscale x 4 x double> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x double> %a
 }
@@ -582,7 +582,7 @@ declare <vscale x 1 x half> @llvm.riscv.vfnmsub.mask.nxv1f16.f16(
   half,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x half> @intrinsic_vfnmsub_mask_vf_nxv1f16_f16_nxv1f16(<vscale x 1 x half> %0, half %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vf_nxv1f16_f16_nxv1f16:
@@ -596,7 +596,7 @@ entry:
     half %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x half> %a
 }
@@ -628,7 +628,7 @@ declare <vscale x 2 x half> @llvm.riscv.vfnmsub.mask.nxv2f16.f16(
   half,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x half> @intrinsic_vfnmsub_mask_vf_nxv2f16_f16_nxv2f16(<vscale x 2 x half> %0, half %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vf_nxv2f16_f16_nxv2f16:
@@ -642,7 +642,7 @@ entry:
     half %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x half> %a
 }
@@ -674,7 +674,7 @@ declare <vscale x 4 x half> @llvm.riscv.vfnmsub.mask.nxv4f16.f16(
   half,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x half> @intrinsic_vfnmsub_mask_vf_nxv4f16_f16_nxv4f16(<vscale x 4 x half> %0, half %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vf_nxv4f16_f16_nxv4f16:
@@ -688,7 +688,7 @@ entry:
     half %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x half> %a
 }
@@ -720,7 +720,7 @@ declare <vscale x 8 x half> @llvm.riscv.vfnmsub.mask.nxv8f16.f16(
   half,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x half> @intrinsic_vfnmsub_mask_vf_nxv8f16_f16_nxv8f16(<vscale x 8 x half> %0, half %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vf_nxv8f16_f16_nxv8f16:
@@ -734,7 +734,7 @@ entry:
     half %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x half> %a
 }
@@ -766,7 +766,7 @@ declare <vscale x 16 x half> @llvm.riscv.vfnmsub.mask.nxv16f16.f16(
   half,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x half> @intrinsic_vfnmsub_mask_vf_nxv16f16_f16_nxv16f16(<vscale x 16 x half> %0, half %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vf_nxv16f16_f16_nxv16f16:
@@ -780,7 +780,7 @@ entry:
     half %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 16 x half> %a
 }
@@ -812,7 +812,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfnmsub.mask.nxv1f32.f32(
   float,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float> @intrinsic_vfnmsub_mask_vf_nxv1f32_f32_nxv1f32(<vscale x 1 x float> %0, float %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vf_nxv1f32_f32_nxv1f32:
@@ -826,7 +826,7 @@ entry:
     float %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x float> %a
 }
@@ -858,7 +858,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfnmsub.mask.nxv2f32.f32(
   float,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float> @intrinsic_vfnmsub_mask_vf_nxv2f32_f32_nxv2f32(<vscale x 2 x float> %0, float %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vf_nxv2f32_f32_nxv2f32:
@@ -872,7 +872,7 @@ entry:
     float %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x float> %a
 }
@@ -904,7 +904,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfnmsub.mask.nxv4f32.f32(
   float,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float> @intrinsic_vfnmsub_mask_vf_nxv4f32_f32_nxv4f32(<vscale x 4 x float> %0, float %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vf_nxv4f32_f32_nxv4f32:
@@ -918,7 +918,7 @@ entry:
     float %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x float> %a
 }
@@ -950,7 +950,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfnmsub.mask.nxv8f32.f32(
   float,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float> @intrinsic_vfnmsub_mask_vf_nxv8f32_f32_nxv8f32(<vscale x 8 x float> %0, float %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vf_nxv8f32_f32_nxv8f32:
@@ -964,7 +964,7 @@ entry:
     float %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x float> %a
 }
@@ -996,7 +996,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfnmsub.mask.nxv1f64.f64(
   double,
   <vscale x 1 x double>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double> @intrinsic_vfnmsub_mask_vf_nxv1f64_f64_nxv1f64(<vscale x 1 x double> %0, double %1, <vscale x 1 x double> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vf_nxv1f64_f64_nxv1f64:
@@ -1010,7 +1010,7 @@ entry:
     double %1,
     <vscale x 1 x double> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x double> %a
 }
@@ -1042,7 +1042,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfnmsub.mask.nxv2f64.f64(
   double,
   <vscale x 2 x double>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double> @intrinsic_vfnmsub_mask_vf_nxv2f64_f64_nxv2f64(<vscale x 2 x double> %0, double %1, <vscale x 2 x double> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vf_nxv2f64_f64_nxv2f64:
@@ -1056,7 +1056,7 @@ entry:
     double %1,
     <vscale x 2 x double> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x double> %a
 }
@@ -1088,7 +1088,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfnmsub.mask.nxv4f64.f64(
   double,
   <vscale x 4 x double>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double> @intrinsic_vfnmsub_mask_vf_nxv4f64_f64_nxv4f64(<vscale x 4 x double> %0, double %1, <vscale x 4 x double> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfnmsub_mask_vf_nxv4f64_f64_nxv4f64:
@@ -1102,7 +1102,7 @@ entry:
     double %1,
     <vscale x 4 x double> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x double> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vfwmacc.ll b/llvm/test/CodeGen/RISCV/rvv/vfwmacc.ll
index f5db61b5e8c7c8..6e9eec795aa353 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vfwmacc.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vfwmacc.ll
@@ -30,7 +30,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfwmacc.mask.nxv1f32.nxv1f16(
   <vscale x 1 x half>,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float>  @intrinsic_vfwmacc_mask_vv_nxv1f32_nxv1f16_nxv1f16(<vscale x 1 x float> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmacc_mask_vv_nxv1f32_nxv1f16_nxv1f16:
@@ -44,7 +44,7 @@ entry:
     <vscale x 1 x half> %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x float> %a
 }
@@ -76,7 +76,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfwmacc.mask.nxv2f32.nxv2f16(
   <vscale x 2 x half>,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float>  @intrinsic_vfwmacc_mask_vv_nxv2f32_nxv2f16_nxv2f16(<vscale x 2 x float> %0, <vscale x 2 x half> %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmacc_mask_vv_nxv2f32_nxv2f16_nxv2f16:
@@ -90,7 +90,7 @@ entry:
     <vscale x 2 x half> %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x float> %a
 }
@@ -122,7 +122,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfwmacc.mask.nxv4f32.nxv4f16(
   <vscale x 4 x half>,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float>  @intrinsic_vfwmacc_mask_vv_nxv4f32_nxv4f16_nxv4f16(<vscale x 4 x float> %0, <vscale x 4 x half> %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmacc_mask_vv_nxv4f32_nxv4f16_nxv4f16:
@@ -136,7 +136,7 @@ entry:
     <vscale x 4 x half> %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x float> %a
 }
@@ -168,7 +168,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfwmacc.mask.nxv8f32.nxv8f16(
   <vscale x 8 x half>,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float>  @intrinsic_vfwmacc_mask_vv_nxv8f32_nxv8f16_nxv8f16(<vscale x 8 x float> %0, <vscale x 8 x half> %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmacc_mask_vv_nxv8f32_nxv8f16_nxv8f16:
@@ -182,7 +182,7 @@ entry:
     <vscale x 8 x half> %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x float> %a
 }
@@ -214,7 +214,7 @@ declare <vscale x 16 x float> @llvm.riscv.vfwmacc.mask.nxv16f32.nxv16f16(
   <vscale x 16 x half>,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x float>  @intrinsic_vfwmacc_mask_vv_nxv16f32_nxv16f16_nxv16f16(<vscale x 16 x float> %0, <vscale x 16 x half> %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmacc_mask_vv_nxv16f32_nxv16f16_nxv16f16:
@@ -228,7 +228,7 @@ entry:
     <vscale x 16 x half> %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 16 x float> %a
 }
@@ -260,7 +260,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfwmacc.mask.nxv1f64.nxv1f32(
   <vscale x 1 x float>,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double>  @intrinsic_vfwmacc_mask_vv_nxv1f64_nxv1f32_nxv1f32(<vscale x 1 x double> %0, <vscale x 1 x float> %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmacc_mask_vv_nxv1f64_nxv1f32_nxv1f32:
@@ -274,7 +274,7 @@ entry:
     <vscale x 1 x float> %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x double> %a
 }
@@ -306,7 +306,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfwmacc.mask.nxv2f64.nxv2f32(
   <vscale x 2 x float>,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double>  @intrinsic_vfwmacc_mask_vv_nxv2f64_nxv2f32_nxv2f32(<vscale x 2 x double> %0, <vscale x 2 x float> %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmacc_mask_vv_nxv2f64_nxv2f32_nxv2f32:
@@ -320,7 +320,7 @@ entry:
     <vscale x 2 x float> %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x double> %a
 }
@@ -352,7 +352,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfwmacc.mask.nxv4f64.nxv4f32(
   <vscale x 4 x float>,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double>  @intrinsic_vfwmacc_mask_vv_nxv4f64_nxv4f32_nxv4f32(<vscale x 4 x double> %0, <vscale x 4 x float> %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmacc_mask_vv_nxv4f64_nxv4f32_nxv4f32:
@@ -366,7 +366,7 @@ entry:
     <vscale x 4 x float> %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x double> %a
 }
@@ -398,7 +398,7 @@ declare <vscale x 8 x double> @llvm.riscv.vfwmacc.mask.nxv8f64.nxv8f32(
   <vscale x 8 x float>,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x double>  @intrinsic_vfwmacc_mask_vv_nxv8f64_nxv8f32_nxv8f32(<vscale x 8 x double> %0, <vscale x 8 x float> %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmacc_mask_vv_nxv8f64_nxv8f32_nxv8f32:
@@ -412,7 +412,7 @@ entry:
     <vscale x 8 x float> %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x double> %a
 }
@@ -444,7 +444,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfwmacc.mask.nxv1f32.f16(
   half,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float> @intrinsic_vfwmacc_mask_vf_nxv1f32_f16_nxv1f16(<vscale x 1 x float> %0, half %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmacc_mask_vf_nxv1f32_f16_nxv1f16:
@@ -458,7 +458,7 @@ entry:
     half %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x float> %a
 }
@@ -490,7 +490,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfwmacc.mask.nxv2f32.f16(
   half,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float> @intrinsic_vfwmacc_mask_vf_nxv2f32_f16_nxv2f16(<vscale x 2 x float> %0, half %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmacc_mask_vf_nxv2f32_f16_nxv2f16:
@@ -504,7 +504,7 @@ entry:
     half %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x float> %a
 }
@@ -536,7 +536,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfwmacc.mask.nxv4f32.f16(
   half,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float> @intrinsic_vfwmacc_mask_vf_nxv4f32_f16_nxv4f16(<vscale x 4 x float> %0, half %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmacc_mask_vf_nxv4f32_f16_nxv4f16:
@@ -550,7 +550,7 @@ entry:
     half %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x float> %a
 }
@@ -582,7 +582,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfwmacc.mask.nxv8f32.f16(
   half,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float> @intrinsic_vfwmacc_mask_vf_nxv8f32_f16_nxv8f16(<vscale x 8 x float> %0, half %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmacc_mask_vf_nxv8f32_f16_nxv8f16:
@@ -596,7 +596,7 @@ entry:
     half %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x float> %a
 }
@@ -628,7 +628,7 @@ declare <vscale x 16 x float> @llvm.riscv.vfwmacc.mask.nxv16f32.f16(
   half,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x float> @intrinsic_vfwmacc_mask_vf_nxv16f32_f16_nxv16f16(<vscale x 16 x float> %0, half %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmacc_mask_vf_nxv16f32_f16_nxv16f16:
@@ -642,7 +642,7 @@ entry:
     half %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 16 x float> %a
 }
@@ -674,7 +674,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfwmacc.mask.nxv1f64.f32(
   float,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double> @intrinsic_vfwmacc_mask_vf_nxv1f64_f32_nxv1f32(<vscale x 1 x double> %0, float %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmacc_mask_vf_nxv1f64_f32_nxv1f32:
@@ -688,7 +688,7 @@ entry:
     float %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x double> %a
 }
@@ -720,7 +720,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfwmacc.mask.nxv2f64.f32(
   float,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double> @intrinsic_vfwmacc_mask_vf_nxv2f64_f32_nxv2f32(<vscale x 2 x double> %0, float %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmacc_mask_vf_nxv2f64_f32_nxv2f32:
@@ -734,7 +734,7 @@ entry:
     float %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x double> %a
 }
@@ -766,7 +766,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfwmacc.mask.nxv4f64.f32(
   float,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double> @intrinsic_vfwmacc_mask_vf_nxv4f64_f32_nxv4f32(<vscale x 4 x double> %0, float %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmacc_mask_vf_nxv4f64_f32_nxv4f32:
@@ -780,7 +780,7 @@ entry:
     float %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x double> %a
 }
@@ -812,7 +812,7 @@ declare <vscale x 8 x double> @llvm.riscv.vfwmacc.mask.nxv8f64.f32(
   float,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x double> @intrinsic_vfwmacc_mask_vf_nxv8f64_f32_nxv8f32(<vscale x 8 x double> %0, float %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmacc_mask_vf_nxv8f64_f32_nxv8f32:
@@ -826,7 +826,7 @@ entry:
     float %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x double> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vfwmsac.ll b/llvm/test/CodeGen/RISCV/rvv/vfwmsac.ll
index 884ee36575b4af..1e9d25d0884d3f 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vfwmsac.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vfwmsac.ll
@@ -30,7 +30,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfwmsac.mask.nxv1f32.nxv1f16(
   <vscale x 1 x half>,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float>  @intrinsic_vfwmsac_mask_vv_nxv1f32_nxv1f16_nxv1f16(<vscale x 1 x float> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmsac_mask_vv_nxv1f32_nxv1f16_nxv1f16:
@@ -44,7 +44,7 @@ entry:
     <vscale x 1 x half> %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x float> %a
 }
@@ -76,7 +76,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfwmsac.mask.nxv2f32.nxv2f16(
   <vscale x 2 x half>,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float>  @intrinsic_vfwmsac_mask_vv_nxv2f32_nxv2f16_nxv2f16(<vscale x 2 x float> %0, <vscale x 2 x half> %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmsac_mask_vv_nxv2f32_nxv2f16_nxv2f16:
@@ -90,7 +90,7 @@ entry:
     <vscale x 2 x half> %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x float> %a
 }
@@ -122,7 +122,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfwmsac.mask.nxv4f32.nxv4f16(
   <vscale x 4 x half>,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float>  @intrinsic_vfwmsac_mask_vv_nxv4f32_nxv4f16_nxv4f16(<vscale x 4 x float> %0, <vscale x 4 x half> %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmsac_mask_vv_nxv4f32_nxv4f16_nxv4f16:
@@ -136,7 +136,7 @@ entry:
     <vscale x 4 x half> %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x float> %a
 }
@@ -168,7 +168,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfwmsac.mask.nxv8f32.nxv8f16(
   <vscale x 8 x half>,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float>  @intrinsic_vfwmsac_mask_vv_nxv8f32_nxv8f16_nxv8f16(<vscale x 8 x float> %0, <vscale x 8 x half> %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmsac_mask_vv_nxv8f32_nxv8f16_nxv8f16:
@@ -182,7 +182,7 @@ entry:
     <vscale x 8 x half> %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x float> %a
 }
@@ -214,7 +214,7 @@ declare <vscale x 16 x float> @llvm.riscv.vfwmsac.mask.nxv16f32.nxv16f16(
   <vscale x 16 x half>,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x float>  @intrinsic_vfwmsac_mask_vv_nxv16f32_nxv16f16_nxv16f16(<vscale x 16 x float> %0, <vscale x 16 x half> %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmsac_mask_vv_nxv16f32_nxv16f16_nxv16f16:
@@ -228,7 +228,7 @@ entry:
     <vscale x 16 x half> %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 16 x float> %a
 }
@@ -260,7 +260,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfwmsac.mask.nxv1f64.nxv1f32(
   <vscale x 1 x float>,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double>  @intrinsic_vfwmsac_mask_vv_nxv1f64_nxv1f32_nxv1f32(<vscale x 1 x double> %0, <vscale x 1 x float> %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmsac_mask_vv_nxv1f64_nxv1f32_nxv1f32:
@@ -274,7 +274,7 @@ entry:
     <vscale x 1 x float> %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x double> %a
 }
@@ -306,7 +306,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfwmsac.mask.nxv2f64.nxv2f32(
   <vscale x 2 x float>,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double>  @intrinsic_vfwmsac_mask_vv_nxv2f64_nxv2f32_nxv2f32(<vscale x 2 x double> %0, <vscale x 2 x float> %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmsac_mask_vv_nxv2f64_nxv2f32_nxv2f32:
@@ -320,7 +320,7 @@ entry:
     <vscale x 2 x float> %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x double> %a
 }
@@ -352,7 +352,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfwmsac.mask.nxv4f64.nxv4f32(
   <vscale x 4 x float>,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double>  @intrinsic_vfwmsac_mask_vv_nxv4f64_nxv4f32_nxv4f32(<vscale x 4 x double> %0, <vscale x 4 x float> %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmsac_mask_vv_nxv4f64_nxv4f32_nxv4f32:
@@ -366,7 +366,7 @@ entry:
     <vscale x 4 x float> %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x double> %a
 }
@@ -398,7 +398,7 @@ declare <vscale x 8 x double> @llvm.riscv.vfwmsac.mask.nxv8f64.nxv8f32(
   <vscale x 8 x float>,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x double>  @intrinsic_vfwmsac_mask_vv_nxv8f64_nxv8f32_nxv8f32(<vscale x 8 x double> %0, <vscale x 8 x float> %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmsac_mask_vv_nxv8f64_nxv8f32_nxv8f32:
@@ -412,7 +412,7 @@ entry:
     <vscale x 8 x float> %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x double> %a
 }
@@ -444,7 +444,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfwmsac.mask.nxv1f32.f16(
   half,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float> @intrinsic_vfwmsac_mask_vf_nxv1f32_f16_nxv1f16(<vscale x 1 x float> %0, half %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmsac_mask_vf_nxv1f32_f16_nxv1f16:
@@ -458,7 +458,7 @@ entry:
     half %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x float> %a
 }
@@ -490,7 +490,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfwmsac.mask.nxv2f32.f16(
   half,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float> @intrinsic_vfwmsac_mask_vf_nxv2f32_f16_nxv2f16(<vscale x 2 x float> %0, half %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmsac_mask_vf_nxv2f32_f16_nxv2f16:
@@ -504,7 +504,7 @@ entry:
     half %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x float> %a
 }
@@ -536,7 +536,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfwmsac.mask.nxv4f32.f16(
   half,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float> @intrinsic_vfwmsac_mask_vf_nxv4f32_f16_nxv4f16(<vscale x 4 x float> %0, half %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmsac_mask_vf_nxv4f32_f16_nxv4f16:
@@ -550,7 +550,7 @@ entry:
     half %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x float> %a
 }
@@ -582,7 +582,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfwmsac.mask.nxv8f32.f16(
   half,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float> @intrinsic_vfwmsac_mask_vf_nxv8f32_f16_nxv8f16(<vscale x 8 x float> %0, half %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmsac_mask_vf_nxv8f32_f16_nxv8f16:
@@ -596,7 +596,7 @@ entry:
     half %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x float> %a
 }
@@ -628,7 +628,7 @@ declare <vscale x 16 x float> @llvm.riscv.vfwmsac.mask.nxv16f32.f16(
   half,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x float> @intrinsic_vfwmsac_mask_vf_nxv16f32_f16_nxv16f16(<vscale x 16 x float> %0, half %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmsac_mask_vf_nxv16f32_f16_nxv16f16:
@@ -642,7 +642,7 @@ entry:
     half %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 16 x float> %a
 }
@@ -674,7 +674,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfwmsac.mask.nxv1f64.f32(
   float,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double> @intrinsic_vfwmsac_mask_vf_nxv1f64_f32_nxv1f32(<vscale x 1 x double> %0, float %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmsac_mask_vf_nxv1f64_f32_nxv1f32:
@@ -688,7 +688,7 @@ entry:
     float %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x double> %a
 }
@@ -720,7 +720,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfwmsac.mask.nxv2f64.f32(
   float,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double> @intrinsic_vfwmsac_mask_vf_nxv2f64_f32_nxv2f32(<vscale x 2 x double> %0, float %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmsac_mask_vf_nxv2f64_f32_nxv2f32:
@@ -734,7 +734,7 @@ entry:
     float %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x double> %a
 }
@@ -766,7 +766,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfwmsac.mask.nxv4f64.f32(
   float,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double> @intrinsic_vfwmsac_mask_vf_nxv4f64_f32_nxv4f32(<vscale x 4 x double> %0, float %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmsac_mask_vf_nxv4f64_f32_nxv4f32:
@@ -780,7 +780,7 @@ entry:
     float %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x double> %a
 }
@@ -812,7 +812,7 @@ declare <vscale x 8 x double> @llvm.riscv.vfwmsac.mask.nxv8f64.f32(
   float,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x double> @intrinsic_vfwmsac_mask_vf_nxv8f64_f32_nxv8f32(<vscale x 8 x double> %0, float %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwmsac_mask_vf_nxv8f64_f32_nxv8f32:
@@ -826,7 +826,7 @@ entry:
     float %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x double> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vfwnmacc.ll b/llvm/test/CodeGen/RISCV/rvv/vfwnmacc.ll
index 4ccd0f8c558355..2542be6d4a8130 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vfwnmacc.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vfwnmacc.ll
@@ -30,7 +30,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfwnmacc.mask.nxv1f32.nxv1f16(
   <vscale x 1 x half>,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float>  @intrinsic_vfwnmacc_mask_vv_nxv1f32_nxv1f16_nxv1f16(<vscale x 1 x float> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmacc_mask_vv_nxv1f32_nxv1f16_nxv1f16:
@@ -44,7 +44,7 @@ entry:
     <vscale x 1 x half> %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x float> %a
 }
@@ -76,7 +76,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfwnmacc.mask.nxv2f32.nxv2f16(
   <vscale x 2 x half>,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float>  @intrinsic_vfwnmacc_mask_vv_nxv2f32_nxv2f16_nxv2f16(<vscale x 2 x float> %0, <vscale x 2 x half> %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmacc_mask_vv_nxv2f32_nxv2f16_nxv2f16:
@@ -90,7 +90,7 @@ entry:
     <vscale x 2 x half> %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x float> %a
 }
@@ -122,7 +122,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfwnmacc.mask.nxv4f32.nxv4f16(
   <vscale x 4 x half>,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float>  @intrinsic_vfwnmacc_mask_vv_nxv4f32_nxv4f16_nxv4f16(<vscale x 4 x float> %0, <vscale x 4 x half> %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmacc_mask_vv_nxv4f32_nxv4f16_nxv4f16:
@@ -136,7 +136,7 @@ entry:
     <vscale x 4 x half> %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x float> %a
 }
@@ -168,7 +168,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfwnmacc.mask.nxv8f32.nxv8f16(
   <vscale x 8 x half>,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float>  @intrinsic_vfwnmacc_mask_vv_nxv8f32_nxv8f16_nxv8f16(<vscale x 8 x float> %0, <vscale x 8 x half> %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmacc_mask_vv_nxv8f32_nxv8f16_nxv8f16:
@@ -182,7 +182,7 @@ entry:
     <vscale x 8 x half> %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x float> %a
 }
@@ -214,7 +214,7 @@ declare <vscale x 16 x float> @llvm.riscv.vfwnmacc.mask.nxv16f32.nxv16f16(
   <vscale x 16 x half>,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x float>  @intrinsic_vfwnmacc_mask_vv_nxv16f32_nxv16f16_nxv16f16(<vscale x 16 x float> %0, <vscale x 16 x half> %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmacc_mask_vv_nxv16f32_nxv16f16_nxv16f16:
@@ -228,7 +228,7 @@ entry:
     <vscale x 16 x half> %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 16 x float> %a
 }
@@ -260,7 +260,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfwnmacc.mask.nxv1f64.nxv1f32(
   <vscale x 1 x float>,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double>  @intrinsic_vfwnmacc_mask_vv_nxv1f64_nxv1f32_nxv1f32(<vscale x 1 x double> %0, <vscale x 1 x float> %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmacc_mask_vv_nxv1f64_nxv1f32_nxv1f32:
@@ -274,7 +274,7 @@ entry:
     <vscale x 1 x float> %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x double> %a
 }
@@ -306,7 +306,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfwnmacc.mask.nxv2f64.nxv2f32(
   <vscale x 2 x float>,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double>  @intrinsic_vfwnmacc_mask_vv_nxv2f64_nxv2f32_nxv2f32(<vscale x 2 x double> %0, <vscale x 2 x float> %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmacc_mask_vv_nxv2f64_nxv2f32_nxv2f32:
@@ -320,7 +320,7 @@ entry:
     <vscale x 2 x float> %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x double> %a
 }
@@ -352,7 +352,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfwnmacc.mask.nxv4f64.nxv4f32(
   <vscale x 4 x float>,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double>  @intrinsic_vfwnmacc_mask_vv_nxv4f64_nxv4f32_nxv4f32(<vscale x 4 x double> %0, <vscale x 4 x float> %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmacc_mask_vv_nxv4f64_nxv4f32_nxv4f32:
@@ -366,7 +366,7 @@ entry:
     <vscale x 4 x float> %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x double> %a
 }
@@ -398,7 +398,7 @@ declare <vscale x 8 x double> @llvm.riscv.vfwnmacc.mask.nxv8f64.nxv8f32(
   <vscale x 8 x float>,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x double>  @intrinsic_vfwnmacc_mask_vv_nxv8f64_nxv8f32_nxv8f32(<vscale x 8 x double> %0, <vscale x 8 x float> %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmacc_mask_vv_nxv8f64_nxv8f32_nxv8f32:
@@ -412,7 +412,7 @@ entry:
     <vscale x 8 x float> %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x double> %a
 }
@@ -444,7 +444,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfwnmacc.mask.nxv1f32.f16(
   half,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float> @intrinsic_vfwnmacc_mask_vf_nxv1f32_f16_nxv1f16(<vscale x 1 x float> %0, half %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmacc_mask_vf_nxv1f32_f16_nxv1f16:
@@ -458,7 +458,7 @@ entry:
     half %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x float> %a
 }
@@ -490,7 +490,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfwnmacc.mask.nxv2f32.f16(
   half,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float> @intrinsic_vfwnmacc_mask_vf_nxv2f32_f16_nxv2f16(<vscale x 2 x float> %0, half %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmacc_mask_vf_nxv2f32_f16_nxv2f16:
@@ -504,7 +504,7 @@ entry:
     half %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x float> %a
 }
@@ -536,7 +536,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfwnmacc.mask.nxv4f32.f16(
   half,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float> @intrinsic_vfwnmacc_mask_vf_nxv4f32_f16_nxv4f16(<vscale x 4 x float> %0, half %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmacc_mask_vf_nxv4f32_f16_nxv4f16:
@@ -550,7 +550,7 @@ entry:
     half %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x float> %a
 }
@@ -582,7 +582,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfwnmacc.mask.nxv8f32.f16(
   half,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float> @intrinsic_vfwnmacc_mask_vf_nxv8f32_f16_nxv8f16(<vscale x 8 x float> %0, half %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmacc_mask_vf_nxv8f32_f16_nxv8f16:
@@ -596,7 +596,7 @@ entry:
     half %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x float> %a
 }
@@ -628,7 +628,7 @@ declare <vscale x 16 x float> @llvm.riscv.vfwnmacc.mask.nxv16f32.f16(
   half,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x float> @intrinsic_vfwnmacc_mask_vf_nxv16f32_f16_nxv16f16(<vscale x 16 x float> %0, half %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmacc_mask_vf_nxv16f32_f16_nxv16f16:
@@ -642,7 +642,7 @@ entry:
     half %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 16 x float> %a
 }
@@ -674,7 +674,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfwnmacc.mask.nxv1f64.f32(
   float,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double> @intrinsic_vfwnmacc_mask_vf_nxv1f64_f32_nxv1f32(<vscale x 1 x double> %0, float %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmacc_mask_vf_nxv1f64_f32_nxv1f32:
@@ -688,7 +688,7 @@ entry:
     float %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x double> %a
 }
@@ -720,7 +720,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfwnmacc.mask.nxv2f64.f32(
   float,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double> @intrinsic_vfwnmacc_mask_vf_nxv2f64_f32_nxv2f32(<vscale x 2 x double> %0, float %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmacc_mask_vf_nxv2f64_f32_nxv2f32:
@@ -734,7 +734,7 @@ entry:
     float %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x double> %a
 }
@@ -766,7 +766,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfwnmacc.mask.nxv4f64.f32(
   float,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double> @intrinsic_vfwnmacc_mask_vf_nxv4f64_f32_nxv4f32(<vscale x 4 x double> %0, float %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmacc_mask_vf_nxv4f64_f32_nxv4f32:
@@ -780,7 +780,7 @@ entry:
     float %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x double> %a
 }
@@ -812,7 +812,7 @@ declare <vscale x 8 x double> @llvm.riscv.vfwnmacc.mask.nxv8f64.f32(
   float,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x double> @intrinsic_vfwnmacc_mask_vf_nxv8f64_f32_nxv8f32(<vscale x 8 x double> %0, float %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmacc_mask_vf_nxv8f64_f32_nxv8f32:
@@ -826,7 +826,7 @@ entry:
     float %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x double> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vfwnmsac.ll b/llvm/test/CodeGen/RISCV/rvv/vfwnmsac.ll
index 26fcb06d891672..ad05c7a68496e8 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vfwnmsac.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vfwnmsac.ll
@@ -30,7 +30,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfwnmsac.mask.nxv1f32.nxv1f16(
   <vscale x 1 x half>,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float>  @intrinsic_vfwnmsac_mask_vv_nxv1f32_nxv1f16_nxv1f16(<vscale x 1 x float> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmsac_mask_vv_nxv1f32_nxv1f16_nxv1f16:
@@ -44,7 +44,7 @@ entry:
     <vscale x 1 x half> %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x float> %a
 }
@@ -76,7 +76,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfwnmsac.mask.nxv2f32.nxv2f16(
   <vscale x 2 x half>,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float>  @intrinsic_vfwnmsac_mask_vv_nxv2f32_nxv2f16_nxv2f16(<vscale x 2 x float> %0, <vscale x 2 x half> %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmsac_mask_vv_nxv2f32_nxv2f16_nxv2f16:
@@ -90,7 +90,7 @@ entry:
     <vscale x 2 x half> %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x float> %a
 }
@@ -122,7 +122,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfwnmsac.mask.nxv4f32.nxv4f16(
   <vscale x 4 x half>,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float>  @intrinsic_vfwnmsac_mask_vv_nxv4f32_nxv4f16_nxv4f16(<vscale x 4 x float> %0, <vscale x 4 x half> %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmsac_mask_vv_nxv4f32_nxv4f16_nxv4f16:
@@ -136,7 +136,7 @@ entry:
     <vscale x 4 x half> %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x float> %a
 }
@@ -168,7 +168,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfwnmsac.mask.nxv8f32.nxv8f16(
   <vscale x 8 x half>,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float>  @intrinsic_vfwnmsac_mask_vv_nxv8f32_nxv8f16_nxv8f16(<vscale x 8 x float> %0, <vscale x 8 x half> %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmsac_mask_vv_nxv8f32_nxv8f16_nxv8f16:
@@ -182,7 +182,7 @@ entry:
     <vscale x 8 x half> %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x float> %a
 }
@@ -214,7 +214,7 @@ declare <vscale x 16 x float> @llvm.riscv.vfwnmsac.mask.nxv16f32.nxv16f16(
   <vscale x 16 x half>,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x float>  @intrinsic_vfwnmsac_mask_vv_nxv16f32_nxv16f16_nxv16f16(<vscale x 16 x float> %0, <vscale x 16 x half> %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmsac_mask_vv_nxv16f32_nxv16f16_nxv16f16:
@@ -228,7 +228,7 @@ entry:
     <vscale x 16 x half> %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 16 x float> %a
 }
@@ -260,7 +260,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfwnmsac.mask.nxv1f64.nxv1f32(
   <vscale x 1 x float>,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double>  @intrinsic_vfwnmsac_mask_vv_nxv1f64_nxv1f32_nxv1f32(<vscale x 1 x double> %0, <vscale x 1 x float> %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmsac_mask_vv_nxv1f64_nxv1f32_nxv1f32:
@@ -274,7 +274,7 @@ entry:
     <vscale x 1 x float> %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x double> %a
 }
@@ -306,7 +306,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfwnmsac.mask.nxv2f64.nxv2f32(
   <vscale x 2 x float>,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double>  @intrinsic_vfwnmsac_mask_vv_nxv2f64_nxv2f32_nxv2f32(<vscale x 2 x double> %0, <vscale x 2 x float> %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmsac_mask_vv_nxv2f64_nxv2f32_nxv2f32:
@@ -320,7 +320,7 @@ entry:
     <vscale x 2 x float> %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x double> %a
 }
@@ -352,7 +352,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfwnmsac.mask.nxv4f64.nxv4f32(
   <vscale x 4 x float>,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double>  @intrinsic_vfwnmsac_mask_vv_nxv4f64_nxv4f32_nxv4f32(<vscale x 4 x double> %0, <vscale x 4 x float> %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmsac_mask_vv_nxv4f64_nxv4f32_nxv4f32:
@@ -366,7 +366,7 @@ entry:
     <vscale x 4 x float> %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x double> %a
 }
@@ -398,7 +398,7 @@ declare <vscale x 8 x double> @llvm.riscv.vfwnmsac.mask.nxv8f64.nxv8f32(
   <vscale x 8 x float>,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x double>  @intrinsic_vfwnmsac_mask_vv_nxv8f64_nxv8f32_nxv8f32(<vscale x 8 x double> %0, <vscale x 8 x float> %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmsac_mask_vv_nxv8f64_nxv8f32_nxv8f32:
@@ -412,7 +412,7 @@ entry:
     <vscale x 8 x float> %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x double> %a
 }
@@ -444,7 +444,7 @@ declare <vscale x 1 x float> @llvm.riscv.vfwnmsac.mask.nxv1f32.f16(
   half,
   <vscale x 1 x half>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x float> @intrinsic_vfwnmsac_mask_vf_nxv1f32_f16_nxv1f16(<vscale x 1 x float> %0, half %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmsac_mask_vf_nxv1f32_f16_nxv1f16:
@@ -458,7 +458,7 @@ entry:
     half %1,
     <vscale x 1 x half> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x float> %a
 }
@@ -490,7 +490,7 @@ declare <vscale x 2 x float> @llvm.riscv.vfwnmsac.mask.nxv2f32.f16(
   half,
   <vscale x 2 x half>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x float> @intrinsic_vfwnmsac_mask_vf_nxv2f32_f16_nxv2f16(<vscale x 2 x float> %0, half %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmsac_mask_vf_nxv2f32_f16_nxv2f16:
@@ -504,7 +504,7 @@ entry:
     half %1,
     <vscale x 2 x half> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x float> %a
 }
@@ -536,7 +536,7 @@ declare <vscale x 4 x float> @llvm.riscv.vfwnmsac.mask.nxv4f32.f16(
   half,
   <vscale x 4 x half>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x float> @intrinsic_vfwnmsac_mask_vf_nxv4f32_f16_nxv4f16(<vscale x 4 x float> %0, half %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmsac_mask_vf_nxv4f32_f16_nxv4f16:
@@ -550,7 +550,7 @@ entry:
     half %1,
     <vscale x 4 x half> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x float> %a
 }
@@ -582,7 +582,7 @@ declare <vscale x 8 x float> @llvm.riscv.vfwnmsac.mask.nxv8f32.f16(
   half,
   <vscale x 8 x half>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x float> @intrinsic_vfwnmsac_mask_vf_nxv8f32_f16_nxv8f16(<vscale x 8 x float> %0, half %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmsac_mask_vf_nxv8f32_f16_nxv8f16:
@@ -596,7 +596,7 @@ entry:
     half %1,
     <vscale x 8 x half> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x float> %a
 }
@@ -628,7 +628,7 @@ declare <vscale x 16 x float> @llvm.riscv.vfwnmsac.mask.nxv16f32.f16(
   half,
   <vscale x 16 x half>,
   <vscale x 16 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 16 x float> @intrinsic_vfwnmsac_mask_vf_nxv16f32_f16_nxv16f16(<vscale x 16 x float> %0, half %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmsac_mask_vf_nxv16f32_f16_nxv16f16:
@@ -642,7 +642,7 @@ entry:
     half %1,
     <vscale x 16 x half> %2,
     <vscale x 16 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 16 x float> %a
 }
@@ -674,7 +674,7 @@ declare <vscale x 1 x double> @llvm.riscv.vfwnmsac.mask.nxv1f64.f32(
   float,
   <vscale x 1 x float>,
   <vscale x 1 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 1 x double> @intrinsic_vfwnmsac_mask_vf_nxv1f64_f32_nxv1f32(<vscale x 1 x double> %0, float %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmsac_mask_vf_nxv1f64_f32_nxv1f32:
@@ -688,7 +688,7 @@ entry:
     float %1,
     <vscale x 1 x float> %2,
     <vscale x 1 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 1 x double> %a
 }
@@ -720,7 +720,7 @@ declare <vscale x 2 x double> @llvm.riscv.vfwnmsac.mask.nxv2f64.f32(
   float,
   <vscale x 2 x float>,
   <vscale x 2 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 2 x double> @intrinsic_vfwnmsac_mask_vf_nxv2f64_f32_nxv2f32(<vscale x 2 x double> %0, float %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmsac_mask_vf_nxv2f64_f32_nxv2f32:
@@ -734,7 +734,7 @@ entry:
     float %1,
     <vscale x 2 x float> %2,
     <vscale x 2 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 2 x double> %a
 }
@@ -766,7 +766,7 @@ declare <vscale x 4 x double> @llvm.riscv.vfwnmsac.mask.nxv4f64.f32(
   float,
   <vscale x 4 x float>,
   <vscale x 4 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 4 x double> @intrinsic_vfwnmsac_mask_vf_nxv4f64_f32_nxv4f32(<vscale x 4 x double> %0, float %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmsac_mask_vf_nxv4f64_f32_nxv4f32:
@@ -780,7 +780,7 @@ entry:
     float %1,
     <vscale x 4 x float> %2,
     <vscale x 4 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 4 x double> %a
 }
@@ -812,7 +812,7 @@ declare <vscale x 8 x double> @llvm.riscv.vfwnmsac.mask.nxv8f64.f32(
   float,
   <vscale x 8 x float>,
   <vscale x 8 x i1>,
-  iXLen);
+  iXLen, iXLen);
 
 define <vscale x 8 x double> @intrinsic_vfwnmsac_mask_vf_nxv8f64_f32_nxv8f32(<vscale x 8 x double> %0, float %1, <vscale x 8 x float> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vfwnmsac_mask_vf_nxv8f64_f32_nxv8f32:
@@ -826,7 +826,7 @@ entry:
     float %1,
     <vscale x 8 x float> %2,
     <vscale x 8 x i1> %3,
-    iXLen %4)
+    iXLen %4, iXLen 0);
 
   ret <vscale x 8 x double> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vmacc-rv32.ll b/llvm/test/CodeGen/RISCV/rvv/vmacc-rv32.ll
index 3aeaa11d2ff949..fbdfac8c0cba86 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vmacc-rv32.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vmacc-rv32.ll
@@ -28,7 +28,7 @@ declare <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.nxv1i8(
   <vscale x 1 x i8>,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i8>  @intrinsic_vmacc_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv1i8_nxv1i8_nxv1i8:
@@ -42,7 +42,7 @@ entry:
     <vscale x 1 x i8> %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -74,7 +74,7 @@ declare <vscale x 2 x i8> @llvm.riscv.vmacc.mask.nxv2i8.nxv2i8(
   <vscale x 2 x i8>,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i8>  @intrinsic_vmacc_mask_vv_nxv2i8_nxv2i8_nxv2i8(<vscale x 2 x i8> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv2i8_nxv2i8_nxv2i8:
@@ -88,7 +88,7 @@ entry:
     <vscale x 2 x i8> %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -120,7 +120,7 @@ declare <vscale x 4 x i8> @llvm.riscv.vmacc.mask.nxv4i8.nxv4i8(
   <vscale x 4 x i8>,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i8>  @intrinsic_vmacc_mask_vv_nxv4i8_nxv4i8_nxv4i8(<vscale x 4 x i8> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv4i8_nxv4i8_nxv4i8:
@@ -134,7 +134,7 @@ entry:
     <vscale x 4 x i8> %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -166,7 +166,7 @@ declare <vscale x 8 x i8> @llvm.riscv.vmacc.mask.nxv8i8.nxv8i8(
   <vscale x 8 x i8>,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i8>  @intrinsic_vmacc_mask_vv_nxv8i8_nxv8i8_nxv8i8(<vscale x 8 x i8> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv8i8_nxv8i8_nxv8i8:
@@ -180,7 +180,7 @@ entry:
     <vscale x 8 x i8> %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -212,7 +212,7 @@ declare <vscale x 16 x i8> @llvm.riscv.vmacc.mask.nxv16i8.nxv16i8(
   <vscale x 16 x i8>,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i8>  @intrinsic_vmacc_mask_vv_nxv16i8_nxv16i8_nxv16i8(<vscale x 16 x i8> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv16i8_nxv16i8_nxv16i8:
@@ -226,7 +226,7 @@ entry:
     <vscale x 16 x i8> %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -258,7 +258,7 @@ declare <vscale x 32 x i8> @llvm.riscv.vmacc.mask.nxv32i8.nxv32i8(
   <vscale x 32 x i8>,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 32 x i8>  @intrinsic_vmacc_mask_vv_nxv32i8_nxv32i8_nxv32i8(<vscale x 32 x i8> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv32i8_nxv32i8_nxv32i8:
@@ -272,7 +272,7 @@ entry:
     <vscale x 32 x i8> %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -304,7 +304,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vmacc.mask.nxv1i16.nxv1i16(
   <vscale x 1 x i16>,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i16>  @intrinsic_vmacc_mask_vv_nxv1i16_nxv1i16_nxv1i16(<vscale x 1 x i16> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv1i16_nxv1i16_nxv1i16:
@@ -318,7 +318,7 @@ entry:
     <vscale x 1 x i16> %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -350,7 +350,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vmacc.mask.nxv2i16.nxv2i16(
   <vscale x 2 x i16>,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i16>  @intrinsic_vmacc_mask_vv_nxv2i16_nxv2i16_nxv2i16(<vscale x 2 x i16> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv2i16_nxv2i16_nxv2i16:
@@ -364,7 +364,7 @@ entry:
     <vscale x 2 x i16> %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -396,7 +396,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vmacc.mask.nxv4i16.nxv4i16(
   <vscale x 4 x i16>,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i16>  @intrinsic_vmacc_mask_vv_nxv4i16_nxv4i16_nxv4i16(<vscale x 4 x i16> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv4i16_nxv4i16_nxv4i16:
@@ -410,7 +410,7 @@ entry:
     <vscale x 4 x i16> %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -442,7 +442,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vmacc.mask.nxv8i16.nxv8i16(
   <vscale x 8 x i16>,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i16>  @intrinsic_vmacc_mask_vv_nxv8i16_nxv8i16_nxv8i16(<vscale x 8 x i16> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv8i16_nxv8i16_nxv8i16:
@@ -456,7 +456,7 @@ entry:
     <vscale x 8 x i16> %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -488,7 +488,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vmacc.mask.nxv16i16.nxv16i16(
   <vscale x 16 x i16>,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i16>  @intrinsic_vmacc_mask_vv_nxv16i16_nxv16i16_nxv16i16(<vscale x 16 x i16> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv16i16_nxv16i16_nxv16i16:
@@ -502,7 +502,7 @@ entry:
     <vscale x 16 x i16> %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -534,7 +534,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vmacc.mask.nxv1i32.nxv1i32(
   <vscale x 1 x i32>,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i32>  @intrinsic_vmacc_mask_vv_nxv1i32_nxv1i32_nxv1i32(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv1i32_nxv1i32_nxv1i32:
@@ -548,7 +548,7 @@ entry:
     <vscale x 1 x i32> %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -580,7 +580,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vmacc.mask.nxv2i32.nxv2i32(
   <vscale x 2 x i32>,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i32>  @intrinsic_vmacc_mask_vv_nxv2i32_nxv2i32_nxv2i32(<vscale x 2 x i32> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv2i32_nxv2i32_nxv2i32:
@@ -594,7 +594,7 @@ entry:
     <vscale x 2 x i32> %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -626,7 +626,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vmacc.mask.nxv4i32.nxv4i32(
   <vscale x 4 x i32>,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i32>  @intrinsic_vmacc_mask_vv_nxv4i32_nxv4i32_nxv4i32(<vscale x 4 x i32> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv4i32_nxv4i32_nxv4i32:
@@ -640,7 +640,7 @@ entry:
     <vscale x 4 x i32> %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -672,7 +672,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vmacc.mask.nxv8i32.nxv8i32(
   <vscale x 8 x i32>,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i32>  @intrinsic_vmacc_mask_vv_nxv8i32_nxv8i32_nxv8i32(<vscale x 8 x i32> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv8i32_nxv8i32_nxv8i32:
@@ -686,7 +686,7 @@ entry:
     <vscale x 8 x i32> %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -718,7 +718,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vmacc.mask.nxv1i64.nxv1i64(
   <vscale x 1 x i64>,
   <vscale x 1 x i64>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i64>  @intrinsic_vmacc_mask_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv1i64_nxv1i64_nxv1i64:
@@ -732,7 +732,7 @@ entry:
     <vscale x 1 x i64> %1,
     <vscale x 1 x i64> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -764,7 +764,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vmacc.mask.nxv2i64.nxv2i64(
   <vscale x 2 x i64>,
   <vscale x 2 x i64>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i64>  @intrinsic_vmacc_mask_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv2i64_nxv2i64_nxv2i64:
@@ -778,7 +778,7 @@ entry:
     <vscale x 2 x i64> %1,
     <vscale x 2 x i64> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -810,7 +810,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vmacc.mask.nxv4i64.nxv4i64(
   <vscale x 4 x i64>,
   <vscale x 4 x i64>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i64>  @intrinsic_vmacc_mask_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv4i64_nxv4i64_nxv4i64:
@@ -824,7 +824,7 @@ entry:
     <vscale x 4 x i64> %1,
     <vscale x 4 x i64> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -856,7 +856,7 @@ declare <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.i8(
   i8,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i8> @intrinsic_vmacc_mask_vx_nxv1i8_i8_nxv1i8(<vscale x 1 x i8> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv1i8_i8_nxv1i8:
@@ -870,7 +870,7 @@ entry:
     i8 %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -902,7 +902,7 @@ declare <vscale x 2 x i8> @llvm.riscv.vmacc.mask.nxv2i8.i8(
   i8,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i8> @intrinsic_vmacc_mask_vx_nxv2i8_i8_nxv2i8(<vscale x 2 x i8> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv2i8_i8_nxv2i8:
@@ -916,7 +916,7 @@ entry:
     i8 %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -948,7 +948,7 @@ declare <vscale x 4 x i8> @llvm.riscv.vmacc.mask.nxv4i8.i8(
   i8,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i8> @intrinsic_vmacc_mask_vx_nxv4i8_i8_nxv4i8(<vscale x 4 x i8> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv4i8_i8_nxv4i8:
@@ -962,7 +962,7 @@ entry:
     i8 %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -994,7 +994,7 @@ declare <vscale x 8 x i8> @llvm.riscv.vmacc.mask.nxv8i8.i8(
   i8,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i8> @intrinsic_vmacc_mask_vx_nxv8i8_i8_nxv8i8(<vscale x 8 x i8> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv8i8_i8_nxv8i8:
@@ -1008,7 +1008,7 @@ entry:
     i8 %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -1040,7 +1040,7 @@ declare <vscale x 16 x i8> @llvm.riscv.vmacc.mask.nxv16i8.i8(
   i8,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i8> @intrinsic_vmacc_mask_vx_nxv16i8_i8_nxv16i8(<vscale x 16 x i8> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv16i8_i8_nxv16i8:
@@ -1054,7 +1054,7 @@ entry:
     i8 %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -1086,7 +1086,7 @@ declare <vscale x 32 x i8> @llvm.riscv.vmacc.mask.nxv32i8.i8(
   i8,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 32 x i8> @intrinsic_vmacc_mask_vx_nxv32i8_i8_nxv32i8(<vscale x 32 x i8> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv32i8_i8_nxv32i8:
@@ -1100,7 +1100,7 @@ entry:
     i8 %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -1132,7 +1132,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vmacc.mask.nxv1i16.i16(
   i16,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i16> @intrinsic_vmacc_mask_vx_nxv1i16_i16_nxv1i16(<vscale x 1 x i16> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv1i16_i16_nxv1i16:
@@ -1146,7 +1146,7 @@ entry:
     i16 %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -1178,7 +1178,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vmacc.mask.nxv2i16.i16(
   i16,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i16> @intrinsic_vmacc_mask_vx_nxv2i16_i16_nxv2i16(<vscale x 2 x i16> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv2i16_i16_nxv2i16:
@@ -1192,7 +1192,7 @@ entry:
     i16 %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -1224,7 +1224,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vmacc.mask.nxv4i16.i16(
   i16,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i16> @intrinsic_vmacc_mask_vx_nxv4i16_i16_nxv4i16(<vscale x 4 x i16> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv4i16_i16_nxv4i16:
@@ -1238,7 +1238,7 @@ entry:
     i16 %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -1270,7 +1270,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vmacc.mask.nxv8i16.i16(
   i16,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i16> @intrinsic_vmacc_mask_vx_nxv8i16_i16_nxv8i16(<vscale x 8 x i16> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv8i16_i16_nxv8i16:
@@ -1284,7 +1284,7 @@ entry:
     i16 %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -1316,7 +1316,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vmacc.mask.nxv16i16.i16(
   i16,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i16> @intrinsic_vmacc_mask_vx_nxv16i16_i16_nxv16i16(<vscale x 16 x i16> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv16i16_i16_nxv16i16:
@@ -1330,7 +1330,7 @@ entry:
     i16 %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -1362,7 +1362,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vmacc.mask.nxv1i32.i32(
   i32,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i32> @intrinsic_vmacc_mask_vx_nxv1i32_i32_nxv1i32(<vscale x 1 x i32> %0, i32 %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv1i32_i32_nxv1i32:
@@ -1376,7 +1376,7 @@ entry:
     i32 %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -1408,7 +1408,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vmacc.mask.nxv2i32.i32(
   i32,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i32> @intrinsic_vmacc_mask_vx_nxv2i32_i32_nxv2i32(<vscale x 2 x i32> %0, i32 %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv2i32_i32_nxv2i32:
@@ -1422,7 +1422,7 @@ entry:
     i32 %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -1454,7 +1454,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vmacc.mask.nxv4i32.i32(
   i32,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i32> @intrinsic_vmacc_mask_vx_nxv4i32_i32_nxv4i32(<vscale x 4 x i32> %0, i32 %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv4i32_i32_nxv4i32:
@@ -1468,7 +1468,7 @@ entry:
     i32 %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -1500,7 +1500,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vmacc.mask.nxv8i32.i32(
   i32,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i32> @intrinsic_vmacc_mask_vx_nxv8i32_i32_nxv8i32(<vscale x 8 x i32> %0, i32 %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv8i32_i32_nxv8i32:
@@ -1514,7 +1514,7 @@ entry:
     i32 %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -1553,7 +1553,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vmacc.mask.nxv1i64.i64(
   i64,
   <vscale x 1 x i64>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i64> @intrinsic_vmacc_mask_vx_nxv1i64_i64_nxv1i64(<vscale x 1 x i64> %0, i64 %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv1i64_i64_nxv1i64:
@@ -1574,7 +1574,7 @@ entry:
     i64 %1,
     <vscale x 1 x i64> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -1613,7 +1613,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vmacc.mask.nxv2i64.i64(
   i64,
   <vscale x 2 x i64>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i64> @intrinsic_vmacc_mask_vx_nxv2i64_i64_nxv2i64(<vscale x 2 x i64> %0, i64 %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv2i64_i64_nxv2i64:
@@ -1634,7 +1634,7 @@ entry:
     i64 %1,
     <vscale x 2 x i64> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -1673,7 +1673,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vmacc.mask.nxv4i64.i64(
   i64,
   <vscale x 4 x i64>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i64> @intrinsic_vmacc_mask_vx_nxv4i64_i64_nxv4i64(<vscale x 4 x i64> %0, i64 %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv4i64_i64_nxv4i64:
@@ -1694,7 +1694,7 @@ entry:
     i64 %1,
     <vscale x 4 x i64> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i64> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vmacc-rv64.ll b/llvm/test/CodeGen/RISCV/rvv/vmacc-rv64.ll
index cfb3a432b92f03..16ea00ad789595 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vmacc-rv64.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vmacc-rv64.ll
@@ -28,7 +28,7 @@ declare <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.nxv1i8(
   <vscale x 1 x i8>,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i8>  @intrinsic_vmacc_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv1i8_nxv1i8_nxv1i8:
@@ -42,7 +42,7 @@ entry:
     <vscale x 1 x i8> %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -74,7 +74,7 @@ declare <vscale x 2 x i8> @llvm.riscv.vmacc.mask.nxv2i8.nxv2i8(
   <vscale x 2 x i8>,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i8>  @intrinsic_vmacc_mask_vv_nxv2i8_nxv2i8_nxv2i8(<vscale x 2 x i8> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv2i8_nxv2i8_nxv2i8:
@@ -88,7 +88,7 @@ entry:
     <vscale x 2 x i8> %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -120,7 +120,7 @@ declare <vscale x 4 x i8> @llvm.riscv.vmacc.mask.nxv4i8.nxv4i8(
   <vscale x 4 x i8>,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i8>  @intrinsic_vmacc_mask_vv_nxv4i8_nxv4i8_nxv4i8(<vscale x 4 x i8> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv4i8_nxv4i8_nxv4i8:
@@ -134,7 +134,7 @@ entry:
     <vscale x 4 x i8> %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -166,7 +166,7 @@ declare <vscale x 8 x i8> @llvm.riscv.vmacc.mask.nxv8i8.nxv8i8(
   <vscale x 8 x i8>,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i8>  @intrinsic_vmacc_mask_vv_nxv8i8_nxv8i8_nxv8i8(<vscale x 8 x i8> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv8i8_nxv8i8_nxv8i8:
@@ -180,7 +180,7 @@ entry:
     <vscale x 8 x i8> %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -212,7 +212,7 @@ declare <vscale x 16 x i8> @llvm.riscv.vmacc.mask.nxv16i8.nxv16i8(
   <vscale x 16 x i8>,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i8>  @intrinsic_vmacc_mask_vv_nxv16i8_nxv16i8_nxv16i8(<vscale x 16 x i8> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv16i8_nxv16i8_nxv16i8:
@@ -226,7 +226,7 @@ entry:
     <vscale x 16 x i8> %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -258,7 +258,7 @@ declare <vscale x 32 x i8> @llvm.riscv.vmacc.mask.nxv32i8.nxv32i8(
   <vscale x 32 x i8>,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 32 x i8>  @intrinsic_vmacc_mask_vv_nxv32i8_nxv32i8_nxv32i8(<vscale x 32 x i8> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv32i8_nxv32i8_nxv32i8:
@@ -272,7 +272,7 @@ entry:
     <vscale x 32 x i8> %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -304,7 +304,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vmacc.mask.nxv1i16.nxv1i16(
   <vscale x 1 x i16>,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i16>  @intrinsic_vmacc_mask_vv_nxv1i16_nxv1i16_nxv1i16(<vscale x 1 x i16> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv1i16_nxv1i16_nxv1i16:
@@ -318,7 +318,7 @@ entry:
     <vscale x 1 x i16> %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -350,7 +350,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vmacc.mask.nxv2i16.nxv2i16(
   <vscale x 2 x i16>,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i16>  @intrinsic_vmacc_mask_vv_nxv2i16_nxv2i16_nxv2i16(<vscale x 2 x i16> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv2i16_nxv2i16_nxv2i16:
@@ -364,7 +364,7 @@ entry:
     <vscale x 2 x i16> %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -396,7 +396,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vmacc.mask.nxv4i16.nxv4i16(
   <vscale x 4 x i16>,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i16>  @intrinsic_vmacc_mask_vv_nxv4i16_nxv4i16_nxv4i16(<vscale x 4 x i16> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv4i16_nxv4i16_nxv4i16:
@@ -410,7 +410,7 @@ entry:
     <vscale x 4 x i16> %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -442,7 +442,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vmacc.mask.nxv8i16.nxv8i16(
   <vscale x 8 x i16>,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i16>  @intrinsic_vmacc_mask_vv_nxv8i16_nxv8i16_nxv8i16(<vscale x 8 x i16> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv8i16_nxv8i16_nxv8i16:
@@ -456,7 +456,7 @@ entry:
     <vscale x 8 x i16> %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -488,7 +488,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vmacc.mask.nxv16i16.nxv16i16(
   <vscale x 16 x i16>,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i16>  @intrinsic_vmacc_mask_vv_nxv16i16_nxv16i16_nxv16i16(<vscale x 16 x i16> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv16i16_nxv16i16_nxv16i16:
@@ -502,7 +502,7 @@ entry:
     <vscale x 16 x i16> %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -534,7 +534,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vmacc.mask.nxv1i32.nxv1i32(
   <vscale x 1 x i32>,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i32>  @intrinsic_vmacc_mask_vv_nxv1i32_nxv1i32_nxv1i32(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv1i32_nxv1i32_nxv1i32:
@@ -548,7 +548,7 @@ entry:
     <vscale x 1 x i32> %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -580,7 +580,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vmacc.mask.nxv2i32.nxv2i32(
   <vscale x 2 x i32>,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i32>  @intrinsic_vmacc_mask_vv_nxv2i32_nxv2i32_nxv2i32(<vscale x 2 x i32> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv2i32_nxv2i32_nxv2i32:
@@ -594,7 +594,7 @@ entry:
     <vscale x 2 x i32> %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -626,7 +626,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vmacc.mask.nxv4i32.nxv4i32(
   <vscale x 4 x i32>,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i32>  @intrinsic_vmacc_mask_vv_nxv4i32_nxv4i32_nxv4i32(<vscale x 4 x i32> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv4i32_nxv4i32_nxv4i32:
@@ -640,7 +640,7 @@ entry:
     <vscale x 4 x i32> %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -672,7 +672,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vmacc.mask.nxv8i32.nxv8i32(
   <vscale x 8 x i32>,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i32>  @intrinsic_vmacc_mask_vv_nxv8i32_nxv8i32_nxv8i32(<vscale x 8 x i32> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv8i32_nxv8i32_nxv8i32:
@@ -686,7 +686,7 @@ entry:
     <vscale x 8 x i32> %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -718,7 +718,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vmacc.mask.nxv1i64.nxv1i64(
   <vscale x 1 x i64>,
   <vscale x 1 x i64>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i64>  @intrinsic_vmacc_mask_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv1i64_nxv1i64_nxv1i64:
@@ -732,7 +732,7 @@ entry:
     <vscale x 1 x i64> %1,
     <vscale x 1 x i64> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -764,7 +764,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vmacc.mask.nxv2i64.nxv2i64(
   <vscale x 2 x i64>,
   <vscale x 2 x i64>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i64>  @intrinsic_vmacc_mask_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv2i64_nxv2i64_nxv2i64:
@@ -778,7 +778,7 @@ entry:
     <vscale x 2 x i64> %1,
     <vscale x 2 x i64> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -810,7 +810,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vmacc.mask.nxv4i64.nxv4i64(
   <vscale x 4 x i64>,
   <vscale x 4 x i64>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i64>  @intrinsic_vmacc_mask_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vv_nxv4i64_nxv4i64_nxv4i64:
@@ -824,7 +824,7 @@ entry:
     <vscale x 4 x i64> %1,
     <vscale x 4 x i64> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -856,7 +856,7 @@ declare <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.i8(
   i8,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i8> @intrinsic_vmacc_mask_vx_nxv1i8_i8_nxv1i8(<vscale x 1 x i8> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv1i8_i8_nxv1i8:
@@ -870,7 +870,7 @@ entry:
     i8 %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -902,7 +902,7 @@ declare <vscale x 2 x i8> @llvm.riscv.vmacc.mask.nxv2i8.i8(
   i8,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i8> @intrinsic_vmacc_mask_vx_nxv2i8_i8_nxv2i8(<vscale x 2 x i8> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv2i8_i8_nxv2i8:
@@ -916,7 +916,7 @@ entry:
     i8 %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -948,7 +948,7 @@ declare <vscale x 4 x i8> @llvm.riscv.vmacc.mask.nxv4i8.i8(
   i8,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i8> @intrinsic_vmacc_mask_vx_nxv4i8_i8_nxv4i8(<vscale x 4 x i8> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv4i8_i8_nxv4i8:
@@ -962,7 +962,7 @@ entry:
     i8 %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -994,7 +994,7 @@ declare <vscale x 8 x i8> @llvm.riscv.vmacc.mask.nxv8i8.i8(
   i8,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i8> @intrinsic_vmacc_mask_vx_nxv8i8_i8_nxv8i8(<vscale x 8 x i8> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv8i8_i8_nxv8i8:
@@ -1008,7 +1008,7 @@ entry:
     i8 %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -1040,7 +1040,7 @@ declare <vscale x 16 x i8> @llvm.riscv.vmacc.mask.nxv16i8.i8(
   i8,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i8> @intrinsic_vmacc_mask_vx_nxv16i8_i8_nxv16i8(<vscale x 16 x i8> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv16i8_i8_nxv16i8:
@@ -1054,7 +1054,7 @@ entry:
     i8 %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -1086,7 +1086,7 @@ declare <vscale x 32 x i8> @llvm.riscv.vmacc.mask.nxv32i8.i8(
   i8,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 32 x i8> @intrinsic_vmacc_mask_vx_nxv32i8_i8_nxv32i8(<vscale x 32 x i8> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv32i8_i8_nxv32i8:
@@ -1100,7 +1100,7 @@ entry:
     i8 %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -1132,7 +1132,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vmacc.mask.nxv1i16.i16(
   i16,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i16> @intrinsic_vmacc_mask_vx_nxv1i16_i16_nxv1i16(<vscale x 1 x i16> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv1i16_i16_nxv1i16:
@@ -1146,7 +1146,7 @@ entry:
     i16 %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -1178,7 +1178,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vmacc.mask.nxv2i16.i16(
   i16,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i16> @intrinsic_vmacc_mask_vx_nxv2i16_i16_nxv2i16(<vscale x 2 x i16> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv2i16_i16_nxv2i16:
@@ -1192,7 +1192,7 @@ entry:
     i16 %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -1224,7 +1224,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vmacc.mask.nxv4i16.i16(
   i16,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i16> @intrinsic_vmacc_mask_vx_nxv4i16_i16_nxv4i16(<vscale x 4 x i16> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv4i16_i16_nxv4i16:
@@ -1238,7 +1238,7 @@ entry:
     i16 %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -1270,7 +1270,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vmacc.mask.nxv8i16.i16(
   i16,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i16> @intrinsic_vmacc_mask_vx_nxv8i16_i16_nxv8i16(<vscale x 8 x i16> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv8i16_i16_nxv8i16:
@@ -1284,7 +1284,7 @@ entry:
     i16 %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -1316,7 +1316,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vmacc.mask.nxv16i16.i16(
   i16,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i16> @intrinsic_vmacc_mask_vx_nxv16i16_i16_nxv16i16(<vscale x 16 x i16> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv16i16_i16_nxv16i16:
@@ -1330,7 +1330,7 @@ entry:
     i16 %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -1362,7 +1362,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vmacc.mask.nxv1i32.i32(
   i32,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i32> @intrinsic_vmacc_mask_vx_nxv1i32_i32_nxv1i32(<vscale x 1 x i32> %0, i32 %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv1i32_i32_nxv1i32:
@@ -1376,7 +1376,7 @@ entry:
     i32 %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -1408,7 +1408,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vmacc.mask.nxv2i32.i32(
   i32,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i32> @intrinsic_vmacc_mask_vx_nxv2i32_i32_nxv2i32(<vscale x 2 x i32> %0, i32 %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv2i32_i32_nxv2i32:
@@ -1422,7 +1422,7 @@ entry:
     i32 %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -1454,7 +1454,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vmacc.mask.nxv4i32.i32(
   i32,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i32> @intrinsic_vmacc_mask_vx_nxv4i32_i32_nxv4i32(<vscale x 4 x i32> %0, i32 %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv4i32_i32_nxv4i32:
@@ -1468,7 +1468,7 @@ entry:
     i32 %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -1500,7 +1500,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vmacc.mask.nxv8i32.i32(
   i32,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i32> @intrinsic_vmacc_mask_vx_nxv8i32_i32_nxv8i32(<vscale x 8 x i32> %0, i32 %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv8i32_i32_nxv8i32:
@@ -1514,7 +1514,7 @@ entry:
     i32 %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -1546,7 +1546,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vmacc.mask.nxv1i64.i64(
   i64,
   <vscale x 1 x i64>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i64> @intrinsic_vmacc_mask_vx_nxv1i64_i64_nxv1i64(<vscale x 1 x i64> %0, i64 %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv1i64_i64_nxv1i64:
@@ -1560,7 +1560,7 @@ entry:
     i64 %1,
     <vscale x 1 x i64> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -1592,7 +1592,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vmacc.mask.nxv2i64.i64(
   i64,
   <vscale x 2 x i64>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i64> @intrinsic_vmacc_mask_vx_nxv2i64_i64_nxv2i64(<vscale x 2 x i64> %0, i64 %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv2i64_i64_nxv2i64:
@@ -1606,7 +1606,7 @@ entry:
     i64 %1,
     <vscale x 2 x i64> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -1638,7 +1638,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vmacc.mask.nxv4i64.i64(
   i64,
   <vscale x 4 x i64>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i64> @intrinsic_vmacc_mask_vx_nxv4i64_i64_nxv4i64(<vscale x 4 x i64> %0, i64 %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmacc_mask_vx_nxv4i64_i64_nxv4i64:
@@ -1652,7 +1652,7 @@ entry:
     i64 %1,
     <vscale x 4 x i64> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i64> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vmadd-rv32.ll b/llvm/test/CodeGen/RISCV/rvv/vmadd-rv32.ll
index fa46e6e8932a47..e93f3bffe82a0f 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vmadd-rv32.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vmadd-rv32.ll
@@ -28,7 +28,7 @@ declare <vscale x 1 x i8> @llvm.riscv.vmadd.mask.nxv1i8.nxv1i8(
   <vscale x 1 x i8>,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i8>  @intrinsic_vmadd_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
@@ -42,7 +42,7 @@ entry:
     <vscale x 1 x i8> %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -74,7 +74,7 @@ declare <vscale x 2 x i8> @llvm.riscv.vmadd.mask.nxv2i8.nxv2i8(
   <vscale x 2 x i8>,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i8>  @intrinsic_vmadd_mask_vv_nxv2i8_nxv2i8_nxv2i8(<vscale x 2 x i8> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv2i8_nxv2i8_nxv2i8:
@@ -88,7 +88,7 @@ entry:
     <vscale x 2 x i8> %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -120,7 +120,7 @@ declare <vscale x 4 x i8> @llvm.riscv.vmadd.mask.nxv4i8.nxv4i8(
   <vscale x 4 x i8>,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i8>  @intrinsic_vmadd_mask_vv_nxv4i8_nxv4i8_nxv4i8(<vscale x 4 x i8> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv4i8_nxv4i8_nxv4i8:
@@ -134,7 +134,7 @@ entry:
     <vscale x 4 x i8> %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -166,7 +166,7 @@ declare <vscale x 8 x i8> @llvm.riscv.vmadd.mask.nxv8i8.nxv8i8(
   <vscale x 8 x i8>,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i8>  @intrinsic_vmadd_mask_vv_nxv8i8_nxv8i8_nxv8i8(<vscale x 8 x i8> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv8i8_nxv8i8_nxv8i8:
@@ -180,7 +180,7 @@ entry:
     <vscale x 8 x i8> %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -212,7 +212,7 @@ declare <vscale x 16 x i8> @llvm.riscv.vmadd.mask.nxv16i8.nxv16i8(
   <vscale x 16 x i8>,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i8>  @intrinsic_vmadd_mask_vv_nxv16i8_nxv16i8_nxv16i8(<vscale x 16 x i8> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv16i8_nxv16i8_nxv16i8:
@@ -226,7 +226,7 @@ entry:
     <vscale x 16 x i8> %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -258,7 +258,7 @@ declare <vscale x 32 x i8> @llvm.riscv.vmadd.mask.nxv32i8.nxv32i8(
   <vscale x 32 x i8>,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 32 x i8>  @intrinsic_vmadd_mask_vv_nxv32i8_nxv32i8_nxv32i8(<vscale x 32 x i8> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv32i8_nxv32i8_nxv32i8:
@@ -272,7 +272,7 @@ entry:
     <vscale x 32 x i8> %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -304,7 +304,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vmadd.mask.nxv1i16.nxv1i16(
   <vscale x 1 x i16>,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i16>  @intrinsic_vmadd_mask_vv_nxv1i16_nxv1i16_nxv1i16(<vscale x 1 x i16> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv1i16_nxv1i16_nxv1i16:
@@ -318,7 +318,7 @@ entry:
     <vscale x 1 x i16> %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -350,7 +350,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vmadd.mask.nxv2i16.nxv2i16(
   <vscale x 2 x i16>,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i16>  @intrinsic_vmadd_mask_vv_nxv2i16_nxv2i16_nxv2i16(<vscale x 2 x i16> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv2i16_nxv2i16_nxv2i16:
@@ -364,7 +364,7 @@ entry:
     <vscale x 2 x i16> %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -396,7 +396,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vmadd.mask.nxv4i16.nxv4i16(
   <vscale x 4 x i16>,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i16>  @intrinsic_vmadd_mask_vv_nxv4i16_nxv4i16_nxv4i16(<vscale x 4 x i16> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv4i16_nxv4i16_nxv4i16:
@@ -410,7 +410,7 @@ entry:
     <vscale x 4 x i16> %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -442,7 +442,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vmadd.mask.nxv8i16.nxv8i16(
   <vscale x 8 x i16>,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i16>  @intrinsic_vmadd_mask_vv_nxv8i16_nxv8i16_nxv8i16(<vscale x 8 x i16> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv8i16_nxv8i16_nxv8i16:
@@ -456,7 +456,7 @@ entry:
     <vscale x 8 x i16> %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -488,7 +488,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vmadd.mask.nxv16i16.nxv16i16(
   <vscale x 16 x i16>,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i16>  @intrinsic_vmadd_mask_vv_nxv16i16_nxv16i16_nxv16i16(<vscale x 16 x i16> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv16i16_nxv16i16_nxv16i16:
@@ -502,7 +502,7 @@ entry:
     <vscale x 16 x i16> %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -534,7 +534,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vmadd.mask.nxv1i32.nxv1i32(
   <vscale x 1 x i32>,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i32>  @intrinsic_vmadd_mask_vv_nxv1i32_nxv1i32_nxv1i32(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv1i32_nxv1i32_nxv1i32:
@@ -548,7 +548,7 @@ entry:
     <vscale x 1 x i32> %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -580,7 +580,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vmadd.mask.nxv2i32.nxv2i32(
   <vscale x 2 x i32>,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i32>  @intrinsic_vmadd_mask_vv_nxv2i32_nxv2i32_nxv2i32(<vscale x 2 x i32> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv2i32_nxv2i32_nxv2i32:
@@ -594,7 +594,7 @@ entry:
     <vscale x 2 x i32> %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -626,7 +626,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vmadd.mask.nxv4i32.nxv4i32(
   <vscale x 4 x i32>,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i32>  @intrinsic_vmadd_mask_vv_nxv4i32_nxv4i32_nxv4i32(<vscale x 4 x i32> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv4i32_nxv4i32_nxv4i32:
@@ -640,7 +640,7 @@ entry:
     <vscale x 4 x i32> %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -672,7 +672,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vmadd.mask.nxv8i32.nxv8i32(
   <vscale x 8 x i32>,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i32>  @intrinsic_vmadd_mask_vv_nxv8i32_nxv8i32_nxv8i32(<vscale x 8 x i32> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv8i32_nxv8i32_nxv8i32:
@@ -686,7 +686,7 @@ entry:
     <vscale x 8 x i32> %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -718,7 +718,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vmadd.mask.nxv1i64.nxv1i64(
   <vscale x 1 x i64>,
   <vscale x 1 x i64>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i64>  @intrinsic_vmadd_mask_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv1i64_nxv1i64_nxv1i64:
@@ -732,7 +732,7 @@ entry:
     <vscale x 1 x i64> %1,
     <vscale x 1 x i64> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -764,7 +764,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vmadd.mask.nxv2i64.nxv2i64(
   <vscale x 2 x i64>,
   <vscale x 2 x i64>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i64>  @intrinsic_vmadd_mask_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv2i64_nxv2i64_nxv2i64:
@@ -778,7 +778,7 @@ entry:
     <vscale x 2 x i64> %1,
     <vscale x 2 x i64> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -810,7 +810,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vmadd.mask.nxv4i64.nxv4i64(
   <vscale x 4 x i64>,
   <vscale x 4 x i64>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i64>  @intrinsic_vmadd_mask_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv4i64_nxv4i64_nxv4i64:
@@ -824,7 +824,7 @@ entry:
     <vscale x 4 x i64> %1,
     <vscale x 4 x i64> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -856,7 +856,7 @@ declare <vscale x 1 x i8> @llvm.riscv.vmadd.mask.nxv1i8.i8(
   i8,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i8> @intrinsic_vmadd_mask_vx_nxv1i8_i8_nxv1i8(<vscale x 1 x i8> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv1i8_i8_nxv1i8:
@@ -870,7 +870,7 @@ entry:
     i8 %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -902,7 +902,7 @@ declare <vscale x 2 x i8> @llvm.riscv.vmadd.mask.nxv2i8.i8(
   i8,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i8> @intrinsic_vmadd_mask_vx_nxv2i8_i8_nxv2i8(<vscale x 2 x i8> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv2i8_i8_nxv2i8:
@@ -916,7 +916,7 @@ entry:
     i8 %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -948,7 +948,7 @@ declare <vscale x 4 x i8> @llvm.riscv.vmadd.mask.nxv4i8.i8(
   i8,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i8> @intrinsic_vmadd_mask_vx_nxv4i8_i8_nxv4i8(<vscale x 4 x i8> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv4i8_i8_nxv4i8:
@@ -962,7 +962,7 @@ entry:
     i8 %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -994,7 +994,7 @@ declare <vscale x 8 x i8> @llvm.riscv.vmadd.mask.nxv8i8.i8(
   i8,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i8> @intrinsic_vmadd_mask_vx_nxv8i8_i8_nxv8i8(<vscale x 8 x i8> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv8i8_i8_nxv8i8:
@@ -1008,7 +1008,7 @@ entry:
     i8 %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -1040,7 +1040,7 @@ declare <vscale x 16 x i8> @llvm.riscv.vmadd.mask.nxv16i8.i8(
   i8,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i8> @intrinsic_vmadd_mask_vx_nxv16i8_i8_nxv16i8(<vscale x 16 x i8> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv16i8_i8_nxv16i8:
@@ -1054,7 +1054,7 @@ entry:
     i8 %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -1086,7 +1086,7 @@ declare <vscale x 32 x i8> @llvm.riscv.vmadd.mask.nxv32i8.i8(
   i8,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 32 x i8> @intrinsic_vmadd_mask_vx_nxv32i8_i8_nxv32i8(<vscale x 32 x i8> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv32i8_i8_nxv32i8:
@@ -1100,7 +1100,7 @@ entry:
     i8 %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -1132,7 +1132,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vmadd.mask.nxv1i16.i16(
   i16,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i16> @intrinsic_vmadd_mask_vx_nxv1i16_i16_nxv1i16(<vscale x 1 x i16> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv1i16_i16_nxv1i16:
@@ -1146,7 +1146,7 @@ entry:
     i16 %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -1178,7 +1178,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vmadd.mask.nxv2i16.i16(
   i16,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i16> @intrinsic_vmadd_mask_vx_nxv2i16_i16_nxv2i16(<vscale x 2 x i16> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv2i16_i16_nxv2i16:
@@ -1192,7 +1192,7 @@ entry:
     i16 %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -1224,7 +1224,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vmadd.mask.nxv4i16.i16(
   i16,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i16> @intrinsic_vmadd_mask_vx_nxv4i16_i16_nxv4i16(<vscale x 4 x i16> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv4i16_i16_nxv4i16:
@@ -1238,7 +1238,7 @@ entry:
     i16 %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -1270,7 +1270,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vmadd.mask.nxv8i16.i16(
   i16,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i16> @intrinsic_vmadd_mask_vx_nxv8i16_i16_nxv8i16(<vscale x 8 x i16> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv8i16_i16_nxv8i16:
@@ -1284,7 +1284,7 @@ entry:
     i16 %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -1316,7 +1316,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vmadd.mask.nxv16i16.i16(
   i16,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i16> @intrinsic_vmadd_mask_vx_nxv16i16_i16_nxv16i16(<vscale x 16 x i16> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv16i16_i16_nxv16i16:
@@ -1330,7 +1330,7 @@ entry:
     i16 %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -1362,7 +1362,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vmadd.mask.nxv1i32.i32(
   i32,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i32> @intrinsic_vmadd_mask_vx_nxv1i32_i32_nxv1i32(<vscale x 1 x i32> %0, i32 %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv1i32_i32_nxv1i32:
@@ -1376,7 +1376,7 @@ entry:
     i32 %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -1408,7 +1408,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vmadd.mask.nxv2i32.i32(
   i32,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i32> @intrinsic_vmadd_mask_vx_nxv2i32_i32_nxv2i32(<vscale x 2 x i32> %0, i32 %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv2i32_i32_nxv2i32:
@@ -1422,7 +1422,7 @@ entry:
     i32 %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -1454,7 +1454,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vmadd.mask.nxv4i32.i32(
   i32,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i32> @intrinsic_vmadd_mask_vx_nxv4i32_i32_nxv4i32(<vscale x 4 x i32> %0, i32 %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv4i32_i32_nxv4i32:
@@ -1468,7 +1468,7 @@ entry:
     i32 %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -1500,7 +1500,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vmadd.mask.nxv8i32.i32(
   i32,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i32> @intrinsic_vmadd_mask_vx_nxv8i32_i32_nxv8i32(<vscale x 8 x i32> %0, i32 %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv8i32_i32_nxv8i32:
@@ -1514,7 +1514,7 @@ entry:
     i32 %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -1553,7 +1553,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vmadd.mask.nxv1i64.i64(
   i64,
   <vscale x 1 x i64>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i64> @intrinsic_vmadd_mask_vx_nxv1i64_i64_nxv1i64(<vscale x 1 x i64> %0, i64 %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv1i64_i64_nxv1i64:
@@ -1574,7 +1574,7 @@ entry:
     i64 %1,
     <vscale x 1 x i64> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -1613,7 +1613,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vmadd.mask.nxv2i64.i64(
   i64,
   <vscale x 2 x i64>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i64> @intrinsic_vmadd_mask_vx_nxv2i64_i64_nxv2i64(<vscale x 2 x i64> %0, i64 %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv2i64_i64_nxv2i64:
@@ -1634,7 +1634,7 @@ entry:
     i64 %1,
     <vscale x 2 x i64> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -1673,7 +1673,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vmadd.mask.nxv4i64.i64(
   i64,
   <vscale x 4 x i64>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i64> @intrinsic_vmadd_mask_vx_nxv4i64_i64_nxv4i64(<vscale x 4 x i64> %0, i64 %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv4i64_i64_nxv4i64:
@@ -1694,7 +1694,7 @@ entry:
     i64 %1,
     <vscale x 4 x i64> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i64> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vmadd-rv64.ll b/llvm/test/CodeGen/RISCV/rvv/vmadd-rv64.ll
index fc401f3cba8197..63695f673f3020 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vmadd-rv64.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vmadd-rv64.ll
@@ -28,7 +28,7 @@ declare <vscale x 1 x i8> @llvm.riscv.vmadd.mask.nxv1i8.nxv1i8(
   <vscale x 1 x i8>,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i8>  @intrinsic_vmadd_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
@@ -42,7 +42,7 @@ entry:
     <vscale x 1 x i8> %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -74,7 +74,7 @@ declare <vscale x 2 x i8> @llvm.riscv.vmadd.mask.nxv2i8.nxv2i8(
   <vscale x 2 x i8>,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i8>  @intrinsic_vmadd_mask_vv_nxv2i8_nxv2i8_nxv2i8(<vscale x 2 x i8> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv2i8_nxv2i8_nxv2i8:
@@ -88,7 +88,7 @@ entry:
     <vscale x 2 x i8> %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -120,7 +120,7 @@ declare <vscale x 4 x i8> @llvm.riscv.vmadd.mask.nxv4i8.nxv4i8(
   <vscale x 4 x i8>,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i8>  @intrinsic_vmadd_mask_vv_nxv4i8_nxv4i8_nxv4i8(<vscale x 4 x i8> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv4i8_nxv4i8_nxv4i8:
@@ -134,7 +134,7 @@ entry:
     <vscale x 4 x i8> %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -166,7 +166,7 @@ declare <vscale x 8 x i8> @llvm.riscv.vmadd.mask.nxv8i8.nxv8i8(
   <vscale x 8 x i8>,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i8>  @intrinsic_vmadd_mask_vv_nxv8i8_nxv8i8_nxv8i8(<vscale x 8 x i8> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv8i8_nxv8i8_nxv8i8:
@@ -180,7 +180,7 @@ entry:
     <vscale x 8 x i8> %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -212,7 +212,7 @@ declare <vscale x 16 x i8> @llvm.riscv.vmadd.mask.nxv16i8.nxv16i8(
   <vscale x 16 x i8>,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i8>  @intrinsic_vmadd_mask_vv_nxv16i8_nxv16i8_nxv16i8(<vscale x 16 x i8> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv16i8_nxv16i8_nxv16i8:
@@ -226,7 +226,7 @@ entry:
     <vscale x 16 x i8> %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -258,7 +258,7 @@ declare <vscale x 32 x i8> @llvm.riscv.vmadd.mask.nxv32i8.nxv32i8(
   <vscale x 32 x i8>,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 32 x i8>  @intrinsic_vmadd_mask_vv_nxv32i8_nxv32i8_nxv32i8(<vscale x 32 x i8> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv32i8_nxv32i8_nxv32i8:
@@ -272,7 +272,7 @@ entry:
     <vscale x 32 x i8> %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -304,7 +304,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vmadd.mask.nxv1i16.nxv1i16(
   <vscale x 1 x i16>,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i16>  @intrinsic_vmadd_mask_vv_nxv1i16_nxv1i16_nxv1i16(<vscale x 1 x i16> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv1i16_nxv1i16_nxv1i16:
@@ -318,7 +318,7 @@ entry:
     <vscale x 1 x i16> %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -350,7 +350,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vmadd.mask.nxv2i16.nxv2i16(
   <vscale x 2 x i16>,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i16>  @intrinsic_vmadd_mask_vv_nxv2i16_nxv2i16_nxv2i16(<vscale x 2 x i16> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv2i16_nxv2i16_nxv2i16:
@@ -364,7 +364,7 @@ entry:
     <vscale x 2 x i16> %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -396,7 +396,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vmadd.mask.nxv4i16.nxv4i16(
   <vscale x 4 x i16>,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i16>  @intrinsic_vmadd_mask_vv_nxv4i16_nxv4i16_nxv4i16(<vscale x 4 x i16> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv4i16_nxv4i16_nxv4i16:
@@ -410,7 +410,7 @@ entry:
     <vscale x 4 x i16> %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -442,7 +442,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vmadd.mask.nxv8i16.nxv8i16(
   <vscale x 8 x i16>,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i16>  @intrinsic_vmadd_mask_vv_nxv8i16_nxv8i16_nxv8i16(<vscale x 8 x i16> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv8i16_nxv8i16_nxv8i16:
@@ -456,7 +456,7 @@ entry:
     <vscale x 8 x i16> %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -488,7 +488,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vmadd.mask.nxv16i16.nxv16i16(
   <vscale x 16 x i16>,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i16>  @intrinsic_vmadd_mask_vv_nxv16i16_nxv16i16_nxv16i16(<vscale x 16 x i16> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv16i16_nxv16i16_nxv16i16:
@@ -502,7 +502,7 @@ entry:
     <vscale x 16 x i16> %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -534,7 +534,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vmadd.mask.nxv1i32.nxv1i32(
   <vscale x 1 x i32>,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i32>  @intrinsic_vmadd_mask_vv_nxv1i32_nxv1i32_nxv1i32(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv1i32_nxv1i32_nxv1i32:
@@ -548,7 +548,7 @@ entry:
     <vscale x 1 x i32> %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -580,7 +580,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vmadd.mask.nxv2i32.nxv2i32(
   <vscale x 2 x i32>,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i32>  @intrinsic_vmadd_mask_vv_nxv2i32_nxv2i32_nxv2i32(<vscale x 2 x i32> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv2i32_nxv2i32_nxv2i32:
@@ -594,7 +594,7 @@ entry:
     <vscale x 2 x i32> %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -626,7 +626,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vmadd.mask.nxv4i32.nxv4i32(
   <vscale x 4 x i32>,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i32>  @intrinsic_vmadd_mask_vv_nxv4i32_nxv4i32_nxv4i32(<vscale x 4 x i32> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv4i32_nxv4i32_nxv4i32:
@@ -640,7 +640,7 @@ entry:
     <vscale x 4 x i32> %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -672,7 +672,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vmadd.mask.nxv8i32.nxv8i32(
   <vscale x 8 x i32>,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i32>  @intrinsic_vmadd_mask_vv_nxv8i32_nxv8i32_nxv8i32(<vscale x 8 x i32> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv8i32_nxv8i32_nxv8i32:
@@ -686,7 +686,7 @@ entry:
     <vscale x 8 x i32> %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -718,7 +718,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vmadd.mask.nxv1i64.nxv1i64(
   <vscale x 1 x i64>,
   <vscale x 1 x i64>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i64>  @intrinsic_vmadd_mask_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv1i64_nxv1i64_nxv1i64:
@@ -732,7 +732,7 @@ entry:
     <vscale x 1 x i64> %1,
     <vscale x 1 x i64> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -764,7 +764,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vmadd.mask.nxv2i64.nxv2i64(
   <vscale x 2 x i64>,
   <vscale x 2 x i64>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i64>  @intrinsic_vmadd_mask_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv2i64_nxv2i64_nxv2i64:
@@ -778,7 +778,7 @@ entry:
     <vscale x 2 x i64> %1,
     <vscale x 2 x i64> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -810,7 +810,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vmadd.mask.nxv4i64.nxv4i64(
   <vscale x 4 x i64>,
   <vscale x 4 x i64>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i64>  @intrinsic_vmadd_mask_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vv_nxv4i64_nxv4i64_nxv4i64:
@@ -824,7 +824,7 @@ entry:
     <vscale x 4 x i64> %1,
     <vscale x 4 x i64> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -856,7 +856,7 @@ declare <vscale x 1 x i8> @llvm.riscv.vmadd.mask.nxv1i8.i8(
   i8,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i8> @intrinsic_vmadd_mask_vx_nxv1i8_i8_nxv1i8(<vscale x 1 x i8> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv1i8_i8_nxv1i8:
@@ -870,7 +870,7 @@ entry:
     i8 %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -902,7 +902,7 @@ declare <vscale x 2 x i8> @llvm.riscv.vmadd.mask.nxv2i8.i8(
   i8,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i8> @intrinsic_vmadd_mask_vx_nxv2i8_i8_nxv2i8(<vscale x 2 x i8> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv2i8_i8_nxv2i8:
@@ -916,7 +916,7 @@ entry:
     i8 %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -948,7 +948,7 @@ declare <vscale x 4 x i8> @llvm.riscv.vmadd.mask.nxv4i8.i8(
   i8,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i8> @intrinsic_vmadd_mask_vx_nxv4i8_i8_nxv4i8(<vscale x 4 x i8> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv4i8_i8_nxv4i8:
@@ -962,7 +962,7 @@ entry:
     i8 %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -994,7 +994,7 @@ declare <vscale x 8 x i8> @llvm.riscv.vmadd.mask.nxv8i8.i8(
   i8,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i8> @intrinsic_vmadd_mask_vx_nxv8i8_i8_nxv8i8(<vscale x 8 x i8> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv8i8_i8_nxv8i8:
@@ -1008,7 +1008,7 @@ entry:
     i8 %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -1040,7 +1040,7 @@ declare <vscale x 16 x i8> @llvm.riscv.vmadd.mask.nxv16i8.i8(
   i8,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i8> @intrinsic_vmadd_mask_vx_nxv16i8_i8_nxv16i8(<vscale x 16 x i8> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv16i8_i8_nxv16i8:
@@ -1054,7 +1054,7 @@ entry:
     i8 %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -1086,7 +1086,7 @@ declare <vscale x 32 x i8> @llvm.riscv.vmadd.mask.nxv32i8.i8(
   i8,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 32 x i8> @intrinsic_vmadd_mask_vx_nxv32i8_i8_nxv32i8(<vscale x 32 x i8> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv32i8_i8_nxv32i8:
@@ -1100,7 +1100,7 @@ entry:
     i8 %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -1132,7 +1132,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vmadd.mask.nxv1i16.i16(
   i16,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i16> @intrinsic_vmadd_mask_vx_nxv1i16_i16_nxv1i16(<vscale x 1 x i16> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv1i16_i16_nxv1i16:
@@ -1146,7 +1146,7 @@ entry:
     i16 %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -1178,7 +1178,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vmadd.mask.nxv2i16.i16(
   i16,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i16> @intrinsic_vmadd_mask_vx_nxv2i16_i16_nxv2i16(<vscale x 2 x i16> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv2i16_i16_nxv2i16:
@@ -1192,7 +1192,7 @@ entry:
     i16 %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -1224,7 +1224,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vmadd.mask.nxv4i16.i16(
   i16,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i16> @intrinsic_vmadd_mask_vx_nxv4i16_i16_nxv4i16(<vscale x 4 x i16> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv4i16_i16_nxv4i16:
@@ -1238,7 +1238,7 @@ entry:
     i16 %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -1270,7 +1270,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vmadd.mask.nxv8i16.i16(
   i16,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i16> @intrinsic_vmadd_mask_vx_nxv8i16_i16_nxv8i16(<vscale x 8 x i16> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv8i16_i16_nxv8i16:
@@ -1284,7 +1284,7 @@ entry:
     i16 %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -1316,7 +1316,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vmadd.mask.nxv16i16.i16(
   i16,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i16> @intrinsic_vmadd_mask_vx_nxv16i16_i16_nxv16i16(<vscale x 16 x i16> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv16i16_i16_nxv16i16:
@@ -1330,7 +1330,7 @@ entry:
     i16 %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -1362,7 +1362,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vmadd.mask.nxv1i32.i32(
   i32,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i32> @intrinsic_vmadd_mask_vx_nxv1i32_i32_nxv1i32(<vscale x 1 x i32> %0, i32 %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv1i32_i32_nxv1i32:
@@ -1376,7 +1376,7 @@ entry:
     i32 %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -1408,7 +1408,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vmadd.mask.nxv2i32.i32(
   i32,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i32> @intrinsic_vmadd_mask_vx_nxv2i32_i32_nxv2i32(<vscale x 2 x i32> %0, i32 %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv2i32_i32_nxv2i32:
@@ -1422,7 +1422,7 @@ entry:
     i32 %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -1454,7 +1454,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vmadd.mask.nxv4i32.i32(
   i32,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i32> @intrinsic_vmadd_mask_vx_nxv4i32_i32_nxv4i32(<vscale x 4 x i32> %0, i32 %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv4i32_i32_nxv4i32:
@@ -1468,7 +1468,7 @@ entry:
     i32 %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -1500,7 +1500,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vmadd.mask.nxv8i32.i32(
   i32,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i32> @intrinsic_vmadd_mask_vx_nxv8i32_i32_nxv8i32(<vscale x 8 x i32> %0, i32 %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv8i32_i32_nxv8i32:
@@ -1514,7 +1514,7 @@ entry:
     i32 %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -1546,7 +1546,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vmadd.mask.nxv1i64.i64(
   i64,
   <vscale x 1 x i64>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i64> @intrinsic_vmadd_mask_vx_nxv1i64_i64_nxv1i64(<vscale x 1 x i64> %0, i64 %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv1i64_i64_nxv1i64:
@@ -1560,7 +1560,7 @@ entry:
     i64 %1,
     <vscale x 1 x i64> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -1592,7 +1592,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vmadd.mask.nxv2i64.i64(
   i64,
   <vscale x 2 x i64>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i64> @intrinsic_vmadd_mask_vx_nxv2i64_i64_nxv2i64(<vscale x 2 x i64> %0, i64 %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv2i64_i64_nxv2i64:
@@ -1606,7 +1606,7 @@ entry:
     i64 %1,
     <vscale x 2 x i64> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -1638,7 +1638,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vmadd.mask.nxv4i64.i64(
   i64,
   <vscale x 4 x i64>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i64> @intrinsic_vmadd_mask_vx_nxv4i64_i64_nxv4i64(<vscale x 4 x i64> %0, i64 %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmadd_mask_vx_nxv4i64_i64_nxv4i64:
@@ -1652,7 +1652,7 @@ entry:
     i64 %1,
     <vscale x 4 x i64> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i64> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vnmsac-rv32.ll b/llvm/test/CodeGen/RISCV/rvv/vnmsac-rv32.ll
index 7c1c3e6fcc4f08..45a6d986a2abca 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vnmsac-rv32.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vnmsac-rv32.ll
@@ -28,7 +28,7 @@ declare <vscale x 1 x i8> @llvm.riscv.vnmsac.mask.nxv1i8.nxv1i8(
   <vscale x 1 x i8>,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i8>  @intrinsic_vnmsac_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv1i8_nxv1i8_nxv1i8:
@@ -42,7 +42,7 @@ entry:
     <vscale x 1 x i8> %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -74,7 +74,7 @@ declare <vscale x 2 x i8> @llvm.riscv.vnmsac.mask.nxv2i8.nxv2i8(
   <vscale x 2 x i8>,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i8>  @intrinsic_vnmsac_mask_vv_nxv2i8_nxv2i8_nxv2i8(<vscale x 2 x i8> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv2i8_nxv2i8_nxv2i8:
@@ -88,7 +88,7 @@ entry:
     <vscale x 2 x i8> %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -120,7 +120,7 @@ declare <vscale x 4 x i8> @llvm.riscv.vnmsac.mask.nxv4i8.nxv4i8(
   <vscale x 4 x i8>,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i8>  @intrinsic_vnmsac_mask_vv_nxv4i8_nxv4i8_nxv4i8(<vscale x 4 x i8> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv4i8_nxv4i8_nxv4i8:
@@ -134,7 +134,7 @@ entry:
     <vscale x 4 x i8> %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -166,7 +166,7 @@ declare <vscale x 8 x i8> @llvm.riscv.vnmsac.mask.nxv8i8.nxv8i8(
   <vscale x 8 x i8>,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i8>  @intrinsic_vnmsac_mask_vv_nxv8i8_nxv8i8_nxv8i8(<vscale x 8 x i8> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv8i8_nxv8i8_nxv8i8:
@@ -180,7 +180,7 @@ entry:
     <vscale x 8 x i8> %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -212,7 +212,7 @@ declare <vscale x 16 x i8> @llvm.riscv.vnmsac.mask.nxv16i8.nxv16i8(
   <vscale x 16 x i8>,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i8>  @intrinsic_vnmsac_mask_vv_nxv16i8_nxv16i8_nxv16i8(<vscale x 16 x i8> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv16i8_nxv16i8_nxv16i8:
@@ -226,7 +226,7 @@ entry:
     <vscale x 16 x i8> %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -258,7 +258,7 @@ declare <vscale x 32 x i8> @llvm.riscv.vnmsac.mask.nxv32i8.nxv32i8(
   <vscale x 32 x i8>,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 32 x i8>  @intrinsic_vnmsac_mask_vv_nxv32i8_nxv32i8_nxv32i8(<vscale x 32 x i8> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv32i8_nxv32i8_nxv32i8:
@@ -272,7 +272,7 @@ entry:
     <vscale x 32 x i8> %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -304,7 +304,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vnmsac.mask.nxv1i16.nxv1i16(
   <vscale x 1 x i16>,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i16>  @intrinsic_vnmsac_mask_vv_nxv1i16_nxv1i16_nxv1i16(<vscale x 1 x i16> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv1i16_nxv1i16_nxv1i16:
@@ -318,7 +318,7 @@ entry:
     <vscale x 1 x i16> %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -350,7 +350,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vnmsac.mask.nxv2i16.nxv2i16(
   <vscale x 2 x i16>,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i16>  @intrinsic_vnmsac_mask_vv_nxv2i16_nxv2i16_nxv2i16(<vscale x 2 x i16> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv2i16_nxv2i16_nxv2i16:
@@ -364,7 +364,7 @@ entry:
     <vscale x 2 x i16> %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -396,7 +396,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vnmsac.mask.nxv4i16.nxv4i16(
   <vscale x 4 x i16>,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i16>  @intrinsic_vnmsac_mask_vv_nxv4i16_nxv4i16_nxv4i16(<vscale x 4 x i16> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv4i16_nxv4i16_nxv4i16:
@@ -410,7 +410,7 @@ entry:
     <vscale x 4 x i16> %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -442,7 +442,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vnmsac.mask.nxv8i16.nxv8i16(
   <vscale x 8 x i16>,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i16>  @intrinsic_vnmsac_mask_vv_nxv8i16_nxv8i16_nxv8i16(<vscale x 8 x i16> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv8i16_nxv8i16_nxv8i16:
@@ -456,7 +456,7 @@ entry:
     <vscale x 8 x i16> %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -488,7 +488,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vnmsac.mask.nxv16i16.nxv16i16(
   <vscale x 16 x i16>,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i16>  @intrinsic_vnmsac_mask_vv_nxv16i16_nxv16i16_nxv16i16(<vscale x 16 x i16> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv16i16_nxv16i16_nxv16i16:
@@ -502,7 +502,7 @@ entry:
     <vscale x 16 x i16> %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -534,7 +534,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vnmsac.mask.nxv1i32.nxv1i32(
   <vscale x 1 x i32>,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i32>  @intrinsic_vnmsac_mask_vv_nxv1i32_nxv1i32_nxv1i32(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv1i32_nxv1i32_nxv1i32:
@@ -548,7 +548,7 @@ entry:
     <vscale x 1 x i32> %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -580,7 +580,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vnmsac.mask.nxv2i32.nxv2i32(
   <vscale x 2 x i32>,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i32>  @intrinsic_vnmsac_mask_vv_nxv2i32_nxv2i32_nxv2i32(<vscale x 2 x i32> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv2i32_nxv2i32_nxv2i32:
@@ -594,7 +594,7 @@ entry:
     <vscale x 2 x i32> %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -626,7 +626,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vnmsac.mask.nxv4i32.nxv4i32(
   <vscale x 4 x i32>,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i32>  @intrinsic_vnmsac_mask_vv_nxv4i32_nxv4i32_nxv4i32(<vscale x 4 x i32> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv4i32_nxv4i32_nxv4i32:
@@ -640,7 +640,7 @@ entry:
     <vscale x 4 x i32> %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -672,7 +672,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vnmsac.mask.nxv8i32.nxv8i32(
   <vscale x 8 x i32>,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i32>  @intrinsic_vnmsac_mask_vv_nxv8i32_nxv8i32_nxv8i32(<vscale x 8 x i32> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv8i32_nxv8i32_nxv8i32:
@@ -686,7 +686,7 @@ entry:
     <vscale x 8 x i32> %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -718,7 +718,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vnmsac.mask.nxv1i64.nxv1i64(
   <vscale x 1 x i64>,
   <vscale x 1 x i64>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i64>  @intrinsic_vnmsac_mask_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv1i64_nxv1i64_nxv1i64:
@@ -732,7 +732,7 @@ entry:
     <vscale x 1 x i64> %1,
     <vscale x 1 x i64> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -764,7 +764,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vnmsac.mask.nxv2i64.nxv2i64(
   <vscale x 2 x i64>,
   <vscale x 2 x i64>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i64>  @intrinsic_vnmsac_mask_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv2i64_nxv2i64_nxv2i64:
@@ -778,7 +778,7 @@ entry:
     <vscale x 2 x i64> %1,
     <vscale x 2 x i64> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -810,7 +810,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vnmsac.mask.nxv4i64.nxv4i64(
   <vscale x 4 x i64>,
   <vscale x 4 x i64>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i64>  @intrinsic_vnmsac_mask_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv4i64_nxv4i64_nxv4i64:
@@ -824,7 +824,7 @@ entry:
     <vscale x 4 x i64> %1,
     <vscale x 4 x i64> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -856,7 +856,7 @@ declare <vscale x 1 x i8> @llvm.riscv.vnmsac.mask.nxv1i8.i8(
   i8,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i8> @intrinsic_vnmsac_mask_vx_nxv1i8_i8_nxv1i8(<vscale x 1 x i8> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv1i8_i8_nxv1i8:
@@ -870,7 +870,7 @@ entry:
     i8 %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -902,7 +902,7 @@ declare <vscale x 2 x i8> @llvm.riscv.vnmsac.mask.nxv2i8.i8(
   i8,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i8> @intrinsic_vnmsac_mask_vx_nxv2i8_i8_nxv2i8(<vscale x 2 x i8> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv2i8_i8_nxv2i8:
@@ -916,7 +916,7 @@ entry:
     i8 %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -948,7 +948,7 @@ declare <vscale x 4 x i8> @llvm.riscv.vnmsac.mask.nxv4i8.i8(
   i8,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i8> @intrinsic_vnmsac_mask_vx_nxv4i8_i8_nxv4i8(<vscale x 4 x i8> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv4i8_i8_nxv4i8:
@@ -962,7 +962,7 @@ entry:
     i8 %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -994,7 +994,7 @@ declare <vscale x 8 x i8> @llvm.riscv.vnmsac.mask.nxv8i8.i8(
   i8,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i8> @intrinsic_vnmsac_mask_vx_nxv8i8_i8_nxv8i8(<vscale x 8 x i8> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv8i8_i8_nxv8i8:
@@ -1008,7 +1008,7 @@ entry:
     i8 %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -1040,7 +1040,7 @@ declare <vscale x 16 x i8> @llvm.riscv.vnmsac.mask.nxv16i8.i8(
   i8,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i8> @intrinsic_vnmsac_mask_vx_nxv16i8_i8_nxv16i8(<vscale x 16 x i8> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv16i8_i8_nxv16i8:
@@ -1054,7 +1054,7 @@ entry:
     i8 %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -1086,7 +1086,7 @@ declare <vscale x 32 x i8> @llvm.riscv.vnmsac.mask.nxv32i8.i8(
   i8,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 32 x i8> @intrinsic_vnmsac_mask_vx_nxv32i8_i8_nxv32i8(<vscale x 32 x i8> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv32i8_i8_nxv32i8:
@@ -1100,7 +1100,7 @@ entry:
     i8 %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -1132,7 +1132,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vnmsac.mask.nxv1i16.i16(
   i16,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i16> @intrinsic_vnmsac_mask_vx_nxv1i16_i16_nxv1i16(<vscale x 1 x i16> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv1i16_i16_nxv1i16:
@@ -1146,7 +1146,7 @@ entry:
     i16 %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -1178,7 +1178,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vnmsac.mask.nxv2i16.i16(
   i16,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i16> @intrinsic_vnmsac_mask_vx_nxv2i16_i16_nxv2i16(<vscale x 2 x i16> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv2i16_i16_nxv2i16:
@@ -1192,7 +1192,7 @@ entry:
     i16 %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -1224,7 +1224,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vnmsac.mask.nxv4i16.i16(
   i16,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i16> @intrinsic_vnmsac_mask_vx_nxv4i16_i16_nxv4i16(<vscale x 4 x i16> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv4i16_i16_nxv4i16:
@@ -1238,7 +1238,7 @@ entry:
     i16 %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -1270,7 +1270,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vnmsac.mask.nxv8i16.i16(
   i16,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i16> @intrinsic_vnmsac_mask_vx_nxv8i16_i16_nxv8i16(<vscale x 8 x i16> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv8i16_i16_nxv8i16:
@@ -1284,7 +1284,7 @@ entry:
     i16 %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -1316,7 +1316,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vnmsac.mask.nxv16i16.i16(
   i16,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i16> @intrinsic_vnmsac_mask_vx_nxv16i16_i16_nxv16i16(<vscale x 16 x i16> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv16i16_i16_nxv16i16:
@@ -1330,7 +1330,7 @@ entry:
     i16 %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -1362,7 +1362,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vnmsac.mask.nxv1i32.i32(
   i32,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i32> @intrinsic_vnmsac_mask_vx_nxv1i32_i32_nxv1i32(<vscale x 1 x i32> %0, i32 %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv1i32_i32_nxv1i32:
@@ -1376,7 +1376,7 @@ entry:
     i32 %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -1408,7 +1408,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vnmsac.mask.nxv2i32.i32(
   i32,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i32> @intrinsic_vnmsac_mask_vx_nxv2i32_i32_nxv2i32(<vscale x 2 x i32> %0, i32 %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv2i32_i32_nxv2i32:
@@ -1422,7 +1422,7 @@ entry:
     i32 %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -1454,7 +1454,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vnmsac.mask.nxv4i32.i32(
   i32,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i32> @intrinsic_vnmsac_mask_vx_nxv4i32_i32_nxv4i32(<vscale x 4 x i32> %0, i32 %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv4i32_i32_nxv4i32:
@@ -1468,7 +1468,7 @@ entry:
     i32 %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -1500,7 +1500,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vnmsac.mask.nxv8i32.i32(
   i32,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i32> @intrinsic_vnmsac_mask_vx_nxv8i32_i32_nxv8i32(<vscale x 8 x i32> %0, i32 %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv8i32_i32_nxv8i32:
@@ -1514,7 +1514,7 @@ entry:
     i32 %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -1553,7 +1553,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vnmsac.mask.nxv1i64.i64(
   i64,
   <vscale x 1 x i64>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i64> @intrinsic_vnmsac_mask_vx_nxv1i64_i64_nxv1i64(<vscale x 1 x i64> %0, i64 %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv1i64_i64_nxv1i64:
@@ -1574,7 +1574,7 @@ entry:
     i64 %1,
     <vscale x 1 x i64> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -1613,7 +1613,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vnmsac.mask.nxv2i64.i64(
   i64,
   <vscale x 2 x i64>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i64> @intrinsic_vnmsac_mask_vx_nxv2i64_i64_nxv2i64(<vscale x 2 x i64> %0, i64 %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv2i64_i64_nxv2i64:
@@ -1634,7 +1634,7 @@ entry:
     i64 %1,
     <vscale x 2 x i64> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -1673,7 +1673,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vnmsac.mask.nxv4i64.i64(
   i64,
   <vscale x 4 x i64>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i64> @intrinsic_vnmsac_mask_vx_nxv4i64_i64_nxv4i64(<vscale x 4 x i64> %0, i64 %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv4i64_i64_nxv4i64:
@@ -1694,7 +1694,7 @@ entry:
     i64 %1,
     <vscale x 4 x i64> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i64> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vnmsac-rv64.ll b/llvm/test/CodeGen/RISCV/rvv/vnmsac-rv64.ll
index f9ba1e3016eabb..a7173a53263add 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vnmsac-rv64.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vnmsac-rv64.ll
@@ -28,7 +28,7 @@ declare <vscale x 1 x i8> @llvm.riscv.vnmsac.mask.nxv1i8.nxv1i8(
   <vscale x 1 x i8>,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i8>  @intrinsic_vnmsac_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv1i8_nxv1i8_nxv1i8:
@@ -42,7 +42,7 @@ entry:
     <vscale x 1 x i8> %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -74,7 +74,7 @@ declare <vscale x 2 x i8> @llvm.riscv.vnmsac.mask.nxv2i8.nxv2i8(
   <vscale x 2 x i8>,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i8>  @intrinsic_vnmsac_mask_vv_nxv2i8_nxv2i8_nxv2i8(<vscale x 2 x i8> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv2i8_nxv2i8_nxv2i8:
@@ -88,7 +88,7 @@ entry:
     <vscale x 2 x i8> %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -120,7 +120,7 @@ declare <vscale x 4 x i8> @llvm.riscv.vnmsac.mask.nxv4i8.nxv4i8(
   <vscale x 4 x i8>,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i8>  @intrinsic_vnmsac_mask_vv_nxv4i8_nxv4i8_nxv4i8(<vscale x 4 x i8> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv4i8_nxv4i8_nxv4i8:
@@ -134,7 +134,7 @@ entry:
     <vscale x 4 x i8> %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -166,7 +166,7 @@ declare <vscale x 8 x i8> @llvm.riscv.vnmsac.mask.nxv8i8.nxv8i8(
   <vscale x 8 x i8>,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i8>  @intrinsic_vnmsac_mask_vv_nxv8i8_nxv8i8_nxv8i8(<vscale x 8 x i8> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv8i8_nxv8i8_nxv8i8:
@@ -180,7 +180,7 @@ entry:
     <vscale x 8 x i8> %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -212,7 +212,7 @@ declare <vscale x 16 x i8> @llvm.riscv.vnmsac.mask.nxv16i8.nxv16i8(
   <vscale x 16 x i8>,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i8>  @intrinsic_vnmsac_mask_vv_nxv16i8_nxv16i8_nxv16i8(<vscale x 16 x i8> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv16i8_nxv16i8_nxv16i8:
@@ -226,7 +226,7 @@ entry:
     <vscale x 16 x i8> %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -258,7 +258,7 @@ declare <vscale x 32 x i8> @llvm.riscv.vnmsac.mask.nxv32i8.nxv32i8(
   <vscale x 32 x i8>,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 32 x i8>  @intrinsic_vnmsac_mask_vv_nxv32i8_nxv32i8_nxv32i8(<vscale x 32 x i8> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv32i8_nxv32i8_nxv32i8:
@@ -272,7 +272,7 @@ entry:
     <vscale x 32 x i8> %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -304,7 +304,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vnmsac.mask.nxv1i16.nxv1i16(
   <vscale x 1 x i16>,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i16>  @intrinsic_vnmsac_mask_vv_nxv1i16_nxv1i16_nxv1i16(<vscale x 1 x i16> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv1i16_nxv1i16_nxv1i16:
@@ -318,7 +318,7 @@ entry:
     <vscale x 1 x i16> %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -350,7 +350,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vnmsac.mask.nxv2i16.nxv2i16(
   <vscale x 2 x i16>,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i16>  @intrinsic_vnmsac_mask_vv_nxv2i16_nxv2i16_nxv2i16(<vscale x 2 x i16> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv2i16_nxv2i16_nxv2i16:
@@ -364,7 +364,7 @@ entry:
     <vscale x 2 x i16> %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -396,7 +396,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vnmsac.mask.nxv4i16.nxv4i16(
   <vscale x 4 x i16>,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i16>  @intrinsic_vnmsac_mask_vv_nxv4i16_nxv4i16_nxv4i16(<vscale x 4 x i16> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv4i16_nxv4i16_nxv4i16:
@@ -410,7 +410,7 @@ entry:
     <vscale x 4 x i16> %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -442,7 +442,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vnmsac.mask.nxv8i16.nxv8i16(
   <vscale x 8 x i16>,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i16>  @intrinsic_vnmsac_mask_vv_nxv8i16_nxv8i16_nxv8i16(<vscale x 8 x i16> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv8i16_nxv8i16_nxv8i16:
@@ -456,7 +456,7 @@ entry:
     <vscale x 8 x i16> %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -488,7 +488,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vnmsac.mask.nxv16i16.nxv16i16(
   <vscale x 16 x i16>,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i16>  @intrinsic_vnmsac_mask_vv_nxv16i16_nxv16i16_nxv16i16(<vscale x 16 x i16> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv16i16_nxv16i16_nxv16i16:
@@ -502,7 +502,7 @@ entry:
     <vscale x 16 x i16> %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -534,7 +534,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vnmsac.mask.nxv1i32.nxv1i32(
   <vscale x 1 x i32>,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i32>  @intrinsic_vnmsac_mask_vv_nxv1i32_nxv1i32_nxv1i32(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv1i32_nxv1i32_nxv1i32:
@@ -548,7 +548,7 @@ entry:
     <vscale x 1 x i32> %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -580,7 +580,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vnmsac.mask.nxv2i32.nxv2i32(
   <vscale x 2 x i32>,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i32>  @intrinsic_vnmsac_mask_vv_nxv2i32_nxv2i32_nxv2i32(<vscale x 2 x i32> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv2i32_nxv2i32_nxv2i32:
@@ -594,7 +594,7 @@ entry:
     <vscale x 2 x i32> %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -626,7 +626,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vnmsac.mask.nxv4i32.nxv4i32(
   <vscale x 4 x i32>,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i32>  @intrinsic_vnmsac_mask_vv_nxv4i32_nxv4i32_nxv4i32(<vscale x 4 x i32> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv4i32_nxv4i32_nxv4i32:
@@ -640,7 +640,7 @@ entry:
     <vscale x 4 x i32> %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -672,7 +672,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vnmsac.mask.nxv8i32.nxv8i32(
   <vscale x 8 x i32>,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i32>  @intrinsic_vnmsac_mask_vv_nxv8i32_nxv8i32_nxv8i32(<vscale x 8 x i32> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv8i32_nxv8i32_nxv8i32:
@@ -686,7 +686,7 @@ entry:
     <vscale x 8 x i32> %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -718,7 +718,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vnmsac.mask.nxv1i64.nxv1i64(
   <vscale x 1 x i64>,
   <vscale x 1 x i64>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i64>  @intrinsic_vnmsac_mask_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv1i64_nxv1i64_nxv1i64:
@@ -732,7 +732,7 @@ entry:
     <vscale x 1 x i64> %1,
     <vscale x 1 x i64> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -764,7 +764,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vnmsac.mask.nxv2i64.nxv2i64(
   <vscale x 2 x i64>,
   <vscale x 2 x i64>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i64>  @intrinsic_vnmsac_mask_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv2i64_nxv2i64_nxv2i64:
@@ -778,7 +778,7 @@ entry:
     <vscale x 2 x i64> %1,
     <vscale x 2 x i64> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -810,7 +810,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vnmsac.mask.nxv4i64.nxv4i64(
   <vscale x 4 x i64>,
   <vscale x 4 x i64>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i64>  @intrinsic_vnmsac_mask_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vv_nxv4i64_nxv4i64_nxv4i64:
@@ -824,7 +824,7 @@ entry:
     <vscale x 4 x i64> %1,
     <vscale x 4 x i64> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -856,7 +856,7 @@ declare <vscale x 1 x i8> @llvm.riscv.vnmsac.mask.nxv1i8.i8(
   i8,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i8> @intrinsic_vnmsac_mask_vx_nxv1i8_i8_nxv1i8(<vscale x 1 x i8> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv1i8_i8_nxv1i8:
@@ -870,7 +870,7 @@ entry:
     i8 %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -902,7 +902,7 @@ declare <vscale x 2 x i8> @llvm.riscv.vnmsac.mask.nxv2i8.i8(
   i8,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i8> @intrinsic_vnmsac_mask_vx_nxv2i8_i8_nxv2i8(<vscale x 2 x i8> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv2i8_i8_nxv2i8:
@@ -916,7 +916,7 @@ entry:
     i8 %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -948,7 +948,7 @@ declare <vscale x 4 x i8> @llvm.riscv.vnmsac.mask.nxv4i8.i8(
   i8,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i8> @intrinsic_vnmsac_mask_vx_nxv4i8_i8_nxv4i8(<vscale x 4 x i8> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv4i8_i8_nxv4i8:
@@ -962,7 +962,7 @@ entry:
     i8 %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -994,7 +994,7 @@ declare <vscale x 8 x i8> @llvm.riscv.vnmsac.mask.nxv8i8.i8(
   i8,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i8> @intrinsic_vnmsac_mask_vx_nxv8i8_i8_nxv8i8(<vscale x 8 x i8> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv8i8_i8_nxv8i8:
@@ -1008,7 +1008,7 @@ entry:
     i8 %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -1040,7 +1040,7 @@ declare <vscale x 16 x i8> @llvm.riscv.vnmsac.mask.nxv16i8.i8(
   i8,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i8> @intrinsic_vnmsac_mask_vx_nxv16i8_i8_nxv16i8(<vscale x 16 x i8> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv16i8_i8_nxv16i8:
@@ -1054,7 +1054,7 @@ entry:
     i8 %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -1086,7 +1086,7 @@ declare <vscale x 32 x i8> @llvm.riscv.vnmsac.mask.nxv32i8.i8(
   i8,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 32 x i8> @intrinsic_vnmsac_mask_vx_nxv32i8_i8_nxv32i8(<vscale x 32 x i8> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv32i8_i8_nxv32i8:
@@ -1100,7 +1100,7 @@ entry:
     i8 %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -1132,7 +1132,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vnmsac.mask.nxv1i16.i16(
   i16,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i16> @intrinsic_vnmsac_mask_vx_nxv1i16_i16_nxv1i16(<vscale x 1 x i16> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv1i16_i16_nxv1i16:
@@ -1146,7 +1146,7 @@ entry:
     i16 %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -1178,7 +1178,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vnmsac.mask.nxv2i16.i16(
   i16,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i16> @intrinsic_vnmsac_mask_vx_nxv2i16_i16_nxv2i16(<vscale x 2 x i16> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv2i16_i16_nxv2i16:
@@ -1192,7 +1192,7 @@ entry:
     i16 %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -1224,7 +1224,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vnmsac.mask.nxv4i16.i16(
   i16,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i16> @intrinsic_vnmsac_mask_vx_nxv4i16_i16_nxv4i16(<vscale x 4 x i16> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv4i16_i16_nxv4i16:
@@ -1238,7 +1238,7 @@ entry:
     i16 %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -1270,7 +1270,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vnmsac.mask.nxv8i16.i16(
   i16,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i16> @intrinsic_vnmsac_mask_vx_nxv8i16_i16_nxv8i16(<vscale x 8 x i16> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv8i16_i16_nxv8i16:
@@ -1284,7 +1284,7 @@ entry:
     i16 %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -1316,7 +1316,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vnmsac.mask.nxv16i16.i16(
   i16,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i16> @intrinsic_vnmsac_mask_vx_nxv16i16_i16_nxv16i16(<vscale x 16 x i16> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv16i16_i16_nxv16i16:
@@ -1330,7 +1330,7 @@ entry:
     i16 %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -1362,7 +1362,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vnmsac.mask.nxv1i32.i32(
   i32,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i32> @intrinsic_vnmsac_mask_vx_nxv1i32_i32_nxv1i32(<vscale x 1 x i32> %0, i32 %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv1i32_i32_nxv1i32:
@@ -1376,7 +1376,7 @@ entry:
     i32 %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -1408,7 +1408,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vnmsac.mask.nxv2i32.i32(
   i32,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i32> @intrinsic_vnmsac_mask_vx_nxv2i32_i32_nxv2i32(<vscale x 2 x i32> %0, i32 %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv2i32_i32_nxv2i32:
@@ -1422,7 +1422,7 @@ entry:
     i32 %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -1454,7 +1454,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vnmsac.mask.nxv4i32.i32(
   i32,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i32> @intrinsic_vnmsac_mask_vx_nxv4i32_i32_nxv4i32(<vscale x 4 x i32> %0, i32 %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv4i32_i32_nxv4i32:
@@ -1468,7 +1468,7 @@ entry:
     i32 %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -1500,7 +1500,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vnmsac.mask.nxv8i32.i32(
   i32,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i32> @intrinsic_vnmsac_mask_vx_nxv8i32_i32_nxv8i32(<vscale x 8 x i32> %0, i32 %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv8i32_i32_nxv8i32:
@@ -1514,7 +1514,7 @@ entry:
     i32 %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -1546,7 +1546,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vnmsac.mask.nxv1i64.i64(
   i64,
   <vscale x 1 x i64>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i64> @intrinsic_vnmsac_mask_vx_nxv1i64_i64_nxv1i64(<vscale x 1 x i64> %0, i64 %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv1i64_i64_nxv1i64:
@@ -1560,7 +1560,7 @@ entry:
     i64 %1,
     <vscale x 1 x i64> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -1592,7 +1592,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vnmsac.mask.nxv2i64.i64(
   i64,
   <vscale x 2 x i64>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i64> @intrinsic_vnmsac_mask_vx_nxv2i64_i64_nxv2i64(<vscale x 2 x i64> %0, i64 %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv2i64_i64_nxv2i64:
@@ -1606,7 +1606,7 @@ entry:
     i64 %1,
     <vscale x 2 x i64> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -1638,7 +1638,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vnmsac.mask.nxv4i64.i64(
   i64,
   <vscale x 4 x i64>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i64> @intrinsic_vnmsac_mask_vx_nxv4i64_i64_nxv4i64(<vscale x 4 x i64> %0, i64 %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsac_mask_vx_nxv4i64_i64_nxv4i64:
@@ -1652,7 +1652,7 @@ entry:
     i64 %1,
     <vscale x 4 x i64> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i64> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vnmsub-rv32.ll b/llvm/test/CodeGen/RISCV/rvv/vnmsub-rv32.ll
index 3cf95ae3132572..8440b0610872d8 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vnmsub-rv32.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vnmsub-rv32.ll
@@ -28,7 +28,7 @@ declare <vscale x 1 x i8> @llvm.riscv.vnmsub.mask.nxv1i8.nxv1i8(
   <vscale x 1 x i8>,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i8>  @intrinsic_vnmsub_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv1i8_nxv1i8_nxv1i8:
@@ -42,7 +42,7 @@ entry:
     <vscale x 1 x i8> %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -74,7 +74,7 @@ declare <vscale x 2 x i8> @llvm.riscv.vnmsub.mask.nxv2i8.nxv2i8(
   <vscale x 2 x i8>,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i8>  @intrinsic_vnmsub_mask_vv_nxv2i8_nxv2i8_nxv2i8(<vscale x 2 x i8> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv2i8_nxv2i8_nxv2i8:
@@ -88,7 +88,7 @@ entry:
     <vscale x 2 x i8> %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -120,7 +120,7 @@ declare <vscale x 4 x i8> @llvm.riscv.vnmsub.mask.nxv4i8.nxv4i8(
   <vscale x 4 x i8>,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i8>  @intrinsic_vnmsub_mask_vv_nxv4i8_nxv4i8_nxv4i8(<vscale x 4 x i8> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv4i8_nxv4i8_nxv4i8:
@@ -134,7 +134,7 @@ entry:
     <vscale x 4 x i8> %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -166,7 +166,7 @@ declare <vscale x 8 x i8> @llvm.riscv.vnmsub.mask.nxv8i8.nxv8i8(
   <vscale x 8 x i8>,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i8>  @intrinsic_vnmsub_mask_vv_nxv8i8_nxv8i8_nxv8i8(<vscale x 8 x i8> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv8i8_nxv8i8_nxv8i8:
@@ -180,7 +180,7 @@ entry:
     <vscale x 8 x i8> %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -212,7 +212,7 @@ declare <vscale x 16 x i8> @llvm.riscv.vnmsub.mask.nxv16i8.nxv16i8(
   <vscale x 16 x i8>,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i8>  @intrinsic_vnmsub_mask_vv_nxv16i8_nxv16i8_nxv16i8(<vscale x 16 x i8> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv16i8_nxv16i8_nxv16i8:
@@ -226,7 +226,7 @@ entry:
     <vscale x 16 x i8> %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -258,7 +258,7 @@ declare <vscale x 32 x i8> @llvm.riscv.vnmsub.mask.nxv32i8.nxv32i8(
   <vscale x 32 x i8>,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 32 x i8>  @intrinsic_vnmsub_mask_vv_nxv32i8_nxv32i8_nxv32i8(<vscale x 32 x i8> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv32i8_nxv32i8_nxv32i8:
@@ -272,7 +272,7 @@ entry:
     <vscale x 32 x i8> %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -304,7 +304,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vnmsub.mask.nxv1i16.nxv1i16(
   <vscale x 1 x i16>,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i16>  @intrinsic_vnmsub_mask_vv_nxv1i16_nxv1i16_nxv1i16(<vscale x 1 x i16> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv1i16_nxv1i16_nxv1i16:
@@ -318,7 +318,7 @@ entry:
     <vscale x 1 x i16> %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -350,7 +350,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vnmsub.mask.nxv2i16.nxv2i16(
   <vscale x 2 x i16>,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i16>  @intrinsic_vnmsub_mask_vv_nxv2i16_nxv2i16_nxv2i16(<vscale x 2 x i16> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv2i16_nxv2i16_nxv2i16:
@@ -364,7 +364,7 @@ entry:
     <vscale x 2 x i16> %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -396,7 +396,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vnmsub.mask.nxv4i16.nxv4i16(
   <vscale x 4 x i16>,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i16>  @intrinsic_vnmsub_mask_vv_nxv4i16_nxv4i16_nxv4i16(<vscale x 4 x i16> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv4i16_nxv4i16_nxv4i16:
@@ -410,7 +410,7 @@ entry:
     <vscale x 4 x i16> %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -442,7 +442,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vnmsub.mask.nxv8i16.nxv8i16(
   <vscale x 8 x i16>,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i16>  @intrinsic_vnmsub_mask_vv_nxv8i16_nxv8i16_nxv8i16(<vscale x 8 x i16> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv8i16_nxv8i16_nxv8i16:
@@ -456,7 +456,7 @@ entry:
     <vscale x 8 x i16> %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -488,7 +488,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vnmsub.mask.nxv16i16.nxv16i16(
   <vscale x 16 x i16>,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i16>  @intrinsic_vnmsub_mask_vv_nxv16i16_nxv16i16_nxv16i16(<vscale x 16 x i16> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv16i16_nxv16i16_nxv16i16:
@@ -502,7 +502,7 @@ entry:
     <vscale x 16 x i16> %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -534,7 +534,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vnmsub.mask.nxv1i32.nxv1i32(
   <vscale x 1 x i32>,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i32>  @intrinsic_vnmsub_mask_vv_nxv1i32_nxv1i32_nxv1i32(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv1i32_nxv1i32_nxv1i32:
@@ -548,7 +548,7 @@ entry:
     <vscale x 1 x i32> %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -580,7 +580,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vnmsub.mask.nxv2i32.nxv2i32(
   <vscale x 2 x i32>,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i32>  @intrinsic_vnmsub_mask_vv_nxv2i32_nxv2i32_nxv2i32(<vscale x 2 x i32> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv2i32_nxv2i32_nxv2i32:
@@ -594,7 +594,7 @@ entry:
     <vscale x 2 x i32> %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -626,7 +626,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vnmsub.mask.nxv4i32.nxv4i32(
   <vscale x 4 x i32>,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i32>  @intrinsic_vnmsub_mask_vv_nxv4i32_nxv4i32_nxv4i32(<vscale x 4 x i32> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv4i32_nxv4i32_nxv4i32:
@@ -640,7 +640,7 @@ entry:
     <vscale x 4 x i32> %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -672,7 +672,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vnmsub.mask.nxv8i32.nxv8i32(
   <vscale x 8 x i32>,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i32>  @intrinsic_vnmsub_mask_vv_nxv8i32_nxv8i32_nxv8i32(<vscale x 8 x i32> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv8i32_nxv8i32_nxv8i32:
@@ -686,7 +686,7 @@ entry:
     <vscale x 8 x i32> %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -718,7 +718,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vnmsub.mask.nxv1i64.nxv1i64(
   <vscale x 1 x i64>,
   <vscale x 1 x i64>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i64>  @intrinsic_vnmsub_mask_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv1i64_nxv1i64_nxv1i64:
@@ -732,7 +732,7 @@ entry:
     <vscale x 1 x i64> %1,
     <vscale x 1 x i64> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -764,7 +764,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vnmsub.mask.nxv2i64.nxv2i64(
   <vscale x 2 x i64>,
   <vscale x 2 x i64>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i64>  @intrinsic_vnmsub_mask_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv2i64_nxv2i64_nxv2i64:
@@ -778,7 +778,7 @@ entry:
     <vscale x 2 x i64> %1,
     <vscale x 2 x i64> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -810,7 +810,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vnmsub.mask.nxv4i64.nxv4i64(
   <vscale x 4 x i64>,
   <vscale x 4 x i64>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i64>  @intrinsic_vnmsub_mask_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv4i64_nxv4i64_nxv4i64:
@@ -824,7 +824,7 @@ entry:
     <vscale x 4 x i64> %1,
     <vscale x 4 x i64> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -856,7 +856,7 @@ declare <vscale x 1 x i8> @llvm.riscv.vnmsub.mask.nxv1i8.i8(
   i8,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i8> @intrinsic_vnmsub_mask_vx_nxv1i8_i8_nxv1i8(<vscale x 1 x i8> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv1i8_i8_nxv1i8:
@@ -870,7 +870,7 @@ entry:
     i8 %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -902,7 +902,7 @@ declare <vscale x 2 x i8> @llvm.riscv.vnmsub.mask.nxv2i8.i8(
   i8,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i8> @intrinsic_vnmsub_mask_vx_nxv2i8_i8_nxv2i8(<vscale x 2 x i8> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv2i8_i8_nxv2i8:
@@ -916,7 +916,7 @@ entry:
     i8 %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -948,7 +948,7 @@ declare <vscale x 4 x i8> @llvm.riscv.vnmsub.mask.nxv4i8.i8(
   i8,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i8> @intrinsic_vnmsub_mask_vx_nxv4i8_i8_nxv4i8(<vscale x 4 x i8> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv4i8_i8_nxv4i8:
@@ -962,7 +962,7 @@ entry:
     i8 %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -994,7 +994,7 @@ declare <vscale x 8 x i8> @llvm.riscv.vnmsub.mask.nxv8i8.i8(
   i8,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i8> @intrinsic_vnmsub_mask_vx_nxv8i8_i8_nxv8i8(<vscale x 8 x i8> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv8i8_i8_nxv8i8:
@@ -1008,7 +1008,7 @@ entry:
     i8 %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -1040,7 +1040,7 @@ declare <vscale x 16 x i8> @llvm.riscv.vnmsub.mask.nxv16i8.i8(
   i8,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i8> @intrinsic_vnmsub_mask_vx_nxv16i8_i8_nxv16i8(<vscale x 16 x i8> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv16i8_i8_nxv16i8:
@@ -1054,7 +1054,7 @@ entry:
     i8 %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -1086,7 +1086,7 @@ declare <vscale x 32 x i8> @llvm.riscv.vnmsub.mask.nxv32i8.i8(
   i8,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 32 x i8> @intrinsic_vnmsub_mask_vx_nxv32i8_i8_nxv32i8(<vscale x 32 x i8> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv32i8_i8_nxv32i8:
@@ -1100,7 +1100,7 @@ entry:
     i8 %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -1132,7 +1132,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vnmsub.mask.nxv1i16.i16(
   i16,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i16> @intrinsic_vnmsub_mask_vx_nxv1i16_i16_nxv1i16(<vscale x 1 x i16> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv1i16_i16_nxv1i16:
@@ -1146,7 +1146,7 @@ entry:
     i16 %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -1178,7 +1178,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vnmsub.mask.nxv2i16.i16(
   i16,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i16> @intrinsic_vnmsub_mask_vx_nxv2i16_i16_nxv2i16(<vscale x 2 x i16> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv2i16_i16_nxv2i16:
@@ -1192,7 +1192,7 @@ entry:
     i16 %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -1224,7 +1224,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vnmsub.mask.nxv4i16.i16(
   i16,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i16> @intrinsic_vnmsub_mask_vx_nxv4i16_i16_nxv4i16(<vscale x 4 x i16> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv4i16_i16_nxv4i16:
@@ -1238,7 +1238,7 @@ entry:
     i16 %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -1270,7 +1270,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vnmsub.mask.nxv8i16.i16(
   i16,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i16> @intrinsic_vnmsub_mask_vx_nxv8i16_i16_nxv8i16(<vscale x 8 x i16> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv8i16_i16_nxv8i16:
@@ -1284,7 +1284,7 @@ entry:
     i16 %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -1316,7 +1316,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vnmsub.mask.nxv16i16.i16(
   i16,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i16> @intrinsic_vnmsub_mask_vx_nxv16i16_i16_nxv16i16(<vscale x 16 x i16> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv16i16_i16_nxv16i16:
@@ -1330,7 +1330,7 @@ entry:
     i16 %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -1362,7 +1362,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vnmsub.mask.nxv1i32.i32(
   i32,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i32> @intrinsic_vnmsub_mask_vx_nxv1i32_i32_nxv1i32(<vscale x 1 x i32> %0, i32 %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv1i32_i32_nxv1i32:
@@ -1376,7 +1376,7 @@ entry:
     i32 %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -1408,7 +1408,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vnmsub.mask.nxv2i32.i32(
   i32,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i32> @intrinsic_vnmsub_mask_vx_nxv2i32_i32_nxv2i32(<vscale x 2 x i32> %0, i32 %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv2i32_i32_nxv2i32:
@@ -1422,7 +1422,7 @@ entry:
     i32 %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -1454,7 +1454,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vnmsub.mask.nxv4i32.i32(
   i32,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i32> @intrinsic_vnmsub_mask_vx_nxv4i32_i32_nxv4i32(<vscale x 4 x i32> %0, i32 %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv4i32_i32_nxv4i32:
@@ -1468,7 +1468,7 @@ entry:
     i32 %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -1500,7 +1500,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vnmsub.mask.nxv8i32.i32(
   i32,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i32> @intrinsic_vnmsub_mask_vx_nxv8i32_i32_nxv8i32(<vscale x 8 x i32> %0, i32 %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv8i32_i32_nxv8i32:
@@ -1514,7 +1514,7 @@ entry:
     i32 %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -1553,7 +1553,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vnmsub.mask.nxv1i64.i64(
   i64,
   <vscale x 1 x i64>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i64> @intrinsic_vnmsub_mask_vx_nxv1i64_i64_nxv1i64(<vscale x 1 x i64> %0, i64 %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv1i64_i64_nxv1i64:
@@ -1574,7 +1574,7 @@ entry:
     i64 %1,
     <vscale x 1 x i64> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -1613,7 +1613,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vnmsub.mask.nxv2i64.i64(
   i64,
   <vscale x 2 x i64>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i64> @intrinsic_vnmsub_mask_vx_nxv2i64_i64_nxv2i64(<vscale x 2 x i64> %0, i64 %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv2i64_i64_nxv2i64:
@@ -1634,7 +1634,7 @@ entry:
     i64 %1,
     <vscale x 2 x i64> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -1673,7 +1673,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vnmsub.mask.nxv4i64.i64(
   i64,
   <vscale x 4 x i64>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i64> @intrinsic_vnmsub_mask_vx_nxv4i64_i64_nxv4i64(<vscale x 4 x i64> %0, i64 %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv4i64_i64_nxv4i64:
@@ -1694,7 +1694,7 @@ entry:
     i64 %1,
     <vscale x 4 x i64> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i64> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vnmsub-rv64.ll b/llvm/test/CodeGen/RISCV/rvv/vnmsub-rv64.ll
index 6519e590aed38d..b18086afca243b 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vnmsub-rv64.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vnmsub-rv64.ll
@@ -28,7 +28,7 @@ declare <vscale x 1 x i8> @llvm.riscv.vnmsub.mask.nxv1i8.nxv1i8(
   <vscale x 1 x i8>,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i8>  @intrinsic_vnmsub_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv1i8_nxv1i8_nxv1i8:
@@ -42,7 +42,7 @@ entry:
     <vscale x 1 x i8> %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -74,7 +74,7 @@ declare <vscale x 2 x i8> @llvm.riscv.vnmsub.mask.nxv2i8.nxv2i8(
   <vscale x 2 x i8>,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i8>  @intrinsic_vnmsub_mask_vv_nxv2i8_nxv2i8_nxv2i8(<vscale x 2 x i8> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv2i8_nxv2i8_nxv2i8:
@@ -88,7 +88,7 @@ entry:
     <vscale x 2 x i8> %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -120,7 +120,7 @@ declare <vscale x 4 x i8> @llvm.riscv.vnmsub.mask.nxv4i8.nxv4i8(
   <vscale x 4 x i8>,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i8>  @intrinsic_vnmsub_mask_vv_nxv4i8_nxv4i8_nxv4i8(<vscale x 4 x i8> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv4i8_nxv4i8_nxv4i8:
@@ -134,7 +134,7 @@ entry:
     <vscale x 4 x i8> %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -166,7 +166,7 @@ declare <vscale x 8 x i8> @llvm.riscv.vnmsub.mask.nxv8i8.nxv8i8(
   <vscale x 8 x i8>,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i8>  @intrinsic_vnmsub_mask_vv_nxv8i8_nxv8i8_nxv8i8(<vscale x 8 x i8> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv8i8_nxv8i8_nxv8i8:
@@ -180,7 +180,7 @@ entry:
     <vscale x 8 x i8> %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -212,7 +212,7 @@ declare <vscale x 16 x i8> @llvm.riscv.vnmsub.mask.nxv16i8.nxv16i8(
   <vscale x 16 x i8>,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i8>  @intrinsic_vnmsub_mask_vv_nxv16i8_nxv16i8_nxv16i8(<vscale x 16 x i8> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv16i8_nxv16i8_nxv16i8:
@@ -226,7 +226,7 @@ entry:
     <vscale x 16 x i8> %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -258,7 +258,7 @@ declare <vscale x 32 x i8> @llvm.riscv.vnmsub.mask.nxv32i8.nxv32i8(
   <vscale x 32 x i8>,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 32 x i8>  @intrinsic_vnmsub_mask_vv_nxv32i8_nxv32i8_nxv32i8(<vscale x 32 x i8> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv32i8_nxv32i8_nxv32i8:
@@ -272,7 +272,7 @@ entry:
     <vscale x 32 x i8> %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -304,7 +304,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vnmsub.mask.nxv1i16.nxv1i16(
   <vscale x 1 x i16>,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i16>  @intrinsic_vnmsub_mask_vv_nxv1i16_nxv1i16_nxv1i16(<vscale x 1 x i16> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv1i16_nxv1i16_nxv1i16:
@@ -318,7 +318,7 @@ entry:
     <vscale x 1 x i16> %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -350,7 +350,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vnmsub.mask.nxv2i16.nxv2i16(
   <vscale x 2 x i16>,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i16>  @intrinsic_vnmsub_mask_vv_nxv2i16_nxv2i16_nxv2i16(<vscale x 2 x i16> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv2i16_nxv2i16_nxv2i16:
@@ -364,7 +364,7 @@ entry:
     <vscale x 2 x i16> %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -396,7 +396,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vnmsub.mask.nxv4i16.nxv4i16(
   <vscale x 4 x i16>,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i16>  @intrinsic_vnmsub_mask_vv_nxv4i16_nxv4i16_nxv4i16(<vscale x 4 x i16> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv4i16_nxv4i16_nxv4i16:
@@ -410,7 +410,7 @@ entry:
     <vscale x 4 x i16> %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -442,7 +442,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vnmsub.mask.nxv8i16.nxv8i16(
   <vscale x 8 x i16>,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i16>  @intrinsic_vnmsub_mask_vv_nxv8i16_nxv8i16_nxv8i16(<vscale x 8 x i16> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv8i16_nxv8i16_nxv8i16:
@@ -456,7 +456,7 @@ entry:
     <vscale x 8 x i16> %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -488,7 +488,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vnmsub.mask.nxv16i16.nxv16i16(
   <vscale x 16 x i16>,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i16>  @intrinsic_vnmsub_mask_vv_nxv16i16_nxv16i16_nxv16i16(<vscale x 16 x i16> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv16i16_nxv16i16_nxv16i16:
@@ -502,7 +502,7 @@ entry:
     <vscale x 16 x i16> %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -534,7 +534,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vnmsub.mask.nxv1i32.nxv1i32(
   <vscale x 1 x i32>,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i32>  @intrinsic_vnmsub_mask_vv_nxv1i32_nxv1i32_nxv1i32(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv1i32_nxv1i32_nxv1i32:
@@ -548,7 +548,7 @@ entry:
     <vscale x 1 x i32> %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -580,7 +580,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vnmsub.mask.nxv2i32.nxv2i32(
   <vscale x 2 x i32>,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i32>  @intrinsic_vnmsub_mask_vv_nxv2i32_nxv2i32_nxv2i32(<vscale x 2 x i32> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv2i32_nxv2i32_nxv2i32:
@@ -594,7 +594,7 @@ entry:
     <vscale x 2 x i32> %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -626,7 +626,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vnmsub.mask.nxv4i32.nxv4i32(
   <vscale x 4 x i32>,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i32>  @intrinsic_vnmsub_mask_vv_nxv4i32_nxv4i32_nxv4i32(<vscale x 4 x i32> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv4i32_nxv4i32_nxv4i32:
@@ -640,7 +640,7 @@ entry:
     <vscale x 4 x i32> %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -672,7 +672,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vnmsub.mask.nxv8i32.nxv8i32(
   <vscale x 8 x i32>,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i32>  @intrinsic_vnmsub_mask_vv_nxv8i32_nxv8i32_nxv8i32(<vscale x 8 x i32> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv8i32_nxv8i32_nxv8i32:
@@ -686,7 +686,7 @@ entry:
     <vscale x 8 x i32> %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -718,7 +718,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vnmsub.mask.nxv1i64.nxv1i64(
   <vscale x 1 x i64>,
   <vscale x 1 x i64>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i64>  @intrinsic_vnmsub_mask_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv1i64_nxv1i64_nxv1i64:
@@ -732,7 +732,7 @@ entry:
     <vscale x 1 x i64> %1,
     <vscale x 1 x i64> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -764,7 +764,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vnmsub.mask.nxv2i64.nxv2i64(
   <vscale x 2 x i64>,
   <vscale x 2 x i64>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i64>  @intrinsic_vnmsub_mask_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv2i64_nxv2i64_nxv2i64:
@@ -778,7 +778,7 @@ entry:
     <vscale x 2 x i64> %1,
     <vscale x 2 x i64> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -810,7 +810,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vnmsub.mask.nxv4i64.nxv4i64(
   <vscale x 4 x i64>,
   <vscale x 4 x i64>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i64>  @intrinsic_vnmsub_mask_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vv_nxv4i64_nxv4i64_nxv4i64:
@@ -824,7 +824,7 @@ entry:
     <vscale x 4 x i64> %1,
     <vscale x 4 x i64> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -856,7 +856,7 @@ declare <vscale x 1 x i8> @llvm.riscv.vnmsub.mask.nxv1i8.i8(
   i8,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i8> @intrinsic_vnmsub_mask_vx_nxv1i8_i8_nxv1i8(<vscale x 1 x i8> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv1i8_i8_nxv1i8:
@@ -870,7 +870,7 @@ entry:
     i8 %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -902,7 +902,7 @@ declare <vscale x 2 x i8> @llvm.riscv.vnmsub.mask.nxv2i8.i8(
   i8,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i8> @intrinsic_vnmsub_mask_vx_nxv2i8_i8_nxv2i8(<vscale x 2 x i8> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv2i8_i8_nxv2i8:
@@ -916,7 +916,7 @@ entry:
     i8 %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -948,7 +948,7 @@ declare <vscale x 4 x i8> @llvm.riscv.vnmsub.mask.nxv4i8.i8(
   i8,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i8> @intrinsic_vnmsub_mask_vx_nxv4i8_i8_nxv4i8(<vscale x 4 x i8> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv4i8_i8_nxv4i8:
@@ -962,7 +962,7 @@ entry:
     i8 %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -994,7 +994,7 @@ declare <vscale x 8 x i8> @llvm.riscv.vnmsub.mask.nxv8i8.i8(
   i8,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i8> @intrinsic_vnmsub_mask_vx_nxv8i8_i8_nxv8i8(<vscale x 8 x i8> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv8i8_i8_nxv8i8:
@@ -1008,7 +1008,7 @@ entry:
     i8 %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -1040,7 +1040,7 @@ declare <vscale x 16 x i8> @llvm.riscv.vnmsub.mask.nxv16i8.i8(
   i8,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i8> @intrinsic_vnmsub_mask_vx_nxv16i8_i8_nxv16i8(<vscale x 16 x i8> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv16i8_i8_nxv16i8:
@@ -1054,7 +1054,7 @@ entry:
     i8 %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -1086,7 +1086,7 @@ declare <vscale x 32 x i8> @llvm.riscv.vnmsub.mask.nxv32i8.i8(
   i8,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 32 x i8> @intrinsic_vnmsub_mask_vx_nxv32i8_i8_nxv32i8(<vscale x 32 x i8> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv32i8_i8_nxv32i8:
@@ -1100,7 +1100,7 @@ entry:
     i8 %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -1132,7 +1132,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vnmsub.mask.nxv1i16.i16(
   i16,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i16> @intrinsic_vnmsub_mask_vx_nxv1i16_i16_nxv1i16(<vscale x 1 x i16> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv1i16_i16_nxv1i16:
@@ -1146,7 +1146,7 @@ entry:
     i16 %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -1178,7 +1178,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vnmsub.mask.nxv2i16.i16(
   i16,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i16> @intrinsic_vnmsub_mask_vx_nxv2i16_i16_nxv2i16(<vscale x 2 x i16> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv2i16_i16_nxv2i16:
@@ -1192,7 +1192,7 @@ entry:
     i16 %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -1224,7 +1224,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vnmsub.mask.nxv4i16.i16(
   i16,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i16> @intrinsic_vnmsub_mask_vx_nxv4i16_i16_nxv4i16(<vscale x 4 x i16> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv4i16_i16_nxv4i16:
@@ -1238,7 +1238,7 @@ entry:
     i16 %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -1270,7 +1270,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vnmsub.mask.nxv8i16.i16(
   i16,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i16> @intrinsic_vnmsub_mask_vx_nxv8i16_i16_nxv8i16(<vscale x 8 x i16> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv8i16_i16_nxv8i16:
@@ -1284,7 +1284,7 @@ entry:
     i16 %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -1316,7 +1316,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vnmsub.mask.nxv16i16.i16(
   i16,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i16> @intrinsic_vnmsub_mask_vx_nxv16i16_i16_nxv16i16(<vscale x 16 x i16> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv16i16_i16_nxv16i16:
@@ -1330,7 +1330,7 @@ entry:
     i16 %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -1362,7 +1362,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vnmsub.mask.nxv1i32.i32(
   i32,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i32> @intrinsic_vnmsub_mask_vx_nxv1i32_i32_nxv1i32(<vscale x 1 x i32> %0, i32 %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv1i32_i32_nxv1i32:
@@ -1376,7 +1376,7 @@ entry:
     i32 %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -1408,7 +1408,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vnmsub.mask.nxv2i32.i32(
   i32,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i32> @intrinsic_vnmsub_mask_vx_nxv2i32_i32_nxv2i32(<vscale x 2 x i32> %0, i32 %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv2i32_i32_nxv2i32:
@@ -1422,7 +1422,7 @@ entry:
     i32 %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -1454,7 +1454,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vnmsub.mask.nxv4i32.i32(
   i32,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i32> @intrinsic_vnmsub_mask_vx_nxv4i32_i32_nxv4i32(<vscale x 4 x i32> %0, i32 %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv4i32_i32_nxv4i32:
@@ -1468,7 +1468,7 @@ entry:
     i32 %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -1500,7 +1500,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vnmsub.mask.nxv8i32.i32(
   i32,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i32> @intrinsic_vnmsub_mask_vx_nxv8i32_i32_nxv8i32(<vscale x 8 x i32> %0, i32 %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv8i32_i32_nxv8i32:
@@ -1514,7 +1514,7 @@ entry:
     i32 %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -1546,7 +1546,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vnmsub.mask.nxv1i64.i64(
   i64,
   <vscale x 1 x i64>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i64> @intrinsic_vnmsub_mask_vx_nxv1i64_i64_nxv1i64(<vscale x 1 x i64> %0, i64 %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv1i64_i64_nxv1i64:
@@ -1560,7 +1560,7 @@ entry:
     i64 %1,
     <vscale x 1 x i64> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -1592,7 +1592,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vnmsub.mask.nxv2i64.i64(
   i64,
   <vscale x 2 x i64>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i64> @intrinsic_vnmsub_mask_vx_nxv2i64_i64_nxv2i64(<vscale x 2 x i64> %0, i64 %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv2i64_i64_nxv2i64:
@@ -1606,7 +1606,7 @@ entry:
     i64 %1,
     <vscale x 2 x i64> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -1638,7 +1638,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vnmsub.mask.nxv4i64.i64(
   i64,
   <vscale x 4 x i64>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i64> @intrinsic_vnmsub_mask_vx_nxv4i64_i64_nxv4i64(<vscale x 4 x i64> %0, i64 %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vnmsub_mask_vx_nxv4i64_i64_nxv4i64:
@@ -1652,7 +1652,7 @@ entry:
     i64 %1,
     <vscale x 4 x i64> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i64> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vslidedown-rv32.ll b/llvm/test/CodeGen/RISCV/rvv/vslidedown-rv32.ll
index 4cd2fd28752ed1..26a4cc73bccbd4 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vslidedown-rv32.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vslidedown-rv32.ll
@@ -28,7 +28,7 @@ declare <vscale x 1 x i8> @llvm.riscv.vslidedown.mask.nxv1i8(
   <vscale x 1 x i8>,
   i32,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i8> @intrinsic_vslidedown_mask_vx_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, i32 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv1i8_nxv1i8:
@@ -42,7 +42,7 @@ entry:
     <vscale x 1 x i8> %1,
     i32 %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -75,7 +75,7 @@ entry:
     <vscale x 1 x i8> %1,
     i32 9,
     <vscale x 1 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -107,7 +107,7 @@ declare <vscale x 2 x i8> @llvm.riscv.vslidedown.mask.nxv2i8(
   <vscale x 2 x i8>,
   i32,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i8> @intrinsic_vslidedown_mask_vx_nxv2i8_nxv2i8(<vscale x 2 x i8> %0, <vscale x 2 x i8> %1, i32 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv2i8_nxv2i8:
@@ -121,7 +121,7 @@ entry:
     <vscale x 2 x i8> %1,
     i32 %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -154,7 +154,7 @@ entry:
     <vscale x 2 x i8> %1,
     i32 9,
     <vscale x 2 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -186,7 +186,7 @@ declare <vscale x 4 x i8> @llvm.riscv.vslidedown.mask.nxv4i8(
   <vscale x 4 x i8>,
   i32,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i8> @intrinsic_vslidedown_mask_vx_nxv4i8_nxv4i8(<vscale x 4 x i8> %0, <vscale x 4 x i8> %1, i32 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv4i8_nxv4i8:
@@ -200,7 +200,7 @@ entry:
     <vscale x 4 x i8> %1,
     i32 %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -233,7 +233,7 @@ entry:
     <vscale x 4 x i8> %1,
     i32 9,
     <vscale x 4 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -265,7 +265,7 @@ declare <vscale x 8 x i8> @llvm.riscv.vslidedown.mask.nxv8i8(
   <vscale x 8 x i8>,
   i32,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i8> @intrinsic_vslidedown_mask_vx_nxv8i8_nxv8i8(<vscale x 8 x i8> %0, <vscale x 8 x i8> %1, i32 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv8i8_nxv8i8:
@@ -279,7 +279,7 @@ entry:
     <vscale x 8 x i8> %1,
     i32 %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -312,7 +312,7 @@ entry:
     <vscale x 8 x i8> %1,
     i32 9,
     <vscale x 8 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -344,7 +344,7 @@ declare <vscale x 16 x i8> @llvm.riscv.vslidedown.mask.nxv16i8(
   <vscale x 16 x i8>,
   i32,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i8> @intrinsic_vslidedown_mask_vx_nxv16i8_nxv16i8(<vscale x 16 x i8> %0, <vscale x 16 x i8> %1, i32 %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv16i8_nxv16i8:
@@ -358,7 +358,7 @@ entry:
     <vscale x 16 x i8> %1,
     i32 %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -391,7 +391,7 @@ entry:
     <vscale x 16 x i8> %1,
     i32 9,
     <vscale x 16 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -423,7 +423,7 @@ declare <vscale x 32 x i8> @llvm.riscv.vslidedown.mask.nxv32i8(
   <vscale x 32 x i8>,
   i32,
   <vscale x 32 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 32 x i8> @intrinsic_vslidedown_mask_vx_nxv32i8_nxv32i8(<vscale x 32 x i8> %0, <vscale x 32 x i8> %1, i32 %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv32i8_nxv32i8:
@@ -437,7 +437,7 @@ entry:
     <vscale x 32 x i8> %1,
     i32 %2,
     <vscale x 32 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -470,7 +470,7 @@ entry:
     <vscale x 32 x i8> %1,
     i32 9,
     <vscale x 32 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -502,7 +502,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vslidedown.mask.nxv1i16(
   <vscale x 1 x i16>,
   i32,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i16> @intrinsic_vslidedown_mask_vx_nxv1i16_nxv1i16(<vscale x 1 x i16> %0, <vscale x 1 x i16> %1, i32 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv1i16_nxv1i16:
@@ -516,7 +516,7 @@ entry:
     <vscale x 1 x i16> %1,
     i32 %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -549,7 +549,7 @@ entry:
     <vscale x 1 x i16> %1,
     i32 9,
     <vscale x 1 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -581,7 +581,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vslidedown.mask.nxv2i16(
   <vscale x 2 x i16>,
   i32,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i16> @intrinsic_vslidedown_mask_vx_nxv2i16_nxv2i16(<vscale x 2 x i16> %0, <vscale x 2 x i16> %1, i32 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv2i16_nxv2i16:
@@ -595,7 +595,7 @@ entry:
     <vscale x 2 x i16> %1,
     i32 %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -628,7 +628,7 @@ entry:
     <vscale x 2 x i16> %1,
     i32 9,
     <vscale x 2 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -660,7 +660,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vslidedown.mask.nxv4i16(
   <vscale x 4 x i16>,
   i32,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i16> @intrinsic_vslidedown_mask_vx_nxv4i16_nxv4i16(<vscale x 4 x i16> %0, <vscale x 4 x i16> %1, i32 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv4i16_nxv4i16:
@@ -674,7 +674,7 @@ entry:
     <vscale x 4 x i16> %1,
     i32 %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -707,7 +707,7 @@ entry:
     <vscale x 4 x i16> %1,
     i32 9,
     <vscale x 4 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -739,7 +739,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vslidedown.mask.nxv8i16(
   <vscale x 8 x i16>,
   i32,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i16> @intrinsic_vslidedown_mask_vx_nxv8i16_nxv8i16(<vscale x 8 x i16> %0, <vscale x 8 x i16> %1, i32 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv8i16_nxv8i16:
@@ -753,7 +753,7 @@ entry:
     <vscale x 8 x i16> %1,
     i32 %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -786,7 +786,7 @@ entry:
     <vscale x 8 x i16> %1,
     i32 9,
     <vscale x 8 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -818,7 +818,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vslidedown.mask.nxv16i16(
   <vscale x 16 x i16>,
   i32,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i16> @intrinsic_vslidedown_mask_vx_nxv16i16_nxv16i16(<vscale x 16 x i16> %0, <vscale x 16 x i16> %1, i32 %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv16i16_nxv16i16:
@@ -832,7 +832,7 @@ entry:
     <vscale x 16 x i16> %1,
     i32 %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -865,7 +865,7 @@ entry:
     <vscale x 16 x i16> %1,
     i32 9,
     <vscale x 16 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -897,7 +897,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vslidedown.mask.nxv1i32(
   <vscale x 1 x i32>,
   i32,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i32> @intrinsic_vslidedown_mask_vx_nxv1i32_nxv1i32(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, i32 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv1i32_nxv1i32:
@@ -911,7 +911,7 @@ entry:
     <vscale x 1 x i32> %1,
     i32 %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -944,7 +944,7 @@ entry:
     <vscale x 1 x i32> %1,
     i32 9,
     <vscale x 1 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -976,7 +976,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vslidedown.mask.nxv2i32(
   <vscale x 2 x i32>,
   i32,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i32> @intrinsic_vslidedown_mask_vx_nxv2i32_nxv2i32(<vscale x 2 x i32> %0, <vscale x 2 x i32> %1, i32 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv2i32_nxv2i32:
@@ -990,7 +990,7 @@ entry:
     <vscale x 2 x i32> %1,
     i32 %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -1023,7 +1023,7 @@ entry:
     <vscale x 2 x i32> %1,
     i32 9,
     <vscale x 2 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -1055,7 +1055,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vslidedown.mask.nxv4i32(
   <vscale x 4 x i32>,
   i32,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i32> @intrinsic_vslidedown_mask_vx_nxv4i32_nxv4i32(<vscale x 4 x i32> %0, <vscale x 4 x i32> %1, i32 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv4i32_nxv4i32:
@@ -1069,7 +1069,7 @@ entry:
     <vscale x 4 x i32> %1,
     i32 %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -1102,7 +1102,7 @@ entry:
     <vscale x 4 x i32> %1,
     i32 9,
     <vscale x 4 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -1134,7 +1134,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vslidedown.mask.nxv8i32(
   <vscale x 8 x i32>,
   i32,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i32> @intrinsic_vslidedown_mask_vx_nxv8i32_nxv8i32(<vscale x 8 x i32> %0, <vscale x 8 x i32> %1, i32 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv8i32_nxv8i32:
@@ -1148,7 +1148,7 @@ entry:
     <vscale x 8 x i32> %1,
     i32 %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -1181,7 +1181,7 @@ entry:
     <vscale x 8 x i32> %1,
     i32 9,
     <vscale x 8 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -1213,7 +1213,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vslidedown.mask.nxv1i64(
   <vscale x 1 x i64>,
   i32,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i64> @intrinsic_vslidedown_mask_vx_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, i32 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv1i64_nxv1i64:
@@ -1227,7 +1227,7 @@ entry:
     <vscale x 1 x i64> %1,
     i32 %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -1260,7 +1260,7 @@ entry:
     <vscale x 1 x i64> %1,
     i32 9,
     <vscale x 1 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -1292,7 +1292,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vslidedown.mask.nxv2i64(
   <vscale x 2 x i64>,
   i32,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i64> @intrinsic_vslidedown_mask_vx_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, i32 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv2i64_nxv2i64:
@@ -1306,7 +1306,7 @@ entry:
     <vscale x 2 x i64> %1,
     i32 %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -1339,7 +1339,7 @@ entry:
     <vscale x 2 x i64> %1,
     i32 9,
     <vscale x 2 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -1371,7 +1371,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vslidedown.mask.nxv4i64(
   <vscale x 4 x i64>,
   i32,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i64> @intrinsic_vslidedown_mask_vx_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, i32 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv4i64_nxv4i64:
@@ -1385,7 +1385,7 @@ entry:
     <vscale x 4 x i64> %1,
     i32 %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -1418,7 +1418,7 @@ entry:
     <vscale x 4 x i64> %1,
     i32 9,
     <vscale x 4 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -1450,7 +1450,7 @@ declare <vscale x 1 x half> @llvm.riscv.vslidedown.mask.nxv1f16(
   <vscale x 1 x half>,
   i32,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x half> @intrinsic_vslidedown_mask_vx_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, i32 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv1f16_nxv1f16:
@@ -1464,7 +1464,7 @@ entry:
     <vscale x 1 x half> %1,
     i32 %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x half> %a
 }
@@ -1497,7 +1497,7 @@ entry:
     <vscale x 1 x half> %1,
     i32 9,
     <vscale x 1 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 1 x half> %a
 }
@@ -1529,7 +1529,7 @@ declare <vscale x 2 x half> @llvm.riscv.vslidedown.mask.nxv2f16(
   <vscale x 2 x half>,
   i32,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x half> @intrinsic_vslidedown_mask_vx_nxv2f16_nxv2f16(<vscale x 2 x half> %0, <vscale x 2 x half> %1, i32 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv2f16_nxv2f16:
@@ -1543,7 +1543,7 @@ entry:
     <vscale x 2 x half> %1,
     i32 %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x half> %a
 }
@@ -1576,7 +1576,7 @@ entry:
     <vscale x 2 x half> %1,
     i32 9,
     <vscale x 2 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 2 x half> %a
 }
@@ -1608,7 +1608,7 @@ declare <vscale x 4 x half> @llvm.riscv.vslidedown.mask.nxv4f16(
   <vscale x 4 x half>,
   i32,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x half> @intrinsic_vslidedown_mask_vx_nxv4f16_nxv4f16(<vscale x 4 x half> %0, <vscale x 4 x half> %1, i32 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv4f16_nxv4f16:
@@ -1622,7 +1622,7 @@ entry:
     <vscale x 4 x half> %1,
     i32 %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x half> %a
 }
@@ -1655,7 +1655,7 @@ entry:
     <vscale x 4 x half> %1,
     i32 9,
     <vscale x 4 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 4 x half> %a
 }
@@ -1687,7 +1687,7 @@ declare <vscale x 8 x half> @llvm.riscv.vslidedown.mask.nxv8f16(
   <vscale x 8 x half>,
   i32,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x half> @intrinsic_vslidedown_mask_vx_nxv8f16_nxv8f16(<vscale x 8 x half> %0, <vscale x 8 x half> %1, i32 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv8f16_nxv8f16:
@@ -1701,7 +1701,7 @@ entry:
     <vscale x 8 x half> %1,
     i32 %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x half> %a
 }
@@ -1734,7 +1734,7 @@ entry:
     <vscale x 8 x half> %1,
     i32 9,
     <vscale x 8 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 8 x half> %a
 }
@@ -1766,7 +1766,7 @@ declare <vscale x 16 x half> @llvm.riscv.vslidedown.mask.nxv16f16(
   <vscale x 16 x half>,
   i32,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x half> @intrinsic_vslidedown_mask_vx_nxv16f16_nxv16f16(<vscale x 16 x half> %0, <vscale x 16 x half> %1, i32 %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv16f16_nxv16f16:
@@ -1780,7 +1780,7 @@ entry:
     <vscale x 16 x half> %1,
     i32 %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x half> %a
 }
@@ -1813,7 +1813,7 @@ entry:
     <vscale x 16 x half> %1,
     i32 9,
     <vscale x 16 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 16 x half> %a
 }
@@ -1845,7 +1845,7 @@ declare <vscale x 1 x float> @llvm.riscv.vslidedown.mask.nxv1f32(
   <vscale x 1 x float>,
   i32,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x float> @intrinsic_vslidedown_mask_vx_nxv1f32_nxv1f32(<vscale x 1 x float> %0, <vscale x 1 x float> %1, i32 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv1f32_nxv1f32:
@@ -1859,7 +1859,7 @@ entry:
     <vscale x 1 x float> %1,
     i32 %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x float> %a
 }
@@ -1892,7 +1892,7 @@ entry:
     <vscale x 1 x float> %1,
     i32 9,
     <vscale x 1 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 1 x float> %a
 }
@@ -1924,7 +1924,7 @@ declare <vscale x 2 x float> @llvm.riscv.vslidedown.mask.nxv2f32(
   <vscale x 2 x float>,
   i32,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x float> @intrinsic_vslidedown_mask_vx_nxv2f32_nxv2f32(<vscale x 2 x float> %0, <vscale x 2 x float> %1, i32 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv2f32_nxv2f32:
@@ -1938,7 +1938,7 @@ entry:
     <vscale x 2 x float> %1,
     i32 %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x float> %a
 }
@@ -1971,7 +1971,7 @@ entry:
     <vscale x 2 x float> %1,
     i32 9,
     <vscale x 2 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 2 x float> %a
 }
@@ -2003,7 +2003,7 @@ declare <vscale x 4 x float> @llvm.riscv.vslidedown.mask.nxv4f32(
   <vscale x 4 x float>,
   i32,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x float> @intrinsic_vslidedown_mask_vx_nxv4f32_nxv4f32(<vscale x 4 x float> %0, <vscale x 4 x float> %1, i32 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv4f32_nxv4f32:
@@ -2017,7 +2017,7 @@ entry:
     <vscale x 4 x float> %1,
     i32 %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x float> %a
 }
@@ -2050,7 +2050,7 @@ entry:
     <vscale x 4 x float> %1,
     i32 9,
     <vscale x 4 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 4 x float> %a
 }
@@ -2082,7 +2082,7 @@ declare <vscale x 8 x float> @llvm.riscv.vslidedown.mask.nxv8f32(
   <vscale x 8 x float>,
   i32,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x float> @intrinsic_vslidedown_mask_vx_nxv8f32_nxv8f32(<vscale x 8 x float> %0, <vscale x 8 x float> %1, i32 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv8f32_nxv8f32:
@@ -2096,7 +2096,7 @@ entry:
     <vscale x 8 x float> %1,
     i32 %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x float> %a
 }
@@ -2129,7 +2129,7 @@ entry:
     <vscale x 8 x float> %1,
     i32 9,
     <vscale x 8 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 8 x float> %a
 }
@@ -2161,7 +2161,7 @@ declare <vscale x 1 x double> @llvm.riscv.vslidedown.mask.nxv1f64(
   <vscale x 1 x double>,
   i32,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x double> @intrinsic_vslidedown_mask_vx_nxv1f64_nxv1f64(<vscale x 1 x double> %0, <vscale x 1 x double> %1, i32 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv1f64_nxv1f64:
@@ -2175,7 +2175,7 @@ entry:
     <vscale x 1 x double> %1,
     i32 %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x double> %a
 }
@@ -2208,7 +2208,7 @@ entry:
     <vscale x 1 x double> %1,
     i32 9,
     <vscale x 1 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 1 x double> %a
 }
@@ -2240,7 +2240,7 @@ declare <vscale x 2 x double> @llvm.riscv.vslidedown.mask.nxv2f64(
   <vscale x 2 x double>,
   i32,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x double> @intrinsic_vslidedown_mask_vx_nxv2f64_nxv2f64(<vscale x 2 x double> %0, <vscale x 2 x double> %1, i32 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv2f64_nxv2f64:
@@ -2254,7 +2254,7 @@ entry:
     <vscale x 2 x double> %1,
     i32 %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x double> %a
 }
@@ -2287,7 +2287,7 @@ entry:
     <vscale x 2 x double> %1,
     i32 9,
     <vscale x 2 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 2 x double> %a
 }
@@ -2319,7 +2319,7 @@ declare <vscale x 4 x double> @llvm.riscv.vslidedown.mask.nxv4f64(
   <vscale x 4 x double>,
   i32,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x double> @intrinsic_vslidedown_mask_vx_nxv4f64_nxv4f64(<vscale x 4 x double> %0, <vscale x 4 x double> %1, i32 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv4f64_nxv4f64:
@@ -2333,7 +2333,7 @@ entry:
     <vscale x 4 x double> %1,
     i32 %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x double> %a
 }
@@ -2366,7 +2366,7 @@ entry:
     <vscale x 4 x double> %1,
     i32 9,
     <vscale x 4 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 4 x double> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vslidedown-rv64.ll b/llvm/test/CodeGen/RISCV/rvv/vslidedown-rv64.ll
index 863c9557531523..4cfe039866a110 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vslidedown-rv64.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vslidedown-rv64.ll
@@ -28,7 +28,7 @@ declare <vscale x 1 x i8> @llvm.riscv.vslidedown.mask.nxv1i8(
   <vscale x 1 x i8>,
   i64,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i8> @intrinsic_vslidedown_mask_vx_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, i64 %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv1i8_nxv1i8:
@@ -42,7 +42,7 @@ entry:
     <vscale x 1 x i8> %1,
     i64 %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -75,7 +75,7 @@ entry:
     <vscale x 1 x i8> %1,
     i64 9,
     <vscale x 1 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -107,7 +107,7 @@ declare <vscale x 2 x i8> @llvm.riscv.vslidedown.mask.nxv2i8(
   <vscale x 2 x i8>,
   i64,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i8> @intrinsic_vslidedown_mask_vx_nxv2i8_nxv2i8(<vscale x 2 x i8> %0, <vscale x 2 x i8> %1, i64 %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv2i8_nxv2i8:
@@ -121,7 +121,7 @@ entry:
     <vscale x 2 x i8> %1,
     i64 %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -154,7 +154,7 @@ entry:
     <vscale x 2 x i8> %1,
     i64 9,
     <vscale x 2 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -186,7 +186,7 @@ declare <vscale x 4 x i8> @llvm.riscv.vslidedown.mask.nxv4i8(
   <vscale x 4 x i8>,
   i64,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i8> @intrinsic_vslidedown_mask_vx_nxv4i8_nxv4i8(<vscale x 4 x i8> %0, <vscale x 4 x i8> %1, i64 %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv4i8_nxv4i8:
@@ -200,7 +200,7 @@ entry:
     <vscale x 4 x i8> %1,
     i64 %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -233,7 +233,7 @@ entry:
     <vscale x 4 x i8> %1,
     i64 9,
     <vscale x 4 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -265,7 +265,7 @@ declare <vscale x 8 x i8> @llvm.riscv.vslidedown.mask.nxv8i8(
   <vscale x 8 x i8>,
   i64,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i8> @intrinsic_vslidedown_mask_vx_nxv8i8_nxv8i8(<vscale x 8 x i8> %0, <vscale x 8 x i8> %1, i64 %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv8i8_nxv8i8:
@@ -279,7 +279,7 @@ entry:
     <vscale x 8 x i8> %1,
     i64 %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -312,7 +312,7 @@ entry:
     <vscale x 8 x i8> %1,
     i64 9,
     <vscale x 8 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -344,7 +344,7 @@ declare <vscale x 16 x i8> @llvm.riscv.vslidedown.mask.nxv16i8(
   <vscale x 16 x i8>,
   i64,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i8> @intrinsic_vslidedown_mask_vx_nxv16i8_nxv16i8(<vscale x 16 x i8> %0, <vscale x 16 x i8> %1, i64 %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv16i8_nxv16i8:
@@ -358,7 +358,7 @@ entry:
     <vscale x 16 x i8> %1,
     i64 %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -391,7 +391,7 @@ entry:
     <vscale x 16 x i8> %1,
     i64 9,
     <vscale x 16 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -423,7 +423,7 @@ declare <vscale x 32 x i8> @llvm.riscv.vslidedown.mask.nxv32i8(
   <vscale x 32 x i8>,
   i64,
   <vscale x 32 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 32 x i8> @intrinsic_vslidedown_mask_vx_nxv32i8_nxv32i8(<vscale x 32 x i8> %0, <vscale x 32 x i8> %1, i64 %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv32i8_nxv32i8:
@@ -437,7 +437,7 @@ entry:
     <vscale x 32 x i8> %1,
     i64 %2,
     <vscale x 32 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -470,7 +470,7 @@ entry:
     <vscale x 32 x i8> %1,
     i64 9,
     <vscale x 32 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -502,7 +502,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vslidedown.mask.nxv1i16(
   <vscale x 1 x i16>,
   i64,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i16> @intrinsic_vslidedown_mask_vx_nxv1i16_nxv1i16(<vscale x 1 x i16> %0, <vscale x 1 x i16> %1, i64 %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv1i16_nxv1i16:
@@ -516,7 +516,7 @@ entry:
     <vscale x 1 x i16> %1,
     i64 %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -549,7 +549,7 @@ entry:
     <vscale x 1 x i16> %1,
     i64 9,
     <vscale x 1 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -581,7 +581,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vslidedown.mask.nxv2i16(
   <vscale x 2 x i16>,
   i64,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i16> @intrinsic_vslidedown_mask_vx_nxv2i16_nxv2i16(<vscale x 2 x i16> %0, <vscale x 2 x i16> %1, i64 %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv2i16_nxv2i16:
@@ -595,7 +595,7 @@ entry:
     <vscale x 2 x i16> %1,
     i64 %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -628,7 +628,7 @@ entry:
     <vscale x 2 x i16> %1,
     i64 9,
     <vscale x 2 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -660,7 +660,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vslidedown.mask.nxv4i16(
   <vscale x 4 x i16>,
   i64,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i16> @intrinsic_vslidedown_mask_vx_nxv4i16_nxv4i16(<vscale x 4 x i16> %0, <vscale x 4 x i16> %1, i64 %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv4i16_nxv4i16:
@@ -674,7 +674,7 @@ entry:
     <vscale x 4 x i16> %1,
     i64 %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -707,7 +707,7 @@ entry:
     <vscale x 4 x i16> %1,
     i64 9,
     <vscale x 4 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -739,7 +739,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vslidedown.mask.nxv8i16(
   <vscale x 8 x i16>,
   i64,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i16> @intrinsic_vslidedown_mask_vx_nxv8i16_nxv8i16(<vscale x 8 x i16> %0, <vscale x 8 x i16> %1, i64 %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv8i16_nxv8i16:
@@ -753,7 +753,7 @@ entry:
     <vscale x 8 x i16> %1,
     i64 %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -786,7 +786,7 @@ entry:
     <vscale x 8 x i16> %1,
     i64 9,
     <vscale x 8 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -818,7 +818,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vslidedown.mask.nxv16i16(
   <vscale x 16 x i16>,
   i64,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i16> @intrinsic_vslidedown_mask_vx_nxv16i16_nxv16i16(<vscale x 16 x i16> %0, <vscale x 16 x i16> %1, i64 %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv16i16_nxv16i16:
@@ -832,7 +832,7 @@ entry:
     <vscale x 16 x i16> %1,
     i64 %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -865,7 +865,7 @@ entry:
     <vscale x 16 x i16> %1,
     i64 9,
     <vscale x 16 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -897,7 +897,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vslidedown.mask.nxv1i32(
   <vscale x 1 x i32>,
   i64,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i32> @intrinsic_vslidedown_mask_vx_nxv1i32_nxv1i32(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, i64 %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv1i32_nxv1i32:
@@ -911,7 +911,7 @@ entry:
     <vscale x 1 x i32> %1,
     i64 %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -944,7 +944,7 @@ entry:
     <vscale x 1 x i32> %1,
     i64 9,
     <vscale x 1 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -976,7 +976,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vslidedown.mask.nxv2i32(
   <vscale x 2 x i32>,
   i64,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i32> @intrinsic_vslidedown_mask_vx_nxv2i32_nxv2i32(<vscale x 2 x i32> %0, <vscale x 2 x i32> %1, i64 %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv2i32_nxv2i32:
@@ -990,7 +990,7 @@ entry:
     <vscale x 2 x i32> %1,
     i64 %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -1023,7 +1023,7 @@ entry:
     <vscale x 2 x i32> %1,
     i64 9,
     <vscale x 2 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -1055,7 +1055,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vslidedown.mask.nxv4i32(
   <vscale x 4 x i32>,
   i64,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i32> @intrinsic_vslidedown_mask_vx_nxv4i32_nxv4i32(<vscale x 4 x i32> %0, <vscale x 4 x i32> %1, i64 %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv4i32_nxv4i32:
@@ -1069,7 +1069,7 @@ entry:
     <vscale x 4 x i32> %1,
     i64 %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -1102,7 +1102,7 @@ entry:
     <vscale x 4 x i32> %1,
     i64 9,
     <vscale x 4 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -1134,7 +1134,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vslidedown.mask.nxv8i32(
   <vscale x 8 x i32>,
   i64,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i32> @intrinsic_vslidedown_mask_vx_nxv8i32_nxv8i32(<vscale x 8 x i32> %0, <vscale x 8 x i32> %1, i64 %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv8i32_nxv8i32:
@@ -1148,7 +1148,7 @@ entry:
     <vscale x 8 x i32> %1,
     i64 %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -1181,7 +1181,7 @@ entry:
     <vscale x 8 x i32> %1,
     i64 9,
     <vscale x 8 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -1213,7 +1213,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vslidedown.mask.nxv1i64(
   <vscale x 1 x i64>,
   i64,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i64> @intrinsic_vslidedown_mask_vx_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, i64 %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv1i64_nxv1i64:
@@ -1227,7 +1227,7 @@ entry:
     <vscale x 1 x i64> %1,
     i64 %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -1260,7 +1260,7 @@ entry:
     <vscale x 1 x i64> %1,
     i64 9,
     <vscale x 1 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -1292,7 +1292,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vslidedown.mask.nxv2i64(
   <vscale x 2 x i64>,
   i64,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i64> @intrinsic_vslidedown_mask_vx_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, i64 %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv2i64_nxv2i64:
@@ -1306,7 +1306,7 @@ entry:
     <vscale x 2 x i64> %1,
     i64 %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -1339,7 +1339,7 @@ entry:
     <vscale x 2 x i64> %1,
     i64 9,
     <vscale x 2 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -1371,7 +1371,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vslidedown.mask.nxv4i64(
   <vscale x 4 x i64>,
   i64,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i64> @intrinsic_vslidedown_mask_vx_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, i64 %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv4i64_nxv4i64:
@@ -1385,7 +1385,7 @@ entry:
     <vscale x 4 x i64> %1,
     i64 %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -1418,7 +1418,7 @@ entry:
     <vscale x 4 x i64> %1,
     i64 9,
     <vscale x 4 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -1450,7 +1450,7 @@ declare <vscale x 1 x half> @llvm.riscv.vslidedown.mask.nxv1f16(
   <vscale x 1 x half>,
   i64,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x half> @intrinsic_vslidedown_mask_vx_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, i64 %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv1f16_nxv1f16:
@@ -1464,7 +1464,7 @@ entry:
     <vscale x 1 x half> %1,
     i64 %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x half> %a
 }
@@ -1497,7 +1497,7 @@ entry:
     <vscale x 1 x half> %1,
     i64 9,
     <vscale x 1 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 1 x half> %a
 }
@@ -1529,7 +1529,7 @@ declare <vscale x 2 x half> @llvm.riscv.vslidedown.mask.nxv2f16(
   <vscale x 2 x half>,
   i64,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x half> @intrinsic_vslidedown_mask_vx_nxv2f16_nxv2f16(<vscale x 2 x half> %0, <vscale x 2 x half> %1, i64 %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv2f16_nxv2f16:
@@ -1543,7 +1543,7 @@ entry:
     <vscale x 2 x half> %1,
     i64 %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x half> %a
 }
@@ -1576,7 +1576,7 @@ entry:
     <vscale x 2 x half> %1,
     i64 9,
     <vscale x 2 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 2 x half> %a
 }
@@ -1608,7 +1608,7 @@ declare <vscale x 4 x half> @llvm.riscv.vslidedown.mask.nxv4f16(
   <vscale x 4 x half>,
   i64,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x half> @intrinsic_vslidedown_mask_vx_nxv4f16_nxv4f16(<vscale x 4 x half> %0, <vscale x 4 x half> %1, i64 %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv4f16_nxv4f16:
@@ -1622,7 +1622,7 @@ entry:
     <vscale x 4 x half> %1,
     i64 %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x half> %a
 }
@@ -1655,7 +1655,7 @@ entry:
     <vscale x 4 x half> %1,
     i64 9,
     <vscale x 4 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 4 x half> %a
 }
@@ -1687,7 +1687,7 @@ declare <vscale x 8 x half> @llvm.riscv.vslidedown.mask.nxv8f16(
   <vscale x 8 x half>,
   i64,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x half> @intrinsic_vslidedown_mask_vx_nxv8f16_nxv8f16(<vscale x 8 x half> %0, <vscale x 8 x half> %1, i64 %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv8f16_nxv8f16:
@@ -1701,7 +1701,7 @@ entry:
     <vscale x 8 x half> %1,
     i64 %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x half> %a
 }
@@ -1734,7 +1734,7 @@ entry:
     <vscale x 8 x half> %1,
     i64 9,
     <vscale x 8 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 8 x half> %a
 }
@@ -1766,7 +1766,7 @@ declare <vscale x 16 x half> @llvm.riscv.vslidedown.mask.nxv16f16(
   <vscale x 16 x half>,
   i64,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x half> @intrinsic_vslidedown_mask_vx_nxv16f16_nxv16f16(<vscale x 16 x half> %0, <vscale x 16 x half> %1, i64 %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv16f16_nxv16f16:
@@ -1780,7 +1780,7 @@ entry:
     <vscale x 16 x half> %1,
     i64 %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x half> %a
 }
@@ -1813,7 +1813,7 @@ entry:
     <vscale x 16 x half> %1,
     i64 9,
     <vscale x 16 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 16 x half> %a
 }
@@ -1845,7 +1845,7 @@ declare <vscale x 1 x float> @llvm.riscv.vslidedown.mask.nxv1f32(
   <vscale x 1 x float>,
   i64,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x float> @intrinsic_vslidedown_mask_vx_nxv1f32_nxv1f32(<vscale x 1 x float> %0, <vscale x 1 x float> %1, i64 %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv1f32_nxv1f32:
@@ -1859,7 +1859,7 @@ entry:
     <vscale x 1 x float> %1,
     i64 %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x float> %a
 }
@@ -1892,7 +1892,7 @@ entry:
     <vscale x 1 x float> %1,
     i64 9,
     <vscale x 1 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 1 x float> %a
 }
@@ -1924,7 +1924,7 @@ declare <vscale x 2 x float> @llvm.riscv.vslidedown.mask.nxv2f32(
   <vscale x 2 x float>,
   i64,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x float> @intrinsic_vslidedown_mask_vx_nxv2f32_nxv2f32(<vscale x 2 x float> %0, <vscale x 2 x float> %1, i64 %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv2f32_nxv2f32:
@@ -1938,7 +1938,7 @@ entry:
     <vscale x 2 x float> %1,
     i64 %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x float> %a
 }
@@ -1971,7 +1971,7 @@ entry:
     <vscale x 2 x float> %1,
     i64 9,
     <vscale x 2 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 2 x float> %a
 }
@@ -2003,7 +2003,7 @@ declare <vscale x 4 x float> @llvm.riscv.vslidedown.mask.nxv4f32(
   <vscale x 4 x float>,
   i64,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x float> @intrinsic_vslidedown_mask_vx_nxv4f32_nxv4f32(<vscale x 4 x float> %0, <vscale x 4 x float> %1, i64 %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv4f32_nxv4f32:
@@ -2017,7 +2017,7 @@ entry:
     <vscale x 4 x float> %1,
     i64 %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x float> %a
 }
@@ -2050,7 +2050,7 @@ entry:
     <vscale x 4 x float> %1,
     i64 9,
     <vscale x 4 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 4 x float> %a
 }
@@ -2082,7 +2082,7 @@ declare <vscale x 8 x float> @llvm.riscv.vslidedown.mask.nxv8f32(
   <vscale x 8 x float>,
   i64,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x float> @intrinsic_vslidedown_mask_vx_nxv8f32_nxv8f32(<vscale x 8 x float> %0, <vscale x 8 x float> %1, i64 %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv8f32_nxv8f32:
@@ -2096,7 +2096,7 @@ entry:
     <vscale x 8 x float> %1,
     i64 %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x float> %a
 }
@@ -2129,7 +2129,7 @@ entry:
     <vscale x 8 x float> %1,
     i64 9,
     <vscale x 8 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 8 x float> %a
 }
@@ -2161,7 +2161,7 @@ declare <vscale x 1 x double> @llvm.riscv.vslidedown.mask.nxv1f64(
   <vscale x 1 x double>,
   i64,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x double> @intrinsic_vslidedown_mask_vx_nxv1f64_nxv1f64(<vscale x 1 x double> %0, <vscale x 1 x double> %1, i64 %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv1f64_nxv1f64:
@@ -2175,7 +2175,7 @@ entry:
     <vscale x 1 x double> %1,
     i64 %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x double> %a
 }
@@ -2208,7 +2208,7 @@ entry:
     <vscale x 1 x double> %1,
     i64 9,
     <vscale x 1 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 1 x double> %a
 }
@@ -2240,7 +2240,7 @@ declare <vscale x 2 x double> @llvm.riscv.vslidedown.mask.nxv2f64(
   <vscale x 2 x double>,
   i64,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x double> @intrinsic_vslidedown_mask_vx_nxv2f64_nxv2f64(<vscale x 2 x double> %0, <vscale x 2 x double> %1, i64 %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv2f64_nxv2f64:
@@ -2254,7 +2254,7 @@ entry:
     <vscale x 2 x double> %1,
     i64 %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x double> %a
 }
@@ -2287,7 +2287,7 @@ entry:
     <vscale x 2 x double> %1,
     i64 9,
     <vscale x 2 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 2 x double> %a
 }
@@ -2319,7 +2319,7 @@ declare <vscale x 4 x double> @llvm.riscv.vslidedown.mask.nxv4f64(
   <vscale x 4 x double>,
   i64,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x double> @intrinsic_vslidedown_mask_vx_nxv4f64_nxv4f64(<vscale x 4 x double> %0, <vscale x 4 x double> %1, i64 %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslidedown_mask_vx_nxv4f64_nxv4f64:
@@ -2333,7 +2333,7 @@ entry:
     <vscale x 4 x double> %1,
     i64 %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x double> %a
 }
@@ -2366,7 +2366,7 @@ entry:
     <vscale x 4 x double> %1,
     i64 9,
     <vscale x 4 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 4 x double> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vslideup-rv32.ll b/llvm/test/CodeGen/RISCV/rvv/vslideup-rv32.ll
index 3cdb4afac9692a..9704fbb93e7107 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vslideup-rv32.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vslideup-rv32.ll
@@ -28,7 +28,7 @@ declare <vscale x 1 x i8> @llvm.riscv.vslideup.mask.nxv1i8(
   <vscale x 1 x i8>,
   i32,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i8> @intrinsic_vslideup_mask_vx_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, i32 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv1i8_nxv1i8:
@@ -42,7 +42,7 @@ entry:
     <vscale x 1 x i8> %1,
     i32 %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -75,7 +75,7 @@ entry:
     <vscale x 1 x i8> %1,
     i32 9,
     <vscale x 1 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -107,7 +107,7 @@ declare <vscale x 2 x i8> @llvm.riscv.vslideup.mask.nxv2i8(
   <vscale x 2 x i8>,
   i32,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i8> @intrinsic_vslideup_mask_vx_nxv2i8_nxv2i8(<vscale x 2 x i8> %0, <vscale x 2 x i8> %1, i32 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv2i8_nxv2i8:
@@ -121,7 +121,7 @@ entry:
     <vscale x 2 x i8> %1,
     i32 %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -154,7 +154,7 @@ entry:
     <vscale x 2 x i8> %1,
     i32 9,
     <vscale x 2 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -186,7 +186,7 @@ declare <vscale x 4 x i8> @llvm.riscv.vslideup.mask.nxv4i8(
   <vscale x 4 x i8>,
   i32,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i8> @intrinsic_vslideup_mask_vx_nxv4i8_nxv4i8(<vscale x 4 x i8> %0, <vscale x 4 x i8> %1, i32 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv4i8_nxv4i8:
@@ -200,7 +200,7 @@ entry:
     <vscale x 4 x i8> %1,
     i32 %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -233,7 +233,7 @@ entry:
     <vscale x 4 x i8> %1,
     i32 9,
     <vscale x 4 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -265,7 +265,7 @@ declare <vscale x 8 x i8> @llvm.riscv.vslideup.mask.nxv8i8(
   <vscale x 8 x i8>,
   i32,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i8> @intrinsic_vslideup_mask_vx_nxv8i8_nxv8i8(<vscale x 8 x i8> %0, <vscale x 8 x i8> %1, i32 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv8i8_nxv8i8:
@@ -279,7 +279,7 @@ entry:
     <vscale x 8 x i8> %1,
     i32 %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -312,7 +312,7 @@ entry:
     <vscale x 8 x i8> %1,
     i32 9,
     <vscale x 8 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -344,7 +344,7 @@ declare <vscale x 16 x i8> @llvm.riscv.vslideup.mask.nxv16i8(
   <vscale x 16 x i8>,
   i32,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i8> @intrinsic_vslideup_mask_vx_nxv16i8_nxv16i8(<vscale x 16 x i8> %0, <vscale x 16 x i8> %1, i32 %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv16i8_nxv16i8:
@@ -358,7 +358,7 @@ entry:
     <vscale x 16 x i8> %1,
     i32 %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -391,7 +391,7 @@ entry:
     <vscale x 16 x i8> %1,
     i32 9,
     <vscale x 16 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -423,7 +423,7 @@ declare <vscale x 32 x i8> @llvm.riscv.vslideup.mask.nxv32i8(
   <vscale x 32 x i8>,
   i32,
   <vscale x 32 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 32 x i8> @intrinsic_vslideup_mask_vx_nxv32i8_nxv32i8(<vscale x 32 x i8> %0, <vscale x 32 x i8> %1, i32 %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv32i8_nxv32i8:
@@ -437,7 +437,7 @@ entry:
     <vscale x 32 x i8> %1,
     i32 %2,
     <vscale x 32 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -470,7 +470,7 @@ entry:
     <vscale x 32 x i8> %1,
     i32 9,
     <vscale x 32 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -502,7 +502,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vslideup.mask.nxv1i16(
   <vscale x 1 x i16>,
   i32,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i16> @intrinsic_vslideup_mask_vx_nxv1i16_nxv1i16(<vscale x 1 x i16> %0, <vscale x 1 x i16> %1, i32 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv1i16_nxv1i16:
@@ -516,7 +516,7 @@ entry:
     <vscale x 1 x i16> %1,
     i32 %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -549,7 +549,7 @@ entry:
     <vscale x 1 x i16> %1,
     i32 9,
     <vscale x 1 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -581,7 +581,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vslideup.mask.nxv2i16(
   <vscale x 2 x i16>,
   i32,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i16> @intrinsic_vslideup_mask_vx_nxv2i16_nxv2i16(<vscale x 2 x i16> %0, <vscale x 2 x i16> %1, i32 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv2i16_nxv2i16:
@@ -595,7 +595,7 @@ entry:
     <vscale x 2 x i16> %1,
     i32 %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -628,7 +628,7 @@ entry:
     <vscale x 2 x i16> %1,
     i32 9,
     <vscale x 2 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -660,7 +660,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vslideup.mask.nxv4i16(
   <vscale x 4 x i16>,
   i32,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i16> @intrinsic_vslideup_mask_vx_nxv4i16_nxv4i16(<vscale x 4 x i16> %0, <vscale x 4 x i16> %1, i32 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv4i16_nxv4i16:
@@ -674,7 +674,7 @@ entry:
     <vscale x 4 x i16> %1,
     i32 %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -707,7 +707,7 @@ entry:
     <vscale x 4 x i16> %1,
     i32 9,
     <vscale x 4 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -739,7 +739,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vslideup.mask.nxv8i16(
   <vscale x 8 x i16>,
   i32,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i16> @intrinsic_vslideup_mask_vx_nxv8i16_nxv8i16(<vscale x 8 x i16> %0, <vscale x 8 x i16> %1, i32 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv8i16_nxv8i16:
@@ -753,7 +753,7 @@ entry:
     <vscale x 8 x i16> %1,
     i32 %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -786,7 +786,7 @@ entry:
     <vscale x 8 x i16> %1,
     i32 9,
     <vscale x 8 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -818,7 +818,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vslideup.mask.nxv16i16(
   <vscale x 16 x i16>,
   i32,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i16> @intrinsic_vslideup_mask_vx_nxv16i16_nxv16i16(<vscale x 16 x i16> %0, <vscale x 16 x i16> %1, i32 %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv16i16_nxv16i16:
@@ -832,7 +832,7 @@ entry:
     <vscale x 16 x i16> %1,
     i32 %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -865,7 +865,7 @@ entry:
     <vscale x 16 x i16> %1,
     i32 9,
     <vscale x 16 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -897,7 +897,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vslideup.mask.nxv1i32(
   <vscale x 1 x i32>,
   i32,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i32> @intrinsic_vslideup_mask_vx_nxv1i32_nxv1i32(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, i32 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv1i32_nxv1i32:
@@ -911,7 +911,7 @@ entry:
     <vscale x 1 x i32> %1,
     i32 %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -944,7 +944,7 @@ entry:
     <vscale x 1 x i32> %1,
     i32 9,
     <vscale x 1 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -976,7 +976,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vslideup.mask.nxv2i32(
   <vscale x 2 x i32>,
   i32,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i32> @intrinsic_vslideup_mask_vx_nxv2i32_nxv2i32(<vscale x 2 x i32> %0, <vscale x 2 x i32> %1, i32 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv2i32_nxv2i32:
@@ -990,7 +990,7 @@ entry:
     <vscale x 2 x i32> %1,
     i32 %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -1023,7 +1023,7 @@ entry:
     <vscale x 2 x i32> %1,
     i32 9,
     <vscale x 2 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -1055,7 +1055,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vslideup.mask.nxv4i32(
   <vscale x 4 x i32>,
   i32,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i32> @intrinsic_vslideup_mask_vx_nxv4i32_nxv4i32(<vscale x 4 x i32> %0, <vscale x 4 x i32> %1, i32 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv4i32_nxv4i32:
@@ -1069,7 +1069,7 @@ entry:
     <vscale x 4 x i32> %1,
     i32 %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -1102,7 +1102,7 @@ entry:
     <vscale x 4 x i32> %1,
     i32 9,
     <vscale x 4 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -1134,7 +1134,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vslideup.mask.nxv8i32(
   <vscale x 8 x i32>,
   i32,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i32> @intrinsic_vslideup_mask_vx_nxv8i32_nxv8i32(<vscale x 8 x i32> %0, <vscale x 8 x i32> %1, i32 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv8i32_nxv8i32:
@@ -1148,7 +1148,7 @@ entry:
     <vscale x 8 x i32> %1,
     i32 %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -1181,7 +1181,7 @@ entry:
     <vscale x 8 x i32> %1,
     i32 9,
     <vscale x 8 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -1213,7 +1213,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vslideup.mask.nxv1i64(
   <vscale x 1 x i64>,
   i32,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i64> @intrinsic_vslideup_mask_vx_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, i32 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv1i64_nxv1i64:
@@ -1227,7 +1227,7 @@ entry:
     <vscale x 1 x i64> %1,
     i32 %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -1260,7 +1260,7 @@ entry:
     <vscale x 1 x i64> %1,
     i32 9,
     <vscale x 1 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -1292,7 +1292,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vslideup.mask.nxv2i64(
   <vscale x 2 x i64>,
   i32,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i64> @intrinsic_vslideup_mask_vx_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, i32 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv2i64_nxv2i64:
@@ -1306,7 +1306,7 @@ entry:
     <vscale x 2 x i64> %1,
     i32 %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -1339,7 +1339,7 @@ entry:
     <vscale x 2 x i64> %1,
     i32 9,
     <vscale x 2 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -1371,7 +1371,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vslideup.mask.nxv4i64(
   <vscale x 4 x i64>,
   i32,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i64> @intrinsic_vslideup_mask_vx_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, i32 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv4i64_nxv4i64:
@@ -1385,7 +1385,7 @@ entry:
     <vscale x 4 x i64> %1,
     i32 %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -1418,7 +1418,7 @@ entry:
     <vscale x 4 x i64> %1,
     i32 9,
     <vscale x 4 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -1450,7 +1450,7 @@ declare <vscale x 1 x half> @llvm.riscv.vslideup.mask.nxv1f16(
   <vscale x 1 x half>,
   i32,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x half> @intrinsic_vslideup_mask_vx_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, i32 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv1f16_nxv1f16:
@@ -1464,7 +1464,7 @@ entry:
     <vscale x 1 x half> %1,
     i32 %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x half> %a
 }
@@ -1497,7 +1497,7 @@ entry:
     <vscale x 1 x half> %1,
     i32 9,
     <vscale x 1 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 1 x half> %a
 }
@@ -1529,7 +1529,7 @@ declare <vscale x 2 x half> @llvm.riscv.vslideup.mask.nxv2f16(
   <vscale x 2 x half>,
   i32,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x half> @intrinsic_vslideup_mask_vx_nxv2f16_nxv2f16(<vscale x 2 x half> %0, <vscale x 2 x half> %1, i32 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv2f16_nxv2f16:
@@ -1543,7 +1543,7 @@ entry:
     <vscale x 2 x half> %1,
     i32 %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x half> %a
 }
@@ -1576,7 +1576,7 @@ entry:
     <vscale x 2 x half> %1,
     i32 9,
     <vscale x 2 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 2 x half> %a
 }
@@ -1608,7 +1608,7 @@ declare <vscale x 4 x half> @llvm.riscv.vslideup.mask.nxv4f16(
   <vscale x 4 x half>,
   i32,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x half> @intrinsic_vslideup_mask_vx_nxv4f16_nxv4f16(<vscale x 4 x half> %0, <vscale x 4 x half> %1, i32 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv4f16_nxv4f16:
@@ -1622,7 +1622,7 @@ entry:
     <vscale x 4 x half> %1,
     i32 %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x half> %a
 }
@@ -1655,7 +1655,7 @@ entry:
     <vscale x 4 x half> %1,
     i32 9,
     <vscale x 4 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 4 x half> %a
 }
@@ -1687,7 +1687,7 @@ declare <vscale x 8 x half> @llvm.riscv.vslideup.mask.nxv8f16(
   <vscale x 8 x half>,
   i32,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x half> @intrinsic_vslideup_mask_vx_nxv8f16_nxv8f16(<vscale x 8 x half> %0, <vscale x 8 x half> %1, i32 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv8f16_nxv8f16:
@@ -1701,7 +1701,7 @@ entry:
     <vscale x 8 x half> %1,
     i32 %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x half> %a
 }
@@ -1734,7 +1734,7 @@ entry:
     <vscale x 8 x half> %1,
     i32 9,
     <vscale x 8 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 8 x half> %a
 }
@@ -1766,7 +1766,7 @@ declare <vscale x 16 x half> @llvm.riscv.vslideup.mask.nxv16f16(
   <vscale x 16 x half>,
   i32,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x half> @intrinsic_vslideup_mask_vx_nxv16f16_nxv16f16(<vscale x 16 x half> %0, <vscale x 16 x half> %1, i32 %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv16f16_nxv16f16:
@@ -1780,7 +1780,7 @@ entry:
     <vscale x 16 x half> %1,
     i32 %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x half> %a
 }
@@ -1813,7 +1813,7 @@ entry:
     <vscale x 16 x half> %1,
     i32 9,
     <vscale x 16 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 16 x half> %a
 }
@@ -1845,7 +1845,7 @@ declare <vscale x 1 x float> @llvm.riscv.vslideup.mask.nxv1f32(
   <vscale x 1 x float>,
   i32,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x float> @intrinsic_vslideup_mask_vx_nxv1f32_nxv1f32(<vscale x 1 x float> %0, <vscale x 1 x float> %1, i32 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv1f32_nxv1f32:
@@ -1859,7 +1859,7 @@ entry:
     <vscale x 1 x float> %1,
     i32 %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x float> %a
 }
@@ -1892,7 +1892,7 @@ entry:
     <vscale x 1 x float> %1,
     i32 9,
     <vscale x 1 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 1 x float> %a
 }
@@ -1924,7 +1924,7 @@ declare <vscale x 2 x float> @llvm.riscv.vslideup.mask.nxv2f32(
   <vscale x 2 x float>,
   i32,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x float> @intrinsic_vslideup_mask_vx_nxv2f32_nxv2f32(<vscale x 2 x float> %0, <vscale x 2 x float> %1, i32 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv2f32_nxv2f32:
@@ -1938,7 +1938,7 @@ entry:
     <vscale x 2 x float> %1,
     i32 %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x float> %a
 }
@@ -1971,7 +1971,7 @@ entry:
     <vscale x 2 x float> %1,
     i32 9,
     <vscale x 2 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 2 x float> %a
 }
@@ -2003,7 +2003,7 @@ declare <vscale x 4 x float> @llvm.riscv.vslideup.mask.nxv4f32(
   <vscale x 4 x float>,
   i32,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x float> @intrinsic_vslideup_mask_vx_nxv4f32_nxv4f32(<vscale x 4 x float> %0, <vscale x 4 x float> %1, i32 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv4f32_nxv4f32:
@@ -2017,7 +2017,7 @@ entry:
     <vscale x 4 x float> %1,
     i32 %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x float> %a
 }
@@ -2050,7 +2050,7 @@ entry:
     <vscale x 4 x float> %1,
     i32 9,
     <vscale x 4 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 4 x float> %a
 }
@@ -2082,7 +2082,7 @@ declare <vscale x 8 x float> @llvm.riscv.vslideup.mask.nxv8f32(
   <vscale x 8 x float>,
   i32,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x float> @intrinsic_vslideup_mask_vx_nxv8f32_nxv8f32(<vscale x 8 x float> %0, <vscale x 8 x float> %1, i32 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv8f32_nxv8f32:
@@ -2096,7 +2096,7 @@ entry:
     <vscale x 8 x float> %1,
     i32 %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x float> %a
 }
@@ -2129,7 +2129,7 @@ entry:
     <vscale x 8 x float> %1,
     i32 9,
     <vscale x 8 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 8 x float> %a
 }
@@ -2161,7 +2161,7 @@ declare <vscale x 1 x double> @llvm.riscv.vslideup.mask.nxv1f64(
   <vscale x 1 x double>,
   i32,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x double> @intrinsic_vslideup_mask_vx_nxv1f64_nxv1f64(<vscale x 1 x double> %0, <vscale x 1 x double> %1, i32 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv1f64_nxv1f64:
@@ -2175,7 +2175,7 @@ entry:
     <vscale x 1 x double> %1,
     i32 %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x double> %a
 }
@@ -2208,7 +2208,7 @@ entry:
     <vscale x 1 x double> %1,
     i32 9,
     <vscale x 1 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 1 x double> %a
 }
@@ -2240,7 +2240,7 @@ declare <vscale x 2 x double> @llvm.riscv.vslideup.mask.nxv2f64(
   <vscale x 2 x double>,
   i32,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x double> @intrinsic_vslideup_mask_vx_nxv2f64_nxv2f64(<vscale x 2 x double> %0, <vscale x 2 x double> %1, i32 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv2f64_nxv2f64:
@@ -2254,7 +2254,7 @@ entry:
     <vscale x 2 x double> %1,
     i32 %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x double> %a
 }
@@ -2287,7 +2287,7 @@ entry:
     <vscale x 2 x double> %1,
     i32 9,
     <vscale x 2 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 2 x double> %a
 }
@@ -2319,7 +2319,7 @@ declare <vscale x 4 x double> @llvm.riscv.vslideup.mask.nxv4f64(
   <vscale x 4 x double>,
   i32,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x double> @intrinsic_vslideup_mask_vx_nxv4f64_nxv4f64(<vscale x 4 x double> %0, <vscale x 4 x double> %1, i32 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv4f64_nxv4f64:
@@ -2333,7 +2333,7 @@ entry:
     <vscale x 4 x double> %1,
     i32 %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x double> %a
 }
@@ -2366,7 +2366,7 @@ entry:
     <vscale x 4 x double> %1,
     i32 9,
     <vscale x 4 x i1> %2,
-    i32 %3)
+    i32 %3, i32 0)
 
   ret <vscale x 4 x double> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vslideup-rv64.ll b/llvm/test/CodeGen/RISCV/rvv/vslideup-rv64.ll
index a9a166c1056e96..aa26b208f122bf 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vslideup-rv64.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vslideup-rv64.ll
@@ -28,7 +28,7 @@ declare <vscale x 1 x i8> @llvm.riscv.vslideup.mask.nxv1i8(
   <vscale x 1 x i8>,
   i64,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i8> @intrinsic_vslideup_mask_vx_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, i64 %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv1i8_nxv1i8:
@@ -42,7 +42,7 @@ entry:
     <vscale x 1 x i8> %1,
     i64 %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -75,7 +75,7 @@ entry:
     <vscale x 1 x i8> %1,
     i64 9,
     <vscale x 1 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 1 x i8> %a
 }
@@ -107,7 +107,7 @@ declare <vscale x 2 x i8> @llvm.riscv.vslideup.mask.nxv2i8(
   <vscale x 2 x i8>,
   i64,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i8> @intrinsic_vslideup_mask_vx_nxv2i8_nxv2i8(<vscale x 2 x i8> %0, <vscale x 2 x i8> %1, i64 %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv2i8_nxv2i8:
@@ -121,7 +121,7 @@ entry:
     <vscale x 2 x i8> %1,
     i64 %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -154,7 +154,7 @@ entry:
     <vscale x 2 x i8> %1,
     i64 9,
     <vscale x 2 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 2 x i8> %a
 }
@@ -186,7 +186,7 @@ declare <vscale x 4 x i8> @llvm.riscv.vslideup.mask.nxv4i8(
   <vscale x 4 x i8>,
   i64,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i8> @intrinsic_vslideup_mask_vx_nxv4i8_nxv4i8(<vscale x 4 x i8> %0, <vscale x 4 x i8> %1, i64 %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv4i8_nxv4i8:
@@ -200,7 +200,7 @@ entry:
     <vscale x 4 x i8> %1,
     i64 %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -233,7 +233,7 @@ entry:
     <vscale x 4 x i8> %1,
     i64 9,
     <vscale x 4 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 4 x i8> %a
 }
@@ -265,7 +265,7 @@ declare <vscale x 8 x i8> @llvm.riscv.vslideup.mask.nxv8i8(
   <vscale x 8 x i8>,
   i64,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i8> @intrinsic_vslideup_mask_vx_nxv8i8_nxv8i8(<vscale x 8 x i8> %0, <vscale x 8 x i8> %1, i64 %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv8i8_nxv8i8:
@@ -279,7 +279,7 @@ entry:
     <vscale x 8 x i8> %1,
     i64 %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -312,7 +312,7 @@ entry:
     <vscale x 8 x i8> %1,
     i64 9,
     <vscale x 8 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 8 x i8> %a
 }
@@ -344,7 +344,7 @@ declare <vscale x 16 x i8> @llvm.riscv.vslideup.mask.nxv16i8(
   <vscale x 16 x i8>,
   i64,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i8> @intrinsic_vslideup_mask_vx_nxv16i8_nxv16i8(<vscale x 16 x i8> %0, <vscale x 16 x i8> %1, i64 %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv16i8_nxv16i8:
@@ -358,7 +358,7 @@ entry:
     <vscale x 16 x i8> %1,
     i64 %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -391,7 +391,7 @@ entry:
     <vscale x 16 x i8> %1,
     i64 9,
     <vscale x 16 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 16 x i8> %a
 }
@@ -423,7 +423,7 @@ declare <vscale x 32 x i8> @llvm.riscv.vslideup.mask.nxv32i8(
   <vscale x 32 x i8>,
   i64,
   <vscale x 32 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 32 x i8> @intrinsic_vslideup_mask_vx_nxv32i8_nxv32i8(<vscale x 32 x i8> %0, <vscale x 32 x i8> %1, i64 %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv32i8_nxv32i8:
@@ -437,7 +437,7 @@ entry:
     <vscale x 32 x i8> %1,
     i64 %2,
     <vscale x 32 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -470,7 +470,7 @@ entry:
     <vscale x 32 x i8> %1,
     i64 9,
     <vscale x 32 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 32 x i8> %a
 }
@@ -502,7 +502,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vslideup.mask.nxv1i16(
   <vscale x 1 x i16>,
   i64,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i16> @intrinsic_vslideup_mask_vx_nxv1i16_nxv1i16(<vscale x 1 x i16> %0, <vscale x 1 x i16> %1, i64 %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv1i16_nxv1i16:
@@ -516,7 +516,7 @@ entry:
     <vscale x 1 x i16> %1,
     i64 %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -549,7 +549,7 @@ entry:
     <vscale x 1 x i16> %1,
     i64 9,
     <vscale x 1 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -581,7 +581,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vslideup.mask.nxv2i16(
   <vscale x 2 x i16>,
   i64,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i16> @intrinsic_vslideup_mask_vx_nxv2i16_nxv2i16(<vscale x 2 x i16> %0, <vscale x 2 x i16> %1, i64 %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv2i16_nxv2i16:
@@ -595,7 +595,7 @@ entry:
     <vscale x 2 x i16> %1,
     i64 %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -628,7 +628,7 @@ entry:
     <vscale x 2 x i16> %1,
     i64 9,
     <vscale x 2 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -660,7 +660,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vslideup.mask.nxv4i16(
   <vscale x 4 x i16>,
   i64,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i16> @intrinsic_vslideup_mask_vx_nxv4i16_nxv4i16(<vscale x 4 x i16> %0, <vscale x 4 x i16> %1, i64 %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv4i16_nxv4i16:
@@ -674,7 +674,7 @@ entry:
     <vscale x 4 x i16> %1,
     i64 %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -707,7 +707,7 @@ entry:
     <vscale x 4 x i16> %1,
     i64 9,
     <vscale x 4 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -739,7 +739,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vslideup.mask.nxv8i16(
   <vscale x 8 x i16>,
   i64,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i16> @intrinsic_vslideup_mask_vx_nxv8i16_nxv8i16(<vscale x 8 x i16> %0, <vscale x 8 x i16> %1, i64 %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv8i16_nxv8i16:
@@ -753,7 +753,7 @@ entry:
     <vscale x 8 x i16> %1,
     i64 %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -786,7 +786,7 @@ entry:
     <vscale x 8 x i16> %1,
     i64 9,
     <vscale x 8 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -818,7 +818,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vslideup.mask.nxv16i16(
   <vscale x 16 x i16>,
   i64,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i16> @intrinsic_vslideup_mask_vx_nxv16i16_nxv16i16(<vscale x 16 x i16> %0, <vscale x 16 x i16> %1, i64 %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv16i16_nxv16i16:
@@ -832,7 +832,7 @@ entry:
     <vscale x 16 x i16> %1,
     i64 %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -865,7 +865,7 @@ entry:
     <vscale x 16 x i16> %1,
     i64 9,
     <vscale x 16 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -897,7 +897,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vslideup.mask.nxv1i32(
   <vscale x 1 x i32>,
   i64,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i32> @intrinsic_vslideup_mask_vx_nxv1i32_nxv1i32(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, i64 %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv1i32_nxv1i32:
@@ -911,7 +911,7 @@ entry:
     <vscale x 1 x i32> %1,
     i64 %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -944,7 +944,7 @@ entry:
     <vscale x 1 x i32> %1,
     i64 9,
     <vscale x 1 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -976,7 +976,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vslideup.mask.nxv2i32(
   <vscale x 2 x i32>,
   i64,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i32> @intrinsic_vslideup_mask_vx_nxv2i32_nxv2i32(<vscale x 2 x i32> %0, <vscale x 2 x i32> %1, i64 %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv2i32_nxv2i32:
@@ -990,7 +990,7 @@ entry:
     <vscale x 2 x i32> %1,
     i64 %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -1023,7 +1023,7 @@ entry:
     <vscale x 2 x i32> %1,
     i64 9,
     <vscale x 2 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -1055,7 +1055,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vslideup.mask.nxv4i32(
   <vscale x 4 x i32>,
   i64,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i32> @intrinsic_vslideup_mask_vx_nxv4i32_nxv4i32(<vscale x 4 x i32> %0, <vscale x 4 x i32> %1, i64 %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv4i32_nxv4i32:
@@ -1069,7 +1069,7 @@ entry:
     <vscale x 4 x i32> %1,
     i64 %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -1102,7 +1102,7 @@ entry:
     <vscale x 4 x i32> %1,
     i64 9,
     <vscale x 4 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -1134,7 +1134,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vslideup.mask.nxv8i32(
   <vscale x 8 x i32>,
   i64,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i32> @intrinsic_vslideup_mask_vx_nxv8i32_nxv8i32(<vscale x 8 x i32> %0, <vscale x 8 x i32> %1, i64 %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv8i32_nxv8i32:
@@ -1148,7 +1148,7 @@ entry:
     <vscale x 8 x i32> %1,
     i64 %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -1181,7 +1181,7 @@ entry:
     <vscale x 8 x i32> %1,
     i64 9,
     <vscale x 8 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -1213,7 +1213,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vslideup.mask.nxv1i64(
   <vscale x 1 x i64>,
   i64,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i64> @intrinsic_vslideup_mask_vx_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, i64 %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv1i64_nxv1i64:
@@ -1227,7 +1227,7 @@ entry:
     <vscale x 1 x i64> %1,
     i64 %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -1260,7 +1260,7 @@ entry:
     <vscale x 1 x i64> %1,
     i64 9,
     <vscale x 1 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -1292,7 +1292,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vslideup.mask.nxv2i64(
   <vscale x 2 x i64>,
   i64,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i64> @intrinsic_vslideup_mask_vx_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, i64 %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv2i64_nxv2i64:
@@ -1306,7 +1306,7 @@ entry:
     <vscale x 2 x i64> %1,
     i64 %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -1339,7 +1339,7 @@ entry:
     <vscale x 2 x i64> %1,
     i64 9,
     <vscale x 2 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -1371,7 +1371,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vslideup.mask.nxv4i64(
   <vscale x 4 x i64>,
   i64,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i64> @intrinsic_vslideup_mask_vx_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, i64 %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv4i64_nxv4i64:
@@ -1385,7 +1385,7 @@ entry:
     <vscale x 4 x i64> %1,
     i64 %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -1418,7 +1418,7 @@ entry:
     <vscale x 4 x i64> %1,
     i64 9,
     <vscale x 4 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -1450,7 +1450,7 @@ declare <vscale x 1 x half> @llvm.riscv.vslideup.mask.nxv1f16(
   <vscale x 1 x half>,
   i64,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x half> @intrinsic_vslideup_mask_vx_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, i64 %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv1f16_nxv1f16:
@@ -1464,7 +1464,7 @@ entry:
     <vscale x 1 x half> %1,
     i64 %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x half> %a
 }
@@ -1497,7 +1497,7 @@ entry:
     <vscale x 1 x half> %1,
     i64 9,
     <vscale x 1 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 1 x half> %a
 }
@@ -1529,7 +1529,7 @@ declare <vscale x 2 x half> @llvm.riscv.vslideup.mask.nxv2f16(
   <vscale x 2 x half>,
   i64,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x half> @intrinsic_vslideup_mask_vx_nxv2f16_nxv2f16(<vscale x 2 x half> %0, <vscale x 2 x half> %1, i64 %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv2f16_nxv2f16:
@@ -1543,7 +1543,7 @@ entry:
     <vscale x 2 x half> %1,
     i64 %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x half> %a
 }
@@ -1576,7 +1576,7 @@ entry:
     <vscale x 2 x half> %1,
     i64 9,
     <vscale x 2 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 2 x half> %a
 }
@@ -1608,7 +1608,7 @@ declare <vscale x 4 x half> @llvm.riscv.vslideup.mask.nxv4f16(
   <vscale x 4 x half>,
   i64,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x half> @intrinsic_vslideup_mask_vx_nxv4f16_nxv4f16(<vscale x 4 x half> %0, <vscale x 4 x half> %1, i64 %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv4f16_nxv4f16:
@@ -1622,7 +1622,7 @@ entry:
     <vscale x 4 x half> %1,
     i64 %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x half> %a
 }
@@ -1655,7 +1655,7 @@ entry:
     <vscale x 4 x half> %1,
     i64 9,
     <vscale x 4 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 4 x half> %a
 }
@@ -1687,7 +1687,7 @@ declare <vscale x 8 x half> @llvm.riscv.vslideup.mask.nxv8f16(
   <vscale x 8 x half>,
   i64,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x half> @intrinsic_vslideup_mask_vx_nxv8f16_nxv8f16(<vscale x 8 x half> %0, <vscale x 8 x half> %1, i64 %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv8f16_nxv8f16:
@@ -1701,7 +1701,7 @@ entry:
     <vscale x 8 x half> %1,
     i64 %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x half> %a
 }
@@ -1734,7 +1734,7 @@ entry:
     <vscale x 8 x half> %1,
     i64 9,
     <vscale x 8 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 8 x half> %a
 }
@@ -1766,7 +1766,7 @@ declare <vscale x 16 x half> @llvm.riscv.vslideup.mask.nxv16f16(
   <vscale x 16 x half>,
   i64,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x half> @intrinsic_vslideup_mask_vx_nxv16f16_nxv16f16(<vscale x 16 x half> %0, <vscale x 16 x half> %1, i64 %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv16f16_nxv16f16:
@@ -1780,7 +1780,7 @@ entry:
     <vscale x 16 x half> %1,
     i64 %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x half> %a
 }
@@ -1813,7 +1813,7 @@ entry:
     <vscale x 16 x half> %1,
     i64 9,
     <vscale x 16 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 16 x half> %a
 }
@@ -1845,7 +1845,7 @@ declare <vscale x 1 x float> @llvm.riscv.vslideup.mask.nxv1f32(
   <vscale x 1 x float>,
   i64,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x float> @intrinsic_vslideup_mask_vx_nxv1f32_nxv1f32(<vscale x 1 x float> %0, <vscale x 1 x float> %1, i64 %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv1f32_nxv1f32:
@@ -1859,7 +1859,7 @@ entry:
     <vscale x 1 x float> %1,
     i64 %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x float> %a
 }
@@ -1892,7 +1892,7 @@ entry:
     <vscale x 1 x float> %1,
     i64 9,
     <vscale x 1 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 1 x float> %a
 }
@@ -1924,7 +1924,7 @@ declare <vscale x 2 x float> @llvm.riscv.vslideup.mask.nxv2f32(
   <vscale x 2 x float>,
   i64,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x float> @intrinsic_vslideup_mask_vx_nxv2f32_nxv2f32(<vscale x 2 x float> %0, <vscale x 2 x float> %1, i64 %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv2f32_nxv2f32:
@@ -1938,7 +1938,7 @@ entry:
     <vscale x 2 x float> %1,
     i64 %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x float> %a
 }
@@ -1971,7 +1971,7 @@ entry:
     <vscale x 2 x float> %1,
     i64 9,
     <vscale x 2 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 2 x float> %a
 }
@@ -2003,7 +2003,7 @@ declare <vscale x 4 x float> @llvm.riscv.vslideup.mask.nxv4f32(
   <vscale x 4 x float>,
   i64,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x float> @intrinsic_vslideup_mask_vx_nxv4f32_nxv4f32(<vscale x 4 x float> %0, <vscale x 4 x float> %1, i64 %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv4f32_nxv4f32:
@@ -2017,7 +2017,7 @@ entry:
     <vscale x 4 x float> %1,
     i64 %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x float> %a
 }
@@ -2050,7 +2050,7 @@ entry:
     <vscale x 4 x float> %1,
     i64 9,
     <vscale x 4 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 4 x float> %a
 }
@@ -2082,7 +2082,7 @@ declare <vscale x 8 x float> @llvm.riscv.vslideup.mask.nxv8f32(
   <vscale x 8 x float>,
   i64,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x float> @intrinsic_vslideup_mask_vx_nxv8f32_nxv8f32(<vscale x 8 x float> %0, <vscale x 8 x float> %1, i64 %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv8f32_nxv8f32:
@@ -2096,7 +2096,7 @@ entry:
     <vscale x 8 x float> %1,
     i64 %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x float> %a
 }
@@ -2129,7 +2129,7 @@ entry:
     <vscale x 8 x float> %1,
     i64 9,
     <vscale x 8 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 8 x float> %a
 }
@@ -2161,7 +2161,7 @@ declare <vscale x 1 x double> @llvm.riscv.vslideup.mask.nxv1f64(
   <vscale x 1 x double>,
   i64,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x double> @intrinsic_vslideup_mask_vx_nxv1f64_nxv1f64(<vscale x 1 x double> %0, <vscale x 1 x double> %1, i64 %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv1f64_nxv1f64:
@@ -2175,7 +2175,7 @@ entry:
     <vscale x 1 x double> %1,
     i64 %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x double> %a
 }
@@ -2208,7 +2208,7 @@ entry:
     <vscale x 1 x double> %1,
     i64 9,
     <vscale x 1 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 1 x double> %a
 }
@@ -2240,7 +2240,7 @@ declare <vscale x 2 x double> @llvm.riscv.vslideup.mask.nxv2f64(
   <vscale x 2 x double>,
   i64,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x double> @intrinsic_vslideup_mask_vx_nxv2f64_nxv2f64(<vscale x 2 x double> %0, <vscale x 2 x double> %1, i64 %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv2f64_nxv2f64:
@@ -2254,7 +2254,7 @@ entry:
     <vscale x 2 x double> %1,
     i64 %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x double> %a
 }
@@ -2287,7 +2287,7 @@ entry:
     <vscale x 2 x double> %1,
     i64 9,
     <vscale x 2 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 2 x double> %a
 }
@@ -2319,7 +2319,7 @@ declare <vscale x 4 x double> @llvm.riscv.vslideup.mask.nxv4f64(
   <vscale x 4 x double>,
   i64,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x double> @intrinsic_vslideup_mask_vx_nxv4f64_nxv4f64(<vscale x 4 x double> %0, <vscale x 4 x double> %1, i64 %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vslideup_mask_vx_nxv4f64_nxv4f64:
@@ -2333,7 +2333,7 @@ entry:
     <vscale x 4 x double> %1,
     i64 %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x double> %a
 }
@@ -2366,7 +2366,7 @@ entry:
     <vscale x 4 x double> %1,
     i64 9,
     <vscale x 4 x i1> %2,
-    i64 %3)
+    i64 %3, i64 0)
 
   ret <vscale x 4 x double> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vwmacc-rv32.ll b/llvm/test/CodeGen/RISCV/rvv/vwmacc-rv32.ll
index a68c26ba9180c6..5fb3a63b137fb2 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vwmacc-rv32.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vwmacc-rv32.ll
@@ -28,7 +28,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.nxv1i8(
   <vscale x 1 x i8>,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i16>  @intrinsic_vwmacc_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv1i16_nxv1i8_nxv1i8:
@@ -42,7 +42,7 @@ entry:
     <vscale x 1 x i8> %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -74,7 +74,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vwmacc.mask.nxv2i16.nxv2i8(
   <vscale x 2 x i8>,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i16>  @intrinsic_vwmacc_mask_vv_nxv2i16_nxv2i8_nxv2i8(<vscale x 2 x i16> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv2i16_nxv2i8_nxv2i8:
@@ -88,7 +88,7 @@ entry:
     <vscale x 2 x i8> %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -120,7 +120,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vwmacc.mask.nxv4i16.nxv4i8(
   <vscale x 4 x i8>,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i16>  @intrinsic_vwmacc_mask_vv_nxv4i16_nxv4i8_nxv4i8(<vscale x 4 x i16> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv4i16_nxv4i8_nxv4i8:
@@ -134,7 +134,7 @@ entry:
     <vscale x 4 x i8> %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -166,7 +166,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vwmacc.mask.nxv8i16.nxv8i8(
   <vscale x 8 x i8>,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i16>  @intrinsic_vwmacc_mask_vv_nxv8i16_nxv8i8_nxv8i8(<vscale x 8 x i16> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv8i16_nxv8i8_nxv8i8:
@@ -180,7 +180,7 @@ entry:
     <vscale x 8 x i8> %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -212,7 +212,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vwmacc.mask.nxv16i16.nxv16i8(
   <vscale x 16 x i8>,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i16>  @intrinsic_vwmacc_mask_vv_nxv16i16_nxv16i8_nxv16i8(<vscale x 16 x i16> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv16i16_nxv16i8_nxv16i8:
@@ -226,7 +226,7 @@ entry:
     <vscale x 16 x i8> %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -258,7 +258,7 @@ declare <vscale x 32 x i16> @llvm.riscv.vwmacc.mask.nxv32i16.nxv32i8(
   <vscale x 32 x i8>,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 32 x i16>  @intrinsic_vwmacc_mask_vv_nxv32i16_nxv32i8_nxv32i8(<vscale x 32 x i16> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv32i16_nxv32i8_nxv32i8:
@@ -272,7 +272,7 @@ entry:
     <vscale x 32 x i8> %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 32 x i16> %a
 }
@@ -304,7 +304,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vwmacc.mask.nxv1i32.nxv1i16(
   <vscale x 1 x i16>,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i32>  @intrinsic_vwmacc_mask_vv_nxv1i32_nxv1i16_nxv1i16(<vscale x 1 x i32> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv1i32_nxv1i16_nxv1i16:
@@ -318,7 +318,7 @@ entry:
     <vscale x 1 x i16> %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -350,7 +350,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vwmacc.mask.nxv2i32.nxv2i16(
   <vscale x 2 x i16>,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i32>  @intrinsic_vwmacc_mask_vv_nxv2i32_nxv2i16_nxv2i16(<vscale x 2 x i32> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv2i32_nxv2i16_nxv2i16:
@@ -364,7 +364,7 @@ entry:
     <vscale x 2 x i16> %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -396,7 +396,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vwmacc.mask.nxv4i32.nxv4i16(
   <vscale x 4 x i16>,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i32>  @intrinsic_vwmacc_mask_vv_nxv4i32_nxv4i16_nxv4i16(<vscale x 4 x i32> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv4i32_nxv4i16_nxv4i16:
@@ -410,7 +410,7 @@ entry:
     <vscale x 4 x i16> %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -442,7 +442,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vwmacc.mask.nxv8i32.nxv8i16(
   <vscale x 8 x i16>,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i32>  @intrinsic_vwmacc_mask_vv_nxv8i32_nxv8i16_nxv8i16(<vscale x 8 x i32> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv8i32_nxv8i16_nxv8i16:
@@ -456,7 +456,7 @@ entry:
     <vscale x 8 x i16> %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -488,7 +488,7 @@ declare <vscale x 16 x i32> @llvm.riscv.vwmacc.mask.nxv16i32.nxv16i16(
   <vscale x 16 x i16>,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i32>  @intrinsic_vwmacc_mask_vv_nxv16i32_nxv16i16_nxv16i16(<vscale x 16 x i32> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv16i32_nxv16i16_nxv16i16:
@@ -502,7 +502,7 @@ entry:
     <vscale x 16 x i16> %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i32> %a
 }
@@ -534,7 +534,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vwmacc.mask.nxv1i64.nxv1i32(
   <vscale x 1 x i32>,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i64>  @intrinsic_vwmacc_mask_vv_nxv1i64_nxv1i32_nxv1i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv1i64_nxv1i32_nxv1i32:
@@ -548,7 +548,7 @@ entry:
     <vscale x 1 x i32> %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -580,7 +580,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vwmacc.mask.nxv2i64.nxv2i32(
   <vscale x 2 x i32>,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i64>  @intrinsic_vwmacc_mask_vv_nxv2i64_nxv2i32_nxv2i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv2i64_nxv2i32_nxv2i32:
@@ -594,7 +594,7 @@ entry:
     <vscale x 2 x i32> %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -626,7 +626,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vwmacc.mask.nxv4i64.nxv4i32(
   <vscale x 4 x i32>,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i64>  @intrinsic_vwmacc_mask_vv_nxv4i64_nxv4i32_nxv4i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv4i64_nxv4i32_nxv4i32:
@@ -640,7 +640,7 @@ entry:
     <vscale x 4 x i32> %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -672,7 +672,7 @@ declare <vscale x 8 x i64> @llvm.riscv.vwmacc.mask.nxv8i64.nxv8i32(
   <vscale x 8 x i32>,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i64>  @intrinsic_vwmacc_mask_vv_nxv8i64_nxv8i32_nxv8i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv8i64_nxv8i32_nxv8i32:
@@ -686,7 +686,7 @@ entry:
     <vscale x 8 x i32> %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i64> %a
 }
@@ -718,7 +718,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.i8(
   i8,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i16> @intrinsic_vwmacc_mask_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv1i16_i8_nxv1i8:
@@ -732,7 +732,7 @@ entry:
     i8 %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -764,7 +764,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vwmacc.mask.nxv2i16.i8(
   i8,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i16> @intrinsic_vwmacc_mask_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv2i16_i8_nxv2i8:
@@ -778,7 +778,7 @@ entry:
     i8 %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -810,7 +810,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vwmacc.mask.nxv4i16.i8(
   i8,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i16> @intrinsic_vwmacc_mask_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv4i16_i8_nxv4i8:
@@ -824,7 +824,7 @@ entry:
     i8 %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -856,7 +856,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vwmacc.mask.nxv8i16.i8(
   i8,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i16> @intrinsic_vwmacc_mask_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv8i16_i8_nxv8i8:
@@ -870,7 +870,7 @@ entry:
     i8 %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -902,7 +902,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vwmacc.mask.nxv16i16.i8(
   i8,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i16> @intrinsic_vwmacc_mask_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv16i16_i8_nxv16i8:
@@ -916,7 +916,7 @@ entry:
     i8 %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -948,7 +948,7 @@ declare <vscale x 32 x i16> @llvm.riscv.vwmacc.mask.nxv32i16.i8(
   i8,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 32 x i16> @intrinsic_vwmacc_mask_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv32i16_i8_nxv32i8:
@@ -962,7 +962,7 @@ entry:
     i8 %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 32 x i16> %a
 }
@@ -994,7 +994,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vwmacc.mask.nxv1i32.i16(
   i16,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i32> @intrinsic_vwmacc_mask_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv1i32_i16_nxv1i16:
@@ -1008,7 +1008,7 @@ entry:
     i16 %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -1040,7 +1040,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vwmacc.mask.nxv2i32.i16(
   i16,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i32> @intrinsic_vwmacc_mask_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv2i32_i16_nxv2i16:
@@ -1054,7 +1054,7 @@ entry:
     i16 %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -1086,7 +1086,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vwmacc.mask.nxv4i32.i16(
   i16,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i32> @intrinsic_vwmacc_mask_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv4i32_i16_nxv4i16:
@@ -1100,7 +1100,7 @@ entry:
     i16 %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -1132,7 +1132,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vwmacc.mask.nxv8i32.i16(
   i16,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i32> @intrinsic_vwmacc_mask_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv8i32_i16_nxv8i16:
@@ -1146,7 +1146,7 @@ entry:
     i16 %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -1178,7 +1178,7 @@ declare <vscale x 16 x i32> @llvm.riscv.vwmacc.mask.nxv16i32.i16(
   i16,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i32> @intrinsic_vwmacc_mask_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv16i32_i16_nxv16i16:
@@ -1192,7 +1192,7 @@ entry:
     i16 %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i32> %a
 }
@@ -1224,7 +1224,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vwmacc.mask.nxv1i64.i32(
   i32,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i64> @intrinsic_vwmacc_mask_vx_nxv1i64_i32_nxv1i32(<vscale x 1 x i64> %0, i32 %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv1i64_i32_nxv1i32:
@@ -1238,7 +1238,7 @@ entry:
     i32 %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -1270,7 +1270,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vwmacc.mask.nxv2i64.i32(
   i32,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i64> @intrinsic_vwmacc_mask_vx_nxv2i64_i32_nxv2i32(<vscale x 2 x i64> %0, i32 %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv2i64_i32_nxv2i32:
@@ -1284,7 +1284,7 @@ entry:
     i32 %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -1316,7 +1316,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vwmacc.mask.nxv4i64.i32(
   i32,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i64> @intrinsic_vwmacc_mask_vx_nxv4i64_i32_nxv4i32(<vscale x 4 x i64> %0, i32 %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv4i64_i32_nxv4i32:
@@ -1330,7 +1330,7 @@ entry:
     i32 %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -1362,7 +1362,7 @@ declare <vscale x 8 x i64> @llvm.riscv.vwmacc.mask.nxv8i64.i32(
   i32,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i64> @intrinsic_vwmacc_mask_vx_nxv8i64_i32_nxv8i32(<vscale x 8 x i64> %0, i32 %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv8i64_i32_nxv8i32:
@@ -1376,7 +1376,7 @@ entry:
     i32 %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i64> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vwmacc-rv64.ll b/llvm/test/CodeGen/RISCV/rvv/vwmacc-rv64.ll
index b3922ccb1ce43e..2b547165170f86 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vwmacc-rv64.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vwmacc-rv64.ll
@@ -28,7 +28,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.nxv1i8(
   <vscale x 1 x i8>,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i16>  @intrinsic_vwmacc_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv1i16_nxv1i8_nxv1i8:
@@ -42,7 +42,7 @@ entry:
     <vscale x 1 x i8> %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -74,7 +74,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vwmacc.mask.nxv2i16.nxv2i8(
   <vscale x 2 x i8>,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i16>  @intrinsic_vwmacc_mask_vv_nxv2i16_nxv2i8_nxv2i8(<vscale x 2 x i16> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv2i16_nxv2i8_nxv2i8:
@@ -88,7 +88,7 @@ entry:
     <vscale x 2 x i8> %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -120,7 +120,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vwmacc.mask.nxv4i16.nxv4i8(
   <vscale x 4 x i8>,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i16>  @intrinsic_vwmacc_mask_vv_nxv4i16_nxv4i8_nxv4i8(<vscale x 4 x i16> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv4i16_nxv4i8_nxv4i8:
@@ -134,7 +134,7 @@ entry:
     <vscale x 4 x i8> %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -166,7 +166,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vwmacc.mask.nxv8i16.nxv8i8(
   <vscale x 8 x i8>,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i16>  @intrinsic_vwmacc_mask_vv_nxv8i16_nxv8i8_nxv8i8(<vscale x 8 x i16> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv8i16_nxv8i8_nxv8i8:
@@ -180,7 +180,7 @@ entry:
     <vscale x 8 x i8> %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -212,7 +212,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vwmacc.mask.nxv16i16.nxv16i8(
   <vscale x 16 x i8>,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i16>  @intrinsic_vwmacc_mask_vv_nxv16i16_nxv16i8_nxv16i8(<vscale x 16 x i16> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv16i16_nxv16i8_nxv16i8:
@@ -226,7 +226,7 @@ entry:
     <vscale x 16 x i8> %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -258,7 +258,7 @@ declare <vscale x 32 x i16> @llvm.riscv.vwmacc.mask.nxv32i16.nxv32i8(
   <vscale x 32 x i8>,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 32 x i16>  @intrinsic_vwmacc_mask_vv_nxv32i16_nxv32i8_nxv32i8(<vscale x 32 x i16> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv32i16_nxv32i8_nxv32i8:
@@ -272,7 +272,7 @@ entry:
     <vscale x 32 x i8> %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 32 x i16> %a
 }
@@ -304,7 +304,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vwmacc.mask.nxv1i32.nxv1i16(
   <vscale x 1 x i16>,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i32>  @intrinsic_vwmacc_mask_vv_nxv1i32_nxv1i16_nxv1i16(<vscale x 1 x i32> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv1i32_nxv1i16_nxv1i16:
@@ -318,7 +318,7 @@ entry:
     <vscale x 1 x i16> %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -350,7 +350,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vwmacc.mask.nxv2i32.nxv2i16(
   <vscale x 2 x i16>,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i32>  @intrinsic_vwmacc_mask_vv_nxv2i32_nxv2i16_nxv2i16(<vscale x 2 x i32> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv2i32_nxv2i16_nxv2i16:
@@ -364,7 +364,7 @@ entry:
     <vscale x 2 x i16> %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -396,7 +396,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vwmacc.mask.nxv4i32.nxv4i16(
   <vscale x 4 x i16>,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i32>  @intrinsic_vwmacc_mask_vv_nxv4i32_nxv4i16_nxv4i16(<vscale x 4 x i32> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv4i32_nxv4i16_nxv4i16:
@@ -410,7 +410,7 @@ entry:
     <vscale x 4 x i16> %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -442,7 +442,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vwmacc.mask.nxv8i32.nxv8i16(
   <vscale x 8 x i16>,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i32>  @intrinsic_vwmacc_mask_vv_nxv8i32_nxv8i16_nxv8i16(<vscale x 8 x i32> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv8i32_nxv8i16_nxv8i16:
@@ -456,7 +456,7 @@ entry:
     <vscale x 8 x i16> %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -488,7 +488,7 @@ declare <vscale x 16 x i32> @llvm.riscv.vwmacc.mask.nxv16i32.nxv16i16(
   <vscale x 16 x i16>,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i32>  @intrinsic_vwmacc_mask_vv_nxv16i32_nxv16i16_nxv16i16(<vscale x 16 x i32> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv16i32_nxv16i16_nxv16i16:
@@ -502,7 +502,7 @@ entry:
     <vscale x 16 x i16> %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i32> %a
 }
@@ -534,7 +534,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vwmacc.mask.nxv1i64.nxv1i32(
   <vscale x 1 x i32>,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i64>  @intrinsic_vwmacc_mask_vv_nxv1i64_nxv1i32_nxv1i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv1i64_nxv1i32_nxv1i32:
@@ -548,7 +548,7 @@ entry:
     <vscale x 1 x i32> %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -580,7 +580,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vwmacc.mask.nxv2i64.nxv2i32(
   <vscale x 2 x i32>,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i64>  @intrinsic_vwmacc_mask_vv_nxv2i64_nxv2i32_nxv2i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv2i64_nxv2i32_nxv2i32:
@@ -594,7 +594,7 @@ entry:
     <vscale x 2 x i32> %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -626,7 +626,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vwmacc.mask.nxv4i64.nxv4i32(
   <vscale x 4 x i32>,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i64>  @intrinsic_vwmacc_mask_vv_nxv4i64_nxv4i32_nxv4i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv4i64_nxv4i32_nxv4i32:
@@ -640,7 +640,7 @@ entry:
     <vscale x 4 x i32> %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -672,7 +672,7 @@ declare <vscale x 8 x i64> @llvm.riscv.vwmacc.mask.nxv8i64.nxv8i32(
   <vscale x 8 x i32>,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i64>  @intrinsic_vwmacc_mask_vv_nxv8i64_nxv8i32_nxv8i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv8i64_nxv8i32_nxv8i32:
@@ -686,7 +686,7 @@ entry:
     <vscale x 8 x i32> %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i64> %a
 }
@@ -718,7 +718,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.i8(
   i8,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i16> @intrinsic_vwmacc_mask_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv1i16_i8_nxv1i8:
@@ -732,7 +732,7 @@ entry:
     i8 %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -764,7 +764,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vwmacc.mask.nxv2i16.i8(
   i8,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i16> @intrinsic_vwmacc_mask_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv2i16_i8_nxv2i8:
@@ -778,7 +778,7 @@ entry:
     i8 %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -810,7 +810,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vwmacc.mask.nxv4i16.i8(
   i8,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i16> @intrinsic_vwmacc_mask_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv4i16_i8_nxv4i8:
@@ -824,7 +824,7 @@ entry:
     i8 %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -856,7 +856,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vwmacc.mask.nxv8i16.i8(
   i8,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i16> @intrinsic_vwmacc_mask_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv8i16_i8_nxv8i8:
@@ -870,7 +870,7 @@ entry:
     i8 %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -902,7 +902,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vwmacc.mask.nxv16i16.i8(
   i8,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i16> @intrinsic_vwmacc_mask_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv16i16_i8_nxv16i8:
@@ -916,7 +916,7 @@ entry:
     i8 %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -948,7 +948,7 @@ declare <vscale x 32 x i16> @llvm.riscv.vwmacc.mask.nxv32i16.i8(
   i8,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 32 x i16> @intrinsic_vwmacc_mask_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv32i16_i8_nxv32i8:
@@ -962,7 +962,7 @@ entry:
     i8 %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 32 x i16> %a
 }
@@ -994,7 +994,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vwmacc.mask.nxv1i32.i16(
   i16,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i32> @intrinsic_vwmacc_mask_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv1i32_i16_nxv1i16:
@@ -1008,7 +1008,7 @@ entry:
     i16 %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -1040,7 +1040,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vwmacc.mask.nxv2i32.i16(
   i16,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i32> @intrinsic_vwmacc_mask_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv2i32_i16_nxv2i16:
@@ -1054,7 +1054,7 @@ entry:
     i16 %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -1086,7 +1086,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vwmacc.mask.nxv4i32.i16(
   i16,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i32> @intrinsic_vwmacc_mask_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv4i32_i16_nxv4i16:
@@ -1100,7 +1100,7 @@ entry:
     i16 %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -1132,7 +1132,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vwmacc.mask.nxv8i32.i16(
   i16,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i32> @intrinsic_vwmacc_mask_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv8i32_i16_nxv8i16:
@@ -1146,7 +1146,7 @@ entry:
     i16 %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -1178,7 +1178,7 @@ declare <vscale x 16 x i32> @llvm.riscv.vwmacc.mask.nxv16i32.i16(
   i16,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i32> @intrinsic_vwmacc_mask_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv16i32_i16_nxv16i16:
@@ -1192,7 +1192,7 @@ entry:
     i16 %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i32> %a
 }
@@ -1224,7 +1224,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vwmacc.mask.nxv1i64.i32(
   i32,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i64> @intrinsic_vwmacc_mask_vx_nxv1i64_i32_nxv1i32(<vscale x 1 x i64> %0, i32 %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv1i64_i32_nxv1i32:
@@ -1238,7 +1238,7 @@ entry:
     i32 %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -1270,7 +1270,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vwmacc.mask.nxv2i64.i32(
   i32,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i64> @intrinsic_vwmacc_mask_vx_nxv2i64_i32_nxv2i32(<vscale x 2 x i64> %0, i32 %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv2i64_i32_nxv2i32:
@@ -1284,7 +1284,7 @@ entry:
     i32 %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -1316,7 +1316,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vwmacc.mask.nxv4i64.i32(
   i32,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i64> @intrinsic_vwmacc_mask_vx_nxv4i64_i32_nxv4i32(<vscale x 4 x i64> %0, i32 %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv4i64_i32_nxv4i32:
@@ -1330,7 +1330,7 @@ entry:
     i32 %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -1362,7 +1362,7 @@ declare <vscale x 8 x i64> @llvm.riscv.vwmacc.mask.nxv8i64.i32(
   i32,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i64> @intrinsic_vwmacc_mask_vx_nxv8i64_i32_nxv8i32(<vscale x 8 x i64> %0, i32 %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv8i64_i32_nxv8i32:
@@ -1376,7 +1376,7 @@ entry:
     i32 %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i64> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vwmaccsu-rv32.ll b/llvm/test/CodeGen/RISCV/rvv/vwmaccsu-rv32.ll
index 3b3ed3ce683ab8..1522eb1e86b21f 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vwmaccsu-rv32.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vwmaccsu-rv32.ll
@@ -28,7 +28,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vwmaccsu.mask.nxv1i16.nxv1i8(
   <vscale x 1 x i8>,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv1i16_nxv1i8_nxv1i8:
@@ -42,7 +42,7 @@ entry:
     <vscale x 1 x i8> %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -74,7 +74,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vwmaccsu.mask.nxv2i16.nxv2i8(
   <vscale x 2 x i8>,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv2i16_nxv2i8_nxv2i8(<vscale x 2 x i16> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv2i16_nxv2i8_nxv2i8:
@@ -88,7 +88,7 @@ entry:
     <vscale x 2 x i8> %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -120,7 +120,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vwmaccsu.mask.nxv4i16.nxv4i8(
   <vscale x 4 x i8>,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv4i16_nxv4i8_nxv4i8(<vscale x 4 x i16> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv4i16_nxv4i8_nxv4i8:
@@ -134,7 +134,7 @@ entry:
     <vscale x 4 x i8> %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -166,7 +166,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vwmaccsu.mask.nxv8i16.nxv8i8(
   <vscale x 8 x i8>,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv8i16_nxv8i8_nxv8i8(<vscale x 8 x i16> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv8i16_nxv8i8_nxv8i8:
@@ -180,7 +180,7 @@ entry:
     <vscale x 8 x i8> %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -212,7 +212,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vwmaccsu.mask.nxv16i16.nxv16i8(
   <vscale x 16 x i8>,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv16i16_nxv16i8_nxv16i8(<vscale x 16 x i16> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv16i16_nxv16i8_nxv16i8:
@@ -226,7 +226,7 @@ entry:
     <vscale x 16 x i8> %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -258,7 +258,7 @@ declare <vscale x 32 x i16> @llvm.riscv.vwmaccsu.mask.nxv32i16.nxv32i8(
   <vscale x 32 x i8>,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 32 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv32i16_nxv32i8_nxv32i8(<vscale x 32 x i16> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv32i16_nxv32i8_nxv32i8:
@@ -272,7 +272,7 @@ entry:
     <vscale x 32 x i8> %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 32 x i16> %a
 }
@@ -304,7 +304,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vwmaccsu.mask.nxv1i32.nxv1i16(
   <vscale x 1 x i16>,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i32>  @intrinsic_vwmaccsu_mask_vv_nxv1i32_nxv1i16_nxv1i16(<vscale x 1 x i32> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv1i32_nxv1i16_nxv1i16:
@@ -318,7 +318,7 @@ entry:
     <vscale x 1 x i16> %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -350,7 +350,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vwmaccsu.mask.nxv2i32.nxv2i16(
   <vscale x 2 x i16>,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i32>  @intrinsic_vwmaccsu_mask_vv_nxv2i32_nxv2i16_nxv2i16(<vscale x 2 x i32> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv2i32_nxv2i16_nxv2i16:
@@ -364,7 +364,7 @@ entry:
     <vscale x 2 x i16> %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -396,7 +396,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vwmaccsu.mask.nxv4i32.nxv4i16(
   <vscale x 4 x i16>,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i32>  @intrinsic_vwmaccsu_mask_vv_nxv4i32_nxv4i16_nxv4i16(<vscale x 4 x i32> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv4i32_nxv4i16_nxv4i16:
@@ -410,7 +410,7 @@ entry:
     <vscale x 4 x i16> %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -442,7 +442,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vwmaccsu.mask.nxv8i32.nxv8i16(
   <vscale x 8 x i16>,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i32>  @intrinsic_vwmaccsu_mask_vv_nxv8i32_nxv8i16_nxv8i16(<vscale x 8 x i32> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv8i32_nxv8i16_nxv8i16:
@@ -456,7 +456,7 @@ entry:
     <vscale x 8 x i16> %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -488,7 +488,7 @@ declare <vscale x 16 x i32> @llvm.riscv.vwmaccsu.mask.nxv16i32.nxv16i16(
   <vscale x 16 x i16>,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i32>  @intrinsic_vwmaccsu_mask_vv_nxv16i32_nxv16i16_nxv16i16(<vscale x 16 x i32> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv16i32_nxv16i16_nxv16i16:
@@ -502,7 +502,7 @@ entry:
     <vscale x 16 x i16> %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i32> %a
 }
@@ -534,7 +534,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vwmaccsu.mask.nxv1i64.nxv1i32(
   <vscale x 1 x i32>,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i64>  @intrinsic_vwmaccsu_mask_vv_nxv1i64_nxv1i32_nxv1i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv1i64_nxv1i32_nxv1i32:
@@ -548,7 +548,7 @@ entry:
     <vscale x 1 x i32> %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -580,7 +580,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vwmaccsu.mask.nxv2i64.nxv2i32(
   <vscale x 2 x i32>,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i64>  @intrinsic_vwmaccsu_mask_vv_nxv2i64_nxv2i32_nxv2i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv2i64_nxv2i32_nxv2i32:
@@ -594,7 +594,7 @@ entry:
     <vscale x 2 x i32> %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -626,7 +626,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vwmaccsu.mask.nxv4i64.nxv4i32(
   <vscale x 4 x i32>,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i64>  @intrinsic_vwmaccsu_mask_vv_nxv4i64_nxv4i32_nxv4i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv4i64_nxv4i32_nxv4i32:
@@ -640,7 +640,7 @@ entry:
     <vscale x 4 x i32> %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -672,7 +672,7 @@ declare <vscale x 8 x i64> @llvm.riscv.vwmaccsu.mask.nxv8i64.nxv8i32(
   <vscale x 8 x i32>,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i64>  @intrinsic_vwmaccsu_mask_vv_nxv8i64_nxv8i32_nxv8i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv8i64_nxv8i32_nxv8i32:
@@ -686,7 +686,7 @@ entry:
     <vscale x 8 x i32> %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i64> %a
 }
@@ -718,7 +718,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vwmaccsu.mask.nxv1i16.i8(
   i8,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i16> @intrinsic_vwmaccsu_mask_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv1i16_i8_nxv1i8:
@@ -732,7 +732,7 @@ entry:
     i8 %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -764,7 +764,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vwmaccsu.mask.nxv2i16.i8(
   i8,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i16> @intrinsic_vwmaccsu_mask_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv2i16_i8_nxv2i8:
@@ -778,7 +778,7 @@ entry:
     i8 %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -810,7 +810,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vwmaccsu.mask.nxv4i16.i8(
   i8,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i16> @intrinsic_vwmaccsu_mask_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv4i16_i8_nxv4i8:
@@ -824,7 +824,7 @@ entry:
     i8 %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -856,7 +856,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vwmaccsu.mask.nxv8i16.i8(
   i8,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i16> @intrinsic_vwmaccsu_mask_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv8i16_i8_nxv8i8:
@@ -870,7 +870,7 @@ entry:
     i8 %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -902,7 +902,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vwmaccsu.mask.nxv16i16.i8(
   i8,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i16> @intrinsic_vwmaccsu_mask_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv16i16_i8_nxv16i8:
@@ -916,7 +916,7 @@ entry:
     i8 %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -948,7 +948,7 @@ declare <vscale x 32 x i16> @llvm.riscv.vwmaccsu.mask.nxv32i16.i8(
   i8,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 32 x i16> @intrinsic_vwmaccsu_mask_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv32i16_i8_nxv32i8:
@@ -962,7 +962,7 @@ entry:
     i8 %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 32 x i16> %a
 }
@@ -994,7 +994,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vwmaccsu.mask.nxv1i32.i16(
   i16,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i32> @intrinsic_vwmaccsu_mask_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv1i32_i16_nxv1i16:
@@ -1008,7 +1008,7 @@ entry:
     i16 %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -1040,7 +1040,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vwmaccsu.mask.nxv2i32.i16(
   i16,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i32> @intrinsic_vwmaccsu_mask_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv2i32_i16_nxv2i16:
@@ -1054,7 +1054,7 @@ entry:
     i16 %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -1086,7 +1086,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vwmaccsu.mask.nxv4i32.i16(
   i16,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i32> @intrinsic_vwmaccsu_mask_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv4i32_i16_nxv4i16:
@@ -1100,7 +1100,7 @@ entry:
     i16 %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -1132,7 +1132,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vwmaccsu.mask.nxv8i32.i16(
   i16,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i32> @intrinsic_vwmaccsu_mask_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv8i32_i16_nxv8i16:
@@ -1146,7 +1146,7 @@ entry:
     i16 %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -1178,7 +1178,7 @@ declare <vscale x 16 x i32> @llvm.riscv.vwmaccsu.mask.nxv16i32.i16(
   i16,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i32> @intrinsic_vwmaccsu_mask_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv16i32_i16_nxv16i16:
@@ -1192,7 +1192,7 @@ entry:
     i16 %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i32> %a
 }
@@ -1224,7 +1224,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vwmaccsu.mask.nxv1i64.i32(
   i32,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i64> @intrinsic_vwmaccsu_mask_vx_nxv1i64_i32_nxv1i32(<vscale x 1 x i64> %0, i32 %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv1i64_i32_nxv1i32:
@@ -1238,7 +1238,7 @@ entry:
     i32 %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -1270,7 +1270,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vwmaccsu.mask.nxv2i64.i32(
   i32,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i64> @intrinsic_vwmaccsu_mask_vx_nxv2i64_i32_nxv2i32(<vscale x 2 x i64> %0, i32 %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv2i64_i32_nxv2i32:
@@ -1284,7 +1284,7 @@ entry:
     i32 %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -1316,7 +1316,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vwmaccsu.mask.nxv4i64.i32(
   i32,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i64> @intrinsic_vwmaccsu_mask_vx_nxv4i64_i32_nxv4i32(<vscale x 4 x i64> %0, i32 %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv4i64_i32_nxv4i32:
@@ -1330,7 +1330,7 @@ entry:
     i32 %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -1362,7 +1362,7 @@ declare <vscale x 8 x i64> @llvm.riscv.vwmaccsu.mask.nxv8i64.i32(
   i32,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i64> @intrinsic_vwmaccsu_mask_vx_nxv8i64_i32_nxv8i32(<vscale x 8 x i64> %0, i32 %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv8i64_i32_nxv8i32:
@@ -1376,7 +1376,7 @@ entry:
     i32 %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i64> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vwmaccsu-rv64.ll b/llvm/test/CodeGen/RISCV/rvv/vwmaccsu-rv64.ll
index 07f11b1ef16154..ab86a96db8bd0e 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vwmaccsu-rv64.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vwmaccsu-rv64.ll
@@ -28,7 +28,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vwmaccsu.mask.nxv1i16.nxv1i8(
   <vscale x 1 x i8>,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv1i16_nxv1i8_nxv1i8:
@@ -42,7 +42,7 @@ entry:
     <vscale x 1 x i8> %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -74,7 +74,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vwmaccsu.mask.nxv2i16.nxv2i8(
   <vscale x 2 x i8>,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv2i16_nxv2i8_nxv2i8(<vscale x 2 x i16> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv2i16_nxv2i8_nxv2i8:
@@ -88,7 +88,7 @@ entry:
     <vscale x 2 x i8> %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -120,7 +120,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vwmaccsu.mask.nxv4i16.nxv4i8(
   <vscale x 4 x i8>,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv4i16_nxv4i8_nxv4i8(<vscale x 4 x i16> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv4i16_nxv4i8_nxv4i8:
@@ -134,7 +134,7 @@ entry:
     <vscale x 4 x i8> %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -166,7 +166,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vwmaccsu.mask.nxv8i16.nxv8i8(
   <vscale x 8 x i8>,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv8i16_nxv8i8_nxv8i8(<vscale x 8 x i16> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv8i16_nxv8i8_nxv8i8:
@@ -180,7 +180,7 @@ entry:
     <vscale x 8 x i8> %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -212,7 +212,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vwmaccsu.mask.nxv16i16.nxv16i8(
   <vscale x 16 x i8>,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv16i16_nxv16i8_nxv16i8(<vscale x 16 x i16> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv16i16_nxv16i8_nxv16i8:
@@ -226,7 +226,7 @@ entry:
     <vscale x 16 x i8> %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -258,7 +258,7 @@ declare <vscale x 32 x i16> @llvm.riscv.vwmaccsu.mask.nxv32i16.nxv32i8(
   <vscale x 32 x i8>,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 32 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv32i16_nxv32i8_nxv32i8(<vscale x 32 x i16> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv32i16_nxv32i8_nxv32i8:
@@ -272,7 +272,7 @@ entry:
     <vscale x 32 x i8> %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 32 x i16> %a
 }
@@ -304,7 +304,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vwmaccsu.mask.nxv1i32.nxv1i16(
   <vscale x 1 x i16>,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i32>  @intrinsic_vwmaccsu_mask_vv_nxv1i32_nxv1i16_nxv1i16(<vscale x 1 x i32> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv1i32_nxv1i16_nxv1i16:
@@ -318,7 +318,7 @@ entry:
     <vscale x 1 x i16> %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -350,7 +350,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vwmaccsu.mask.nxv2i32.nxv2i16(
   <vscale x 2 x i16>,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i32>  @intrinsic_vwmaccsu_mask_vv_nxv2i32_nxv2i16_nxv2i16(<vscale x 2 x i32> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv2i32_nxv2i16_nxv2i16:
@@ -364,7 +364,7 @@ entry:
     <vscale x 2 x i16> %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -396,7 +396,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vwmaccsu.mask.nxv4i32.nxv4i16(
   <vscale x 4 x i16>,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i32>  @intrinsic_vwmaccsu_mask_vv_nxv4i32_nxv4i16_nxv4i16(<vscale x 4 x i32> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv4i32_nxv4i16_nxv4i16:
@@ -410,7 +410,7 @@ entry:
     <vscale x 4 x i16> %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -442,7 +442,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vwmaccsu.mask.nxv8i32.nxv8i16(
   <vscale x 8 x i16>,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i32>  @intrinsic_vwmaccsu_mask_vv_nxv8i32_nxv8i16_nxv8i16(<vscale x 8 x i32> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv8i32_nxv8i16_nxv8i16:
@@ -456,7 +456,7 @@ entry:
     <vscale x 8 x i16> %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -488,7 +488,7 @@ declare <vscale x 16 x i32> @llvm.riscv.vwmaccsu.mask.nxv16i32.nxv16i16(
   <vscale x 16 x i16>,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i32>  @intrinsic_vwmaccsu_mask_vv_nxv16i32_nxv16i16_nxv16i16(<vscale x 16 x i32> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv16i32_nxv16i16_nxv16i16:
@@ -502,7 +502,7 @@ entry:
     <vscale x 16 x i16> %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i32> %a
 }
@@ -534,7 +534,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vwmaccsu.mask.nxv1i64.nxv1i32(
   <vscale x 1 x i32>,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i64>  @intrinsic_vwmaccsu_mask_vv_nxv1i64_nxv1i32_nxv1i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv1i64_nxv1i32_nxv1i32:
@@ -548,7 +548,7 @@ entry:
     <vscale x 1 x i32> %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -580,7 +580,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vwmaccsu.mask.nxv2i64.nxv2i32(
   <vscale x 2 x i32>,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i64>  @intrinsic_vwmaccsu_mask_vv_nxv2i64_nxv2i32_nxv2i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv2i64_nxv2i32_nxv2i32:
@@ -594,7 +594,7 @@ entry:
     <vscale x 2 x i32> %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -626,7 +626,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vwmaccsu.mask.nxv4i64.nxv4i32(
   <vscale x 4 x i32>,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i64>  @intrinsic_vwmaccsu_mask_vv_nxv4i64_nxv4i32_nxv4i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv4i64_nxv4i32_nxv4i32:
@@ -640,7 +640,7 @@ entry:
     <vscale x 4 x i32> %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -672,7 +672,7 @@ declare <vscale x 8 x i64> @llvm.riscv.vwmaccsu.mask.nxv8i64.nxv8i32(
   <vscale x 8 x i32>,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i64>  @intrinsic_vwmaccsu_mask_vv_nxv8i64_nxv8i32_nxv8i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv8i64_nxv8i32_nxv8i32:
@@ -686,7 +686,7 @@ entry:
     <vscale x 8 x i32> %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i64> %a
 }
@@ -718,7 +718,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vwmaccsu.mask.nxv1i16.i8(
   i8,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i16> @intrinsic_vwmaccsu_mask_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv1i16_i8_nxv1i8:
@@ -732,7 +732,7 @@ entry:
     i8 %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -764,7 +764,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vwmaccsu.mask.nxv2i16.i8(
   i8,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i16> @intrinsic_vwmaccsu_mask_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv2i16_i8_nxv2i8:
@@ -778,7 +778,7 @@ entry:
     i8 %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -810,7 +810,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vwmaccsu.mask.nxv4i16.i8(
   i8,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i16> @intrinsic_vwmaccsu_mask_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv4i16_i8_nxv4i8:
@@ -824,7 +824,7 @@ entry:
     i8 %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -856,7 +856,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vwmaccsu.mask.nxv8i16.i8(
   i8,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i16> @intrinsic_vwmaccsu_mask_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv8i16_i8_nxv8i8:
@@ -870,7 +870,7 @@ entry:
     i8 %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -902,7 +902,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vwmaccsu.mask.nxv16i16.i8(
   i8,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i16> @intrinsic_vwmaccsu_mask_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv16i16_i8_nxv16i8:
@@ -916,7 +916,7 @@ entry:
     i8 %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -948,7 +948,7 @@ declare <vscale x 32 x i16> @llvm.riscv.vwmaccsu.mask.nxv32i16.i8(
   i8,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 32 x i16> @intrinsic_vwmaccsu_mask_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv32i16_i8_nxv32i8:
@@ -962,7 +962,7 @@ entry:
     i8 %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 32 x i16> %a
 }
@@ -994,7 +994,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vwmaccsu.mask.nxv1i32.i16(
   i16,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i32> @intrinsic_vwmaccsu_mask_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv1i32_i16_nxv1i16:
@@ -1008,7 +1008,7 @@ entry:
     i16 %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -1040,7 +1040,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vwmaccsu.mask.nxv2i32.i16(
   i16,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i32> @intrinsic_vwmaccsu_mask_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv2i32_i16_nxv2i16:
@@ -1054,7 +1054,7 @@ entry:
     i16 %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -1086,7 +1086,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vwmaccsu.mask.nxv4i32.i16(
   i16,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i32> @intrinsic_vwmaccsu_mask_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv4i32_i16_nxv4i16:
@@ -1100,7 +1100,7 @@ entry:
     i16 %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -1132,7 +1132,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vwmaccsu.mask.nxv8i32.i16(
   i16,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i32> @intrinsic_vwmaccsu_mask_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv8i32_i16_nxv8i16:
@@ -1146,7 +1146,7 @@ entry:
     i16 %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -1178,7 +1178,7 @@ declare <vscale x 16 x i32> @llvm.riscv.vwmaccsu.mask.nxv16i32.i16(
   i16,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i32> @intrinsic_vwmaccsu_mask_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv16i32_i16_nxv16i16:
@@ -1192,7 +1192,7 @@ entry:
     i16 %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i32> %a
 }
@@ -1224,7 +1224,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vwmaccsu.mask.nxv1i64.i32(
   i32,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i64> @intrinsic_vwmaccsu_mask_vx_nxv1i64_i32_nxv1i32(<vscale x 1 x i64> %0, i32 %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv1i64_i32_nxv1i32:
@@ -1238,7 +1238,7 @@ entry:
     i32 %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -1270,7 +1270,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vwmaccsu.mask.nxv2i64.i32(
   i32,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i64> @intrinsic_vwmaccsu_mask_vx_nxv2i64_i32_nxv2i32(<vscale x 2 x i64> %0, i32 %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv2i64_i32_nxv2i32:
@@ -1284,7 +1284,7 @@ entry:
     i32 %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -1316,7 +1316,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vwmaccsu.mask.nxv4i64.i32(
   i32,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i64> @intrinsic_vwmaccsu_mask_vx_nxv4i64_i32_nxv4i32(<vscale x 4 x i64> %0, i32 %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv4i64_i32_nxv4i32:
@@ -1330,7 +1330,7 @@ entry:
     i32 %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -1362,7 +1362,7 @@ declare <vscale x 8 x i64> @llvm.riscv.vwmaccsu.mask.nxv8i64.i32(
   i32,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i64> @intrinsic_vwmaccsu_mask_vx_nxv8i64_i32_nxv8i32(<vscale x 8 x i64> %0, i32 %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv8i64_i32_nxv8i32:
@@ -1376,7 +1376,7 @@ entry:
     i32 %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i64> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vwmaccu-rv32.ll b/llvm/test/CodeGen/RISCV/rvv/vwmaccu-rv32.ll
index 4d7057a8f81a21..183fd36188a908 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vwmaccu-rv32.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vwmaccu-rv32.ll
@@ -28,7 +28,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vwmaccu.mask.nxv1i16.nxv1i8(
   <vscale x 1 x i8>,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i16>  @intrinsic_vwmaccu_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv1i16_nxv1i8_nxv1i8:
@@ -42,7 +42,7 @@ entry:
     <vscale x 1 x i8> %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -74,7 +74,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vwmaccu.mask.nxv2i16.nxv2i8(
   <vscale x 2 x i8>,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i16>  @intrinsic_vwmaccu_mask_vv_nxv2i16_nxv2i8_nxv2i8(<vscale x 2 x i16> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv2i16_nxv2i8_nxv2i8:
@@ -88,7 +88,7 @@ entry:
     <vscale x 2 x i8> %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -120,7 +120,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vwmaccu.mask.nxv4i16.nxv4i8(
   <vscale x 4 x i8>,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i16>  @intrinsic_vwmaccu_mask_vv_nxv4i16_nxv4i8_nxv4i8(<vscale x 4 x i16> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv4i16_nxv4i8_nxv4i8:
@@ -134,7 +134,7 @@ entry:
     <vscale x 4 x i8> %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -166,7 +166,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vwmaccu.mask.nxv8i16.nxv8i8(
   <vscale x 8 x i8>,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i16>  @intrinsic_vwmaccu_mask_vv_nxv8i16_nxv8i8_nxv8i8(<vscale x 8 x i16> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv8i16_nxv8i8_nxv8i8:
@@ -180,7 +180,7 @@ entry:
     <vscale x 8 x i8> %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -212,7 +212,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vwmaccu.mask.nxv16i16.nxv16i8(
   <vscale x 16 x i8>,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i16>  @intrinsic_vwmaccu_mask_vv_nxv16i16_nxv16i8_nxv16i8(<vscale x 16 x i16> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv16i16_nxv16i8_nxv16i8:
@@ -226,7 +226,7 @@ entry:
     <vscale x 16 x i8> %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -258,7 +258,7 @@ declare <vscale x 32 x i16> @llvm.riscv.vwmaccu.mask.nxv32i16.nxv32i8(
   <vscale x 32 x i8>,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 32 x i16>  @intrinsic_vwmaccu_mask_vv_nxv32i16_nxv32i8_nxv32i8(<vscale x 32 x i16> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv32i16_nxv32i8_nxv32i8:
@@ -272,7 +272,7 @@ entry:
     <vscale x 32 x i8> %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 32 x i16> %a
 }
@@ -304,7 +304,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vwmaccu.mask.nxv1i32.nxv1i16(
   <vscale x 1 x i16>,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i32>  @intrinsic_vwmaccu_mask_vv_nxv1i32_nxv1i16_nxv1i16(<vscale x 1 x i32> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv1i32_nxv1i16_nxv1i16:
@@ -318,7 +318,7 @@ entry:
     <vscale x 1 x i16> %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -350,7 +350,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vwmaccu.mask.nxv2i32.nxv2i16(
   <vscale x 2 x i16>,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i32>  @intrinsic_vwmaccu_mask_vv_nxv2i32_nxv2i16_nxv2i16(<vscale x 2 x i32> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv2i32_nxv2i16_nxv2i16:
@@ -364,7 +364,7 @@ entry:
     <vscale x 2 x i16> %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -396,7 +396,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vwmaccu.mask.nxv4i32.nxv4i16(
   <vscale x 4 x i16>,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i32>  @intrinsic_vwmaccu_mask_vv_nxv4i32_nxv4i16_nxv4i16(<vscale x 4 x i32> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv4i32_nxv4i16_nxv4i16:
@@ -410,7 +410,7 @@ entry:
     <vscale x 4 x i16> %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -442,7 +442,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vwmaccu.mask.nxv8i32.nxv8i16(
   <vscale x 8 x i16>,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i32>  @intrinsic_vwmaccu_mask_vv_nxv8i32_nxv8i16_nxv8i16(<vscale x 8 x i32> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv8i32_nxv8i16_nxv8i16:
@@ -456,7 +456,7 @@ entry:
     <vscale x 8 x i16> %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -488,7 +488,7 @@ declare <vscale x 16 x i32> @llvm.riscv.vwmaccu.mask.nxv16i32.nxv16i16(
   <vscale x 16 x i16>,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i32>  @intrinsic_vwmaccu_mask_vv_nxv16i32_nxv16i16_nxv16i16(<vscale x 16 x i32> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv16i32_nxv16i16_nxv16i16:
@@ -502,7 +502,7 @@ entry:
     <vscale x 16 x i16> %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i32> %a
 }
@@ -534,7 +534,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vwmaccu.mask.nxv1i64.nxv1i32(
   <vscale x 1 x i32>,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i64>  @intrinsic_vwmaccu_mask_vv_nxv1i64_nxv1i32_nxv1i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv1i64_nxv1i32_nxv1i32:
@@ -548,7 +548,7 @@ entry:
     <vscale x 1 x i32> %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -580,7 +580,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vwmaccu.mask.nxv2i64.nxv2i32(
   <vscale x 2 x i32>,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i64>  @intrinsic_vwmaccu_mask_vv_nxv2i64_nxv2i32_nxv2i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv2i64_nxv2i32_nxv2i32:
@@ -594,7 +594,7 @@ entry:
     <vscale x 2 x i32> %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -626,7 +626,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vwmaccu.mask.nxv4i64.nxv4i32(
   <vscale x 4 x i32>,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i64>  @intrinsic_vwmaccu_mask_vv_nxv4i64_nxv4i32_nxv4i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv4i64_nxv4i32_nxv4i32:
@@ -640,7 +640,7 @@ entry:
     <vscale x 4 x i32> %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -672,7 +672,7 @@ declare <vscale x 8 x i64> @llvm.riscv.vwmaccu.mask.nxv8i64.nxv8i32(
   <vscale x 8 x i32>,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i64>  @intrinsic_vwmaccu_mask_vv_nxv8i64_nxv8i32_nxv8i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv8i64_nxv8i32_nxv8i32:
@@ -686,7 +686,7 @@ entry:
     <vscale x 8 x i32> %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i64> %a
 }
@@ -718,7 +718,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vwmaccu.mask.nxv1i16.i8(
   i8,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i16> @intrinsic_vwmaccu_mask_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv1i16_i8_nxv1i8:
@@ -732,7 +732,7 @@ entry:
     i8 %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -764,7 +764,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vwmaccu.mask.nxv2i16.i8(
   i8,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i16> @intrinsic_vwmaccu_mask_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv2i16_i8_nxv2i8:
@@ -778,7 +778,7 @@ entry:
     i8 %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -810,7 +810,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vwmaccu.mask.nxv4i16.i8(
   i8,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i16> @intrinsic_vwmaccu_mask_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv4i16_i8_nxv4i8:
@@ -824,7 +824,7 @@ entry:
     i8 %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -856,7 +856,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vwmaccu.mask.nxv8i16.i8(
   i8,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i16> @intrinsic_vwmaccu_mask_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv8i16_i8_nxv8i8:
@@ -870,7 +870,7 @@ entry:
     i8 %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -902,7 +902,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vwmaccu.mask.nxv16i16.i8(
   i8,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i16> @intrinsic_vwmaccu_mask_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv16i16_i8_nxv16i8:
@@ -916,7 +916,7 @@ entry:
     i8 %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -948,7 +948,7 @@ declare <vscale x 32 x i16> @llvm.riscv.vwmaccu.mask.nxv32i16.i8(
   i8,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 32 x i16> @intrinsic_vwmaccu_mask_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv32i16_i8_nxv32i8:
@@ -962,7 +962,7 @@ entry:
     i8 %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 32 x i16> %a
 }
@@ -994,7 +994,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vwmaccu.mask.nxv1i32.i16(
   i16,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i32> @intrinsic_vwmaccu_mask_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv1i32_i16_nxv1i16:
@@ -1008,7 +1008,7 @@ entry:
     i16 %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -1040,7 +1040,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vwmaccu.mask.nxv2i32.i16(
   i16,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i32> @intrinsic_vwmaccu_mask_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv2i32_i16_nxv2i16:
@@ -1054,7 +1054,7 @@ entry:
     i16 %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -1086,7 +1086,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vwmaccu.mask.nxv4i32.i16(
   i16,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i32> @intrinsic_vwmaccu_mask_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv4i32_i16_nxv4i16:
@@ -1100,7 +1100,7 @@ entry:
     i16 %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -1132,7 +1132,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vwmaccu.mask.nxv8i32.i16(
   i16,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i32> @intrinsic_vwmaccu_mask_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv8i32_i16_nxv8i16:
@@ -1146,7 +1146,7 @@ entry:
     i16 %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -1178,7 +1178,7 @@ declare <vscale x 16 x i32> @llvm.riscv.vwmaccu.mask.nxv16i32.i16(
   i16,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i32> @intrinsic_vwmaccu_mask_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv16i32_i16_nxv16i16:
@@ -1192,7 +1192,7 @@ entry:
     i16 %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i32> %a
 }
@@ -1224,7 +1224,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vwmaccu.mask.nxv1i64.i32(
   i32,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i64> @intrinsic_vwmaccu_mask_vx_nxv1i64_i32_nxv1i32(<vscale x 1 x i64> %0, i32 %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv1i64_i32_nxv1i32:
@@ -1238,7 +1238,7 @@ entry:
     i32 %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -1270,7 +1270,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vwmaccu.mask.nxv2i64.i32(
   i32,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i64> @intrinsic_vwmaccu_mask_vx_nxv2i64_i32_nxv2i32(<vscale x 2 x i64> %0, i32 %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv2i64_i32_nxv2i32:
@@ -1284,7 +1284,7 @@ entry:
     i32 %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -1316,7 +1316,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vwmaccu.mask.nxv4i64.i32(
   i32,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i64> @intrinsic_vwmaccu_mask_vx_nxv4i64_i32_nxv4i32(<vscale x 4 x i64> %0, i32 %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv4i64_i32_nxv4i32:
@@ -1330,7 +1330,7 @@ entry:
     i32 %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -1362,7 +1362,7 @@ declare <vscale x 8 x i64> @llvm.riscv.vwmaccu.mask.nxv8i64.i32(
   i32,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i64> @intrinsic_vwmaccu_mask_vx_nxv8i64_i32_nxv8i32(<vscale x 8 x i64> %0, i32 %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv8i64_i32_nxv8i32:
@@ -1376,7 +1376,7 @@ entry:
     i32 %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i64> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vwmaccu-rv64.ll b/llvm/test/CodeGen/RISCV/rvv/vwmaccu-rv64.ll
index 623e00c03aadbd..49b2fec00220ac 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vwmaccu-rv64.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vwmaccu-rv64.ll
@@ -28,7 +28,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vwmaccu.mask.nxv1i16.nxv1i8(
   <vscale x 1 x i8>,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i16>  @intrinsic_vwmaccu_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv1i16_nxv1i8_nxv1i8:
@@ -42,7 +42,7 @@ entry:
     <vscale x 1 x i8> %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -74,7 +74,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vwmaccu.mask.nxv2i16.nxv2i8(
   <vscale x 2 x i8>,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i16>  @intrinsic_vwmaccu_mask_vv_nxv2i16_nxv2i8_nxv2i8(<vscale x 2 x i16> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv2i16_nxv2i8_nxv2i8:
@@ -88,7 +88,7 @@ entry:
     <vscale x 2 x i8> %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -120,7 +120,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vwmaccu.mask.nxv4i16.nxv4i8(
   <vscale x 4 x i8>,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i16>  @intrinsic_vwmaccu_mask_vv_nxv4i16_nxv4i8_nxv4i8(<vscale x 4 x i16> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv4i16_nxv4i8_nxv4i8:
@@ -134,7 +134,7 @@ entry:
     <vscale x 4 x i8> %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -166,7 +166,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vwmaccu.mask.nxv8i16.nxv8i8(
   <vscale x 8 x i8>,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i16>  @intrinsic_vwmaccu_mask_vv_nxv8i16_nxv8i8_nxv8i8(<vscale x 8 x i16> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv8i16_nxv8i8_nxv8i8:
@@ -180,7 +180,7 @@ entry:
     <vscale x 8 x i8> %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -212,7 +212,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vwmaccu.mask.nxv16i16.nxv16i8(
   <vscale x 16 x i8>,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i16>  @intrinsic_vwmaccu_mask_vv_nxv16i16_nxv16i8_nxv16i8(<vscale x 16 x i16> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv16i16_nxv16i8_nxv16i8:
@@ -226,7 +226,7 @@ entry:
     <vscale x 16 x i8> %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -258,7 +258,7 @@ declare <vscale x 32 x i16> @llvm.riscv.vwmaccu.mask.nxv32i16.nxv32i8(
   <vscale x 32 x i8>,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 32 x i16>  @intrinsic_vwmaccu_mask_vv_nxv32i16_nxv32i8_nxv32i8(<vscale x 32 x i16> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv32i16_nxv32i8_nxv32i8:
@@ -272,7 +272,7 @@ entry:
     <vscale x 32 x i8> %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 32 x i16> %a
 }
@@ -304,7 +304,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vwmaccu.mask.nxv1i32.nxv1i16(
   <vscale x 1 x i16>,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i32>  @intrinsic_vwmaccu_mask_vv_nxv1i32_nxv1i16_nxv1i16(<vscale x 1 x i32> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv1i32_nxv1i16_nxv1i16:
@@ -318,7 +318,7 @@ entry:
     <vscale x 1 x i16> %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -350,7 +350,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vwmaccu.mask.nxv2i32.nxv2i16(
   <vscale x 2 x i16>,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i32>  @intrinsic_vwmaccu_mask_vv_nxv2i32_nxv2i16_nxv2i16(<vscale x 2 x i32> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv2i32_nxv2i16_nxv2i16:
@@ -364,7 +364,7 @@ entry:
     <vscale x 2 x i16> %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -396,7 +396,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vwmaccu.mask.nxv4i32.nxv4i16(
   <vscale x 4 x i16>,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i32>  @intrinsic_vwmaccu_mask_vv_nxv4i32_nxv4i16_nxv4i16(<vscale x 4 x i32> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv4i32_nxv4i16_nxv4i16:
@@ -410,7 +410,7 @@ entry:
     <vscale x 4 x i16> %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -442,7 +442,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vwmaccu.mask.nxv8i32.nxv8i16(
   <vscale x 8 x i16>,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i32>  @intrinsic_vwmaccu_mask_vv_nxv8i32_nxv8i16_nxv8i16(<vscale x 8 x i32> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv8i32_nxv8i16_nxv8i16:
@@ -456,7 +456,7 @@ entry:
     <vscale x 8 x i16> %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -488,7 +488,7 @@ declare <vscale x 16 x i32> @llvm.riscv.vwmaccu.mask.nxv16i32.nxv16i16(
   <vscale x 16 x i16>,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i32>  @intrinsic_vwmaccu_mask_vv_nxv16i32_nxv16i16_nxv16i16(<vscale x 16 x i32> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv16i32_nxv16i16_nxv16i16:
@@ -502,7 +502,7 @@ entry:
     <vscale x 16 x i16> %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i32> %a
 }
@@ -534,7 +534,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vwmaccu.mask.nxv1i64.nxv1i32(
   <vscale x 1 x i32>,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i64>  @intrinsic_vwmaccu_mask_vv_nxv1i64_nxv1i32_nxv1i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv1i64_nxv1i32_nxv1i32:
@@ -548,7 +548,7 @@ entry:
     <vscale x 1 x i32> %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -580,7 +580,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vwmaccu.mask.nxv2i64.nxv2i32(
   <vscale x 2 x i32>,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i64>  @intrinsic_vwmaccu_mask_vv_nxv2i64_nxv2i32_nxv2i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv2i64_nxv2i32_nxv2i32:
@@ -594,7 +594,7 @@ entry:
     <vscale x 2 x i32> %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -626,7 +626,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vwmaccu.mask.nxv4i64.nxv4i32(
   <vscale x 4 x i32>,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i64>  @intrinsic_vwmaccu_mask_vv_nxv4i64_nxv4i32_nxv4i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv4i64_nxv4i32_nxv4i32:
@@ -640,7 +640,7 @@ entry:
     <vscale x 4 x i32> %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -672,7 +672,7 @@ declare <vscale x 8 x i64> @llvm.riscv.vwmaccu.mask.nxv8i64.nxv8i32(
   <vscale x 8 x i32>,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i64>  @intrinsic_vwmaccu_mask_vv_nxv8i64_nxv8i32_nxv8i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv8i64_nxv8i32_nxv8i32:
@@ -686,7 +686,7 @@ entry:
     <vscale x 8 x i32> %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i64> %a
 }
@@ -718,7 +718,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vwmaccu.mask.nxv1i16.i8(
   i8,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i16> @intrinsic_vwmaccu_mask_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv1i16_i8_nxv1i8:
@@ -732,7 +732,7 @@ entry:
     i8 %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -764,7 +764,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vwmaccu.mask.nxv2i16.i8(
   i8,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i16> @intrinsic_vwmaccu_mask_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv2i16_i8_nxv2i8:
@@ -778,7 +778,7 @@ entry:
     i8 %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -810,7 +810,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vwmaccu.mask.nxv4i16.i8(
   i8,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i16> @intrinsic_vwmaccu_mask_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv4i16_i8_nxv4i8:
@@ -824,7 +824,7 @@ entry:
     i8 %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -856,7 +856,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vwmaccu.mask.nxv8i16.i8(
   i8,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i16> @intrinsic_vwmaccu_mask_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv8i16_i8_nxv8i8:
@@ -870,7 +870,7 @@ entry:
     i8 %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -902,7 +902,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vwmaccu.mask.nxv16i16.i8(
   i8,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i16> @intrinsic_vwmaccu_mask_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv16i16_i8_nxv16i8:
@@ -916,7 +916,7 @@ entry:
     i8 %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -948,7 +948,7 @@ declare <vscale x 32 x i16> @llvm.riscv.vwmaccu.mask.nxv32i16.i8(
   i8,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 32 x i16> @intrinsic_vwmaccu_mask_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv32i16_i8_nxv32i8:
@@ -962,7 +962,7 @@ entry:
     i8 %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 32 x i16> %a
 }
@@ -994,7 +994,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vwmaccu.mask.nxv1i32.i16(
   i16,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i32> @intrinsic_vwmaccu_mask_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv1i32_i16_nxv1i16:
@@ -1008,7 +1008,7 @@ entry:
     i16 %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -1040,7 +1040,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vwmaccu.mask.nxv2i32.i16(
   i16,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i32> @intrinsic_vwmaccu_mask_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv2i32_i16_nxv2i16:
@@ -1054,7 +1054,7 @@ entry:
     i16 %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -1086,7 +1086,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vwmaccu.mask.nxv4i32.i16(
   i16,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i32> @intrinsic_vwmaccu_mask_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv4i32_i16_nxv4i16:
@@ -1100,7 +1100,7 @@ entry:
     i16 %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -1132,7 +1132,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vwmaccu.mask.nxv8i32.i16(
   i16,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i32> @intrinsic_vwmaccu_mask_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv8i32_i16_nxv8i16:
@@ -1146,7 +1146,7 @@ entry:
     i16 %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -1178,7 +1178,7 @@ declare <vscale x 16 x i32> @llvm.riscv.vwmaccu.mask.nxv16i32.i16(
   i16,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i32> @intrinsic_vwmaccu_mask_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv16i32_i16_nxv16i16:
@@ -1192,7 +1192,7 @@ entry:
     i16 %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i32> %a
 }
@@ -1224,7 +1224,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vwmaccu.mask.nxv1i64.i32(
   i32,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i64> @intrinsic_vwmaccu_mask_vx_nxv1i64_i32_nxv1i32(<vscale x 1 x i64> %0, i32 %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv1i64_i32_nxv1i32:
@@ -1238,7 +1238,7 @@ entry:
     i32 %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -1270,7 +1270,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vwmaccu.mask.nxv2i64.i32(
   i32,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i64> @intrinsic_vwmaccu_mask_vx_nxv2i64_i32_nxv2i32(<vscale x 2 x i64> %0, i32 %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv2i64_i32_nxv2i32:
@@ -1284,7 +1284,7 @@ entry:
     i32 %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -1316,7 +1316,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vwmaccu.mask.nxv4i64.i32(
   i32,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i64> @intrinsic_vwmaccu_mask_vx_nxv4i64_i32_nxv4i32(<vscale x 4 x i64> %0, i32 %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv4i64_i32_nxv4i32:
@@ -1330,7 +1330,7 @@ entry:
     i32 %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -1362,7 +1362,7 @@ declare <vscale x 8 x i64> @llvm.riscv.vwmaccu.mask.nxv8i64.i32(
   i32,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i64> @intrinsic_vwmaccu_mask_vx_nxv8i64_i32_nxv8i32(<vscale x 8 x i64> %0, i32 %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv8i64_i32_nxv8i32:
@@ -1376,7 +1376,7 @@ entry:
     i32 %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i64> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vwmaccus-rv32.ll b/llvm/test/CodeGen/RISCV/rvv/vwmaccus-rv32.ll
index 6a3aabba1be277..ccde867da07c64 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vwmaccus-rv32.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vwmaccus-rv32.ll
@@ -28,7 +28,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vwmaccus.mask.nxv1i16.i8(
   i8,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i16> @intrinsic_vwmaccus_mask_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv1i16_i8_nxv1i8:
@@ -42,7 +42,7 @@ entry:
     i8 %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -74,7 +74,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vwmaccus.mask.nxv2i16.i8(
   i8,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i16> @intrinsic_vwmaccus_mask_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv2i16_i8_nxv2i8:
@@ -88,7 +88,7 @@ entry:
     i8 %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -120,7 +120,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vwmaccus.mask.nxv4i16.i8(
   i8,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i16> @intrinsic_vwmaccus_mask_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv4i16_i8_nxv4i8:
@@ -134,7 +134,7 @@ entry:
     i8 %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -166,7 +166,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vwmaccus.mask.nxv8i16.i8(
   i8,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i16> @intrinsic_vwmaccus_mask_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv8i16_i8_nxv8i8:
@@ -180,7 +180,7 @@ entry:
     i8 %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -212,7 +212,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vwmaccus.mask.nxv16i16.i8(
   i8,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i16> @intrinsic_vwmaccus_mask_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv16i16_i8_nxv16i8:
@@ -226,7 +226,7 @@ entry:
     i8 %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -258,7 +258,7 @@ declare <vscale x 32 x i16> @llvm.riscv.vwmaccus.mask.nxv32i16.i8(
   i8,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 32 x i16> @intrinsic_vwmaccus_mask_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv32i16_i8_nxv32i8:
@@ -272,7 +272,7 @@ entry:
     i8 %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 32 x i16> %a
 }
@@ -304,7 +304,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vwmaccus.mask.nxv1i32.i16(
   i16,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i32> @intrinsic_vwmaccus_mask_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv1i32_i16_nxv1i16:
@@ -318,7 +318,7 @@ entry:
     i16 %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -350,7 +350,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vwmaccus.mask.nxv2i32.i16(
   i16,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i32> @intrinsic_vwmaccus_mask_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv2i32_i16_nxv2i16:
@@ -364,7 +364,7 @@ entry:
     i16 %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -396,7 +396,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vwmaccus.mask.nxv4i32.i16(
   i16,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i32> @intrinsic_vwmaccus_mask_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv4i32_i16_nxv4i16:
@@ -410,7 +410,7 @@ entry:
     i16 %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -442,7 +442,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vwmaccus.mask.nxv8i32.i16(
   i16,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i32> @intrinsic_vwmaccus_mask_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv8i32_i16_nxv8i16:
@@ -456,7 +456,7 @@ entry:
     i16 %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -488,7 +488,7 @@ declare <vscale x 16 x i32> @llvm.riscv.vwmaccus.mask.nxv16i32.i16(
   i16,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 16 x i32> @intrinsic_vwmaccus_mask_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv16i32_i16_nxv16i16:
@@ -502,7 +502,7 @@ entry:
     i16 %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 16 x i32> %a
 }
@@ -534,7 +534,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vwmaccus.mask.nxv1i64.i32(
   i32,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 1 x i64> @intrinsic_vwmaccus_mask_vx_nxv1i64_i32_nxv1i32(<vscale x 1 x i64> %0, i32 %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv1i64_i32_nxv1i32:
@@ -548,7 +548,7 @@ entry:
     i32 %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -580,7 +580,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vwmaccus.mask.nxv2i64.i32(
   i32,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 2 x i64> @intrinsic_vwmaccus_mask_vx_nxv2i64_i32_nxv2i32(<vscale x 2 x i64> %0, i32 %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv2i64_i32_nxv2i32:
@@ -594,7 +594,7 @@ entry:
     i32 %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -626,7 +626,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vwmaccus.mask.nxv4i64.i32(
   i32,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 4 x i64> @intrinsic_vwmaccus_mask_vx_nxv4i64_i32_nxv4i32(<vscale x 4 x i64> %0, i32 %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv4i64_i32_nxv4i32:
@@ -640,7 +640,7 @@ entry:
     i32 %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -672,7 +672,7 @@ declare <vscale x 8 x i64> @llvm.riscv.vwmaccus.mask.nxv8i64.i32(
   i32,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i32);
+  i32, i32);
 
 define <vscale x 8 x i64> @intrinsic_vwmaccus_mask_vx_nxv8i64_i32_nxv8i32(<vscale x 8 x i64> %0, i32 %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv8i64_i32_nxv8i32:
@@ -686,7 +686,7 @@ entry:
     i32 %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i32 %4)
+    i32 %4, i32 0)
 
   ret <vscale x 8 x i64> %a
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vwmaccus-rv64.ll b/llvm/test/CodeGen/RISCV/rvv/vwmaccus-rv64.ll
index 6786754645fdcf..95b164270ced5c 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vwmaccus-rv64.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vwmaccus-rv64.ll
@@ -28,7 +28,7 @@ declare <vscale x 1 x i16> @llvm.riscv.vwmaccus.mask.nxv1i16.i8(
   i8,
   <vscale x 1 x i8>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i16> @intrinsic_vwmaccus_mask_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv1i16_i8_nxv1i8:
@@ -42,7 +42,7 @@ entry:
     i8 %1,
     <vscale x 1 x i8> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i16> %a
 }
@@ -74,7 +74,7 @@ declare <vscale x 2 x i16> @llvm.riscv.vwmaccus.mask.nxv2i16.i8(
   i8,
   <vscale x 2 x i8>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i16> @intrinsic_vwmaccus_mask_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv2i16_i8_nxv2i8:
@@ -88,7 +88,7 @@ entry:
     i8 %1,
     <vscale x 2 x i8> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i16> %a
 }
@@ -120,7 +120,7 @@ declare <vscale x 4 x i16> @llvm.riscv.vwmaccus.mask.nxv4i16.i8(
   i8,
   <vscale x 4 x i8>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i16> @intrinsic_vwmaccus_mask_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv4i16_i8_nxv4i8:
@@ -134,7 +134,7 @@ entry:
     i8 %1,
     <vscale x 4 x i8> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i16> %a
 }
@@ -166,7 +166,7 @@ declare <vscale x 8 x i16> @llvm.riscv.vwmaccus.mask.nxv8i16.i8(
   i8,
   <vscale x 8 x i8>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i16> @intrinsic_vwmaccus_mask_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv8i16_i8_nxv8i8:
@@ -180,7 +180,7 @@ entry:
     i8 %1,
     <vscale x 8 x i8> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i16> %a
 }
@@ -212,7 +212,7 @@ declare <vscale x 16 x i16> @llvm.riscv.vwmaccus.mask.nxv16i16.i8(
   i8,
   <vscale x 16 x i8>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i16> @intrinsic_vwmaccus_mask_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv16i16_i8_nxv16i8:
@@ -226,7 +226,7 @@ entry:
     i8 %1,
     <vscale x 16 x i8> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i16> %a
 }
@@ -258,7 +258,7 @@ declare <vscale x 32 x i16> @llvm.riscv.vwmaccus.mask.nxv32i16.i8(
   i8,
   <vscale x 32 x i8>,
   <vscale x 32 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 32 x i16> @intrinsic_vwmaccus_mask_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv32i16_i8_nxv32i8:
@@ -272,7 +272,7 @@ entry:
     i8 %1,
     <vscale x 32 x i8> %2,
     <vscale x 32 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 32 x i16> %a
 }
@@ -304,7 +304,7 @@ declare <vscale x 1 x i32> @llvm.riscv.vwmaccus.mask.nxv1i32.i16(
   i16,
   <vscale x 1 x i16>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i32> @intrinsic_vwmaccus_mask_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv1i32_i16_nxv1i16:
@@ -318,7 +318,7 @@ entry:
     i16 %1,
     <vscale x 1 x i16> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i32> %a
 }
@@ -350,7 +350,7 @@ declare <vscale x 2 x i32> @llvm.riscv.vwmaccus.mask.nxv2i32.i16(
   i16,
   <vscale x 2 x i16>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i32> @intrinsic_vwmaccus_mask_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv2i32_i16_nxv2i16:
@@ -364,7 +364,7 @@ entry:
     i16 %1,
     <vscale x 2 x i16> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i32> %a
 }
@@ -396,7 +396,7 @@ declare <vscale x 4 x i32> @llvm.riscv.vwmaccus.mask.nxv4i32.i16(
   i16,
   <vscale x 4 x i16>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i32> @intrinsic_vwmaccus_mask_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv4i32_i16_nxv4i16:
@@ -410,7 +410,7 @@ entry:
     i16 %1,
     <vscale x 4 x i16> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i32> %a
 }
@@ -442,7 +442,7 @@ declare <vscale x 8 x i32> @llvm.riscv.vwmaccus.mask.nxv8i32.i16(
   i16,
   <vscale x 8 x i16>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i32> @intrinsic_vwmaccus_mask_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv8i32_i16_nxv8i16:
@@ -456,7 +456,7 @@ entry:
     i16 %1,
     <vscale x 8 x i16> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i32> %a
 }
@@ -488,7 +488,7 @@ declare <vscale x 16 x i32> @llvm.riscv.vwmaccus.mask.nxv16i32.i16(
   i16,
   <vscale x 16 x i16>,
   <vscale x 16 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 16 x i32> @intrinsic_vwmaccus_mask_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv16i32_i16_nxv16i16:
@@ -502,7 +502,7 @@ entry:
     i16 %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 16 x i32> %a
 }
@@ -534,7 +534,7 @@ declare <vscale x 1 x i64> @llvm.riscv.vwmaccus.mask.nxv1i64.i32(
   i32,
   <vscale x 1 x i32>,
   <vscale x 1 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 1 x i64> @intrinsic_vwmaccus_mask_vx_nxv1i64_i32_nxv1i32(<vscale x 1 x i64> %0, i32 %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv1i64_i32_nxv1i32:
@@ -548,7 +548,7 @@ entry:
     i32 %1,
     <vscale x 1 x i32> %2,
     <vscale x 1 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 1 x i64> %a
 }
@@ -580,7 +580,7 @@ declare <vscale x 2 x i64> @llvm.riscv.vwmaccus.mask.nxv2i64.i32(
   i32,
   <vscale x 2 x i32>,
   <vscale x 2 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 2 x i64> @intrinsic_vwmaccus_mask_vx_nxv2i64_i32_nxv2i32(<vscale x 2 x i64> %0, i32 %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv2i64_i32_nxv2i32:
@@ -594,7 +594,7 @@ entry:
     i32 %1,
     <vscale x 2 x i32> %2,
     <vscale x 2 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 2 x i64> %a
 }
@@ -626,7 +626,7 @@ declare <vscale x 4 x i64> @llvm.riscv.vwmaccus.mask.nxv4i64.i32(
   i32,
   <vscale x 4 x i32>,
   <vscale x 4 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 4 x i64> @intrinsic_vwmaccus_mask_vx_nxv4i64_i32_nxv4i32(<vscale x 4 x i64> %0, i32 %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv4i64_i32_nxv4i32:
@@ -640,7 +640,7 @@ entry:
     i32 %1,
     <vscale x 4 x i32> %2,
     <vscale x 4 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 4 x i64> %a
 }
@@ -672,7 +672,7 @@ declare <vscale x 8 x i64> @llvm.riscv.vwmaccus.mask.nxv8i64.i32(
   i32,
   <vscale x 8 x i32>,
   <vscale x 8 x i1>,
-  i64);
+  i64, i64);
 
 define <vscale x 8 x i64> @intrinsic_vwmaccus_mask_vx_nxv8i64_i32_nxv8i32(<vscale x 8 x i64> %0, i32 %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
 ; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv8i64_i32_nxv8i32:
@@ -686,7 +686,7 @@ entry:
     i32 %1,
     <vscale x 8 x i32> %2,
     <vscale x 8 x i1> %3,
-    i64 %4)
+    i64 %4, i64 0)
 
   ret <vscale x 8 x i64> %a
 }