Author: Simon Pilgrim Date: 2024-10-07T15:57:22+01:00 New Revision: aa4d94839e6bb30389deb3339fb0d0e4aa8baa4e
URL: https://github.com/llvm/llvm-project/commit/aa4d94839e6bb30389deb3339fb0d0e4aa8baa4e DIFF: https://github.com/llvm/llvm-project/commit/aa4d94839e6bb30389deb3339fb0d0e4aa8baa4e.diff LOG: [clang][x86] Update FMA/FMA4 intrinsic tests for both C/C++ and 32/64-bit targets Requires some call instructions to handle additional markers Added: Modified: clang/test/CodeGen/X86/fma-builtins-constrained.c clang/test/CodeGen/X86/fma-builtins.c clang/test/CodeGen/X86/fma4-builtins.c Removed: ################################################################################ diff --git a/clang/test/CodeGen/X86/fma-builtins-constrained.c b/clang/test/CodeGen/X86/fma-builtins-constrained.c index 74177ef7d1c9d6..019dde2e025149 100644 --- a/clang/test/CodeGen/X86/fma-builtins-constrained.c +++ b/clang/test/CodeGen/X86/fma-builtins-constrained.c @@ -1,8 +1,12 @@ // REQUIRES: x86-registered-target -// RUN: %clang_cc1 -ffreestanding %s -triple=x86_64-unknown-linux-gnu -target-feature +fma -O2 -emit-llvm -o - | FileCheck %s --check-prefixes=COMMON,COMMONIR,UNCONSTRAINED -// RUN: %clang_cc1 -ffreestanding %s -triple=x86_64-unknown-linux-gnu -target-feature +fma -ffp-exception-behavior=maytrap -DSTRICT=1 -O2 -emit-llvm -o - | FileCheck %s --check-prefixes=COMMON,COMMONIR,CONSTRAINED -// RUN: %clang_cc1 -ffreestanding %s -triple=x86_64-unknown-linux-gnu -target-feature +fma -O2 -S -o - | FileCheck %s --check-prefixes=COMMON,CHECK-ASM -// RUN: %clang_cc1 -ffreestanding %s -triple=x86_64-unknown-linux-gnu -target-feature +fma -O2 -ffp-exception-behavior=maytrap -DSTRICT=1 -S -o - | FileCheck %s --check-prefixes=COMMON,CHECK-ASM +// RUN: %clang_cc1 -x c -ffreestanding %s -triple=x86_64-unknown-linux-gnu -target-feature +fma -O2 -emit-llvm -o - | FileCheck %s --check-prefixes=COMMON,COMMONIR,UNCONSTRAINED +// RUN: %clang_cc1 -x c -ffreestanding %s -triple=x86_64-unknown-linux-gnu -target-feature +fma -ffp-exception-behavior=maytrap -DSTRICT=1 -O2 -emit-llvm -o - | FileCheck %s --check-prefixes=COMMON,COMMONIR,CONSTRAINED +// RUN: %clang_cc1 -x c -ffreestanding %s -triple=x86_64-unknown-linux-gnu -target-feature +fma -O2 -S -o - | FileCheck %s --check-prefixes=COMMON,CHECK-ASM +// RUN: %clang_cc1 -x c -ffreestanding %s -triple=x86_64-unknown-linux-gnu -target-feature +fma -O2 -ffp-exception-behavior=maytrap -DSTRICT=1 -S -o - | FileCheck %s --check-prefixes=COMMON,CHECK-ASM +// RUN: %clang_cc1 -x c++ -ffreestanding %s -triple=x86_64-unknown-linux-gnu -target-feature +fma -O2 -emit-llvm -o - | FileCheck %s --check-prefixes=COMMON,COMMONIR,UNCONSTRAINED +// RUN: %clang_cc1 -x c++ -ffreestanding %s -triple=x86_64-unknown-linux-gnu -target-feature +fma -ffp-exception-behavior=maytrap -DSTRICT=1 -O2 -emit-llvm -o - | FileCheck %s --check-prefixes=COMMON,COMMONIR,CONSTRAINED +// RUN: %clang_cc1 -x c++ -ffreestanding %s -triple=x86_64-unknown-linux-gnu -target-feature +fma -O2 -S -o - | FileCheck %s --check-prefixes=COMMON,CHECK-ASM +// RUN: %clang_cc1 -x c++ -ffreestanding %s -triple=x86_64-unknown-linux-gnu -target-feature +fma -O2 -ffp-exception-behavior=maytrap -DSTRICT=1 -S -o - | FileCheck %s --check-prefixes=COMMON,CHECK-ASM #ifdef STRICT // Test that the constrained intrinsics are picking up the exception @@ -15,16 +19,16 @@ __m128 test_mm_fmadd_ps(__m128 a, __m128 b, __m128 c) { // COMMON-LABEL: test_mm_fmadd_ps - // UNCONSTRAINED: call <4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) - // CONSTRAINED: call <4 x float> @llvm.experimental.constrained.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}, metadata !{{.*}}) + // UNCONSTRAINED: call {{.*}}<4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) + // CONSTRAINED: call {{.*}}<4 x float> @llvm.experimental.constrained.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}, metadata !{{.*}}) // CHECK-ASM: vfmadd213ps return _mm_fmadd_ps(a, b, c); } __m128d test_mm_fmadd_pd(__m128d a, __m128d b, __m128d c) { // COMMON-LABEL: test_mm_fmadd_pd - // UNCONSTRAINED: call <2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) - // CONSTRAINED: call <2 x double> @llvm.experimental.constrained.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}, metadata !{{.*}}) + // UNCONSTRAINED: call {{.*}}<2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) + // CONSTRAINED: call {{.*}}<2 x double> @llvm.experimental.constrained.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}, metadata !{{.*}}) // CHECK-ASM: vfmadd213pd return _mm_fmadd_pd(a, b, c); } @@ -56,8 +60,8 @@ __m128d test_mm_fmadd_sd(__m128d a, __m128d b, __m128d c) { __m128 test_mm_fmsub_ps(__m128 a, __m128 b, __m128 c) { // COMMON-LABEL: test_mm_fmsub_ps // COMMONIR: [[NEG:%.+]] = fneg <4 x float> %{{.+}} - // UNCONSTRAINED: call <4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) - // CONSTRAINED: call <4 x float> @llvm.experimental.constrained.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}, metadata !{{.*}}) + // UNCONSTRAINED: call {{.*}}<4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) + // CONSTRAINED: call {{.*}}<4 x float> @llvm.experimental.constrained.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}, metadata !{{.*}}) // CHECK-ASM: vfmsub213ps return _mm_fmsub_ps(a, b, c); } @@ -65,8 +69,8 @@ __m128 test_mm_fmsub_ps(__m128 a, __m128 b, __m128 c) { __m128d test_mm_fmsub_pd(__m128d a, __m128d b, __m128d c) { // COMMON-LABEL: test_mm_fmsub_pd // COMMONIR: [[NEG:%.+]] = fneg <2 x double> %{{.+}} - // UNCONSTRAINED: call <2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) - // CONSTRAINED: call <2 x double> @llvm.experimental.constrained.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}, metadata !{{.*}}) + // UNCONSTRAINED: call {{.*}}<2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) + // CONSTRAINED: call {{.*}}<2 x double> @llvm.experimental.constrained.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}, metadata !{{.*}}) // CHECK-ASM: vfmsub213pd return _mm_fmsub_pd(a, b, c); } @@ -100,8 +104,8 @@ __m128d test_mm_fmsub_sd(__m128d a, __m128d b, __m128d c) { __m128 test_mm_fnmadd_ps(__m128 a, __m128 b, __m128 c) { // COMMON-LABEL: test_mm_fnmadd_ps // COMMONIR: [[NEG:%.+]] = fneg <4 x float> %{{.+}} - // UNCONSTRAINED: call <4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) - // CONSTRAINED: call <4 x float> @llvm.experimental.constrained.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}, metadata !{{.*}}) + // UNCONSTRAINED: call {{.*}}<4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) + // CONSTRAINED: call {{.*}}<4 x float> @llvm.experimental.constrained.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}, metadata !{{.*}}) // CHECK-ASM: vfnmadd213ps return _mm_fnmadd_ps(a, b, c); } @@ -109,8 +113,8 @@ __m128 test_mm_fnmadd_ps(__m128 a, __m128 b, __m128 c) { __m128d test_mm_fnmadd_pd(__m128d a, __m128d b, __m128d c) { // COMMON-LABEL: test_mm_fnmadd_pd // COMMONIR: [[NEG:%.+]] = fneg <2 x double> %{{.+}} - // UNCONSTRAINED: call <2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) - // CONSTRAINED: call <2 x double> @llvm.experimental.constrained.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}, metadata !{{.*}}) + // UNCONSTRAINED: call {{.*}}<2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) + // CONSTRAINED: call {{.*}}<2 x double> @llvm.experimental.constrained.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}, metadata !{{.*}}) // CHECK-ASM: vfnmadd213pd return _mm_fnmadd_pd(a, b, c); } @@ -145,8 +149,8 @@ __m128 test_mm_fnmsub_ps(__m128 a, __m128 b, __m128 c) { // COMMON-LABEL: test_mm_fnmsub_ps // COMMONIR: [[NEG:%.+]] = fneg <4 x float> %{{.+}} // COMMONIR: [[NEG2:%.+]] = fneg <4 x float> %{{.+}} - // UNCONSTRAINED: call <4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) - // CONSTRAINED: call <4 x float> @llvm.experimental.constrained.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}, metadata !{{.*}}) + // UNCONSTRAINED: call {{.*}}<4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) + // CONSTRAINED: call {{.*}}<4 x float> @llvm.experimental.constrained.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}, metadata !{{.*}}) // CHECK-ASM: vfnmsub213ps return _mm_fnmsub_ps(a, b, c); } @@ -155,8 +159,8 @@ __m128d test_mm_fnmsub_pd(__m128d a, __m128d b, __m128d c) { // COMMON-LABEL: test_mm_fnmsub_pd // COMMONIR: [[NEG:%.+]] = fneg <2 x double> %{{.+}} // COMMONIR: [[NEG2:%.+]] = fneg <2 x double> %{{.+}} - // UNCONSTRAINED: call <2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) - // CONSTRAINED: call <2 x double> @llvm.experimental.constrained.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}, metadata !{{.*}}) + // UNCONSTRAINED: call {{.*}}<2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) + // CONSTRAINED: call {{.*}}<2 x double> @llvm.experimental.constrained.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}, metadata !{{.*}}) // CHECK-ASM: vfnmsub213pd return _mm_fnmsub_pd(a, b, c); } @@ -192,7 +196,7 @@ __m128d test_mm_fnmsub_sd(__m128d a, __m128d b, __m128d c) { __m128 test_mm_fmaddsub_ps(__m128 a, __m128 b, __m128 c) { // COMMON-LABEL: test_mm_fmaddsub_ps // COMMONIR-NOT: fneg - // COMMONIR: tail call <4 x float> @llvm.x86.fma.vfmaddsub.ps(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) + // COMMONIR: tail call {{.*}}<4 x float> @llvm.x86.fma.vfmaddsub.ps(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) // CHECK-ASM: vfmaddsub213ps return _mm_fmaddsub_ps(a, b, c); } @@ -200,7 +204,7 @@ __m128 test_mm_fmaddsub_ps(__m128 a, __m128 b, __m128 c) { __m128d test_mm_fmaddsub_pd(__m128d a, __m128d b, __m128d c) { // COMMON-LABEL: test_mm_fmaddsub_pd // COMMONIR-NOT: fneg - // COMMONIR: tail call <2 x double> @llvm.x86.fma.vfmaddsub.pd(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) + // COMMONIR: tail call {{.*}}<2 x double> @llvm.x86.fma.vfmaddsub.pd(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) // CHECK-ASM: vfmaddsub213pd return _mm_fmaddsub_pd(a, b, c); } @@ -208,7 +212,7 @@ __m128d test_mm_fmaddsub_pd(__m128d a, __m128d b, __m128d c) { __m128 test_mm_fmsubadd_ps(__m128 a, __m128 b, __m128 c) { // COMMON-LABEL: test_mm_fmsubadd_ps // COMMONIR: [[FNEG:%.+]] = fneg <4 x float> %{{.*}} - // COMMONIR: tail call <4 x float> @llvm.x86.fma.vfmaddsub.ps(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> [[FNEG]]) + // COMMONIR: tail call {{.*}}<4 x float> @llvm.x86.fma.vfmaddsub.ps(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> [[FNEG]]) // CHECK-ASM: vfmsubadd213ps return _mm_fmsubadd_ps(a, b, c); } @@ -216,23 +220,23 @@ __m128 test_mm_fmsubadd_ps(__m128 a, __m128 b, __m128 c) { __m128d test_mm_fmsubadd_pd(__m128d a, __m128d b, __m128d c) { // COMMON-LABEL: test_mm_fmsubadd_pd // COMMONIR: [[FNEG:%.+]] = fneg <2 x double> %{{.*}} - // COMMONIR: tail call <2 x double> @llvm.x86.fma.vfmaddsub.pd(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> [[FNEG]]) + // COMMONIR: tail call {{.*}}<2 x double> @llvm.x86.fma.vfmaddsub.pd(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> [[FNEG]]) // CHECK-ASM: vfmsubadd213pd return _mm_fmsubadd_pd(a, b, c); } __m256 test_mm256_fmadd_ps(__m256 a, __m256 b, __m256 c) { // COMMON-LABEL: test_mm256_fmadd_ps - // UNCONSTRAINED: call <8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) - // CONSTRAINED: call <8 x float> @llvm.experimental.constrained.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}, metadata !{{.*}}) + // UNCONSTRAINED: call {{.*}}<8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) + // CONSTRAINED: call {{.*}}<8 x float> @llvm.experimental.constrained.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}, metadata !{{.*}}) // CHECK-ASM: vfmadd213ps return _mm256_fmadd_ps(a, b, c); } __m256d test_mm256_fmadd_pd(__m256d a, __m256d b, __m256d c) { // COMMON-LABEL: test_mm256_fmadd_pd - // UNCONSTRAINED: call <4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) - // CONSTRAINED: call <4 x double> @llvm.experimental.constrained.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}, metadata !{{.*}}) + // UNCONSTRAINED: call {{.*}}<4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) + // CONSTRAINED: call {{.*}}<4 x double> @llvm.experimental.constrained.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}, metadata !{{.*}}) // CHECK-ASM: vfmadd213pd return _mm256_fmadd_pd(a, b, c); } @@ -240,8 +244,8 @@ __m256d test_mm256_fmadd_pd(__m256d a, __m256d b, __m256d c) { __m256 test_mm256_fmsub_ps(__m256 a, __m256 b, __m256 c) { // COMMON-LABEL: test_mm256_fmsub_ps // COMMONIR: [[NEG:%.+]] = fneg <8 x float> %{{.*}} - // UNCONSTRAINED: call <8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) - // CONSTRAINED: call <8 x float> @llvm.experimental.constrained.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}, metadata !{{.*}}) + // UNCONSTRAINED: call {{.*}}<8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) + // CONSTRAINED: call {{.*}}<8 x float> @llvm.experimental.constrained.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}, metadata !{{.*}}) // CHECK-ASM: vfmsub213ps return _mm256_fmsub_ps(a, b, c); } @@ -249,8 +253,8 @@ __m256 test_mm256_fmsub_ps(__m256 a, __m256 b, __m256 c) { __m256d test_mm256_fmsub_pd(__m256d a, __m256d b, __m256d c) { // COMMON-LABEL: test_mm256_fmsub_pd // COMMONIR: [[NEG:%.+]] = fneg <4 x double> %{{.+}} - // UNCONSTRAINED: call <4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) - // CONSTRAINED: call <4 x double> @llvm.experimental.constrained.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}, metadata !{{.*}}) + // UNCONSTRAINED: call {{.*}}<4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) + // CONSTRAINED: call {{.*}}<4 x double> @llvm.experimental.constrained.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}, metadata !{{.*}}) // CHECK-ASM: vfmsub213pd return _mm256_fmsub_pd(a, b, c); } @@ -258,8 +262,8 @@ __m256d test_mm256_fmsub_pd(__m256d a, __m256d b, __m256d c) { __m256 test_mm256_fnmadd_ps(__m256 a, __m256 b, __m256 c) { // COMMON-LABEL: test_mm256_fnmadd_ps // COMMONIR: [[NEG:%.+]] = fneg <8 x float> %{{.*}} - // UNCONSTRAINED: call <8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) - // CONSTRAINED: call <8 x float> @llvm.experimental.constrained.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}, metadata !{{.*}}) + // UNCONSTRAINED: call {{.*}}<8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) + // CONSTRAINED: call {{.*}}<8 x float> @llvm.experimental.constrained.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}, metadata !{{.*}}) // CHECK-ASM: vfnmadd213ps return _mm256_fnmadd_ps(a, b, c); } @@ -267,8 +271,8 @@ __m256 test_mm256_fnmadd_ps(__m256 a, __m256 b, __m256 c) { __m256d test_mm256_fnmadd_pd(__m256d a, __m256d b, __m256d c) { // COMMON-LABEL: test_mm256_fnmadd_pd // COMMONIR: [[NEG:%.+]] = fneg <4 x double> %{{.+}} - // UNCONSTRAINED: call <4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) - // CONSTRAINED: call <4 x double> @llvm.experimental.constrained.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}, metadata !{{.*}}) + // UNCONSTRAINED: call {{.*}}<4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) + // CONSTRAINED: call {{.*}}<4 x double> @llvm.experimental.constrained.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}, metadata !{{.*}}) // CHECK-ASM: vfnmadd213pd return _mm256_fnmadd_pd(a, b, c); } @@ -277,8 +281,8 @@ __m256 test_mm256_fnmsub_ps(__m256 a, __m256 b, __m256 c) { // COMMON-LABEL: test_mm256_fnmsub_ps // COMMONIR: [[NEG:%.+]] = fneg <8 x float> %{{.*}} // COMMONIR: [[NEG2:%.+]] = fneg <8 x float> %{{.*}} - // UNCONSTRAINED: call <8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) - // CONSTRAINED: call <8 x float> @llvm.experimental.constrained.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}, metadata !{{.*}}) + // UNCONSTRAINED: call {{.*}}<8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) + // CONSTRAINED: call {{.*}}<8 x float> @llvm.experimental.constrained.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}, metadata !{{.*}}) // CHECK-ASM: vfnmsub213ps return _mm256_fnmsub_ps(a, b, c); } @@ -287,8 +291,8 @@ __m256d test_mm256_fnmsub_pd(__m256d a, __m256d b, __m256d c) { // COMMON-LABEL: test_mm256_fnmsub_pd // COMMONIR: [[NEG:%.+]] = fneg <4 x double> %{{.+}} // COMMONIR: [[NEG2:%.+]] = fneg <4 x double> %{{.+}} - // UNCONSTRAINED: call <4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) - // CONSTRAINED: call <4 x double> @llvm.experimental.constrained.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}, metadata !{{.*}}) + // UNCONSTRAINED: call {{.*}}<4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) + // CONSTRAINED: call {{.*}}<4 x double> @llvm.experimental.constrained.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}, metadata !{{.*}}) // CHECK-ASM: vfnmsub213pd return _mm256_fnmsub_pd(a, b, c); } @@ -296,7 +300,7 @@ __m256d test_mm256_fnmsub_pd(__m256d a, __m256d b, __m256d c) { __m256 test_mm256_fmaddsub_ps(__m256 a, __m256 b, __m256 c) { // COMMON-LABEL: test_mm256_fmaddsub_ps // COMMONIR-NOT: fneg - // COMMONIR: tail call <8 x float> @llvm.x86.fma.vfmaddsub.ps.256(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) + // COMMONIR: tail call {{.*}}<8 x float> @llvm.x86.fma.vfmaddsub.ps.256(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) // CHECK-ASM: vfmaddsub213ps return _mm256_fmaddsub_ps(a, b, c); } @@ -304,7 +308,7 @@ __m256 test_mm256_fmaddsub_ps(__m256 a, __m256 b, __m256 c) { __m256d test_mm256_fmaddsub_pd(__m256d a, __m256d b, __m256d c) { // COMMON-LABEL: test_mm256_fmaddsub_pd // COMMONIR-NOT: fneg - // COMMONIR: tail call <4 x double> @llvm.x86.fma.vfmaddsub.pd.256(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) + // COMMONIR: tail call {{.*}}<4 x double> @llvm.x86.fma.vfmaddsub.pd.256(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) // CHECK-ASM: vfmaddsub213pd return _mm256_fmaddsub_pd(a, b, c); } @@ -312,7 +316,7 @@ __m256d test_mm256_fmaddsub_pd(__m256d a, __m256d b, __m256d c) { __m256 test_mm256_fmsubadd_ps(__m256 a, __m256 b, __m256 c) { // COMMON-LABEL: test_mm256_fmsubadd_ps // COMMONIR: [[FNEG:%.+]] = fneg <8 x float> %{{.*}} - // COMMONIR: tail call <8 x float> @llvm.x86.fma.vfmaddsub.ps.256(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> [[FNEG]]) + // COMMONIR: tail call {{.*}}<8 x float> @llvm.x86.fma.vfmaddsub.ps.256(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> [[FNEG]]) // CHECK-ASM: vfmsubadd213ps return _mm256_fmsubadd_ps(a, b, c); } @@ -320,7 +324,7 @@ __m256 test_mm256_fmsubadd_ps(__m256 a, __m256 b, __m256 c) { __m256d test_mm256_fmsubadd_pd(__m256d a, __m256d b, __m256d c) { // COMMON-LABEL: test_mm256_fmsubadd_pd // COMMONIR: [[FNEG:%.+]] = fneg <4 x double> %{{.*}} - // COMMONIR: tail call <4 x double> @llvm.x86.fma.vfmaddsub.pd.256(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> [[FNEG]]) + // COMMONIR: tail call {{.*}}<4 x double> @llvm.x86.fma.vfmaddsub.pd.256(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> [[FNEG]]) // CHECK-ASM: vfmsubadd213pd return _mm256_fmsubadd_pd(a, b, c); } diff --git a/clang/test/CodeGen/X86/fma-builtins.c b/clang/test/CodeGen/X86/fma-builtins.c index 4686b3686a4f8f..aa17dcc62fbc06 100644 --- a/clang/test/CodeGen/X86/fma-builtins.c +++ b/clang/test/CodeGen/X86/fma-builtins.c @@ -1,17 +1,20 @@ -// RUN: %clang_cc1 -flax-vector-conversions=none -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +fma -emit-llvm -o - | FileCheck %s +// RUN: %clang_cc1 -x c -flax-vector-conversions=none -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +fma -emit-llvm -o - | FileCheck %s +// RUN: %clang_cc1 -x c -flax-vector-conversions=none -ffreestanding %s -triple=i386-apple-darwin -target-feature +fma -emit-llvm -o - | FileCheck %s +// RUN: %clang_cc1 -x c++ -flax-vector-conversions=none -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +fma -emit-llvm -o - | FileCheck %s +// RUN: %clang_cc1 -x c++ -flax-vector-conversions=none -ffreestanding %s -triple=i386-apple-darwin -target-feature +fma -emit-llvm -o - | FileCheck %s #include <immintrin.h> __m128 test_mm_fmadd_ps(__m128 a, __m128 b, __m128 c) { // CHECK-LABEL: test_mm_fmadd_ps - // CHECK: call <4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) + // CHECK: call {{.*}}<4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) return _mm_fmadd_ps(a, b, c); } __m128d test_mm_fmadd_pd(__m128d a, __m128d b, __m128d c) { // CHECK-LABEL: test_mm_fmadd_pd - // CHECK: call <2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) + // CHECK: call {{.*}}<2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) return _mm_fmadd_pd(a, b, c); } @@ -38,14 +41,14 @@ __m128d test_mm_fmadd_sd(__m128d a, __m128d b, __m128d c) { __m128 test_mm_fmsub_ps(__m128 a, __m128 b, __m128 c) { // CHECK-LABEL: test_mm_fmsub_ps // CHECK: [[NEG:%.+]] = fneg <4 x float> %{{.+}} - // CHECK: call <4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) + // CHECK: call {{.*}}<4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) return _mm_fmsub_ps(a, b, c); } __m128d test_mm_fmsub_pd(__m128d a, __m128d b, __m128d c) { // CHECK-LABEL: test_mm_fmsub_pd // CHECK: [[NEG:%.+]] = fneg <2 x double> %{{.+}} - // CHECK: call <2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) + // CHECK: call {{.*}}<2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) return _mm_fmsub_pd(a, b, c); } @@ -74,14 +77,14 @@ __m128d test_mm_fmsub_sd(__m128d a, __m128d b, __m128d c) { __m128 test_mm_fnmadd_ps(__m128 a, __m128 b, __m128 c) { // CHECK-LABEL: test_mm_fnmadd_ps // CHECK: [[NEG:%.+]] = fneg <4 x float> %{{.+}} - // CHECK: call <4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) + // CHECK: call {{.*}}<4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) return _mm_fnmadd_ps(a, b, c); } __m128d test_mm_fnmadd_pd(__m128d a, __m128d b, __m128d c) { // CHECK-LABEL: test_mm_fnmadd_pd // CHECK: [[NEG:%.+]] = fneg <2 x double> %{{.+}} - // CHECK: call <2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) + // CHECK: call {{.*}}<2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) return _mm_fnmadd_pd(a, b, c); } @@ -111,7 +114,7 @@ __m128 test_mm_fnmsub_ps(__m128 a, __m128 b, __m128 c) { // CHECK-LABEL: test_mm_fnmsub_ps // CHECK: [[NEG:%.+]] = fneg <4 x float> %{{.+}} // CHECK: [[NEG2:%.+]] = fneg <4 x float> %{{.+}} - // CHECK: call <4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) + // CHECK: call {{.*}}<4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) return _mm_fnmsub_ps(a, b, c); } @@ -119,7 +122,7 @@ __m128d test_mm_fnmsub_pd(__m128d a, __m128d b, __m128d c) { // CHECK-LABEL: test_mm_fnmsub_pd // CHECK: [[NEG:%.+]] = fneg <2 x double> %{{.+}} // CHECK: [[NEG2:%.+]] = fneg <2 x double> %{{.+}} - // CHECK: call <2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) + // CHECK: call {{.*}}<2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) return _mm_fnmsub_pd(a, b, c); } @@ -150,68 +153,68 @@ __m128d test_mm_fnmsub_sd(__m128d a, __m128d b, __m128d c) { __m128 test_mm_fmaddsub_ps(__m128 a, __m128 b, __m128 c) { // CHECK-LABEL: test_mm_fmaddsub_ps // CHECK-NOT: fneg - // CHECK: call <4 x float> @llvm.x86.fma.vfmaddsub.ps(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) + // CHECK: call {{.*}}<4 x float> @llvm.x86.fma.vfmaddsub.ps(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) return _mm_fmaddsub_ps(a, b, c); } __m128d test_mm_fmaddsub_pd(__m128d a, __m128d b, __m128d c) { // CHECK-LABEL: test_mm_fmaddsub_pd // CHECK-NOT: fneg - // CHECK: call <2 x double> @llvm.x86.fma.vfmaddsub.pd(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) + // CHECK: call {{.*}}<2 x double> @llvm.x86.fma.vfmaddsub.pd(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) return _mm_fmaddsub_pd(a, b, c); } __m128 test_mm_fmsubadd_ps(__m128 a, __m128 b, __m128 c) { // CHECK-LABEL: test_mm_fmsubadd_ps // CHECK: [[NEG:%.+]] = fneg <4 x float> %{{.+}} - // CHECK: call <4 x float> @llvm.x86.fma.vfmaddsub.ps(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> [[NEG]]) + // CHECK: call {{.*}}<4 x float> @llvm.x86.fma.vfmaddsub.ps(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> [[NEG]]) return _mm_fmsubadd_ps(a, b, c); } __m128d test_mm_fmsubadd_pd(__m128d a, __m128d b, __m128d c) { // CHECK-LABEL: test_mm_fmsubadd_pd // CHECK: [[NEG:%.+]] = fneg <2 x double> %{{.+}} - // CHECK: call <2 x double> @llvm.x86.fma.vfmaddsub.pd(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> [[NEG]]) + // CHECK: call {{.*}}<2 x double> @llvm.x86.fma.vfmaddsub.pd(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> [[NEG]]) return _mm_fmsubadd_pd(a, b, c); } __m256 test_mm256_fmadd_ps(__m256 a, __m256 b, __m256 c) { // CHECK-LABEL: test_mm256_fmadd_ps - // CHECK: call <8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) + // CHECK: call {{.*}}<8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) return _mm256_fmadd_ps(a, b, c); } __m256d test_mm256_fmadd_pd(__m256d a, __m256d b, __m256d c) { // CHECK-LABEL: test_mm256_fmadd_pd - // CHECK: call <4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) + // CHECK: call {{.*}}<4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) return _mm256_fmadd_pd(a, b, c); } __m256 test_mm256_fmsub_ps(__m256 a, __m256 b, __m256 c) { // CHECK-LABEL: test_mm256_fmsub_ps // CHECK: [[NEG:%.+]] = fneg <8 x float> %{{.*}} - // CHECK: call <8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) + // CHECK: call {{.*}}<8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) return _mm256_fmsub_ps(a, b, c); } __m256d test_mm256_fmsub_pd(__m256d a, __m256d b, __m256d c) { // CHECK-LABEL: test_mm256_fmsub_pd // CHECK: [[NEG:%.+]] = fneg <4 x double> %{{.+}} - // CHECK: call <4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) + // CHECK: call {{.*}}<4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) return _mm256_fmsub_pd(a, b, c); } __m256 test_mm256_fnmadd_ps(__m256 a, __m256 b, __m256 c) { // CHECK-LABEL: test_mm256_fnmadd_ps // CHECK: [[NEG:%.+]] = fneg <8 x float> %{{.*}} - // CHECK: call <8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) + // CHECK: call {{.*}}<8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) return _mm256_fnmadd_ps(a, b, c); } __m256d test_mm256_fnmadd_pd(__m256d a, __m256d b, __m256d c) { // CHECK-LABEL: test_mm256_fnmadd_pd // CHECK: [[NEG:%.+]] = fneg <4 x double> %{{.+}} - // CHECK: call <4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) + // CHECK: call {{.*}}<4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) return _mm256_fnmadd_pd(a, b, c); } @@ -219,7 +222,7 @@ __m256 test_mm256_fnmsub_ps(__m256 a, __m256 b, __m256 c) { // CHECK-LABEL: test_mm256_fnmsub_ps // CHECK: [[NEG:%.+]] = fneg <8 x float> %{{.*}} // CHECK: [[NEG2:%.+]] = fneg <8 x float> %{{.*}} - // CHECK: call <8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) + // CHECK: call {{.*}}<8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) return _mm256_fnmsub_ps(a, b, c); } @@ -227,34 +230,34 @@ __m256d test_mm256_fnmsub_pd(__m256d a, __m256d b, __m256d c) { // CHECK-LABEL: test_mm256_fnmsub_pd // CHECK: [[NEG:%.+]] = fneg <4 x double> %{{.+}} // CHECK: [[NEG2:%.+]] = fneg <4 x double> %{{.+}} - // CHECK: call <4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) + // CHECK: call {{.*}}<4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) return _mm256_fnmsub_pd(a, b, c); } __m256 test_mm256_fmaddsub_ps(__m256 a, __m256 b, __m256 c) { // CHECK-LABEL: test_mm256_fmaddsub_ps // CHECK-NOT: fneg - // CHECK: call <8 x float> @llvm.x86.fma.vfmaddsub.ps.256(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) + // CHECK: call {{.*}}<8 x float> @llvm.x86.fma.vfmaddsub.ps.256(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) return _mm256_fmaddsub_ps(a, b, c); } __m256d test_mm256_fmaddsub_pd(__m256d a, __m256d b, __m256d c) { // CHECK-LABEL: test_mm256_fmaddsub_pd // CHECK-NOT: fneg - // CHECK: call <4 x double> @llvm.x86.fma.vfmaddsub.pd.256(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) + // CHECK: call {{.*}}<4 x double> @llvm.x86.fma.vfmaddsub.pd.256(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) return _mm256_fmaddsub_pd(a, b, c); } __m256 test_mm256_fmsubadd_ps(__m256 a, __m256 b, __m256 c) { // CHECK-LABEL: test_mm256_fmsubadd_ps // CHECK: [[NEG:%.+]] = fneg <8 x float> %{{.+}} - // CHECK: call <8 x float> @llvm.x86.fma.vfmaddsub.ps.256(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> [[NEG]]) + // CHECK: call {{.*}}<8 x float> @llvm.x86.fma.vfmaddsub.ps.256(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> [[NEG]]) return _mm256_fmsubadd_ps(a, b, c); } __m256d test_mm256_fmsubadd_pd(__m256d a, __m256d b, __m256d c) { // CHECK-LABEL: test_mm256_fmsubadd_pd // CHECK: [[NEG:%.+]] = fneg <4 x double> %{{.+}} - // CHECK: call <4 x double> @llvm.x86.fma.vfmaddsub.pd.256(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> [[NEG]]) + // CHECK: call {{.*}}<4 x double> @llvm.x86.fma.vfmaddsub.pd.256(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> [[NEG]]) return _mm256_fmsubadd_pd(a, b, c); } diff --git a/clang/test/CodeGen/X86/fma4-builtins.c b/clang/test/CodeGen/X86/fma4-builtins.c index 94dcaf61f2681a..ccdba8fea87b50 100644 --- a/clang/test/CodeGen/X86/fma4-builtins.c +++ b/clang/test/CodeGen/X86/fma4-builtins.c @@ -1,18 +1,20 @@ -// RUN: %clang_cc1 -flax-vector-conversions=none -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +fma4 -emit-llvm -o - -Wall -Werror | FileCheck %s -// RUN: %clang_cc1 -flax-vector-conversions=none -ffreestanding %s -triple=i386-apple-darwin -target-feature +fma4 -emit-llvm -o - -Wall -Werror | FileCheck %s +// RUN: %clang_cc1 -x c -flax-vector-conversions=none -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +fma4 -emit-llvm -o - -Wall -Werror | FileCheck %s +// RUN: %clang_cc1 -x c -flax-vector-conversions=none -ffreestanding %s -triple=i386-apple-darwin -target-feature +fma4 -emit-llvm -o - -Wall -Werror | FileCheck %s +// RUN: %clang_cc1 -x c++ -flax-vector-conversions=none -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +fma4 -emit-llvm -o - -Wall -Werror | FileCheck %s +// RUN: %clang_cc1 -x c++ -flax-vector-conversions=none -ffreestanding %s -triple=i386-apple-darwin -target-feature +fma4 -emit-llvm -o - -Wall -Werror | FileCheck %s #include <x86intrin.h> __m128 test_mm_macc_ps(__m128 a, __m128 b, __m128 c) { // CHECK-LABEL: test_mm_macc_ps - // CHECK: call <4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) + // CHECK: call {{.*}}<4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) return _mm_macc_ps(a, b, c); } __m128d test_mm_macc_pd(__m128d a, __m128d b, __m128d c) { // CHECK-LABEL: test_mm_macc_pd - // CHECK: call <2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) + // CHECK: call {{.*}}<2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) return _mm_macc_pd(a, b, c); } @@ -39,14 +41,14 @@ __m128d test_mm_macc_sd(__m128d a, __m128d b, __m128d c) { __m128 test_mm_msub_ps(__m128 a, __m128 b, __m128 c) { // CHECK-LABEL: test_mm_msub_ps // CHECK: [[NEG:%.+]] = fneg <4 x float> %{{.+}} - // CHECK: call <4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) + // CHECK: call {{.*}}<4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) return _mm_msub_ps(a, b, c); } __m128d test_mm_msub_pd(__m128d a, __m128d b, __m128d c) { // CHECK-LABEL: test_mm_msub_pd // CHECK: [[NEG:%.+]] = fneg <2 x double> %{{.+}} - // CHECK: call <2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) + // CHECK: call {{.*}}<2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) return _mm_msub_pd(a, b, c); } @@ -75,14 +77,14 @@ __m128d test_mm_msub_sd(__m128d a, __m128d b, __m128d c) { __m128 test_mm_nmacc_ps(__m128 a, __m128 b, __m128 c) { // CHECK-LABEL: test_mm_nmacc_ps // CHECK: [[NEG:%.+]] = fneg <4 x float> %{{.+}} - // CHECK: call <4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) + // CHECK: call {{.*}}<4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) return _mm_nmacc_ps(a, b, c); } __m128d test_mm_nmacc_pd(__m128d a, __m128d b, __m128d c) { // CHECK-LABEL: test_mm_nmacc_pd // CHECK: [[NEG:%.+]] = fneg <2 x double> %{{.+}} - // CHECK: call <2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) + // CHECK: call {{.*}}<2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) return _mm_nmacc_pd(a, b, c); } @@ -112,7 +114,7 @@ __m128 test_mm_nmsub_ps(__m128 a, __m128 b, __m128 c) { // CHECK-LABEL: test_mm_nmsub_ps // CHECK: [[NEG:%.+]] = fneg <4 x float> %{{.+}} // CHECK: [[NEG2:%.+]] = fneg <4 x float> %{{.+}} - // CHECK: call <4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) + // CHECK: call {{.*}}<4 x float> @llvm.fma.v4f32(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) return _mm_nmsub_ps(a, b, c); } @@ -120,7 +122,7 @@ __m128d test_mm_nmsub_pd(__m128d a, __m128d b, __m128d c) { // CHECK-LABEL: test_mm_nmsub_pd // CHECK: [[NEG:%.+]] = fneg <2 x double> %{{.+}} // CHECK: [[NEG2:%.+]] = fneg <2 x double> %{{.+}} - // CHECK: call <2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) + // CHECK: call {{.*}}<2 x double> @llvm.fma.v2f64(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) return _mm_nmsub_pd(a, b, c); } @@ -151,68 +153,68 @@ __m128d test_mm_nmsub_sd(__m128d a, __m128d b, __m128d c) { __m128 test_mm_maddsub_ps(__m128 a, __m128 b, __m128 c) { // CHECK-LABEL: test_mm_maddsub_ps // CHECK-NOT: fneg - // CHECK: call <4 x float> @llvm.x86.fma.vfmaddsub.ps(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) + // CHECK: call {{.*}}<4 x float> @llvm.x86.fma.vfmaddsub.ps(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> %{{.*}}) return _mm_maddsub_ps(a, b, c); } __m128d test_mm_maddsub_pd(__m128d a, __m128d b, __m128d c) { // CHECK-LABEL: test_mm_maddsub_pd // CHECK-NOT: fneg - // CHECK: call <2 x double> @llvm.x86.fma.vfmaddsub.pd(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) + // CHECK: call {{.*}}<2 x double> @llvm.x86.fma.vfmaddsub.pd(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}) return _mm_maddsub_pd(a, b, c); } __m128 test_mm_msubadd_ps(__m128 a, __m128 b, __m128 c) { // CHECK-LABEL: test_mm_msubadd_ps // CHECK: [[NEG:%.+]] = fneg <4 x float> %{{.+}} - // CHECK: call <4 x float> @llvm.x86.fma.vfmaddsub.ps(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> [[NEG]]) + // CHECK: call {{.*}}<4 x float> @llvm.x86.fma.vfmaddsub.ps(<4 x float> %{{.*}}, <4 x float> %{{.*}}, <4 x float> [[NEG]]) return _mm_msubadd_ps(a, b, c); } __m128d test_mm_msubadd_pd(__m128d a, __m128d b, __m128d c) { // CHECK-LABEL: test_mm_msubadd_pd // CHECK: [[NEG:%.+]] = fneg <2 x double> %{{.+}} - // CHECK: call <2 x double> @llvm.x86.fma.vfmaddsub.pd(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> [[NEG]]) + // CHECK: call {{.*}}<2 x double> @llvm.x86.fma.vfmaddsub.pd(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> [[NEG]]) return _mm_msubadd_pd(a, b, c); } __m256 test_mm256_macc_ps(__m256 a, __m256 b, __m256 c) { // CHECK-LABEL: test_mm256_macc_ps - // CHECK: call <8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) + // CHECK: call {{.*}}<8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) return _mm256_macc_ps(a, b, c); } __m256d test_mm256_macc_pd(__m256d a, __m256d b, __m256d c) { // CHECK-LABEL: test_mm256_macc_pd - // CHECK: call <4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) + // CHECK: call {{.*}}<4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) return _mm256_macc_pd(a, b, c); } __m256 test_mm256_msub_ps(__m256 a, __m256 b, __m256 c) { // CHECK-LABEL: test_mm256_msub_ps // CHECK: [[NEG:%.+]] = fneg <8 x float> %{{.*}} - // CHECK: call <8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) + // CHECK: call {{.*}}<8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) return _mm256_msub_ps(a, b, c); } __m256d test_mm256_msub_pd(__m256d a, __m256d b, __m256d c) { // CHECK-LABEL: test_mm256_msub_pd // CHECK: [[NEG:%.+]] = fneg <4 x double> %{{.+}} - // CHECK: call <4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) + // CHECK: call {{.*}}<4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) return _mm256_msub_pd(a, b, c); } __m256 test_mm256_nmacc_ps(__m256 a, __m256 b, __m256 c) { // CHECK-LABEL: test_mm256_nmacc_ps // CHECK: [[NEG:%.+]] = fneg <8 x float> %{{.*}} - // CHECK: call <8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) + // CHECK: call {{.*}}<8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) return _mm256_nmacc_ps(a, b, c); } __m256d test_mm256_nmacc_pd(__m256d a, __m256d b, __m256d c) { // CHECK-LABEL: test_mm256_nmacc_pd // CHECK: [[NEG:%.+]] = fneg <4 x double> %{{.+}} - // CHECK: call <4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) + // CHECK: call {{.*}}<4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) return _mm256_nmacc_pd(a, b, c); } @@ -220,7 +222,7 @@ __m256 test_mm256_nmsub_ps(__m256 a, __m256 b, __m256 c) { // CHECK-LABEL: test_mm256_nmsub_ps // CHECK: [[NEG:%.+]] = fneg <8 x float> %{{.*}} // CHECK: [[NEG2:%.+]] = fneg <8 x float> %{{.*}} - // CHECK: call <8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) + // CHECK: call {{.*}}<8 x float> @llvm.fma.v8f32(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) return _mm256_nmsub_ps(a, b, c); } @@ -228,34 +230,34 @@ __m256d test_mm256_nmsub_pd(__m256d a, __m256d b, __m256d c) { // CHECK-LABEL: test_mm256_nmsub_pd // CHECK: [[NEG:%.+]] = fneg <4 x double> %{{.+}} // CHECK: [[NEG2:%.+]] = fneg <4 x double> %{{.+}} - // CHECK: call <4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) + // CHECK: call {{.*}}<4 x double> @llvm.fma.v4f64(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) return _mm256_nmsub_pd(a, b, c); } __m256 test_mm256_maddsub_ps(__m256 a, __m256 b, __m256 c) { // CHECK-LABEL: test_mm256_maddsub_ps // CHECK-NOT: fneg - // CHECK: call <8 x float> @llvm.x86.fma.vfmaddsub.ps.256(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) + // CHECK: call {{.*}}<8 x float> @llvm.x86.fma.vfmaddsub.ps.256(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}}) return _mm256_maddsub_ps(a, b, c); } __m256d test_mm256_maddsub_pd(__m256d a, __m256d b, __m256d c) { // CHECK-LABEL: test_mm256_maddsub_pd // CHECK-NOT: fneg - // CHECK: call <4 x double> @llvm.x86.fma.vfmaddsub.pd.256(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) + // CHECK: call {{.*}}<4 x double> @llvm.x86.fma.vfmaddsub.pd.256(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> %{{.*}}) return _mm256_maddsub_pd(a, b, c); } __m256 test_mm256_msubadd_ps(__m256 a, __m256 b, __m256 c) { // CHECK-LABEL: test_mm256_msubadd_ps // CHECK: [[NEG:%.+]] = fneg <8 x float> %{{.*}} - // CHECK: call <8 x float> @llvm.x86.fma.vfmaddsub.ps.256(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> [[NEG]]) + // CHECK: call {{.*}}<8 x float> @llvm.x86.fma.vfmaddsub.ps.256(<8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x float> [[NEG]]) return _mm256_msubadd_ps(a, b, c); } __m256d test_mm256_msubadd_pd(__m256d a, __m256d b, __m256d c) { // CHECK-LABEL: test_mm256_msubadd_pd // CHECK: [[NEG:%.+]] = fneg <4 x double> {{.+}} - // CHECK: call <4 x double> @llvm.x86.fma.vfmaddsub.pd.256(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> [[NEG]]) + // CHECK: call {{.*}}<4 x double> @llvm.x86.fma.vfmaddsub.pd.256(<4 x double> %{{.*}}, <4 x double> %{{.*}}, <4 x double> [[NEG]]) return _mm256_msubadd_pd(a, b, c); } _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
