diff options
Diffstat (limited to 'test/CodeGen/sse4a-builtins.c')
-rw-r--r-- | test/CodeGen/sse4a-builtins.c | 43 |
1 files changed, 23 insertions, 20 deletions
diff --git a/test/CodeGen/sse4a-builtins.c b/test/CodeGen/sse4a-builtins.c index e1d7e8f..9a408b8 100644 --- a/test/CodeGen/sse4a-builtins.c +++ b/test/CodeGen/sse4a-builtins.c @@ -1,39 +1,42 @@ -// RUN: %clang_cc1 -ffreestanding -triple i386-apple-darwin9 -target-cpu pentium4 -target-feature +sse4a -g -emit-llvm %s -o - | FileCheck %s +// RUN: %clang_cc1 %s -triple=x86_64-apple-darwin -target-feature +sse4a -emit-llvm -o - -Werror | FileCheck %s -#include <ammintrin.h> +// Don't include mm_malloc.h, it's system specific. +#define __MM_MALLOC_H -__m128i test_extracti_si64(__m128i x) { +#include <x86intrin.h> + +__m128i test_mm_extracti_si64(__m128i x) { + // CHECK-LABEL: test_mm_extracti_si64 + // CHECK: call <2 x i64> @llvm.x86.sse4a.extrqi(<2 x i64> %{{[^,]+}}, i8 3, i8 2) return _mm_extracti_si64(x, 3, 2); -// CHECK: @test_extracti_si64 -// CHECK: @llvm.x86.sse4a.extrqi(<2 x i64> %{{[^,]+}}, i8 3, i8 2) } -__m128i test_extract_si64(__m128i x, __m128i y) { +__m128i test_mm_extract_si64(__m128i x, __m128i y) { + // CHECK-LABEL: test_mm_extract_si64 + // CHECK: call <2 x i64> @llvm.x86.sse4a.extrq(<2 x i64> %{{[^,]+}}, <16 x i8> %{{[^,]+}}) return _mm_extract_si64(x, y); -// CHECK: @test_extract_si64 -// CHECK: @llvm.x86.sse4a.extrq(<2 x i64> %{{[^,]+}}, <16 x i8> %{{[^,]+}}) } -__m128i test_inserti_si64(__m128i x, __m128i y) { +__m128i test_mm_inserti_si64(__m128i x, __m128i y) { + // CHECK-LABEL: test_mm_inserti_si64 + // CHECK: call <2 x i64> @llvm.x86.sse4a.insertqi(<2 x i64> %{{[^,]+}}, <2 x i64> %{{[^,]+}}, i8 5, i8 6) return _mm_inserti_si64(x, y, 5, 6); -// CHECK: @test_inserti_si64 -// CHECK: @llvm.x86.sse4a.insertqi(<2 x i64> %{{[^,]+}}, <2 x i64> %{{[^,]+}}, i8 5, i8 6) } -__m128i test_insert_si64(__m128i x, __m128i y) { +__m128i test_mm_insert_si64(__m128i x, __m128i y) { + // CHECK-LABEL: test_mm_insert_si64 + // CHECK: call <2 x i64> @llvm.x86.sse4a.insertq(<2 x i64> %{{[^,]+}}, <2 x i64> %{{[^,]+}}) return _mm_insert_si64(x, y); -// CHECK: @test_insert_si64 -// CHECK: @llvm.x86.sse4a.insertq(<2 x i64> %{{[^,]+}}, <2 x i64> %{{[^,]+}}) } -void test_stream_sd(double *p, __m128d a) { +void test_mm_stream_sd(double *p, __m128d a) { + // CHECK-LABEL: test_mm_stream_sd + // CHECK: call void @llvm.x86.sse4a.movnt.sd(i8* %{{[^,]+}}, <2 x double> %{{[^,]+}}) _mm_stream_sd(p, a); -// CHECK: @test_stream_sd -// CHECK: @llvm.x86.sse4a.movnt.sd(i8* %{{[^,]+}}, <2 x double> %{{[^,]+}}) } -void test_stream_ss(float *p, __m128 a) { +void test_mm_stream_ss(float *p, __m128 a) { + // CHECK-LABEL: test_mm_stream_ss + // CHECK: call void @llvm.x86.sse4a.movnt.ss(i8* %{{[^,]+}}, <4 x float> %{{[^,]+}}) _mm_stream_ss(p, a); -// CHECK: @test_stream_ss -// CHECK: @llvm.x86.sse4a.movnt.ss(i8* %{{[^,]+}}, <4 x float> %{{[^,]+}}) } |