1 // RUN: %clang_cc1 -flax-vector-conversions=none -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +sse4a -emit-llvm -o - -Wall -Werror | FileCheck %s
2 
3 
4 #include <x86intrin.h>
5 
6 // NOTE: This should match the tests in llvm/test/CodeGen/X86/sse4a-intrinsics-fast-isel.ll
7 
test_mm_extracti_si64(__m128i x)8 __m128i test_mm_extracti_si64(__m128i x) {
9   // CHECK-LABEL: test_mm_extracti_si64
10   // CHECK: call <2 x i64> @llvm.x86.sse4a.extrqi(<2 x i64> %{{[^,]+}}, i8 3, i8 2)
11   return _mm_extracti_si64(x, 3, 2);
12 }
13 
test_mm_extract_si64(__m128i x,__m128i y)14 __m128i test_mm_extract_si64(__m128i x, __m128i y) {
15   // CHECK-LABEL: test_mm_extract_si64
16   // CHECK: call <2 x i64> @llvm.x86.sse4a.extrq(<2 x i64> %{{[^,]+}}, <16 x i8> %{{[^,]+}})
17   return _mm_extract_si64(x, y);
18 }
19 
test_mm_inserti_si64(__m128i x,__m128i y)20 __m128i test_mm_inserti_si64(__m128i x, __m128i y) {
21   // CHECK-LABEL: test_mm_inserti_si64
22   // CHECK: call <2 x i64> @llvm.x86.sse4a.insertqi(<2 x i64> %{{[^,]+}}, <2 x i64> %{{[^,]+}}, i8 5, i8 6)
23   return _mm_inserti_si64(x, y, 5, 6);
24 }
25 
test_mm_insert_si64(__m128i x,__m128i y)26 __m128i test_mm_insert_si64(__m128i x, __m128i y) {
27   // CHECK-LABEL: test_mm_insert_si64
28   // CHECK: call <2 x i64> @llvm.x86.sse4a.insertq(<2 x i64> %{{[^,]+}}, <2 x i64> %{{[^,]+}})
29   return _mm_insert_si64(x, y);
30 }
31 
test_mm_stream_sd(double * p,__m128d a)32 void test_mm_stream_sd(double *p, __m128d a) {
33   // CHECK-LABEL: test_mm_stream_sd
34   // CHECK: extractelement <2 x double> %{{.*}}, i64 0
35   // CHECK: store double %{{.*}}, double* %{{.*}}, align 1, !nontemporal
36    _mm_stream_sd(p, a);
37 }
38 
test_mm_stream_ss(float * p,__m128 a)39 void test_mm_stream_ss(float *p, __m128 a) {
40   // CHECK-LABEL: test_mm_stream_ss
41   // CHECK: extractelement <4 x float> %{{.*}}, i64 0
42   // CHECK: store float %{{.*}}, float* %{{.*}}, align 1, !nontemporal
43   _mm_stream_ss(p, a);
44 }
45