replica
/
hanchenye-llvm-project
2
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[x86] Fix bugs of some intrinsic functions in CLANG : _mm512_stream_ps, _mm512_stream_pd, _mm512_stream_si512 

Reviewers: craig.topper, pengfei, LuoYuanke, RKSimon, spatel

Reviewed By: RKSimon

Subscribers: llvm-commits

Patch by Bing Yu (yubing)

Differential Revision: https://reviews.llvm.org/D66786

llvm-svn: 370691
This commit is contained in:
Pengfei Wang 2019-09-03 02:06:15 +00:00
parent ea366122d2
commit dea9cad10e
2 changed files with 20 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
clang/lib/Headers/avx512fintrin.h
View File

@ -8436,7 +8436,7 @@ _store_mask16(__mmask16 *__A, __mmask16 __B) {
}
static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_stream_si512 (__m512i * __P, __m512i __A)
_mm512_stream_si512 (void * __P, __m512i __A)
{
typedef __v8di __v8di_aligned __attribute__((aligned(64)));
__builtin_nontemporal_store((__v8di_aligned)__A, (__v8di_aligned*)__P);
@ -8450,14 +8450,14 @@ _mm512_stream_load_si512 (void const *__P)
}
static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_stream_pd (double *__P, __m512d __A)
_mm512_stream_pd (void *__P, __m512d __A)
{
typedef __v8df __v8df_aligned __attribute__((aligned(64)));
__builtin_nontemporal_store((__v8df_aligned)__A, (__v8df_aligned*)__P);
}
static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_stream_ps (float *__P, __m512 __A)
_mm512_stream_ps (void *__P, __m512 __A)
{
typedef __v16sf __v16sf_aligned __attribute__((aligned(64)));
__builtin_nontemporal_store((__v16sf_aligned)__A, (__v16sf_aligned*)__P);

17
clang/test/CodeGen/avx512f-builtins.c
View File

@ -8575,6 +8575,12 @@ void test_mm512_stream_si512(__m512i * __P, __m512i __A) {
_mm512_stream_si512(__P, __A);
}
void test_mm512_stream_si512_2(void * __P, __m512i __A) {
// CHECK-LABEL: @test_mm512_stream_si512
// CHECK: store <8 x i64> %{{.*}}, <8 x i64>* %{{.*}}, align 64, !nontemporal
_mm512_stream_si512(__P, __A);
}
__m512i test_mm512_stream_load_si512(void *__P) {
// CHECK-LABEL: @test_mm512_stream_load_si512
// CHECK: load <8 x i64>, <8 x i64>* %{{.*}}, align 64, !nontemporal
@ -8593,12 +8599,23 @@ void test_mm512_stream_pd(double *__P, __m512d __A) {
return _mm512_stream_pd(__P, __A);
}
void test_mm512_stream_pd_2(void *__P, __m512d __A) {
// CHECK-LABEL: @test_mm512_stream_pd
// CHECK: store <8 x double> %{{.*}}, <8 x double>* %{{.*}}, align 64, !nontemporal
return _mm512_stream_pd(__P, __A);
}
void test_mm512_stream_ps(float *__P, __m512 __A) {
// CHECK-LABEL: @test_mm512_stream_ps
// CHECK: store <16 x float> %{{.*}}, <16 x float>* %{{.*}}, align 64, !nontemporal
_mm512_stream_ps(__P, __A);
}
void test_mm512_stream_ps_2(void *__P, __m512 __A) {
// CHECK-LABEL: @test_mm512_stream_ps
// CHECK: store <16 x float> %{{.*}}, <16 x float>* %{{.*}}, align 64, !nontemporal
_mm512_stream_ps(__P, __A);
}
__m512d test_mm512_mask_compress_pd(__m512d __W, __mmask8 __U, __m512d __A) {
// CHECK-LABEL: @test_mm512_mask_compress_pd
// CHECK: @llvm.x86.avx512.mask.compress