From d04eb5b33deee850e955748729953769a1a42e52 Mon Sep 17 00:00:00 2001
From: Pengfei Wang <pengfei.wang@intel.com>
Date: Tue, 3 Sep 2019 02:06:15 +0000
Subject: [PATCH] [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

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@370691 91177308-0d34-0410-b5e6-96231b3b80d8
---
 lib/Headers/avx512fintrin.h     |  6 +++---
 test/CodeGen/avx512f-builtins.c | 17 +++++++++++++++++
 2 files changed, 20 insertions(+), 3 deletions(-)

diff --git a/lib/Headers/avx512fintrin.h b/lib/Headers/avx512fintrin.h
index 5748c1ea6e..4e341a10fa 100644
--- a/lib/Headers/avx512fintrin.h
+++ b/lib/Headers/avx512fintrin.h
@@ -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);
diff --git a/test/CodeGen/avx512f-builtins.c b/test/CodeGen/avx512f-builtins.c
index aaf3157c08..2a083e3e5c 100644
--- a/test/CodeGen/avx512f-builtins.c
+++ b/test/CodeGen/avx512f-builtins.c
@@ -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
-- 
2.40.0