r287313 - [AVX-512] Replace masked 16-bit element variable shift builtins with new unmasked versions and selects.
Craig Topper via cfe-commits
cfe-commits at lists.llvm.org
Thu Nov 17 21:04:51 PST 2016
Author: ctopper
Date: Thu Nov 17 23:04:51 2016
New Revision: 287313
URL: http://llvm.org/viewvc/llvm-project?rev=287313&view=rev
Log:
[AVX-512] Replace masked 16-bit element variable shift builtins with new unmasked versions and selects.
Modified:
cfe/trunk/include/clang/Basic/BuiltinsX86.def
cfe/trunk/lib/Headers/avx512bwintrin.h
cfe/trunk/lib/Headers/avx512vlbwintrin.h
cfe/trunk/test/CodeGen/avx512bw-builtins.c
cfe/trunk/test/CodeGen/avx512vlbw-builtins.c
Modified: cfe/trunk/include/clang/Basic/BuiltinsX86.def
URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/include/clang/Basic/BuiltinsX86.def?rev=287313&r1=287312&r2=287313&view=diff
==============================================================================
--- cfe/trunk/include/clang/Basic/BuiltinsX86.def (original)
+++ cfe/trunk/include/clang/Basic/BuiltinsX86.def Thu Nov 17 23:04:51 2016
@@ -1352,21 +1352,21 @@ TARGET_BUILTIN(__builtin_ia32_prorvd128_
TARGET_BUILTIN(__builtin_ia32_prorvd256_mask, "V8iV8iV8iV8iUc","","avx512vl")
TARGET_BUILTIN(__builtin_ia32_prorvq128_mask, "V2LLiV2LLiV2LLiV2LLiUc","","avx512vl")
TARGET_BUILTIN(__builtin_ia32_prorvq256_mask, "V4LLiV4LLiV4LLiV4LLiUc","","avx512vl")
-TARGET_BUILTIN(__builtin_ia32_psllv32hi_mask, "V32sV32sV32sV32sUi","","avx512bw")
+TARGET_BUILTIN(__builtin_ia32_psllv32hi, "V32sV32sV32s","","avx512bw")
TARGET_BUILTIN(__builtin_ia32_psllw512, "V32sV32sV8s","","avx512bw")
TARGET_BUILTIN(__builtin_ia32_psllwi512, "V32sV32si","","avx512bw")
-TARGET_BUILTIN(__builtin_ia32_psllv16hi_mask, "V16sV16sV16sV16sUs","","avx512bw,avx512vl")
-TARGET_BUILTIN(__builtin_ia32_psllv8hi_mask, "V8sV8sV8sV8sUc","","avx512bw,avx512vl")
+TARGET_BUILTIN(__builtin_ia32_psllv16hi, "V16sV16sV16s","","avx512bw,avx512vl")
+TARGET_BUILTIN(__builtin_ia32_psllv8hi, "V8sV8sV8s","","avx512bw,avx512vl")
TARGET_BUILTIN(__builtin_ia32_pslldi512, "V16iV16ii","","avx512f")
TARGET_BUILTIN(__builtin_ia32_psllqi512, "V8LLiV8LLii","","avx512f")
-TARGET_BUILTIN(__builtin_ia32_psrlv32hi_mask, "V32sV32sV32sV32sUi","","avx512bw")
-TARGET_BUILTIN(__builtin_ia32_psrlv16hi_mask, "V16sV16sV16sV16sUs","","avx512bw,avx512vl")
-TARGET_BUILTIN(__builtin_ia32_psrlv8hi_mask, "V8sV8sV8sV8sUc","","avx512bw,avx512vl")
+TARGET_BUILTIN(__builtin_ia32_psrlv32hi, "V32sV32sV32s","","avx512bw")
+TARGET_BUILTIN(__builtin_ia32_psrlv16hi, "V16sV16sV16s","","avx512bw,avx512vl")
+TARGET_BUILTIN(__builtin_ia32_psrlv8hi, "V8sV8sV8s","","avx512bw,avx512vl")
TARGET_BUILTIN(__builtin_ia32_psrldi512, "V16iV16ii","","avx512f")
TARGET_BUILTIN(__builtin_ia32_psrlqi512, "V8LLiV8LLii","","avx512f")
-TARGET_BUILTIN(__builtin_ia32_psrav32hi_mask, "V32sV32sV32sV32sUi","","avx512bw")
-TARGET_BUILTIN(__builtin_ia32_psrav16hi_mask, "V16sV16sV16sV16sUs","","avx512bw,avx512vl")
-TARGET_BUILTIN(__builtin_ia32_psrav8hi_mask, "V8sV8sV8sV8sUc","","avx512bw,avx512vl")
+TARGET_BUILTIN(__builtin_ia32_psrav32hi, "V32sV32sV32s","","avx512bw")
+TARGET_BUILTIN(__builtin_ia32_psrav16hi, "V16sV16sV16s","","avx512bw,avx512vl")
+TARGET_BUILTIN(__builtin_ia32_psrav8hi, "V8sV8sV8s","","avx512bw,avx512vl")
TARGET_BUILTIN(__builtin_ia32_psravq128, "V2LLiV2LLiV2LLi","","avx512vl")
TARGET_BUILTIN(__builtin_ia32_psravq256, "V4LLiV4LLiV4LLi","","avx512vl")
TARGET_BUILTIN(__builtin_ia32_psraw512, "V32sV32sV8s","","avx512bw")
Modified: cfe/trunk/lib/Headers/avx512bwintrin.h
URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Headers/avx512bwintrin.h?rev=287313&r1=287312&r2=287313&view=diff
==============================================================================
--- cfe/trunk/lib/Headers/avx512bwintrin.h (original)
+++ cfe/trunk/lib/Headers/avx512bwintrin.h Thu Nov 17 23:04:51 2016
@@ -1688,33 +1688,25 @@ _mm512_maskz_cvtepu8_epi16(__mmask32 __U
(__v32hi)_mm512_setzero_hi()); })
static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_sllv_epi16 (__m512i __A, __m512i __B)
+_mm512_sllv_epi16(__m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_psllv32hi_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi)
- _mm512_setzero_hi (),
- (__mmask32) -1);
+ return (__m512i)__builtin_ia32_psllv32hi((__v32hi) __A, (__v32hi) __B);
}
static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_sllv_epi16 (__m512i __W, __mmask32 __U, __m512i __A,
- __m512i __B)
+_mm512_mask_sllv_epi16 (__m512i __W, __mmask32 __U, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_psllv32hi_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) __W,
- (__mmask32) __U);
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_sllv_epi16(__A, __B),
+ (__v32hi)__W);
}
static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_sllv_epi16 (__mmask32 __U, __m512i __A, __m512i __B)
+_mm512_maskz_sllv_epi16(__mmask32 __U, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_psllv32hi_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi)
- _mm512_setzero_hi (),
- (__mmask32) __U);
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_sllv_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_hi());
}
static __inline__ __m512i __DEFAULT_FN_ATTRS
@@ -1831,63 +1823,47 @@ _mm512_maskz_slli_epi16(__mmask32 __U, _
((char)(imm)&0xF0) ? 63 : ((char)(imm)>0xF ? 79 : 127) - (char)(imm)); })
static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_srlv_epi16 (__m512i __A, __m512i __B)
+_mm512_srlv_epi16(__m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_psrlv32hi_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi)
- _mm512_setzero_hi (),
- (__mmask32) -1);
+ return (__m512i)__builtin_ia32_psrlv32hi((__v32hi)__A, (__v32hi)__B);
}
static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_srlv_epi16 (__m512i __W, __mmask32 __U, __m512i __A,
- __m512i __B)
+_mm512_mask_srlv_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_psrlv32hi_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) __W,
- (__mmask32) __U);
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_srlv_epi16(__A, __B),
+ (__v32hi)__W);
}
static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_srlv_epi16 (__mmask32 __U, __m512i __A, __m512i __B)
+_mm512_maskz_srlv_epi16(__mmask32 __U, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_psrlv32hi_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi)
- _mm512_setzero_hi (),
- (__mmask32) __U);
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_srlv_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_hi());
}
static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_srav_epi16 (__m512i __A, __m512i __B)
+_mm512_srav_epi16(__m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_psrav32hi_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi)
- _mm512_setzero_hi (),
- (__mmask32) -1);
+ return (__m512i)__builtin_ia32_psrav32hi((__v32hi)__A, (__v32hi)__B);
}
static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_srav_epi16 (__m512i __W, __mmask32 __U, __m512i __A,
- __m512i __B)
+_mm512_mask_srav_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_psrav32hi_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi) __W,
- (__mmask32) __U);
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_srav_epi16(__A, __B),
+ (__v32hi)__W);
}
static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_srav_epi16 (__mmask32 __U, __m512i __A, __m512i __B)
+_mm512_maskz_srav_epi16(__mmask32 __U, __m512i __A, __m512i __B)
{
- return (__m512i) __builtin_ia32_psrav32hi_mask ((__v32hi) __A,
- (__v32hi) __B,
- (__v32hi)
- _mm512_setzero_hi (),
- (__mmask32) __U);
+ return (__m512i)__builtin_ia32_selectw_512((__mmask32)__U,
+ (__v32hi)_mm512_srav_epi16(__A, __B),
+ (__v32hi)_mm512_setzero_hi());
}
static __inline__ __m512i __DEFAULT_FN_ATTRS
Modified: cfe/trunk/lib/Headers/avx512vlbwintrin.h
URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Headers/avx512vlbwintrin.h?rev=287313&r1=287312&r2=287313&view=diff
==============================================================================
--- cfe/trunk/lib/Headers/avx512vlbwintrin.h (original)
+++ cfe/trunk/lib/Headers/avx512vlbwintrin.h Thu Nov 17 23:04:51 2016
@@ -2269,63 +2269,47 @@ _mm256_maskz_cvtepu8_epi16 (__mmask16 __
(__v16hi)_mm256_setzero_si256()); })
static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_sllv_epi16 (__m256i __A, __m256i __B)
+_mm256_sllv_epi16(__m256i __A, __m256i __B)
{
- return (__m256i) __builtin_ia32_psllv16hi_mask ((__v16hi) __A,
- (__v16hi) __B,
- (__v16hi)
- _mm256_setzero_si256 (),
- (__mmask16) -1);
+ return (__m256i)__builtin_ia32_psllv16hi((__v16hi)__A, (__v16hi)__B);
}
static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_sllv_epi16 (__m256i __W, __mmask16 __U, __m256i __A,
- __m256i __B)
+_mm256_mask_sllv_epi16(__m256i __W, __mmask16 __U, __m256i __A, __m256i __B)
{
- return (__m256i) __builtin_ia32_psllv16hi_mask ((__v16hi) __A,
- (__v16hi) __B,
- (__v16hi) __W,
- (__mmask16) __U);
+ return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
+ (__v16hi)_mm256_sllv_epi16(__A, __B),
+ (__v16hi)__W);
}
static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_sllv_epi16 (__mmask16 __U, __m256i __A, __m256i __B)
+_mm256_maskz_sllv_epi16(__mmask16 __U, __m256i __A, __m256i __B)
{
- return (__m256i) __builtin_ia32_psllv16hi_mask ((__v16hi) __A,
- (__v16hi) __B,
- (__v16hi)
- _mm256_setzero_si256 (),
- (__mmask16) __U);
+ return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
+ (__v16hi)_mm256_sllv_epi16(__A, __B),
+ (__v16hi)_mm256_setzero_si256());
}
static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_sllv_epi16 (__m128i __A, __m128i __B)
+_mm_sllv_epi16(__m128i __A, __m128i __B)
{
- return (__m128i) __builtin_ia32_psllv8hi_mask ((__v8hi) __A,
- (__v8hi) __B,
- (__v8hi)
- _mm_setzero_hi (),
- (__mmask8) -1);
+ return (__m128i)__builtin_ia32_psllv8hi((__v8hi)__A, (__v8hi)__B);
}
static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_sllv_epi16 (__m128i __W, __mmask8 __U, __m128i __A,
- __m128i __B)
+_mm_mask_sllv_epi16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
- return (__m128i) __builtin_ia32_psllv8hi_mask ((__v8hi) __A,
- (__v8hi) __B,
- (__v8hi) __W,
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
+ (__v8hi)_mm_sllv_epi16(__A, __B),
+ (__v8hi)__W);
}
static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_sllv_epi16 (__mmask8 __U, __m128i __A, __m128i __B)
+_mm_maskz_sllv_epi16(__mmask8 __U, __m128i __A, __m128i __B)
{
- return (__m128i) __builtin_ia32_psllv8hi_mask ((__v8hi) __A,
- (__v8hi) __B,
- (__v8hi)
- _mm_setzero_si128 (),
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
+ (__v8hi)_mm_sllv_epi16(__A, __B),
+ (__v8hi)_mm_setzero_si128());
}
static __inline__ __m128i __DEFAULT_FN_ATTRS
@@ -2393,123 +2377,91 @@ _mm256_maskz_slli_epi16(__mmask16 __U, _
}
static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_srlv_epi16 (__m256i __A, __m256i __B)
+_mm256_srlv_epi16(__m256i __A, __m256i __B)
{
- return (__m256i) __builtin_ia32_psrlv16hi_mask ((__v16hi) __A,
- (__v16hi) __B,
- (__v16hi)
- _mm256_setzero_si256 (),
- (__mmask16) -1);
+ return (__m256i)__builtin_ia32_psrlv16hi((__v16hi)__A, (__v16hi)__B);
}
static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_srlv_epi16 (__m256i __W, __mmask16 __U, __m256i __A,
- __m256i __B)
+_mm256_mask_srlv_epi16(__m256i __W, __mmask16 __U, __m256i __A, __m256i __B)
{
- return (__m256i) __builtin_ia32_psrlv16hi_mask ((__v16hi) __A,
- (__v16hi) __B,
- (__v16hi) __W,
- (__mmask16) __U);
+ return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
+ (__v16hi)_mm256_srlv_epi16(__A, __B),
+ (__v16hi)__W);
}
static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_srlv_epi16 (__mmask16 __U, __m256i __A, __m256i __B)
+_mm256_maskz_srlv_epi16(__mmask16 __U, __m256i __A, __m256i __B)
{
- return (__m256i) __builtin_ia32_psrlv16hi_mask ((__v16hi) __A,
- (__v16hi) __B,
- (__v16hi)
- _mm256_setzero_si256 (),
- (__mmask16) __U);
+ return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
+ (__v16hi)_mm256_srlv_epi16(__A, __B),
+ (__v16hi)_mm256_setzero_si256());
}
static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_srlv_epi16 (__m128i __A, __m128i __B)
+_mm_srlv_epi16(__m128i __A, __m128i __B)
{
- return (__m128i) __builtin_ia32_psrlv8hi_mask ((__v8hi) __A,
- (__v8hi) __B,
- (__v8hi)
- _mm_setzero_hi (),
- (__mmask8) -1);
+ return (__m128i)__builtin_ia32_psrlv8hi((__v8hi)__A, (__v8hi)__B);
}
static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_srlv_epi16 (__m128i __W, __mmask8 __U, __m128i __A,
- __m128i __B)
+_mm_mask_srlv_epi16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
- return (__m128i) __builtin_ia32_psrlv8hi_mask ((__v8hi) __A,
- (__v8hi) __B,
- (__v8hi) __W,
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
+ (__v8hi)_mm_srlv_epi16(__A, __B),
+ (__v8hi)__W);
}
static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_srlv_epi16 (__mmask8 __U, __m128i __A, __m128i __B)
+_mm_maskz_srlv_epi16(__mmask8 __U, __m128i __A, __m128i __B)
{
- return (__m128i) __builtin_ia32_psrlv8hi_mask ((__v8hi) __A,
- (__v8hi) __B,
- (__v8hi)
- _mm_setzero_si128 (),
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
+ (__v8hi)_mm_srlv_epi16(__A, __B),
+ (__v8hi)_mm_setzero_si128());
}
static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_srav_epi16 (__m256i __A, __m256i __B)
+_mm256_srav_epi16(__m256i __A, __m256i __B)
{
- return (__m256i) __builtin_ia32_psrav16hi_mask ((__v16hi) __A,
- (__v16hi) __B,
- (__v16hi)
- _mm256_setzero_si256 (),
- (__mmask16) -1);
+ return (__m256i)__builtin_ia32_psrav16hi((__v16hi)__A, (__v16hi)__B);
}
static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_mask_srav_epi16 (__m256i __W, __mmask16 __U, __m256i __A,
- __m256i __B)
+_mm256_mask_srav_epi16(__m256i __W, __mmask16 __U, __m256i __A, __m256i __B)
{
- return (__m256i) __builtin_ia32_psrav16hi_mask ((__v16hi) __A,
- (__v16hi) __B,
- (__v16hi) __W,
- (__mmask16) __U);
+ return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
+ (__v16hi)_mm256_srav_epi16(__A, __B),
+ (__v16hi)__W);
}
static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm256_maskz_srav_epi16 (__mmask16 __U, __m256i __A, __m256i __B)
+_mm256_maskz_srav_epi16(__mmask16 __U, __m256i __A, __m256i __B)
{
- return (__m256i) __builtin_ia32_psrav16hi_mask ((__v16hi) __A,
- (__v16hi) __B,
- (__v16hi)
- _mm256_setzero_si256 (),
- (__mmask16) __U);
+ return (__m256i)__builtin_ia32_selectw_256((__mmask16)__U,
+ (__v16hi)_mm256_srav_epi16(__A, __B),
+ (__v16hi)_mm256_setzero_si256());
}
static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_srav_epi16 (__m128i __A, __m128i __B)
+_mm_srav_epi16(__m128i __A, __m128i __B)
{
- return (__m128i) __builtin_ia32_psrav8hi_mask ((__v8hi) __A,
- (__v8hi) __B,
- (__v8hi)
- _mm_setzero_hi (),
- (__mmask8) -1);
+ return (__m128i)__builtin_ia32_psrav8hi((__v8hi)__A, (__v8hi)__B);
}
static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_mask_srav_epi16 (__m128i __W, __mmask8 __U, __m128i __A,
- __m128i __B)
+_mm_mask_srav_epi16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B)
{
- return (__m128i) __builtin_ia32_psrav8hi_mask ((__v8hi) __A,
- (__v8hi) __B,
- (__v8hi) __W,
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
+ (__v8hi)_mm_srav_epi16(__A, __B),
+ (__v8hi)__W);
}
static __inline__ __m128i __DEFAULT_FN_ATTRS
-_mm_maskz_srav_epi16 (__mmask8 __U, __m128i __A, __m128i __B)
+_mm_maskz_srav_epi16(__mmask8 __U, __m128i __A, __m128i __B)
{
- return (__m128i) __builtin_ia32_psrav8hi_mask ((__v8hi) __A,
- (__v8hi) __B,
- (__v8hi)
- _mm_setzero_si128 (),
- (__mmask8) __U);
+ return (__m128i)__builtin_ia32_selectw_128((__mmask8)__U,
+ (__v8hi)_mm_srav_epi16(__A, __B),
+ (__v8hi)_mm_setzero_si128());
}
static __inline__ __m128i __DEFAULT_FN_ATTRS
Modified: cfe/trunk/test/CodeGen/avx512bw-builtins.c
URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/test/CodeGen/avx512bw-builtins.c?rev=287313&r1=287312&r2=287313&view=diff
==============================================================================
--- cfe/trunk/test/CodeGen/avx512bw-builtins.c (original)
+++ cfe/trunk/test/CodeGen/avx512bw-builtins.c Thu Nov 17 23:04:51 2016
@@ -1207,19 +1207,21 @@ __m512i test_mm512_maskz_shufflelo_epi16
__m512i test_mm512_sllv_epi16(__m512i __A, __m512i __B) {
// CHECK-LABEL: @test_mm512_sllv_epi16
- // CHECK: @llvm.x86.avx512.mask.psllv
+ // CHECK: @llvm.x86.avx512.psllv.w.512(
return _mm512_sllv_epi16(__A, __B);
}
__m512i test_mm512_mask_sllv_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m512i __B) {
// CHECK-LABEL: @test_mm512_mask_sllv_epi16
- // CHECK: @llvm.x86.avx512.mask.psllv
+ // CHECK: @llvm.x86.avx512.psllv.w.512(
+ // CHECK: select <32 x i1> %{{.*}}, <32 x i16> %{{.*}}, <32 x i16> %{{.*}}
return _mm512_mask_sllv_epi16(__W, __U, __A, __B);
}
__m512i test_mm512_maskz_sllv_epi16(__mmask32 __U, __m512i __A, __m512i __B) {
// CHECK-LABEL: @test_mm512_maskz_sllv_epi16
- // CHECK: @llvm.x86.avx512.mask.psllv
+ // CHECK: @llvm.x86.avx512.psllv.w.512(
+ // CHECK: select <32 x i1> %{{.*}}, <32 x i16> %{{.*}}, <32 x i16> %{{.*}}
return _mm512_maskz_sllv_epi16(__U, __A, __B);
}
@@ -1271,37 +1273,41 @@ __m512i test_mm512_bslli_epi128(__m512i
__m512i test_mm512_srlv_epi16(__m512i __A, __m512i __B) {
// CHECK-LABEL: @test_mm512_srlv_epi16
- // CHECK: @llvm.x86.avx512.mask.psrlv
+ // CHECK: @llvm.x86.avx512.psrlv.w.512(
return _mm512_srlv_epi16(__A, __B);
}
__m512i test_mm512_mask_srlv_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m512i __B) {
// CHECK-LABEL: @test_mm512_mask_srlv_epi16
- // CHECK: @llvm.x86.avx512.mask.psrlv
+ // CHECK: @llvm.x86.avx512.psrlv.w.512(
+ // CHECK: select <32 x i1> %{{.*}}, <32 x i16> %{{.*}}, <32 x i16> %{{.*}}
return _mm512_mask_srlv_epi16(__W, __U, __A, __B);
}
__m512i test_mm512_maskz_srlv_epi16(__mmask32 __U, __m512i __A, __m512i __B) {
// CHECK-LABEL: @test_mm512_maskz_srlv_epi16
- // CHECK: @llvm.x86.avx512.mask.psrlv
+ // CHECK: @llvm.x86.avx512.psrlv.w.512(
+ // CHECK: select <32 x i1> %{{.*}}, <32 x i16> %{{.*}}, <32 x i16> %{{.*}}
return _mm512_maskz_srlv_epi16(__U, __A, __B);
}
__m512i test_mm512_srav_epi16(__m512i __A, __m512i __B) {
// CHECK-LABEL: @test_mm512_srav_epi16
- // CHECK: @llvm.x86.avx512.mask.psrav
+ // CHECK: @llvm.x86.avx512.psrav.w.512(
return _mm512_srav_epi16(__A, __B);
}
__m512i test_mm512_mask_srav_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m512i __B) {
// CHECK-LABEL: @test_mm512_mask_srav_epi16
- // CHECK: @llvm.x86.avx512.mask.psrav
+ // CHECK: @llvm.x86.avx512.psrav.w.512(
+ // CHECK: select <32 x i1> %{{.*}}, <32 x i16> %{{.*}}, <32 x i16> %{{.*}}
return _mm512_mask_srav_epi16(__W, __U, __A, __B);
}
__m512i test_mm512_maskz_srav_epi16(__mmask32 __U, __m512i __A, __m512i __B) {
// CHECK-LABEL: @test_mm512_maskz_srav_epi16
- // CHECK: @llvm.x86.avx512.mask.psrav
+ // CHECK: @llvm.x86.avx512.psrav.w.512(
+ // CHECK: select <32 x i1> %{{.*}}, <32 x i16> %{{.*}}, <32 x i16> %{{.*}}
return _mm512_maskz_srav_epi16(__U, __A, __B);
}
Modified: cfe/trunk/test/CodeGen/avx512vlbw-builtins.c
URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/test/CodeGen/avx512vlbw-builtins.c?rev=287313&r1=287312&r2=287313&view=diff
==============================================================================
--- cfe/trunk/test/CodeGen/avx512vlbw-builtins.c (original)
+++ cfe/trunk/test/CodeGen/avx512vlbw-builtins.c Thu Nov 17 23:04:51 2016
@@ -2006,37 +2006,41 @@ __m256i test_mm256_maskz_cvtepu8_epi16(_
__m256i test_mm256_sllv_epi16(__m256i __A, __m256i __B) {
// CHECK-LABEL: @test_mm256_sllv_epi16
- // CHECK: @llvm.x86.avx512.mask.psllv
+ // CHECK: @llvm.x86.avx512.psllv.w.256(
return _mm256_sllv_epi16(__A, __B);
}
__m256i test_mm256_mask_sllv_epi16(__m256i __W, __mmask16 __U, __m256i __A, __m256i __B) {
// CHECK-LABEL: @test_mm256_mask_sllv_epi16
- // CHECK: @llvm.x86.avx512.mask.psllv
+ // CHECK: @llvm.x86.avx512.psllv.w.256(
+ // CHECK: select <16 x i1> %{{.*}}, <16 x i16> %{{.*}}, <16 x i16> %{{.*}}
return _mm256_mask_sllv_epi16(__W, __U, __A, __B);
}
__m256i test_mm256_maskz_sllv_epi16(__mmask16 __U, __m256i __A, __m256i __B) {
// CHECK-LABEL: @test_mm256_maskz_sllv_epi16
- // CHECK: @llvm.x86.avx512.mask.psllv
+ // CHECK: @llvm.x86.avx512.psllv.w.256(
+ // CHECK: select <16 x i1> %{{.*}}, <16 x i16> %{{.*}}, <16 x i16> %{{.*}}
return _mm256_maskz_sllv_epi16(__U, __A, __B);
}
__m128i test_mm_sllv_epi16(__m128i __A, __m128i __B) {
// CHECK-LABEL: @test_mm_sllv_epi16
- // CHECK: @llvm.x86.avx512.mask.psllv
+ // CHECK: @llvm.x86.avx512.psllv.w.128(
return _mm_sllv_epi16(__A, __B);
}
__m128i test_mm_mask_sllv_epi16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B) {
// CHECK-LABEL: @test_mm_mask_sllv_epi16
- // CHECK: @llvm.x86.avx512.mask.psllv
+ // CHECK: @llvm.x86.avx512.psllv.w.128(
+ // CHECK: select <8 x i1> %{{.*}}, <8 x i16> %{{.*}}, <8 x i16> %{{.*}}
return _mm_mask_sllv_epi16(__W, __U, __A, __B);
}
__m128i test_mm_maskz_sllv_epi16(__mmask8 __U, __m128i __A, __m128i __B) {
// CHECK-LABEL: @test_mm_maskz_sllv_epi16
- // CHECK: @llvm.x86.avx512.mask.psllv
+ // CHECK: @llvm.x86.avx512.psllv.w.128(
+ // CHECK: select <8 x i1> %{{.*}}, <8 x i16> %{{.*}}, <8 x i16> %{{.*}}
return _mm_maskz_sllv_epi16(__U, __A, __B);
}
@@ -2098,37 +2102,41 @@ __m256i test_mm256_maskz_slli_epi16(__mm
__m256i test_mm256_srlv_epi16(__m256i __A, __m256i __B) {
// CHECK-LABEL: @test_mm256_srlv_epi16
- // CHECK: @llvm.x86.avx512.mask.psrlv
+ // CHECK: @llvm.x86.avx512.psrlv.w.256(
return _mm256_srlv_epi16(__A, __B);
}
__m256i test_mm256_mask_srlv_epi16(__m256i __W, __mmask16 __U, __m256i __A, __m256i __B) {
// CHECK-LABEL: @test_mm256_mask_srlv_epi16
- // CHECK: @llvm.x86.avx512.mask.psrlv
+ // CHECK: @llvm.x86.avx512.psrlv.w.256(
+ // CHECK: select <16 x i1> %{{.*}}, <16 x i16> %{{.*}}, <16 x i16> %{{.*}}
return _mm256_mask_srlv_epi16(__W, __U, __A, __B);
}
__m256i test_mm256_maskz_srlv_epi16(__mmask16 __U, __m256i __A, __m256i __B) {
// CHECK-LABEL: @test_mm256_maskz_srlv_epi16
- // CHECK: @llvm.x86.avx512.mask.psrlv
+ // CHECK: @llvm.x86.avx512.psrlv.w.256(
+ // CHECK: select <16 x i1> %{{.*}}, <16 x i16> %{{.*}}, <16 x i16> %{{.*}}
return _mm256_maskz_srlv_epi16(__U, __A, __B);
}
__m128i test_mm_srlv_epi16(__m128i __A, __m128i __B) {
// CHECK-LABEL: @test_mm_srlv_epi16
- // CHECK: @llvm.x86.avx512.mask.psrlv
+ // CHECK: @llvm.x86.avx512.psrlv.w.128(
return _mm_srlv_epi16(__A, __B);
}
__m128i test_mm_mask_srlv_epi16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B) {
// CHECK-LABEL: @test_mm_mask_srlv_epi16
- // CHECK: @llvm.x86.avx512.mask.psrlv
+ // CHECK: @llvm.x86.avx512.psrlv.w.128(
+ // CHECK: select <8 x i1> %{{.*}}, <8 x i16> %{{.*}}, <8 x i16> %{{.*}}
return _mm_mask_srlv_epi16(__W, __U, __A, __B);
}
__m128i test_mm_maskz_srlv_epi16(__mmask8 __U, __m128i __A, __m128i __B) {
// CHECK-LABEL: @test_mm_maskz_srlv_epi16
- // CHECK: @llvm.x86.avx512.mask.psrlv
+ // CHECK: @llvm.x86.avx512.psrlv.w.128(
+ // CHECK: select <8 x i1> %{{.*}}, <8 x i16> %{{.*}}, <8 x i16> %{{.*}}
return _mm_maskz_srlv_epi16(__U, __A, __B);
}
@@ -2190,37 +2198,41 @@ __m256i test_mm256_maskz_srli_epi16(__mm
__m256i test_mm256_srav_epi16(__m256i __A, __m256i __B) {
// CHECK-LABEL: @test_mm256_srav_epi16
- // CHECK: @llvm.x86.avx512.mask.psrav
+ // CHECK: @llvm.x86.avx512.psrav.w.256(
return _mm256_srav_epi16(__A, __B);
}
__m256i test_mm256_mask_srav_epi16(__m256i __W, __mmask16 __U, __m256i __A, __m256i __B) {
// CHECK-LABEL: @test_mm256_mask_srav_epi16
- // CHECK: @llvm.x86.avx512.mask.psrav
+ // CHECK: @llvm.x86.avx512.psrav.w.256(
+ // CHECK: select <16 x i1> %{{.*}}, <16 x i16> %{{.*}}, <16 x i16> %{{.*}}
return _mm256_mask_srav_epi16(__W, __U, __A, __B);
}
__m256i test_mm256_maskz_srav_epi16(__mmask16 __U, __m256i __A, __m256i __B) {
// CHECK-LABEL: @test_mm256_maskz_srav_epi16
- // CHECK: @llvm.x86.avx512.mask.psrav
+ // CHECK: @llvm.x86.avx512.psrav.w.256(
+ // CHECK: select <16 x i1> %{{.*}}, <16 x i16> %{{.*}}, <16 x i16> %{{.*}}
return _mm256_maskz_srav_epi16(__U, __A, __B);
}
__m128i test_mm_srav_epi16(__m128i __A, __m128i __B) {
// CHECK-LABEL: @test_mm_srav_epi16
- // CHECK: @llvm.x86.avx512.mask.psrav
+ // CHECK: @llvm.x86.avx512.psrav.w.128(
return _mm_srav_epi16(__A, __B);
}
__m128i test_mm_mask_srav_epi16(__m128i __W, __mmask8 __U, __m128i __A, __m128i __B) {
// CHECK-LABEL: @test_mm_mask_srav_epi16
- // CHECK: @llvm.x86.avx512.mask.psrav
+ // CHECK: @llvm.x86.avx512.psrav.w.128(
+ // CHECK: select <8 x i1> %{{.*}}, <8 x i16> %{{.*}}, <8 x i16> %{{.*}}
return _mm_mask_srav_epi16(__W, __U, __A, __B);
}
__m128i test_mm_maskz_srav_epi16(__mmask8 __U, __m128i __A, __m128i __B) {
// CHECK-LABEL: @test_mm_maskz_srav_epi16
- // CHECK: @llvm.x86.avx512.mask.psrav
+ // CHECK: @llvm.x86.avx512.psrav.w.128(
+ // CHECK: select <8 x i1> %{{.*}}, <8 x i16> %{{.*}}, <8 x i16> %{{.*}}
return _mm_maskz_srav_epi16(__U, __A, __B);
}
More information about the cfe-commits
mailing list