This is an archive of the discontinued LLVM Phabricator instance.

Differential D20062

[Clang][AVX512] completing missing intrinsics [vmin/vmax].
ClosedPublic

Authored by m_zuckerman on May 9 2016, 4:00 AM.

Details

Reviewers
AsafBadouh
delena
igorb
Commits
rGf9be3bb1d529: [clang][AVX512] completing missing intrinsics [vmin/vmax].
rC268910: [clang][AVX512] completing missing intrinsics [vmin/vmax].
rL268910: [clang][AVX512] completing missing intrinsics [vmin/vmax].

Diff Detail

Repository
rL LLVM

Event Timeline

m_zuckerman updated this revision to Diff 56547.May 9 2016, 4:00 AM
m_zuckerman retitled this revision from to [Clang][AVX512] completing missing intrinsics [vmin/vmax]..
m_zuckerman updated this object.
m_zuckerman added reviewers: AsafBadouh, igorb, delena.
m_zuckerman added a subscriber: llvm-commits.
AsafBadouh accepted this revision.May 9 2016, 4:08 AM
AsafBadouh edited edge metadata.

LGTM

This revision is now accepted and ready to land.May 9 2016, 4:08 AM
Closed by commit rL268910: [clang][AVX512] completing missing intrinsics [vmin/vmax]. (authored by mzuckerm). · Explain WhyMay 9 2016, 5:44 AM
This revision was automatically updated to reflect the committed changes.

Revision Contents

PathSize
cfe/
trunk/
lib/
Headers/
84 lines
test/
CodeGen/
58 lines

Diff 56561

cfe/trunk/lib/Headers/avx512fintrin.h

  • This file is larger than 256 KB, so syntax highlighting is disabled by default.
Show First 20 Lines • Show All 729 Lines • ▼ Show 20 Lines_mm512_max_pd(__m512d __A, __m512d __B)
return (__m512d) __builtin_ia32_maxpd512_mask ((__v8df) __A, return (__m512d) __builtin_ia32_maxpd512_mask ((__v8df) __A,
(__v8df) __B, (__v8df) __B,
(__v8df) (__v8df)
_mm512_setzero_pd (), _mm512_setzero_pd (),
(__mmask8) -1, (__mmask8) -1,
_MM_FROUND_CUR_DIRECTION); _MM_FROUND_CUR_DIRECTION);
} }
static __inline__ __m512d __DEFAULT_FN_ATTRS
_mm512_mask_max_pd (__m512d __W, __mmask8 __U, __m512d __A, __m512d __B)
{
return (__m512d) __builtin_ia32_maxpd512_mask ((__v8df) __A,
(__v8df) __B,
(__v8df) __W,
(__mmask8) __U,
_MM_FROUND_CUR_DIRECTION);
}
static __inline__ __m512d __DEFAULT_FN_ATTRS
_mm512_maskz_max_pd (__mmask8 __U, __m512d __A, __m512d __B)
{
return (__m512d) __builtin_ia32_maxpd512_mask ((__v8df) __A,
(__v8df) __B,
(__v8df)
_mm512_setzero_pd (),
(__mmask8) __U,
_MM_FROUND_CUR_DIRECTION);
}
static __inline__ __m512 __DEFAULT_FN_ATTRS static __inline__ __m512 __DEFAULT_FN_ATTRS
_mm512_max_ps(__m512 __A, __m512 __B) _mm512_max_ps(__m512 __A, __m512 __B)
{ {
return (__m512) __builtin_ia32_maxps512_mask ((__v16sf) __A, return (__m512) __builtin_ia32_maxps512_mask ((__v16sf) __A,
(__v16sf) __B, (__v16sf) __B,
(__v16sf) (__v16sf)
_mm512_setzero_ps (), _mm512_setzero_ps (),
(__mmask16) -1, (__mmask16) -1,
_MM_FROUND_CUR_DIRECTION); _MM_FROUND_CUR_DIRECTION);
} }
static __inline__ __m512 __DEFAULT_FN_ATTRS
_mm512_mask_max_ps (__m512 __W, __mmask16 __U, __m512 __A, __m512 __B)
{
return (__m512) __builtin_ia32_maxps512_mask ((__v16sf) __A,
(__v16sf) __B,
(__v16sf) __W,
(__mmask16) __U,
_MM_FROUND_CUR_DIRECTION);
}
static __inline__ __m512 __DEFAULT_FN_ATTRS
_mm512_maskz_max_ps (__mmask16 __U, __m512 __A, __m512 __B)
{
return (__m512) __builtin_ia32_maxps512_mask ((__v16sf) __A,
(__v16sf) __B,
(__v16sf)
_mm512_setzero_ps (),
(__mmask16) __U,
_MM_FROUND_CUR_DIRECTION);
}
static __inline__ __m128 __DEFAULT_FN_ATTRS static __inline__ __m128 __DEFAULT_FN_ATTRS
_mm_mask_max_ss(__m128 __W, __mmask8 __U,__m128 __A, __m128 __B) { _mm_mask_max_ss(__m128 __W, __mmask8 __U,__m128 __A, __m128 __B) {
return (__m128) __builtin_ia32_maxss_round_mask ((__v4sf) __A, return (__m128) __builtin_ia32_maxss_round_mask ((__v4sf) __A,
(__v4sf) __B, (__v4sf) __B,
(__v4sf) __W, (__v4sf) __W,
(__mmask8) __U, (__mmask8) __U,
_MM_FROUND_CUR_DIRECTION); _MM_FROUND_CUR_DIRECTION);
} }
▲ Show 20 Lines • Show All 96 Lines • ▼ Show 20 Lines_mm512_min_pd(__m512d __A, __m512d __B)
return (__m512d) __builtin_ia32_minpd512_mask ((__v8df) __A, return (__m512d) __builtin_ia32_minpd512_mask ((__v8df) __A,
(__v8df) __B, (__v8df) __B,
(__v8df) (__v8df)
_mm512_setzero_pd (), _mm512_setzero_pd (),
(__mmask8) -1, (__mmask8) -1,
_MM_FROUND_CUR_DIRECTION); _MM_FROUND_CUR_DIRECTION);
} }
static __inline__ __m512d __DEFAULT_FN_ATTRS
_mm512_mask_min_pd (__m512d __W, __mmask8 __U, __m512d __A, __m512d __B)
{
return (__m512d) __builtin_ia32_minpd512_mask ((__v8df) __A,
(__v8df) __B,
(__v8df) __W,
(__mmask8) __U,
_MM_FROUND_CUR_DIRECTION);
}
static __inline__ __m512d __DEFAULT_FN_ATTRS
_mm512_maskz_min_pd (__mmask8 __U, __m512d __A, __m512d __B)
{
return (__m512d) __builtin_ia32_minpd512_mask ((__v8df) __A,
(__v8df) __B,
(__v8df)
_mm512_setzero_pd (),
(__mmask8) __U,
_MM_FROUND_CUR_DIRECTION);
}
static __inline__ __m512 __DEFAULT_FN_ATTRS static __inline__ __m512 __DEFAULT_FN_ATTRS
_mm512_min_ps(__m512 __A, __m512 __B) _mm512_min_ps(__m512 __A, __m512 __B)
{ {
return (__m512) __builtin_ia32_minps512_mask ((__v16sf) __A, return (__m512) __builtin_ia32_minps512_mask ((__v16sf) __A,
(__v16sf) __B, (__v16sf) __B,
(__v16sf) (__v16sf)
_mm512_setzero_ps (), _mm512_setzero_ps (),
(__mmask16) -1, (__mmask16) -1,
_MM_FROUND_CUR_DIRECTION); _MM_FROUND_CUR_DIRECTION);
} }
static __inline__ __m512 __DEFAULT_FN_ATTRS
_mm512_mask_min_ps (__m512 __W, __mmask16 __U, __m512 __A, __m512 __B)
{
return (__m512) __builtin_ia32_minps512_mask ((__v16sf) __A,
(__v16sf) __B,
(__v16sf) __W,
(__mmask16) __U,
_MM_FROUND_CUR_DIRECTION);
}
static __inline__ __m512 __DEFAULT_FN_ATTRS
_mm512_maskz_min_ps (__mmask16 __U, __m512 __A, __m512 __B)
{
return (__m512) __builtin_ia32_minps512_mask ((__v16sf) __A,
(__v16sf) __B,
(__v16sf)
_mm512_setzero_ps (),
(__mmask16) __U,
_MM_FROUND_CUR_DIRECTION);
}
static __inline__ __m128 __DEFAULT_FN_ATTRS static __inline__ __m128 __DEFAULT_FN_ATTRS
_mm_mask_min_ss(__m128 __W, __mmask8 __U,__m128 __A, __m128 __B) { _mm_mask_min_ss(__m128 __W, __mmask8 __U,__m128 __A, __m128 __B) {
return (__m128) __builtin_ia32_minss_round_mask ((__v4sf) __A, return (__m128) __builtin_ia32_minss_round_mask ((__v4sf) __A,
(__v4sf) __B, (__v4sf) __B,
(__v4sf) __W, (__v4sf) __W,
(__mmask8) __U, (__mmask8) __U,
_MM_FROUND_CUR_DIRECTION); _MM_FROUND_CUR_DIRECTION);
} }
▲ Show 20 Lines • Show All 7,608 Lines • Show Last 20 Lines

cfe/trunk/test/CodeGen/avx512f-builtins.c

Show First 20 Lines • Show All 5,960 Lines • ▼ Show 20 Lines
__m512i test_mm512_mask_cvtps_epu32 (__m512i __W, __mmask16 __U, __m512 __A) __m512i test_mm512_mask_cvtps_epu32 (__m512i __W, __mmask16 __U, __m512 __A)
{ {
// CHECK-LABEL: @test_mm512_mask_cvtps_epu32 // CHECK-LABEL: @test_mm512_mask_cvtps_epu32
// CHECK: @llvm.x86.avx512.mask.cvtps2udq.512 // CHECK: @llvm.x86.avx512.mask.cvtps2udq.512
return _mm512_mask_cvtps_epu32( __W, __U, __A); return _mm512_mask_cvtps_epu32( __W, __U, __A);
} }
__m512d test_mm512_mask_max_pd (__m512d __W, __mmask8 __U, __m512d __A, __m512d __B)
{
// CHECK-LABEL: @test_mm512_mask_max_pd
// CHECK: @llvm.x86.avx512.mask.max.pd.512
return _mm512_mask_max_pd (__W,__U,__A,__B);
}
__m512d test_mm512_maskz_max_pd (__mmask8 __U, __m512d __A, __m512d __B)
{
// CHECK-LABEL: @test_mm512_maskz_max_pd
// CHECK: @llvm.x86.avx512.mask.max.pd.512
return _mm512_maskz_max_pd (__U,__A,__B);
}
__m512 test_mm512_mask_max_ps (__m512 __W, __mmask16 __U, __m512 __A, __m512 __B)
{
// CHECK-LABEL: @test_mm512_mask_max_ps
// CHECK: @llvm.x86.avx512.mask.max.ps.512
return _mm512_mask_max_ps (__W,__U,__A,__B);
}
__m512 test_mm512_maskz_max_ps (__mmask16 __U, __m512 __A, __m512 __B)
{
// CHECK-LABEL: @test_mm512_maskz_max_ps
// CHECK: @llvm.x86.avx512.mask.max.ps.512
return _mm512_maskz_max_ps (__U,__A,__B);
}
__m512d test_mm512_mask_min_pd (__m512d __W, __mmask8 __U, __m512d __A, __m512d __B)
{
// CHECK-LABEL: @test_mm512_mask_min_pd
// CHECK: @llvm.x86.avx512.mask.min.pd.512
return _mm512_mask_min_pd (__W,__U,__A,__B);
}
__m512d test_mm512_maskz_min_pd (__mmask8 __U, __m512d __A, __m512d __B)
{
// CHECK-LABEL: @test_mm512_maskz_min_pd
// CHECK: @llvm.x86.avx512.mask.min.pd.512
return _mm512_maskz_min_pd (__U,__A,__B);
}
__m512 test_mm512_mask_min_ps (__m512 __W, __mmask16 __U, __m512 __A, __m512 __B)
{
// CHECK-LABEL: @test_mm512_mask_min_ps
// CHECK: @llvm.x86.avx512.mask.min.ps.512
return _mm512_mask_min_ps (__W,__U,__A,__B);
}
__m512 test_mm512_maskz_min_ps (__mmask16 __U, __m512 __A, __m512 __B)
{
// CHECK-LABEL: @test_mm512_maskz_min_ps
// CHECK: @llvm.x86.avx512.mask.min.ps.512
return _mm512_maskz_min_ps (__U,__A,__B);
}