This is an archive of the discontinued LLVM Phabricator instance.

Differential D109487

[X86] Support *_set1_pch(Float16 _Complex h)
ClosedPublic

Authored by pengfei on Sep 9 2021, 12:15 AM.

Details

Reviewers
LuoYuanke
FreddyYe
craig.topper
RKSimon
Commits
rG2aaa6466fe00: [X86] Support *_set1_pch(Float16 _Complex h)

Diff Detail

Event Timeline

pengfei requested review of this revision.Sep 9 2021, 12:15 AM
pengfei created this revision.
Herald added a project: Restricted Project. · View Herald TranscriptSep 9 2021, 12:15 AM
Herald added a subscriber: cfe-commits. · View Herald Transcript
Harbormaster completed remote builds in B123166: Diff 371517.Sep 9 2021, 12:52 AM
LuoYuanke accepted this revision.Sep 11 2021, 1:04 AM

LGTM, thanks.

This revision is now accepted and ready to land.Sep 11 2021, 1:04 AM
This revision was landed with ongoing or failed builds.Sep 11 2021, 3:20 AM
Closed by commit rG2aaa6466fe00: [X86] Support *_set1_pch(Float16 _Complex h) (authored by Wang, Pengfei <pengfei.wang@intel.com>). · Explain Why
This revision was automatically updated to reflect the committed changes.
pengfei added a commit: rG2aaa6466fe00: [X86] Support *_set1_pch(Float16 _Complex h).

Revision Contents

PathSize
clang/
lib/
Headers/
5 lines
10 lines
test/
CodeGen/
X86/
23 lines
26 lines

Diff 372069

clang/lib/Headers/avx512fp16intrin.h

Show First 20 Lines • Show All 91 Lines • ▼ Show 20 Lines
#define _mm512_setr_ph(h1, h2, h3, h4, h5, h6, h7, h8, h9, h10, h11, h12, h13, \ #define _mm512_setr_ph(h1, h2, h3, h4, h5, h6, h7, h8, h9, h10, h11, h12, h13, \
h14, h15, h16, h17, h18, h19, h20, h21, h22, h23, h24, \ h14, h15, h16, h17, h18, h19, h20, h21, h22, h23, h24, \
h25, h26, h27, h28, h29, h30, h31, h32) \ h25, h26, h27, h28, h29, h30, h31, h32) \
_mm512_set_ph((h32), (h31), (h30), (h29), (h28), (h27), (h26), (h25), (h24), \ _mm512_set_ph((h32), (h31), (h30), (h29), (h28), (h27), (h26), (h25), (h24), \
(h23), (h22), (h21), (h20), (h19), (h18), (h17), (h16), (h15), \ (h23), (h22), (h21), (h20), (h19), (h18), (h17), (h16), (h15), \
(h14), (h13), (h12), (h11), (h10), (h9), (h8), (h7), (h6), \ (h14), (h13), (h12), (h11), (h10), (h9), (h8), (h7), (h6), \
(h5), (h4), (h3), (h2), (h1)) (h5), (h4), (h3), (h2), (h1))
static __inline __m512h __DEFAULT_FN_ATTRS512
_mm512_set1_pch(_Float16 _Complex h) {
return (__m512h)_mm512_set1_ps(__builtin_bit_cast(float, h));
}
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_castph_ps(__m128h __a) { static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_castph_ps(__m128h __a) {
return (__m128)__a; return (__m128)__a;
} }
static __inline__ __m256 __DEFAULT_FN_ATTRS256 _mm256_castph_ps(__m256h __a) { static __inline__ __m256 __DEFAULT_FN_ATTRS256 _mm256_castph_ps(__m256h __a) {
return (__m256)__a; return (__m256)__a;
} }
▲ Show 20 Lines • Show All 3,216 Lines • Show Last 20 Lines

clang/lib/Headers/avx512vlfp16intrin.h

Show First 20 Lines • Show All 46 Lines • ▼ Show 20 Lines
static __inline __m128h __DEFAULT_FN_ATTRS128 static __inline __m128h __DEFAULT_FN_ATTRS128
_mm_set_ph(_Float16 __h1, _Float16 __h2, _Float16 __h3, _Float16 __h4, _mm_set_ph(_Float16 __h1, _Float16 __h2, _Float16 __h3, _Float16 __h4,
_Float16 __h5, _Float16 __h6, _Float16 __h7, _Float16 __h8) { _Float16 __h5, _Float16 __h6, _Float16 __h7, _Float16 __h8) {
return (__m128h)(__v8hf){__h8, __h7, __h6, __h5, __h4, __h3, __h2, __h1}; return (__m128h)(__v8hf){__h8, __h7, __h6, __h5, __h4, __h3, __h2, __h1};
} }
static __inline __m256h __DEFAULT_FN_ATTRS256 static __inline __m256h __DEFAULT_FN_ATTRS256
_mm256_set1_pch(_Float16 _Complex h) {
return (__m256h)_mm256_set1_ps(__builtin_bit_cast(float, h));
}
static __inline __m128h __DEFAULT_FN_ATTRS128
_mm_set1_pch(_Float16 _Complex h) {
return (__m128h)_mm_set1_ps(__builtin_bit_cast(float, h));
}
static __inline __m256h __DEFAULT_FN_ATTRS256
_mm256_set_ph(_Float16 __h1, _Float16 __h2, _Float16 __h3, _Float16 __h4, _mm256_set_ph(_Float16 __h1, _Float16 __h2, _Float16 __h3, _Float16 __h4,
_Float16 __h5, _Float16 __h6, _Float16 __h7, _Float16 __h8, _Float16 __h5, _Float16 __h6, _Float16 __h7, _Float16 __h8,
_Float16 __h9, _Float16 __h10, _Float16 __h11, _Float16 __h12, _Float16 __h9, _Float16 __h10, _Float16 __h11, _Float16 __h12,
_Float16 __h13, _Float16 __h14, _Float16 __h15, _Float16 __h16) { _Float16 __h13, _Float16 __h14, _Float16 __h15, _Float16 __h16) {
return (__m256h)(__v16hf){__h16, __h15, __h14, __h13, __h12, __h11, return (__m256h)(__v16hf){__h16, __h15, __h14, __h13, __h12, __h11,
__h10, __h9, __h8, __h7, __h6, __h5, __h10, __h9, __h8, __h7, __h6, __h5,
__h4, __h3, __h2, __h1}; __h4, __h3, __h2, __h1};
} }
▲ Show 20 Lines • Show All 1,981 Lines • Show Last 20 Lines

clang/test/CodeGen/X86/avx512fp16-builtins.c

Show First 20 Lines • Show All 75 Lines • ▼ Show 20 Lines__m512h test_mm512_set1_ph(_Float16 h) {
// CHECK: insertelement <32 x half> {{.*}}, i32 27 // CHECK: insertelement <32 x half> {{.*}}, i32 27
// CHECK: insertelement <32 x half> {{.*}}, i32 28 // CHECK: insertelement <32 x half> {{.*}}, i32 28
// CHECK: insertelement <32 x half> {{.*}}, i32 29 // CHECK: insertelement <32 x half> {{.*}}, i32 29
// CHECK: insertelement <32 x half> {{.*}}, i32 30 // CHECK: insertelement <32 x half> {{.*}}, i32 30
// CHECK: insertelement <32 x half> {{.*}}, i32 31 // CHECK: insertelement <32 x half> {{.*}}, i32 31
return _mm512_set1_ph(h); return _mm512_set1_ph(h);
} }
__m512h test_mm512_set1_pch(_Float16 _Complex h) {
// CHECK-LABEL: @test_mm512_set1_pch
// CHECK: bitcast { half, half }{{.*}} to float
// CHECK: insertelement <16 x float> {{.*}}, i32 0
// CHECK: insertelement <16 x float> {{.*}}, i32 1
// CHECK: insertelement <16 x float> {{.*}}, i32 2
// CHECK: insertelement <16 x float> {{.*}}, i32 3
// CHECK: insertelement <16 x float> {{.*}}, i32 4
// CHECK: insertelement <16 x float> {{.*}}, i32 5
// CHECK: insertelement <16 x float> {{.*}}, i32 6
// CHECK: insertelement <16 x float> {{.*}}, i32 7
// CHECK: insertelement <16 x float> {{.*}}, i32 8
// CHECK: insertelement <16 x float> {{.*}}, i32 9
// CHECK: insertelement <16 x float> {{.*}}, i32 10
// CHECK: insertelement <16 x float> {{.*}}, i32 11
// CHECK: insertelement <16 x float> {{.*}}, i32 12
// CHECK: insertelement <16 x float> {{.*}}, i32 13
// CHECK: insertelement <16 x float> {{.*}}, i32 14
// CHECK: insertelement <16 x float> {{.*}}, i32 15
// CHECK: bitcast <16 x float>{{.*}} to <32 x half>
return _mm512_set1_pch(h);
}
__m512h test_mm512_set_ph(_Float16 __h1, _Float16 __h2, _Float16 __h3, _Float16 __h4, __m512h test_mm512_set_ph(_Float16 __h1, _Float16 __h2, _Float16 __h3, _Float16 __h4,
_Float16 __h5, _Float16 __h6, _Float16 __h7, _Float16 __h8, _Float16 __h5, _Float16 __h6, _Float16 __h7, _Float16 __h8,
_Float16 __h9, _Float16 __h10, _Float16 __h11, _Float16 __h12, _Float16 __h9, _Float16 __h10, _Float16 __h11, _Float16 __h12,
_Float16 __h13, _Float16 __h14, _Float16 __h15, _Float16 __h16, _Float16 __h13, _Float16 __h14, _Float16 __h15, _Float16 __h16,
_Float16 __h17, _Float16 __h18, _Float16 __h19, _Float16 __h20, _Float16 __h17, _Float16 __h18, _Float16 __h19, _Float16 __h20,
_Float16 __h21, _Float16 __h22, _Float16 __h23, _Float16 __h24, _Float16 __h21, _Float16 __h22, _Float16 __h23, _Float16 __h24,
_Float16 __h25, _Float16 __h26, _Float16 __h27, _Float16 __h28, _Float16 __h25, _Float16 __h26, _Float16 __h27, _Float16 __h28,
_Float16 __h29, _Float16 __h30, _Float16 __h31, _Float16 __h32) { _Float16 __h29, _Float16 __h30, _Float16 __h31, _Float16 __h32) {
▲ Show 20 Lines • Show All 4,352 Lines • Show Last 20 Lines

clang/test/CodeGen/X86/avx512vlfp16-builtins.c

Show First 20 Lines • Show All 55 Lines • ▼ Show 20 Lines__m256h test_mm256_set1_ph(_Float16 h) {
// CHECK: insertelement <16 x half> {{.*}}, i32 11 // CHECK: insertelement <16 x half> {{.*}}, i32 11
// CHECK: insertelement <16 x half> {{.*}}, i32 12 // CHECK: insertelement <16 x half> {{.*}}, i32 12
// CHECK: insertelement <16 x half> {{.*}}, i32 13 // CHECK: insertelement <16 x half> {{.*}}, i32 13
// CHECK: insertelement <16 x half> {{.*}}, i32 14 // CHECK: insertelement <16 x half> {{.*}}, i32 14
// CHECK: insertelement <16 x half> {{.*}}, i32 15 // CHECK: insertelement <16 x half> {{.*}}, i32 15
return _mm256_set1_ph(h); return _mm256_set1_ph(h);
} }
__m128h test_mm_set1_pch(_Float16 _Complex h) {
// CHECK-LABEL: @test_mm_set1_pch
// CHECK: bitcast { half, half }{{.*}} to float
// CHECK: insertelement <4 x float> {{.*}}, i32 0
// CHECK: insertelement <4 x float> {{.*}}, i32 1
// CHECK: insertelement <4 x float> {{.*}}, i32 2
// CHECK: insertelement <4 x float> {{.*}}, i32 3
// CHECK: bitcast <4 x float>{{.*}} to <8 x half>
return _mm_set1_pch(h);
}
__m256h test_mm256_set1_pch(_Float16 _Complex h) {
// CHECK-LABEL: @test_mm256_set1_pch
// CHECK: bitcast { half, half }{{.*}} to float
// CHECK: insertelement <8 x float> {{.*}}, i32 0
// CHECK: insertelement <8 x float> {{.*}}, i32 1
// CHECK: insertelement <8 x float> {{.*}}, i32 2
// CHECK: insertelement <8 x float> {{.*}}, i32 3
// CHECK: insertelement <8 x float> {{.*}}, i32 4
// CHECK: insertelement <8 x float> {{.*}}, i32 5
// CHECK: insertelement <8 x float> {{.*}}, i32 6
// CHECK: insertelement <8 x float> {{.*}}, i32 7
// CHECK: bitcast <8 x float>{{.*}} to <16 x half>
return _mm256_set1_pch(h);
}
__m128h test_mm_set_ph(_Float16 __h1, _Float16 __h2, _Float16 __h3, _Float16 __h4, __m128h test_mm_set_ph(_Float16 __h1, _Float16 __h2, _Float16 __h3, _Float16 __h4,
_Float16 __h5, _Float16 __h6, _Float16 __h7, _Float16 __h8) { _Float16 __h5, _Float16 __h6, _Float16 __h7, _Float16 __h8) {
// CHECK-LABEL: @test_mm_set_ph // CHECK-LABEL: @test_mm_set_ph
// CHECK: insertelement <8 x half> {{.*}}, i32 0 // CHECK: insertelement <8 x half> {{.*}}, i32 0
// CHECK: insertelement <8 x half> {{.*}}, i32 1 // CHECK: insertelement <8 x half> {{.*}}, i32 1
// CHECK: insertelement <8 x half> {{.*}}, i32 2 // CHECK: insertelement <8 x half> {{.*}}, i32 2
// CHECK: insertelement <8 x half> {{.*}}, i32 3 // CHECK: insertelement <8 x half> {{.*}}, i32 3
// CHECK: insertelement <8 x half> {{.*}}, i32 4 // CHECK: insertelement <8 x half> {{.*}}, i32 4
▲ Show 20 Lines • Show All 3,019 Lines • Show Last 20 Lines