FreddyYe updated this revision to Diff 385443.
FreddyYe added a comment.
address comments. add alias for 36 intrinsics in all.
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D112777/new/
https://reviews.llvm.org/D112777
Files:
clang/lib/Headers/avx512fp16intrin.h
clang/lib/Headers/avx512vlfp16intrin.h
clang/test/CodeGen/X86/avx512fp16-builtins.c
clang/test/CodeGen/X86/avx512vlfp16-builtins.c
Index: clang/test/CodeGen/X86/avx512vlfp16-builtins.c
===================================================================
--- clang/test/CodeGen/X86/avx512vlfp16-builtins.c
+++ clang/test/CodeGen/X86/avx512vlfp16-builtins.c
@@ -3114,3 +3114,76 @@
// CHECK: call nnan half @llvm.vector.reduce.fmax.v8f16(<8 x half> %{{.*}})
return _mm_reduce_max_ph(__W);
}
+
+// tests below are for alias intrinsics.
+__m128h test_mm_mul_pch(__m128h __A, __m128h __B) {
+ // CHECK-LABEL: @test_mm_mul_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfmul.cph.128
+ return _mm_mul_pch(__A, __B);
+}
+
+__m128h test_mm_mask_mul_pch(__m128h __W, __mmask8 __U, __m128h __A, __m128h __B) {
+ // CHECK-LABEL: @test_mm_mask_mul_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfmul.cph.128
+ return _mm_mask_mul_pch(__W, __U, __A, __B);
+}
+
+__m128h test_mm_maskz_mul_pch(__mmask8 __U, __m128h __A, __m128h __B) {
+ // CHECK-LABEL: @test_mm_maskz_mul_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfmul.cph.128
+ return _mm_maskz_mul_pch(__U, __A, __B);
+}
+
+__m256h test_mm256_mul_pch(__m256h __A, __m256h __B) {
+ // CHECK-LABEL: @test_mm256_mul_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfmul.cph.256
+ return _mm256_mul_pch(__A, __B);
+}
+
+__m256h test_mm256_mask_mul_pch(__m256h __W, __mmask8 __U, __m256h __A, __m256h __B) {
+ // CHECK-LABEL: @test_mm256_mask_mul_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfmul.cph.256
+ return _mm256_mask_mul_pch(__W, __U, __A, __B);
+}
+
+__m256h test_mm256_maskz_mul_pch(__mmask8 __U, __m256h __A, __m256h __B) {
+ // CHECK-LABEL: @test_mm256_maskz_mul_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfmul.cph.256
+ return _mm256_maskz_mul_pch(__U, __A, __B);
+}
+
+__m128h test_mm_cmul_pch(__m128h __A, __m128h __B) {
+ // CHECK-LABEL: @test_mm_cmul_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfcmul.cph.128
+ return _mm_cmul_pch(__A, __B);
+}
+
+__m128h test_mm_mask_cmul_pch(__m128h __W, __mmask8 __U, __m128h __A, __m128h __B) {
+ // CHECK-LABEL: @test_mm_mask_cmul_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfcmul.cph.128
+ return _mm_mask_fcmul_pch(__W, __U, __A, __B);
+}
+
+__m128h test_mm_maskz_cmul_pch(__mmask8 __U, __m128h __A, __m128h __B) {
+ // CHECK-LABEL: @test_mm_maskz_cmul_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfcmul.cph.128
+ return _mm_maskz_cmul_pch(__U, __A, __B);
+}
+
+__m256h test_mm256_cmul_pch(__m256h __A, __m256h __B) {
+ // CHECK-LABEL: @test_mm256_cmul_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfcmul.cph.256
+ return _mm256_cmul_pch(__A, __B);
+}
+
+__m256h test_mm256_mask_cmul_pch(__m256h __W, __mmask8 __U, __m256h __A, __m256h __B) {
+ // CHECK-LABEL: @test_mm256_mask_cmul_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfcmul.cph.256
+ return _mm256_mask_cmul_pch(__W, __U, __A, __B);
+}
+
+__m256h test_mm256_maskz_cmul_pch(__mmask8 __U, __m256h __A, __m256h __B) {
+ // CHECK-LABEL: @test_mm256_maskz_cmul_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfcmul.cph.256
+ return _mm256_maskz_cmul_pch(__U, __A, __B);
+}
Index: clang/test/CodeGen/X86/avx512fp16-builtins.c
===================================================================
--- clang/test/CodeGen/X86/avx512fp16-builtins.c
+++ clang/test/CodeGen/X86/avx512fp16-builtins.c
@@ -4482,3 +4482,147 @@
// CHECK: %{{.*}} = bitcast <32 x i16> %{{.*}} to <32 x half>
return _mm512_permutexvar_ph(__A, __B);
}
+
+// tests below are for alias intrinsics.
+__m512h test_mm512_mul_pch(__m512h __A, __m512h __B) {
+ // CHECK-LABEL: @test_mm512_mul_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfmul.cph.512
+ return _mm512_mul_pch(__A, __B);
+}
+
+__m512h test_mm512_mask_mul_pch(__m512h __W, __mmask16 __U, __m512h __A, __m512h __B) {
+ // CHECK-LABEL: @test_mm512_mask_mul_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfmul.cph.512
+ return _mm512_mask_mul_pch(__W, __U, __A, __B);
+}
+
+__m512h test_mm512_maskz_mul_pch(__mmask16 __U, __m512h __A, __m512h __B) {
+ // CHECK-LABEL: @test_mm512_maskz_mul_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfmul.cph.512
+ return _mm512_maskz_mul_pch(__U, __A, __B);
+}
+
+__m512h test_mm512_cmul_pch(__m512h __A, __m512h __B) {
+ // CHECK-LABEL: @test_mm512_cmul_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfcmul.cph.512
+ return _mm512_cmul_pch(__A, __B);
+}
+__m512h test_mm512_mask_cmul_pch(__m512h __W, __mmask16 __U, __m512h __A, __m512h __B) {
+ // CHECK-LABEL: @test_mm512_mask_cmul_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfcmul.cph.512
+ return _mm512_mask_cmul_pch(__W, __U, __A, __B);
+}
+
+__m512h test_mm512_maskz_cmul_pch(__mmask16 __U, __m512h __A, __m512h __B) {
+ // CHECK-LABEL: @test_mm512_maskz_cmul_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfcmul.cph.512
+ return _mm512_maskz_cmul_pch(__U, __A, __B);
+}
+
+__m128h test_mm_mul_sch(__m128h __A, __m128h __B) {
+ // CHECK-LABEL: @test_mm_mul_sch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfmul.csh
+ return _mm_mul_sch(__A, __B);
+}
+
+__m128h test_mm_mask_mul_sch(__m128h __W, __mmask8 __U, __m128h __A, __m128h __B) {
+ // CHECK-LABEL: @test_mm_mask_mul_sch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfmul.csh
+ return _mm_mask_mul_sch(__W, __U, __A, __B);
+}
+
+__m128h test_mm_maskz_mul_sch(__mmask8 __U, __m128h __A, __m128h __B) {
+ // CHECK-LABEL: @test_mm_maskz_mul_sch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfmul.csh
+ return _mm_maskz_mul_sch(__U, __A, __B);
+}
+
+__m128h test_mm_mul_round_sch(__m128h __A, __m128h __B) {
+ // CHECK-LABEL: @test_mm_mul_round_sch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfmul.csh
+ return _mm_mul_round_sch(__A, __B, _MM_FROUND_TO_ZERO | _MM_FROUND_NO_EXC);
+}
+
+__m128h test_mm_mask_mul_round_sch(__m128h __W, __mmask8 __U, __m128h __A, __m128h __B) {
+ // CHECK-LABEL: @test_mm_mask_mul_round_sch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfmul.csh
+ return _mm_mask_mul_round_sch(__W, __U, __A, __B, _MM_FROUND_TO_ZERO | _MM_FROUND_NO_EXC);
+}
+
+__m128h test_mm_maskz_mul_round_sch(__mmask8 __U, __m128h __A, __m128h __B) {
+ // CHECK-LABEL: @test_mm_maskz_mul_round_sch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfmul.csh
+ return _mm_maskz_mul_round_sch(__U, __A, __B, _MM_FROUND_TO_ZERO | _MM_FROUND_NO_EXC);
+}
+
+__m512h test_mm512_mul_round_pch(__m512h __A, __m512h __B) {
+ // CHECK-LABEL: @test_mm512_mul_round_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfmul.cph.512
+ return _mm512_mul_round_pch(__A, __B, _MM_FROUND_TO_ZERO | _MM_FROUND_NO_EXC);
+}
+
+__m512h test_mm512_mask_mul_round_pch(__m512h __W, __mmask16 __U, __m512h __A, __m512h __B) {
+ // CHECK-LABEL: @test_mm512_mask_mul_round_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfmul.cph.512
+ return _mm512_mask_mul_round_pch(__W, __U, __A, __B, _MM_FROUND_TO_ZERO | _MM_FROUND_NO_EXC);
+}
+
+__m512h test_mm512_maskz_mul_round_pch(__mmask16 __U, __m512h __A, __m512h __B) {
+ // CHECK-LABEL: @test_mm512_maskz_mul_round_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfmul.cph.512
+ return _mm512_maskz_mul_round_pch(__U, __A, __B, _MM_FROUND_TO_ZERO | _MM_FROUND_NO_EXC);
+}
+
+__m512h test_mm512_cmul_round_pch(__m512h __A, __m512h __B) {
+ // CHECK-LABEL: @test_mm512_cmul_round_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfcmul.cph.512
+ return _mm512_cmul_round_pch(__A, __B, _MM_FROUND_TO_ZERO | _MM_FROUND_NO_EXC);
+}
+
+__m512h test_mm512_mask_cmul_round_pch(__m512h __W, __mmask16 __U, __m512h __A, __m512h __B) {
+ // CHECK-LABEL: @test_mm512_mask_cmul_round_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfcmul.cph.512
+ return _mm512_mask_cmul_round_pch(__W, __U, __A, __B, _MM_FROUND_TO_ZERO | _MM_FROUND_NO_EXC);
+}
+
+__m512h test_mm512_maskz_cmul_round_pch(__mmask16 __U, __m512h __A, __m512h __B) {
+ // CHECK-LABEL: @test_mm512_maskz_cmul_round_pch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfcmul.cph.512
+ return _mm512_maskz_cmul_round_pch(__U, __A, __B, _MM_FROUND_TO_ZERO | _MM_FROUND_NO_EXC);
+}
+
+__m128h test_mm_cmul_sch(__m128h __A, __m128h __B) {
+ // CHECK-LABEL: @test_mm_cmul_sch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfcmul.csh
+ return _mm_cmul_sch(__A, __B);
+}
+
+__m128h test_mm_mask_cmul_sch(__m128h __W, __mmask8 __U, __m128h __A, __m128h __B) {
+ // CHECK-LABEL: @test_mm_mask_cmul_sch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfcmul.csh
+ return _mm_mask_cmul_sch(__W, __U, __A, __B);
+}
+
+__m128h test_mm_maskz_cmul_sch(__mmask8 __U, __m128h __A, __m128h __B) {
+ // CHECK-LABEL: @test_mm_maskz_cmul_sch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfcmul.csh
+ return _mm_maskz_cmul_sch(__U, __A, __B);
+}
+
+__m128h test_mm_cmul_round_sch(__m128h __A, __m128h __B) {
+ // CHECK-LABEL: @test_mm_cmul_round_sch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfcmul.csh
+ return _mm_cmul_round_sch(__A, __B, _MM_FROUND_TO_ZERO | _MM_FROUND_NO_EXC);
+}
+
+__m128h test_mm_mask_cmul_round_sch(__m128h __W, __mmask8 __U, __m128h __A, __m128h __B) {
+ // CHECK-LABEL: @test_mm_mask_cmul_round_sch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfcmul.csh
+ return _mm_mask_cmul_round_sch(__W, __U, __A, __B, _MM_FROUND_TO_ZERO | _MM_FROUND_NO_EXC);
+}
+
+__m128h test_mm_maskz_cmul_round_sch(__mmask8 __U, __m128h __A, __m128h __B) {
+ // CHECK-LABEL: @test_mm_maskz_cmul_round_sch
+ // CHECK: @llvm.x86.avx512fp16.mask.vfcmul.csh
+ return _mm_maskz_cmul_round_sch(__U, __A, __B, _MM_FROUND_TO_ZERO | _MM_FROUND_NO_EXC);
+}
Index: clang/lib/Headers/avx512vlfp16intrin.h
===================================================================
--- clang/lib/Headers/avx512vlfp16intrin.h
+++ clang/lib/Headers/avx512vlfp16intrin.h
@@ -2047,6 +2047,21 @@
return __builtin_ia32_reduce_fmin_ph128(__V);
}
+// intrinsics below are alias for f*mul_*ch
+#define _mm_mul_pch(A, B) _mm_fmul_pch(A, B)
+#define _mm_mask_mul_pch(W, U, A, B) _mm_mask_fmul_pch(W, U, A, B)
+#define _mm_maskz_mul_pch(U, A, B) _mm_maskz_fmul_pch(U, A, B)
+#define _mm256_mul_pch(A, B) _mm256_fmul_pch(A, B)
+#define _mm256_mask_mul_pch(W, U, A, B) _mm256_mask_fmul_pch(W, U, A, B)
+#define _mm256_maskz_mul_pch(U, A, B) _mm256_maskz_fmul_pch(U, A, B)
+
+#define _mm_cmul_pch(A, B) _mm_fcmul_pch(A, B)
+#define _mm_mask_cmul_pch(W, U, A, B) _mm_mask_fcmul_pch(W, U, A, B)
+#define _mm_maskz_cmul_pch(U, A, B) _mm_maskz_fcmul_pch(U, A, B)
+#define _mm256_cmul_pch(A, B) _mm256_fcmul_pch(A, B)
+#define _mm256_mask_cmul_pch(W, U, A, B) _mm256_mask_fcmul_pch(W, U, A, B)
+#define _mm256_maskz_cmul_pch(U, A, B) _mm256_maskz_fcmul_pch(U, A, B)
+
#undef __DEFAULT_FN_ATTRS128
#undef __DEFAULT_FN_ATTRS256
Index: clang/lib/Headers/avx512fp16intrin.h
===================================================================
--- clang/lib/Headers/avx512fp16intrin.h
+++ clang/lib/Headers/avx512fp16intrin.h
@@ -3306,6 +3306,42 @@
return (__m512h)__builtin_ia32_permvarhi512((__v32hi)__B, (__v32hi)__A);
}
+// intrinsics below are alias for f*mul_*ch
+#define _mm512_mul_pch(A, B) _mm512_fmul_pch(A, B)
+#define _mm512_mask_mul_pch(W, U, A, B) _mm512_mask_fmul_pch(W, U, A, B)
+#define _mm512_maskz_mul_pch(U, A, B) _mm512_maskz_fmul_pch(U, A, B)
+#define _mm512_mul_round_pch(A, B, R) _mm512_fmul_round_pch(A, B, R)
+#define _mm512_mask_mul_round_pch(W, U, A, B, R) \
+ _mm512_mask_fmul_round_pch(W, U, A, B, R)
+#define _mm512_maskz_mul_round_pch(U, A, B, R) \
+ _mm512_maskz_fmul_round_pch(U, A, B, R)
+
+#define _mm512_cmul_pch(A, B) _mm512_fcmul_pch(A, B)
+#define _mm512_mask_cmul_pch(W, U, A, B) _mm512_mask_fcmul_pch(W, U, A, B)
+#define _mm512_maskz_cmul_pch(U, A, B) _mm512_maskz_fcmul_pch(U, A, B)
+#define _mm512_cmul_round_pch(A, B, R) _mm512_fcmul_round_pch(A, B, R)
+#define _mm512_mask_cmul_round_pch(W, U, A, B, R) \
+ _mm512_mask_fcmul_round_pch(W, U, A, B, R)
+#define _mm512_maskz_cmul_round_pch(U, A, B, R) \
+ _mm512_maskz_fcmul_round_pch(U, A, B, R)
+
+#define _mm_mul_sch(A, B) _mm_fmul_sch(A, B)
+#define _mm_mask_mul_sch(W, U, A, B) _mm_mask_fmul_sch(W, U, A, B)
+#define _mm_maskz_mul_sch(U, A, B) _mm_maskz_fmul_sch(U, A, B)
+#define _mm_mul_round_sch(A, B, R) _mm_fmul_round_sch(A, B, R)
+#define _mm_mask_mul_round_sch(W, U, A, B, R) \
+ _mm_mask_fmul_round_sch(W, U, A, B, R)
+#define _mm_maskz_mul_round_sch(U, A, B, R) _mm_maskz_fmul_round_sch(U, A, B, R)
+
+#define _mm_cmul_sch(A, B) _mm_fcmul_sch(A, B)
+#define _mm_mask_cmul_sch(W, U, A, B) _mm_mask_fcmul_sch(W, U, A, B)
+#define _mm_maskz_cmul_sch(U, A, B) _mm_maskz_fcmul_sch(U, A, B)
+#define _mm_cmul_round_sch(A, B, R) _mm_fcmul_round_sch(A, B, R)
+#define _mm_mask_cmul_round_sch(W, U, A, B, R) \
+ _mm_mask_fcmul_round_sch(W, U, A, B, R)
+#define _mm_maskz_cmul_round_sch(U, A, B, R) \
+ _mm_maskz_fcmul_round_sch(U, A, B, R)
+
#undef __DEFAULT_FN_ATTRS128
#undef __DEFAULT_FN_ATTRS256
#undef __DEFAULT_FN_ATTRS512
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
