release/20.x: [AVX10.2] Fix wrong intrinsic names after rename (#126390) #126687
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
@phoebewang What do you think about merging this PR to the release branch? |
llvmbot
added
clang
|
@llvm/pr-subscribers-clang @llvm/pr-subscribers-backend-x86 Author: None (llvmbot) ChangesBackport 161cfc6 Requested by: @phoebewang Full diff: https://github.com/llvm/llvm-project/pull/126687.diff 4 Files Affected:
diff --git a/clang/lib/Headers/avx10_2_512convertintrin.h b/clang/lib/Headers/avx10_2_512convertintrin.h
index 0b5fca5cda5228f..516ccc68672d636 100644
--- a/clang/lib/Headers/avx10_2_512convertintrin.h
+++ b/clang/lib/Headers/avx10_2_512convertintrin.h
@@ -213,19 +213,19 @@ _mm512_maskz_cvts2ph_hf8(__mmask64 __U, __m512h __A, __m512h __B) {
(__v64qi)(__m512i)_mm512_setzero_si512());
}
-static __inline__ __m512h __DEFAULT_FN_ATTRS512 _mm512_cvthf8(__m256i __A) {
+static __inline__ __m512h __DEFAULT_FN_ATTRS512 _mm512_cvthf8_ph(__m256i __A) {
return (__m512h)__builtin_ia32_vcvthf8_2ph512_mask(
(__v32qi)__A, (__v32hf)(__m512h)_mm512_undefined_ph(), (__mmask32)-1);
}
static __inline__ __m512h __DEFAULT_FN_ATTRS512
-_mm512_mask_cvthf8(__m512h __W, __mmask32 __U, __m256i __A) {
+_mm512_mask_cvthf8_ph(__m512h __W, __mmask32 __U, __m256i __A) {
return (__m512h)__builtin_ia32_vcvthf8_2ph512_mask(
(__v32qi)__A, (__v32hf)(__m512h)__W, (__mmask32)__U);
}
static __inline__ __m512h __DEFAULT_FN_ATTRS512
-_mm512_maskz_cvthf8(__mmask32 __U, __m256i __A) {
+_mm512_maskz_cvthf8_ph(__mmask32 __U, __m256i __A) {
return (__m512h)__builtin_ia32_vcvthf8_2ph512_mask(
(__v32qi)__A, (__v32hf)(__m512h)_mm512_setzero_ph(), (__mmask32)__U);
}
diff --git a/clang/lib/Headers/avx10_2convertintrin.h b/clang/lib/Headers/avx10_2convertintrin.h
index c67a5b890f1957d..c419323910f187a 100644
--- a/clang/lib/Headers/avx10_2convertintrin.h
+++ b/clang/lib/Headers/avx10_2convertintrin.h
@@ -381,37 +381,36 @@ _mm256_maskz_cvts2ph_hf8(__mmask32 __U, __m256h __A, __m256h __B) {
(__v32qi)(__m256i)_mm256_setzero_si256());
}
-static __inline__ __m128h __DEFAULT_FN_ATTRS128 _mm_cvthf8(__m128i __A) {
+static __inline__ __m128h __DEFAULT_FN_ATTRS128 _mm_cvthf8_ph(__m128i __A) {
return (__m128h)__builtin_ia32_vcvthf8_2ph128_mask(
(__v16qi)__A, (__v8hf)(__m128h)_mm_undefined_ph(), (__mmask8)-1);
}
-static __inline__ __m128h __DEFAULT_FN_ATTRS128 _mm_mask_cvthf8(__m128h __W,
- __mmask8 __U,
- __m128i __A) {
+static __inline__ __m128h __DEFAULT_FN_ATTRS128
+_mm_mask_cvthf8_ph(__m128h __W, __mmask8 __U, __m128i __A) {
return (__m128h)__builtin_ia32_vcvthf8_2ph128_mask(
(__v16qi)__A, (__v8hf)(__m128h)__W, (__mmask8)__U);
}
-static __inline__ __m128h __DEFAULT_FN_ATTRS128 _mm_maskz_cvthf8(__mmask8 __U,
- __m128i __A) {
+static __inline__ __m128h __DEFAULT_FN_ATTRS128
+_mm_maskz_cvthf8_ph(__mmask8 __U, __m128i __A) {
return (__m128h)__builtin_ia32_vcvthf8_2ph128_mask(
(__v16qi)__A, (__v8hf)(__m128h)_mm_setzero_ph(), (__mmask8)__U);
}
-static __inline__ __m256h __DEFAULT_FN_ATTRS256 _mm256_cvthf8(__m128i __A) {
+static __inline__ __m256h __DEFAULT_FN_ATTRS256 _mm256_cvthf8_ph(__m128i __A) {
return (__m256h)__builtin_ia32_vcvthf8_2ph256_mask(
(__v16qi)__A, (__v16hf)(__m256h)_mm256_undefined_ph(), (__mmask16)-1);
}
static __inline__ __m256h __DEFAULT_FN_ATTRS256
-_mm256_mask_cvthf8(__m256h __W, __mmask16 __U, __m128i __A) {
+_mm256_mask_cvthf8_ph(__m256h __W, __mmask16 __U, __m128i __A) {
return (__m256h)__builtin_ia32_vcvthf8_2ph256_mask(
(__v16qi)__A, (__v16hf)(__m256h)__W, (__mmask16)__U);
}
static __inline__ __m256h __DEFAULT_FN_ATTRS256
-_mm256_maskz_cvthf8(__mmask16 __U, __m128i __A) {
+_mm256_maskz_cvthf8_ph(__mmask16 __U, __m128i __A) {
return (__m256h)__builtin_ia32_vcvthf8_2ph256_mask(
(__v16qi)__A, (__v16hf)(__m256h)_mm256_setzero_ph(), (__mmask16)__U);
}
diff --git a/clang/test/CodeGen/X86/avx10_2_512convert-builtins.c b/clang/test/CodeGen/X86/avx10_2_512convert-builtins.c
index 22503c640a727f2..dcf7bbc005a7c6c 100644
--- a/clang/test/CodeGen/X86/avx10_2_512convert-builtins.c
+++ b/clang/test/CodeGen/X86/avx10_2_512convert-builtins.c
@@ -201,22 +201,22 @@ __m512i test_mm512_maskz_cvts2ph_hf8(__mmask64 __U, __m512h __A, __m512h __B) {
return _mm512_maskz_cvts2ph_hf8(__U, __A, __B);
}
-__m512h test_mm512_cvthf8(__m256i __A) {
- // CHECK-LABEL: @test_mm512_cvthf8(
+__m512h test_mm512_cvthf8_ph(__m256i __A) {
+ // CHECK-LABEL: @test_mm512_cvthf8_ph(
// CHECK: call <32 x half> @llvm.x86.avx10.mask.vcvthf82ph512(
- return _mm512_cvthf8(__A);
+ return _mm512_cvthf8_ph(__A);
}
-__m512h test_mm512_mask_cvthf8(__m512h __A, __mmask32 __B, __m256i __C) {
- // CHECK-LABEL: @test_mm512_mask_cvthf8(
+__m512h test_mm512_mask_cvthf8_ph(__m512h __A, __mmask32 __B, __m256i __C) {
+ // CHECK-LABEL: @test_mm512_mask_cvthf8_ph(
// CHECK: call <32 x half> @llvm.x86.avx10.mask.vcvthf82ph512(
- return _mm512_mask_cvthf8(__A, __B, __C);
+ return _mm512_mask_cvthf8_ph(__A, __B, __C);
}
-__m512h test_mm512_maskz_cvthf8(__mmask32 __A, __m256i __B) {
- // CHECK-LABEL: @test_mm512_maskz_cvthf8(
+__m512h test_mm512_maskz_cvthf8_ph(__mmask32 __A, __m256i __B) {
+ // CHECK-LABEL: @test_mm512_maskz_cvthf8_ph(
// CHECK: call <32 x half> @llvm.x86.avx10.mask.vcvthf82ph512(
- return _mm512_maskz_cvthf8(__A, __B);
+ return _mm512_maskz_cvthf8_ph(__A, __B);
}
__m256i test_mm512_cvtph_bf8(__m512h __A) {
diff --git a/clang/test/CodeGen/X86/avx10_2convert-builtins.c b/clang/test/CodeGen/X86/avx10_2convert-builtins.c
index efd9a31c40875ab..d73a223922ce03d 100644
--- a/clang/test/CodeGen/X86/avx10_2convert-builtins.c
+++ b/clang/test/CodeGen/X86/avx10_2convert-builtins.c
@@ -379,40 +379,40 @@ __m256i test_mm256_maskz_cvts2ph_hf8(__mmask16 __U, __m256h __A, __m256h __B) {
return _mm256_maskz_cvts2ph_hf8(__U, __A, __B);
}
-__m128h test_mm_cvthf8(__m128i __A) {
- // CHECK-LABEL: @test_mm_cvthf8(
+__m128h test_mm_cvthf8_ph(__m128i __A) {
+ // CHECK-LABEL: @test_mm_cvthf8_ph(
// CHECK: call <8 x half> @llvm.x86.avx10.mask.vcvthf82ph128(
- return _mm_cvthf8(__A);
+ return _mm_cvthf8_ph(__A);
}
-__m128h test_mm_mask_cvthf8(__m128h __A, __mmask8 __B, __m128i __C) {
- // CHECK-LABEL: @test_mm_mask_cvthf8(
+__m128h test_mm_mask_cvthf8_ph(__m128h __A, __mmask8 __B, __m128i __C) {
+ // CHECK-LABEL: @test_mm_mask_cvthf8_ph(
// CHECK: call <8 x half> @llvm.x86.avx10.mask.vcvthf82ph128(
- return _mm_mask_cvthf8(__A, __B, __C);
+ return _mm_mask_cvthf8_ph(__A, __B, __C);
}
-__m128h test_mm_maskz_cvthf8(__mmask8 __A, __m128i __B) {
- // CHECK-LABEL: @test_mm_maskz_cvthf8(
+__m128h test_mm_maskz_cvthf8_ph(__mmask8 __A, __m128i __B) {
+ // CHECK-LABEL: @test_mm_maskz_cvthf8_ph(
// CHECK: call <8 x half> @llvm.x86.avx10.mask.vcvthf82ph128(
- return _mm_maskz_cvthf8(__A, __B);
+ return _mm_maskz_cvthf8_ph(__A, __B);
}
-__m256h test_mm256_cvthf8(__m128i __A) {
- // CHECK-LABEL: @test_mm256_cvthf8(
+__m256h test_mm256_cvthf8_ph(__m128i __A) {
+ // CHECK-LABEL: @test_mm256_cvthf8_ph(
// CHECK: call <16 x half> @llvm.x86.avx10.mask.vcvthf82ph256(
- return _mm256_cvthf8(__A);
+ return _mm256_cvthf8_ph(__A);
}
-__m256h test_mm256_mask_cvthf8(__m256h __A, __mmask16 __B, __m128i __C) {
- // CHECK-LABEL: @test_mm256_mask_cvthf8(
+__m256h test_mm256_mask_cvthf8_ph(__m256h __A, __mmask16 __B, __m128i __C) {
+ // CHECK-LABEL: @test_mm256_mask_cvthf8_ph(
// CHECK: call <16 x half> @llvm.x86.avx10.mask.vcvthf82ph256(
- return _mm256_mask_cvthf8(__A, __B, __C);
+ return _mm256_mask_cvthf8_ph(__A, __B, __C);
}
-__m256h test_mm256_maskz_cvthf8(__mmask16 __A, __m128i __B) {
- // CHECK-LABEL: @test_mm256_maskz_cvthf8(
+__m256h test_mm256_maskz_cvthf8_ph(__mmask16 __A, __m128i __B) {
+ // CHECK-LABEL: @test_mm256_maskz_cvthf8_ph(
// CHECK: call <16 x half> @llvm.x86.avx10.mask.vcvthf82ph256(
- return _mm256_maskz_cvthf8(__A, __B);
+ return _mm256_maskz_cvthf8_ph(__A, __B);
}
__m128i test_mm_cvtph_bf8(__m128h __A) {
|
|
This is a bug fix without risk, LGTM. |
In my previous PR (llvm#123656) to update the names of AVX10.2 intrinsics and mnemonics, I have erroneously deleted `_ph` from few intrinsics. This PR corrects this. (cherry picked from commit 161cfc6)
|
@phoebewang (or anyone else). If you would like to add a note about this fix in the release notes (completely optional). Please reply to this comment with a one or two sentence description of the fix. When you are done, please add the release:note label to this PR. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Backport 161cfc6
Requested by: @phoebewang