[X86][AVX10.2] Fix wrong mask bits in cvtpbf8_ph intrinsics #120927
Merged
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
llvmbot
added
clang
Member
|
@llvm/pr-subscribers-backend-x86 @llvm/pr-subscribers-clang Author: Phoebe Wang (phoebewang) ChangesFound during review #120766 Full diff: https://github.com/llvm/llvm-project/pull/120927.diff 4 Files Affected:
diff --git a/clang/lib/Headers/avx10_2_512convertintrin.h b/clang/lib/Headers/avx10_2_512convertintrin.h
index a34e135fa30473..60a5b1ef4548d8 100644
--- a/clang/lib/Headers/avx10_2_512convertintrin.h
+++ b/clang/lib/Headers/avx10_2_512convertintrin.h
@@ -308,13 +308,13 @@ static __inline __m512h __DEFAULT_FN_ATTRS512 _mm512_cvtpbf8_ph(__m256i __A) {
}
static __inline __m512h __DEFAULT_FN_ATTRS512
-_mm512_mask_cvtpbf8_ph(__m512h __S, __mmask16 __U, __m256i __A) {
+_mm512_mask_cvtpbf8_ph(__m512h __S, __mmask32 __U, __m256i __A) {
return _mm512_castsi512_ph(
_mm512_mask_slli_epi16((__m512i)__S, __U, _mm512_cvtepi8_epi16(__A), 8));
}
static __inline __m512h __DEFAULT_FN_ATTRS512
-_mm512_maskz_cvtpbf8_ph(__mmask16 __U, __m256i __A) {
+_mm512_maskz_cvtpbf8_ph(__mmask32 __U, __m256i __A) {
return _mm512_castsi512_ph(
_mm512_slli_epi16(_mm512_maskz_cvtepi8_epi16(__U, __A), 8));
}
diff --git a/clang/lib/Headers/avx10_2convertintrin.h b/clang/lib/Headers/avx10_2convertintrin.h
index 134adb2850c8de..efe8477cbbf9be 100644
--- a/clang/lib/Headers/avx10_2convertintrin.h
+++ b/clang/lib/Headers/avx10_2convertintrin.h
@@ -580,13 +580,13 @@ static __inline__ __m256h __DEFAULT_FN_ATTRS256 _mm256_cvtpbf8_ph(__m128i __A) {
}
static __inline__ __m256h __DEFAULT_FN_ATTRS256
-_mm256_mask_cvtpbf8_ph(__m256h __S, __mmask8 __U, __m128i __A) {
+_mm256_mask_cvtpbf8_ph(__m256h __S, __mmask16 __U, __m128i __A) {
return _mm256_castsi256_ph(
_mm256_mask_slli_epi16((__m256i)__S, __U, _mm256_cvtepi8_epi16(__A), 8));
}
static __inline__ __m256h __DEFAULT_FN_ATTRS256
-_mm256_maskz_cvtpbf8_ph(__mmask8 __U, __m128i __A) {
+_mm256_maskz_cvtpbf8_ph(__mmask16 __U, __m128i __A) {
return _mm256_castsi256_ph(
_mm256_slli_epi16(_mm256_maskz_cvtepi8_epi16(__U, __A), 8));
}
diff --git a/clang/test/CodeGen/X86/avx10_2_512convert-builtins.c b/clang/test/CodeGen/X86/avx10_2_512convert-builtins.c
index e71cc0c9ad6b02..6662e0cbf8a913 100644
--- a/clang/test/CodeGen/X86/avx10_2_512convert-builtins.c
+++ b/clang/test/CodeGen/X86/avx10_2_512convert-builtins.c
@@ -299,7 +299,7 @@ __m512h test_mm512_cvtpbf8_ph(__m256i A) {
return _mm512_cvtpbf8_ph(A);
}
-__m512h test_mm512_mask_cvtpbf8_ph(__m512h S, __mmask16 M, __m256i A) {
+__m512h test_mm512_mask_cvtpbf8_ph(__m512h S, __mmask32 M, __m256i A) {
// CHECK-LABEL: @test_mm512_mask_cvtpbf8_ph
// CHECK: sext <32 x i8> %{{.*}} to <32 x i16>
// CHECK: @llvm.x86.avx512.pslli.w.512
@@ -308,7 +308,7 @@ __m512h test_mm512_mask_cvtpbf8_ph(__m512h S, __mmask16 M, __m256i A) {
return _mm512_mask_cvtpbf8_ph(S, M, A);
}
-__m512h test_mm512_maskz_cvtpbf8_ph(__mmask16 M, __m256i A) {
+__m512h test_mm512_maskz_cvtpbf8_ph(__mmask32 M, __m256i A) {
// CHECK-LABEL: @test_mm512_maskz_cvtpbf8_ph
// CHECK: sext <32 x i8> %{{.*}} to <32 x i16>
// CHECK: select <32 x i1> %{{.*}}, <32 x i16> %{{.*}}, <32 x i16> %{{.*}}
diff --git a/clang/test/CodeGen/X86/avx10_2convert-builtins.c b/clang/test/CodeGen/X86/avx10_2convert-builtins.c
index 8086c1b5d33993..7121b28719c077 100644
--- a/clang/test/CodeGen/X86/avx10_2convert-builtins.c
+++ b/clang/test/CodeGen/X86/avx10_2convert-builtins.c
@@ -593,7 +593,7 @@ __m128h test_mm_cvtpbf8_ph(__m128i A) {
return _mm_cvtpbf8_ph(A);
}
-__m128h test_mm_mask_cvtpbf8_ph(__m128h S, __mmask8 M, __m128i A) {
+__m128h test_mm_mask_cvtpbf8_ph(__m128h S, __mmask16 M, __m128i A) {
// CHECK-LABEL: @test_mm_mask_cvtpbf8_ph
// CHECK: sext <8 x i8> %{{.*}} to <8 x i16>
// CHECK: @llvm.x86.sse2.pslli.w
@@ -602,7 +602,7 @@ __m128h test_mm_mask_cvtpbf8_ph(__m128h S, __mmask8 M, __m128i A) {
return _mm_mask_cvtpbf8_ph(S, M, A);
}
-__m128h test_mm_maskz_cvtpbf8_ph(__mmask8 M, __m128i A) {
+__m128h test_mm_maskz_cvtpbf8_ph(__mmask16 M, __m128i A) {
// CHECK-LABEL: @test_mm_maskz_cvtpbf8_ph
// CHECK: sext <8 x i8> %{{.*}} to <8 x i16>
// CHECK: select <8 x i1> %{{.*}}, <8 x i16> %{{.*}}, <8 x i16> %{{.*}}
|
Found during review llvm#120766
RKSimon
approved these changes
Dec 23, 2024
Collaborator
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/59/builds/10268 Here is the relevant piece of the build log for the reference |
phoebewang
pushed a commit
that referenced
this pull request
Feb 11, 2025
Found during work on #120927. This caused the compiler to silently drop ignore half of the mask in the specific intrinsics.
swift-ci
pushed a commit
to swiftlang/llvm-project
that referenced
this pull request
Feb 11, 2025
) Found during work on llvm#120927. This caused the compiler to silently drop ignore half of the mask in the specific intrinsics. (cherry picked from commit af522c5)
Icohedron
pushed a commit
to Icohedron/llvm-project
that referenced
this pull request
Feb 11, 2025
) Found during work on llvm#120927. This caused the compiler to silently drop ignore half of the mask in the specific intrinsics.
joaosaffran
pushed a commit
to joaosaffran/llvm-project
that referenced
this pull request
Feb 14, 2025
) Found during work on llvm#120927. This caused the compiler to silently drop ignore half of the mask in the specific intrinsics.
sivan-shani
pushed a commit
to sivan-shani/llvm-project
that referenced
this pull request
Feb 24, 2025
) Found during work on llvm#120927. This caused the compiler to silently drop ignore half of the mask in the specific intrinsics.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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.
Found during review #120766