Skip to content
Snippets Groups Projects
Commit b008a9e9 authored by Richard Angersbach's avatar Richard Angersbach
Browse files

Adapt guards for generated avx512 horizontal ops

parent 806dcb6b
No related branches found
No related tags found
1 merge request!438Reduction Support
Pipeline #77425 passed
Stage: Code Quality
Stage: Unit Tests
Stage: legacy_test
Stage: docs
Hide whitespace changes
Inline Side-by-side
src/pystencils/include/simd_horizontal_helpers.h
+ 1
− 1
View file @ b008a9e9
...@@ -120,7 +120,7 @@ inline float _mm256_horizontal_max_ps(float dst, __m256 src) { ...@@ -120,7 +120,7 @@ inline float _mm256_horizontal_max_ps(float dst, __m256 src) {
#endif #endif
#if defined(__AVX512VL__) #if defined(__AVX512F__)
#include <immintrin.h> #include <immintrin.h>
inline double _mm512_horizontal_add_pd(double dst, __m512d src) { inline double _mm512_horizontal_add_pd(double dst, __m512d src) {
......
util/generate_simd_horizontal_op.py
+ 1
− 1
View file @ b008a9e9
...@@ -277,7 +277,7 @@ vtypes_for_instruction_set = { ...@@ -277,7 +277,7 @@ vtypes_for_instruction_set = {
guards_for_instruction_sets = { guards_for_instruction_sets = {
InstructionSets.SSE3: "__SSE3__", InstructionSets.SSE3: "__SSE3__",
InstructionSets.AVX: "__AVX__", InstructionSets.AVX: "__AVX__",
InstructionSets.AVX512: '__AVX512VL__', InstructionSets.AVX512: '__AVX512F__',
InstructionSets.NEON: '_M_ARM64', InstructionSets.NEON: '_M_ARM64',
} }
......
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Please register or to comment