Browse Source

Merge pull request #4204 from martin-frbg/llvm17-2

Work around LLVM17 miscompiling the AVX512 microkernels for CASUM/ZASUM
tags/v0.3.24
Martin Kroeker GitHub 2 years ago
parent
commit
88435104c8
No known key found for this signature in database GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 2 additions and 2 deletions
  1. +1
    -1
      kernel/x86_64/casum_microk_skylakex-2.c
  2. +1
    -1
      kernel/x86_64/zasum_microk_skylakex-2.c

+ 1
- 1
kernel/x86_64/casum_microk_skylakex-2.c View File

@@ -2,7 +2,7 @@
#ifdef __NVCOMPILER #ifdef __NVCOMPILER
#define NVCOMPVERS ( __NVCOMPILER_MAJOR__ * 100 + __NVCOMPILER_MINOR__ ) #define NVCOMPVERS ( __NVCOMPILER_MAJOR__ * 100 + __NVCOMPILER_MINOR__ )
#endif #endif
#if ((( defined(__GNUC__) && __GNUC__ > 6 && defined(__AVX512CD__)) || (defined(__clang__) && __clang_major__ >= 9)) || ( defined(__NVCOMPILER) && NVCOMPVERS >= 2309))
#if ((( defined(__GNUC__) && __GNUC__ > 6 && defined(__AVX512CD__)) || (defined(__clang__) && (__clang_major__ >= 9 &&__clang_major__ !=17)) || ( defined(__NVCOMPILER) && NVCOMPVERS >= 2309)))


#if (!(defined(__NVCOMPILER) && NVCOMPVERS < 2309)) #if (!(defined(__NVCOMPILER) && NVCOMPVERS < 2309))




+ 1
- 1
kernel/x86_64/zasum_microk_skylakex-2.c View File

@@ -2,7 +2,7 @@
#ifdef __NVCOMPILER #ifdef __NVCOMPILER
#define NVCOMPVERS ( __NVCOMPILER_MAJOR__ * 100 + __NVCOMPILER_MINOR__ ) #define NVCOMPVERS ( __NVCOMPILER_MAJOR__ * 100 + __NVCOMPILER_MINOR__ )
#endif #endif
#if ((( defined(__GNUC__) && __GNUC__ > 6 && defined(__AVX512CD__)) || (defined(__clang__) && __clang_major__ >= 9)) || (defined(__NVCOMPILER) && NVCOMPVERS >= 2309))
#if ((( defined(__GNUC__) && __GNUC__ > 6 && defined(__AVX512CD__)) || (defined(__clang__) && ( __clang_major__ >= 9 && __clang_major__ != 17)) || (defined(__NVCOMPILER) && NVCOMPVERS >= 2309)))


#if (!(defined(__NVCOMPILER) && NVCOMPVERS < 2309)) #if (!(defined(__NVCOMPILER) && NVCOMPVERS < 2309))




Loading…
Cancel
Save