Allow VortexM4 on the same fast path only with non-gcc compilers

This commit is contained in:
Martin Kroeker
2025-10-19 13:42:17 -07:00
committed by GitHub
parent 643a0b53b0
commit e01b1094de

View File

@@ -372,10 +372,14 @@ void CNAME(enum CBLAS_ORDER order, enum CBLAS_SIDE Side, enum CBLAS_UPLO Uplo,
}
#if !defined(COMPLEX) && !defined(DOUBLE) && !defined(BFLOAT16) && !defined(HFLOAT16)
#if defined(ARCH_ARM64) && (defined(USE_SSYMM_KERNEL_DIRECT)||defined(DYNAMIC_ARCH))
#if defined(ARCH_ARM64) && (defined(USE_SSYMM_KERNEL_DIRECT)||defined(DYNAMIC_AR
CH))
#if defined(DYNAMIC_ARCH)
if (support_sme1())
if (strcmp(gotoblas_corename(), "armv9sme") == 0
#if defined(__clang__)
|| strcmp(gotoblas_corename(), "vortexm4") == 0
#endif
)
if (args.m == 0 || args.n == 0) return;
if (order == CblasRowMajor && m == lda && n == ldb && n == ldc)
{