diff --git a/Changelog.txt b/Changelog.txt index 60798d2a9..c4e6a8fe2 100644 --- a/Changelog.txt +++ b/Changelog.txt @@ -22,6 +22,7 @@ common: x86/x86_64: * Fixed #28 a wrong result of dsdot on x86_64. * Fixed #32 a SEGFAULT bug of zdotc with gcc-4.6. + * Fixed #33 ztrmm bug on Nehalem. MIPS64: * Fixed #28 a wrong result of dsdot on Loongson3A/MIPS64. diff --git a/kernel/x86_64/zgemm_kernel_1x4_nehalem.S b/kernel/x86_64/zgemm_kernel_1x4_nehalem.S index e72a19c96..4ddfc488b 100644 --- a/kernel/x86_64/zgemm_kernel_1x4_nehalem.S +++ b/kernel/x86_64/zgemm_kernel_1x4_nehalem.S @@ -544,7 +544,7 @@ jg .L11 #if defined(TRMMKERNEL) && !defined(LEFT) - addq $1, KK + addq $4, KK #endif leaq (C, LDC, 4), C @@ -594,7 +594,7 @@ jg .L11 #if defined(TRMMKERNEL) && !defined(LEFT) - addq $1, KK + addq $4, KK #endif leaq (C, LDC, 4), C