floraachy
/
OpenBLAS
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

dgemm: use dgemm_ncopy_8_skylakex.c also for Haswell 

The dgemm_ncopy_8_skylakex.c code is not avx512 specific and gives
a nice performance boost for medium sized matrices
This commit is contained in:
Arjan van de Ven 2018-12-16 23:06:58 +00:00
parent c43331ad0a
commit d321448a63
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
kernel/x86_64/KERNEL.HASWELL
View File

@ -48,7 +48,7 @@ DGEMMKERNEL = dgemm_kernel_4x8_haswell.S
DGEMM_BETA = dgemm_beta_skylakex.c
DGEMMINCOPY = ../generic/gemm_ncopy_4.c
DGEMMITCOPY = ../generic/gemm_tcopy_4.c
DGEMMONCOPY = ../generic/gemm_ncopy_8.c
DGEMMONCOPY = dgemm_ncopy_8_skylakex.c
DGEMMOTCOPY = ../generic/gemm_tcopy_8.c
DGEMMINCOPYOBJ = dgemm_incopy$(TSUFFIX).$(SUFFIX)
DGEMMITCOPYOBJ = dgemm_itcopy$(TSUFFIX).$(SUFFIX)