Import AMD Piledriver DGEMM kernel generated by AUGEM.

So far, this kernel doesn't deal with edge.

AUGEM: Automatically Generate High Performance Dense Linear Algebra
Kernels on x86 CPUs.
Qian Wang, Xianyi Zhang, Yunquan Zhang, and Qing Yi. In the
International Conference for High Performance Computing, Networking,
Storage and Analysis (SC'13). Denver, CO. Nov, 2013.
This commit is contained in:
Zhang Xianyi
2013-08-25 10:16:01 -03:00
parent fe98de2f68
commit 6c4a7d0828
19 changed files with 6121 additions and 10 deletions

View File

@@ -16,15 +16,17 @@ SGEMMINCOPYOBJ = sgemm_incopy$(TSUFFIX).$(SUFFIX)
SGEMMITCOPYOBJ = sgemm_itcopy$(TSUFFIX).$(SUFFIX)
SGEMMONCOPYOBJ = sgemm_oncopy$(TSUFFIX).$(SUFFIX)
SGEMMOTCOPYOBJ = sgemm_otcopy$(TSUFFIX).$(SUFFIX)
DGEMMKERNEL = dgemm_kernel_8x2_bulldozer.S
DGEMMINCOPY = dgemm_ncopy_8_bulldozer.S
DGEMMITCOPY = dgemm_tcopy_8_bulldozer.S
DGEMMONCOPY = gemm_ncopy_2_bulldozer.S
DGEMMOTCOPY = gemm_tcopy_2_bulldozer.S
DGEMMKERNEL = dgemm_kernel_6x4_piledriver.S
DGEMMINCOPY = ../generic/gemm_ncopy_6.c
DGEMMITCOPY = ../generic/gemm_tcopy_6.c
DGEMMONCOPY = ../generic/gemm_ncopy_4.c
DGEMMOTCOPY = ../generic/gemm_tcopy_4.c
DGEMMINCOPYOBJ = dgemm_incopy$(TSUFFIX).$(SUFFIX)
DGEMMITCOPYOBJ = dgemm_itcopy$(TSUFFIX).$(SUFFIX)
DGEMMONCOPYOBJ = dgemm_oncopy$(TSUFFIX).$(SUFFIX)
DGEMMOTCOPYOBJ = dgemm_otcopy$(TSUFFIX).$(SUFFIX)
CGEMMKERNEL = cgemm_kernel_4x2_bulldozer.S
CGEMMINCOPY = ../generic/zgemm_ncopy_4.c
CGEMMITCOPY = ../generic/zgemm_tcopy_4.c

File diff suppressed because it is too large Load Diff