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

added dgemm-, dtrmm-, zgemm- and ztrmm-kernel for power8

Browse Source
This commit is contained in:
Werner Saar
2016-03-01 07:33:56 +01:00
parent 3e8d6ea74f
commit b752858d6c
28 changed files with 14013 additions and 166 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
common_power.h
View File

@@ -236,7 +236,7 @@ static inline int blas_quickdivide(blasint x, blasint y){
#define HAVE_PREFETCH
#endif
#if defined(POWER3) || defined(POWER6) || defined(PPCG4) || defined(CELL)
#if defined(POWER3) || defined(POWER6) || defined(PPCG4) || defined(CELL) || defined(POWER8)
#define DCBT_ARG 0
#else
#define DCBT_ARG 8
@@ -258,6 +258,13 @@ static inline int blas_quickdivide(blasint x, blasint y){
#define L1_PREFETCH dcbtst
#endif
#if defined(POWER8)
#define L1_DUALFETCH
#define L1_PREFETCHSIZE (16 + 128 * 100)
#define L1_PREFETCH dcbtst
#endif
#
#ifndef L1_PREFETCH
#define L1_PREFETCH dcbt
#endif