amd/blis
1
0
Fork 0
mirror of https://github.com/amd/blis.git synced 2026-05-12 18:15:37 +00:00
Code Issues Packages Projects Releases Wiki Activity

small matrix trsm intrinsics optimization code for AX=B and XA'=B

Browse Source 
Change-Id: I90123c4d9adbd314c867995cd19dc975150b448c
This commit is contained in:
sraut
2018-06-06 11:24:33 +05:30
committed by Devrajegowda, Kiran
parent ca6f5b762d
commit d56ca14589
2 changed files with 5418 additions and 13 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
frame/3/trsm/bli_trsm_front.c
View File

@@ -34,7 +34,7 @@
*/
#include "blis.h"
//#define PRINT_SMALL_TRSM_INFO
void bli_trsm_front
(
side_t side,
@@ -47,11 +47,15 @@ void bli_trsm_front
)
{
bli_init_once();
int i, j;
obj_t a_local;
obj_t b_local;
obj_t c_local;
int m = bli_obj_length(*b);
int n = bli_obj_width(*b);
float *L = a->buffer;
float *B = b->buffer;
#ifdef PRINT_SMALL_TRSM_INFO
printf("Side:: %c\n", side ? 'R' : 'L');

5423
kernels/zen/3/bli_trsm_small.c
View File

File diff suppressed because it is too large Load Diff