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

Merge pull request #1825 from brada4/hemv 

Delay _hemv threading in attempt to address #1820
This commit is contained in:
Martin Kroeker 2018-10-21 20:34:05 +02:00 committed by GitHub
parent 71c6deed60 a293bdcd5e
commit e3c262e5cf
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 9 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
interface/zhemv.c
View File

@ -43,6 +43,10 @@
#include "functable.h" #include "functable.h"
#endif #endif
// this is smallest dimension N of square input a to permit threading
// see graph in issue #1820 for explanation
#define MULTI_THREAD_MINIMAL 362
#ifdef XDOUBLE #ifdef XDOUBLE
#define ERROR_NAME "XHEMV " #define ERROR_NAME "XHEMV "
#elif defined(DOUBLE) #elif defined(DOUBLE)
@ -195,7 +199,11 @@ void CNAME(enum CBLAS_ORDER order, enum CBLAS_UPLO Uplo, blasint n, void *VALPHA
buffer = (FLOAT *)blas_memory_alloc(1); buffer = (FLOAT *)blas_memory_alloc(1);
#ifdef SMP #ifdef SMP
nthreads = num_cpu_avail(2); if (n<MULTI_THREAD_MINIMAL) {
nthreads = 1 ;
} else {
nthreads = num_cpu_avail(2);
};
if (nthreads == 1) { if (nthreads == 1) {
#endif #endif