amd/blis
1
0
Fork 0
mirror of https://github.com/amd/blis.git synced 2026-05-11 01:30:00 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'master' into dev

Browse Source
This commit is contained in:
Field G. Van Zee
2018-10-17 18:00:40 -05:00
parent 779d64dc30 3c52725693
commit 49d3f9fcbb
108 changed files with 20250 additions and 646 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
kernels/bgq/1/bli_dotv_bgq_int.c
View File

@@ -34,8 +34,8 @@
#include "blis.h"
void bli_ddotv_bgq_int
(
void bli_ddotv_bgq_int
(
conj_t conjx,
conj_t conjy,
dim_t n,
@@ -44,14 +44,14 @@ void bli_ddotv_bgq_int
double* restrict rho,
cntx_t* restrict cntx
)
{
{
bool_t use_ref = FALSE;
// If the vector lengths are zero, set rho to zero and return.
if ( bli_zero_dim1( n ) ) {
PASTEMAC(d,set0s)( rho );
return;
}
PASTEMAC(d,set0s)( *rho );
return;
}
// If there is anything that would interfere with our use of aligned
// vector loads/stores, call the reference implementation.
if ( incx != 1 || incy != 1 || bli_is_unaligned_to( ( siz_t )x, 32 ) || bli_is_unaligned_to( ( siz_t )y, 32 ) )
@@ -64,7 +64,7 @@ void bli_ddotv_bgq_int
dim_t n_run = n / 4;
dim_t n_left = n % 4;
double rhos = 0.0;
#pragma omp parallel reduction(+:rhos)
{