summaryrefslogtreecommitdiff
path: root/src/kernels/level3/xgemm_part3.opencl
diff options
context:
space:
mode:
authorCedric Nugteren <web@cedricnugteren.nl>2016-10-22 10:50:12 +0200
committerCedric Nugteren <web@cedricnugteren.nl>2016-10-22 10:50:12 +0200
commit9b596820d2dd833648706bff505b459c58f45b4b (patch)
treea7e74fc6da72ee72a684f343755fbcecb250c010 /src/kernels/level3/xgemm_part3.opencl
parentdb17b1fbe9267ccd3ad8a7c419689d7b7b98a763 (diff)
Fixed a bug in the SYRK/SYR2K/HERK/HER2K routines that would occur with specific tuning parameters (2)
Diffstat (limited to 'src/kernels/level3/xgemm_part3.opencl')
-rw-r--r--src/kernels/level3/xgemm_part3.opencl2
1 files changed, 1 insertions, 1 deletions
diff --git a/src/kernels/level3/xgemm_part3.opencl b/src/kernels/level3/xgemm_part3.opencl
index 3b26e943..8ac3a3a8 100644
--- a/src/kernels/level3/xgemm_part3.opencl
+++ b/src/kernels/level3/xgemm_part3.opencl
@@ -113,7 +113,7 @@ void XgemmUpper(const int kSizeN, const int kSizeK,
const real beta = GetRealArg(arg_beta);
// Skip these threads if they do not contain threads contributing to the upper-triangle
- if (GetGroupID1()*NWG < GetGroupID0()*MWG) {
+ if ((GetGroupID1() + 1)*NWG < GetGroupID0()*MWG) {
return;
}