summaryrefslogtreecommitdiff
path: root/src/kernels/level3/xgemm_part3.opencl
diff options
context:
space:
mode:
authorCedric Nugteren <web@cedricnugteren.nl>2016-10-22 10:41:02 +0200
committerCedric Nugteren <web@cedricnugteren.nl>2016-10-22 10:41:02 +0200
commitdb17b1fbe9267ccd3ad8a7c419689d7b7b98a763 (patch)
treea277d044f361596a0febff3cdaa7101b982aac64 /src/kernels/level3/xgemm_part3.opencl
parent597974b40d52d9cd199e5189d6d55cd12c841e63 (diff)
Fixed a bug in the SYRK/SYR2K/HERK/HER2K routines that would occur with specific tuning parameters
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 a5faef5a..3b26e943 100644
--- a/src/kernels/level3/xgemm_part3.opencl
+++ b/src/kernels/level3/xgemm_part3.opencl
@@ -153,7 +153,7 @@ void XgemmLower(const int kSizeN, const int kSizeK,
const real beta = GetRealArg(arg_beta);
// Skip these threads if they do not contain threads contributing to the lower-triangle
- if (GetGroupID1()*NWG > GetGroupID0()*MWG) {
+ if (GetGroupID1()*NWG > (GetGroupID0() + 1)*MWG) {
return;
}