From 9b596820d2dd833648706bff505b459c58f45b4b Mon Sep 17 00:00:00 2001 From: Cedric Nugteren Date: Sat, 22 Oct 2016 10:50:12 +0200 Subject: Fixed a bug in the SYRK/SYR2K/HERK/HER2K routines that would occur with specific tuning parameters (2) --- src/kernels/level3/xgemm_part3.opencl | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'src/kernels') 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; } -- cgit v1.2.3