From db17b1fbe9267ccd3ad8a7c419689d7b7b98a763 Mon Sep 17 00:00:00 2001 From: Cedric Nugteren Date: Sat, 22 Oct 2016 10:41:02 +0200 Subject: Fixed a bug in the SYRK/SYR2K/HERK/HER2K routines that would occur with specific tuning parameters --- 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 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; } -- cgit v1.2.3