cwpearson/astaroth
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Disabled the project-wide maxrregcount flag by default since it is only beneficial for resource-heavy kernels. The maximum register count should be defined per kernel instead if needed.

Browse Source
This commit is contained in:
jpekkila
2019-07-22 12:58:28 +03:00
parent a8caad1ade
commit 0db61dd411
1 changed files with 0 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
src/core/CMakeLists.txt
View File

@@ -19,7 +19,6 @@ set(CUDA_ARCH_FLAGS -gencode arch=compute_37,code=sm_37
-gencode arch=compute_60,code=sm_60 -gencode arch=compute_60,code=sm_60
-gencode arch=compute_61,code=sm_61 -gencode arch=compute_61,code=sm_61
-lineinfo -lineinfo
--maxrregcount=255
-ftz=true -ftz=true
-std=c++11) #--maxrregcount=255 -ftz=true #ftz = flush denormalized floats to zero -std=c++11) #--maxrregcount=255 -ftz=true #ftz = flush denormalized floats to zero
# -Xptxas -dlcm=ca opt-in to cache all global loads to L1/texture cache # -Xptxas -dlcm=ca opt-in to cache all global loads to L1/texture cache