Hi all,

First time poster here. Is there any way I can change global memory size
and max allocable memory size for an OpenCL program? My card has 4GB or
VRAM, but seems like I can only use a quarter of it for OCL, and I can only
use a quarter of the quarter for a single block of allocation. Is there
anyway to lift this limitation? I know I can do that by exporting 2
environment variables when using fglrx, but how to do this with
AMDGPU+Clover?

My GPU is AMD R9 nano, and I have enough host memory (64GB).



Thanks,
Bo Gao
_______________________________________________
mesa-dev mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/mesa-dev

Reply via email to