Closed PhilipFackler closed 4 months ago
@PhilipFackler one quick check is to just use rand
as-is in your kernel function. It should be overloaded by CUDA.jl and AMDGPU.jl as you point out in your link. Meaning that rand would be understood inside the kernel at compilation time as in here for CUDA.jl and here for AMDGPU.jl. Let me know if this worked as I need it for this app as well. Thanks.
@williamfgc you're right. I can't use the version I was using, but r = randn()
works. Thanks.
@PhilipFackler do you think we can close this?
My code is using the version
CUDA.randn!(A::AnyCuArray)
(here) from within a kernel. This corresponds to here. Is this something we could add to JACC or should I just try to wrap this myself? (The random number interface is pretty large. And right now I only need this one function.)