On Tue, Aug 25, 2020 at 9:01 PM Sajid Ali <sajidsyed2...@u.northwestern.edu>
wrote:

> Hi Barry,
>
> Thanks for the explanation! Removing the calls to
> PetscMalloc(Re)SetCUDAHost solved that issue.
>
> Just to clarify, all PetscMalloc(s) happen on the host and there is no
> special PetscMalloc for device memory allocation ? (Say for an operation
> sequence PetscMalloc1(N, &ptr), VecCUDAGetArray(cudavec, &ptr) )
>
> PetscMallocSetCUDAHost() is to instruct petsc to
use cudaHostAlloc() thereafter to allocate non-pagable host memory.
Yes, there is no variant of PetscMalloc doing cudaMalloc.


> Thank You,
> Sajid Ali | PhD Candidate
> Applied Physics
> Northwestern University
> s-sajid-ali.github.io
>

Reply via email to