[petsc-dev] MatMPIAIJGetLocalMat problem with GPUs

Mark Adams mfadams at lbl.gov
Mon Jun 27 19:40:58 CDT 2022

On Mon, Jun 27, 2022 at 2:16 PM Matthew Martineau <mmartineau at nvidia.com>

> Theoretically the access patterns can be worse, but our sparse operations
> output matrices with unordered columns, so the fine matrix being sorted
> shouldn’t impact the overall performance.

And this "unsorted" is really two _sorted_ blocks and the 2nd block is for
off-processor "halo" neighbors, which is a lower order thing.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20220627/9d30360d/attachment.html>

More information about the petsc-dev mailing list