[petsc-dev] problem with MatSeqAIJCUSPARSEILUAnalysisAndCopyToGPU

Mark Adams mfadams at lbl.gov
Tue Dec 22 19:38:02 CST 2020

> Allocation/free of pinned memory is slow, usually on the order of several
> milliseconds. So these numbers look normal. Is there any opportunity to
> reuse the pinned memory in these functions?

Ah, good. So we understand this.

I was thinking of adding a member to the class and allocated it as needed.
I think I will do that, start with this one and see how it goes.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20201222/573d968c/attachment-0001.html>

More information about the petsc-dev mailing list