[petsc-dev] Fwd: Poisson step in GTS

[Jed Brown]

On Sun, Jun 19, 2011 at 22:19, Matthew Knepley <knepley at gmail.com> wrote:

> Isn't there an API for prescribing affinity, or is this too hard to do for
> most things?


Libnuma is very useful, but it is Linux-only.


> It seems like we know what we want in matvec.


You only really know once the matrix is allocated, although if the user does
perfect nnz[] preallocation, then you would know when allocating memory.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20110619/2c434e34/attachment.html>