On Mon, May 16, 2011 at 13:52, Dmitry Karpeev <karpeev at mcs.anl.gov> wrote:
> True, although MatSetValuesLocal (not the default implementation in > the interface) > is a way to extend the mechanism that relies on ISLocalToGlobalMapping. > This can include whatever semantics are appropriate to handle the overlaps, > etc. > I think this is a reasonable splitting of functionality between > PetscLayout and MatSetValuesLocal(). > Yes, I agree that the MatGetLocalSubMatrix()/MatSetValuesLocal() paradigm is a good way to manage overlap. With more general transformations, we could consider extending the interface (or having a new interface that is similar in spirit). -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20110516/b1807792/attachment.html>
