On Jan 5, 2012, at 10:52 PM, Jed Brown wrote: > On Thu, Jan 5, 2012 at 21:34, Mark F. Adams <mark.adams at columbia.edu> > wrote: > This is scary having two communicators running around but the processors that > are dropped out of the new communicator have no rows -- that is why they are > dropped out. > > Okay, now I see the "if (ncrs0 != 0)". I also recall that this is only > running on a sub-communicator because ParMetis happens to crash if you give > it zero vertices on a given proc. I'm inclined to move this communicator > shuffle into MatPartitioningApply_Parmetis() so that user code doesn't have > to mess with it.
Yes it would be much better to move this communicator junk into the parmetis interface. -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120105/e8c9973c/attachment.html>