On Wed, Mar 14, 2012 at 5:11 AM, Jed Brown <jedbrown at mcs.anl.gov> wrote:
> On Wed, Mar 14, 2012 at 02:53, Klaij, Christiaan <C.Klaij at marin.nl> wrote: > >> Thanks, no more problems after PCSetUP(). I am using petsc-dev, >> so are you saying that MatSetNullSpace() on the block matrix [A00 >> A01, A10 A11] will be enough? Can/will the null space of the >> Schur complement be deduced from this? Otherwise one would still >> need to "drill down" to get Schur complement matrix, right? >> > > Unfortunately, no, and I don't think it's possible to do this > automatically. > > Barry/Matt, this is the recurring issue of whether > MatGetSubMatrix()/MatGetSchurComplement() should have the "big" matrix > retain ownership, such that the user could decompose, set null spaces and > auxiliary operators, restore, and have PCFieldSplit and everyone else do > the right thing. > I thought MatSchurComplement() made the original matrix available? I still disagree with this for MatGetSubMatrix() since it will create a nightmare for references. I am still tracking down a reference bug in FS when you use a custom preconditoner for Schur. Matt -- What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead. -- Norbert Wiener -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20120314/f31d4a85/attachment.htm>
