On Tue, Mar 29, 2011 at 12:55, Thomas Witkowski < thomas.witkowski at tu-dresden.de> wrote:
> I have implemented a Schur complement solver based on the fieldsplit > preconditioner, i.e., my code is executed with the option > -pc_fieldsplit_type schur. How can I define a preconditioner Sp for the > inner Schur complement solution? I want to define the preconditioner in an > explicit way, thus build a matrix that should approximate S in some way! See PCFieldSplitSchurPrecondition and (perhaps) PCShell. Perhaps if you give a more specific example, we can give more specific suggestions. -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20110329/6eeae101/attachment.html>