On Thu, Mar 26, 2015 at 10:36 AM, Ataollah Mesgarnejad < [email protected]> wrote:
> Dear all, > > I was wondering if someone can tell me how you get a field's Vec (subVec > of the global Vec) from a distributed DMPlex. It seems that DMCreateFieldIS > works for sequential DMPlex but gives incorrect values for a distributed > DMPlex. > It seems like what you want is: https://bitbucket.org/petsc/petsc/src/fdac1df83b6d73782209c263cf9d6b0b33600136/src/vec/vec/utils/vsection.c?at=master#cl-229 and also it seems like DMCreateFieldIS() is broken somehow. We should fix it. It currently passes all the tests for the FieldSplit solvers, so I am not sure what is going wrong here. Thanks, Matt > Many thanks, > Ata > -- 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
