[petsc-dev] proposed minor PetscPartitioner changes

[Matthew Knepley]

On Mon, Nov 6, 2017 at 8:09 AM, Vaclav Hapla <vaclav.hapla at erdw.ethz.ch>
wrote:

> Hello
>
> The whole PetscPartitioner class sources are part of src/dm/impls/plex/plexpartition.c,
> mixed together with some DMPlex* functions.
> If you don't mind, I would move the PetscPartitioner* stuff into the
> separate file petscpartitioner.c
> (in future, it could be even moved to a separate directory).
>

Yes, that would be fine. I would say that

  src/dm/utils/petscpartitioner.c

would be fine with me.


> I would also like to
> * add PetscPartitioner{Add,Set,Get}OptionsPrefix,
>

Yes.


> * create the partitioner lazily in DMPlexGetPartitioner, followed by
> PetscObjectIncrementTabLevel, PetscLogObjectParent and
> PetscPartitionerSetOptionsPrefix calls,
>

Yes. I am not sure that it helps anything to create the object later.
Waiting the set the type is fine.


> * edit src/dm/impls/plex/examples/tutorials/ex5.c so that it calls
> DMSetFromOptions right after DMPlexCreateFromFile, in order to make the
> partitioner changeable from options.
>

I am not in favor of this one. I prefer you call

    ierr = PetscPartitionerSetFromOptions(part);CHKERRQ(ierr);

there, as in SNES ex12. Refinement happens inside the SetFromOptions(), and
usually people want this after distribution.

  Thanks,

     Matt


> Do you have anything against?
>
> Vaclav




-- 
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

https://www.cse.buffalo.edu/~knepley/ <http://www.caam.rice.edu/~mk51/>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20171106/1af80464/attachment.html>