[petsc-dev] Any plan to add ParMetis-3.1.1?

[Matthew Knepley]

On Tue, Mar 23, 2010 at 11:31 AM, Satish Balay <balay at mcs.anl.gov> wrote:

> On Tue, 23 Mar 2010, Jed Brown wrote:
>
> > On Tue, 23 Mar 2010 11:00:04 -0500, Keita Teranishi <keita at cray.com>
> wrote:
> > > I'd like to know if you have had any problem with 3.1.1 and if you
> > > have any plan to introduce this version in the forthcoming PETSc
> > > release.
> >
> > I have had a local ParMetis-3.1.1 install for more than a year, with
> > both petsc-release 3.0.0 and petsc-dev.  If you have installed
> > ParMetis-3.1.1 yourself, then you should have no problems
> > --with-parmetis-dir=/path/to/parmetis-3.1.1.
> >
> > I didn't realize that --download-parmetis installs ParMetis-3.1.0 (plus
> > some patches), but I expect that there will be no problem updating that
> > to 3.1.1 (this is independent of PETSc's release, the ParMetis API did
> > not change).
>
> Well the parmetis tarball [used by petsc] appears to have quiet a few
> custom patches. Matt would have to go through some of that stuff to
> see if they need to be migrated to ParMetis-3.1.0 or not.
>
> Also - at some point I redid the parmetis build makefiles to make it
> easy for petsc configure to build parmetis. We have to go throught
> that part aswell [and see if its still necessary] before we can
> migrate --download-parmetis to this new version.
>
> So - lot of patches to check. [so I punted it earler when I visited
> this issue].
>
> Satish
>

As I recall the patches are:

  1) George is not careful about declaring routines

  2) He is sloppy with types at some points

  3) There was one actual bug involved with checking bounds

I can do this when I come back from KAUST.

   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-dev/attachments/20100323/95ab1dda/attachment.html>