[Dune] [Dune-Commit] dune-istl r1524 - trunk/m4

Martin Nolte nolte at mathematik.uni-freiburg.de
Sat Feb 11 19:43:00 CET 2012


Hi Christoph,

I would also like the metis / parmetis test to stay in dune-common, because it 
is indeed useful for grid partitioning, too. Truth is, that I thought the test 
would be in dune-istl and was intending to propose moving the test to 
dune-common for this reason.

With respect to ALUGrid, however, I can put your mind at rest: The ALUGrid 
bindings do no longer include metis.h directly or indirectly and the dependency 
to the metis library is handled by ALUGrid's pkg-config file.

Best,

Martin

On 02/11/2012 06:04 PM, Christoph Grüninger wrote:
> Hi there,
> first I want to apologize for any inconvenience my faulty
> commit might caused. Please tell me if a have to communi-
> cate more (or less) before I change code.
>
> I reverted the move of the BLAS test. If more parts have
> to be reverted I'm sorry and will do it.
>
>> are you sure the test wasn't run implcitly by some
>> other test.
>
> I checked but unfortunately I missed the LAPACK test.
> Beside that only petsc.m4 in dune-pdelab calls it but that
> is not a problem because dune-istl is required by PDELab.
>
>> please be cautious with changes to the build-system. It
>> is fragile!
>
> Sadly I'm aware of that fact. Currently I want to clean up
> the tests to make dependencies more obvious. Maybe we can
> throw away some old tests nobody uses anymore or where
> former problems are now solved by newer libraries or
> compilers. For example the LAPACK test is almost 10 years
> old. We should check whether the projects from which we
> copied tests have improved them in the meantime.
>
> Thanks for giving me the heads-up,
> Christoph
>
> _______________________________________________
> Dune mailing list
> Dune at dune-project.org
> http://lists.dune-project.org/mailman/listinfo/dune

-- 
Dr. Martin Nolte <nolte at mathematik.uni-freiburg.de>

Universität Freiburg                                   phone: +49-761-203-5630
Abteilung für angewandte Mathematik                    fax:   +49-761-203-5632
Hermann-Herder-Straße 10
79104 Freiburg, Germany




More information about the Dune mailing list