[dune-pdelab] Parallel vtkwriter hangs during output on multiple processors

[Alexander Schlaich]

Hi Steffen,

with your version r508 I now get following error when compiling example01:

/mnt/work/schlaich/progs/dune/dune-pdelab-2.0snapshot/dune/pdelab/gridfunctionspace/../common/multitypetree.hh:195:
error: 'AlwaysTrue' was not declared in this scope
/mnt/work/schlaich/progs/dune/dune-pdelab-2.0snapshot/dune/pdelab/gridfunctionspace/../common/multitypetree.hh:195:
error: 'integral_constant' was not declared in this scope
/mnt/work/schlaich/progs/dune/dune-pdelab-2.0snapshot/dune/pdelab/gridfunctionspace/../common/multitypetree.hh:195:
error: template argument 1 is invalid
/mnt/work/schlaich/progs/dune/dune-pdelab-2.0snapshot/dune/pdelab/gridfunctionspace/../common/multitypetree.hh:195:
error: expected nested-name-specifier

I just switched to the stable branch because I thought maybe there won"t
be that problem with vtkwriter. Is it a good Idea to switch back to the
trunk version that was working for me (pdelab r758)? Have you got any
hint for the vtkwriter problem there?

Thank you a lot,

Alex


On 12/01/2010 02:32 PM, Steffen Müthing wrote:
> Hi Alex,
>
> I think this discussion should go back to the PDELab mailing list. I think
> this might be related to a fix for the multitype tree in r508 of
> dune-pdelab.
> If you are using the 2.0.1 release of the core modules, could you please
> try replacing dune/pdelab/common/multitypetree.hh with the attached file?
>
> Steffen
>
>
>
>
>
> Am 01.12.2010 um 08:57 schrieb Alexander Schlaich:
>
>> Hi Dune,
>>
>> I'm not sure if this belongs to this mailinglist or the pdelab one, so
>> sorry for cross-posting (see my original mail).
>>
>> I now tried to use Dune-2.0 stable release with pdelab-2.0 branch. For
>> pdelab-howto-2.0 I get compilation errors like:
>>
>> /mnt/work/schlaich/progs/dune/dune-pdelab-2.0snapshot/dune/pdelab/gridfunctionspace/../common/multitypetree.hh:195:
>> error: default template arguments may not be used in function templates
>>
>> So should I better stay with trunk version?
>>
>> Thanks a lot for any reply!
>>
>> Alex
>>
>>
>>
>> On Mon, November 29, 2010 5:24 pm, Alexander Schlaich wrote:
>>> Hi there!
>>>
>>> My name is Alexander Schlaich and I'm working at the Institute for
>>> Computational Physics at University of Stuttgart. For my diploma thesis
>>> I'm
>>> actually implementing a solver for the nonlinear Poisson Boltzmann
>>> equation for which I adapted example 2 of dune-pdelab-howto. Works fine.
>>>
>>> When parallelizing my code and using more than 1 processor vtkwriter
>>> crashes
>>> somehow, One output file is written correctly, a second one is
>>> touched and
>>> my
>>> program hangs up. I orientated on example 6 for parallelization but
>>> don't
>>> see
>>> how to use vtkwriter in another way.
>>>
>>> Does anybody know what the problem is? Any hint would be much
>>> appreciated.
>>>
>>> Many thanks in advance,
>>>
>>>
>>> Alex
>>>
>>> ---
>>> My code is attached.
>>> I'm using:
>>>
>>> dune-common revision: 6204
>>> dune-grid revision: 7038
>>> dune-istl revision: 1323
>>> dune-localfunctions revision: 908
>>> pdelab revision: 758
>>> ---
>>> _______________________________________________
>>> dune-pdelab mailing list
>>> dune-pdelab at dune-project.org
>>> http://lists.dune-project.org/mailman/listinfo/dune-pdelab
>>>
>>
>>
>> <pb_code.tar.gz>_______________________________________________
>> Dune mailing list
>> Dune at dune-project.org
>> http://lists.dune-project.org/mailman/listinfo/dune
>
> Steffen Müthing
> Universität Stuttgart
> Institut für Parallele und Verteilte Systeme
> Universitätsstr. 38
> 70569 Stuttgart
> Tel: +49 711 685 88429
> Fax: +49 711 685 88340
> Email: steffen.muething at ipvs.uni-stuttgart.de
>
>
>
>
>
>
>
> _______________________________________________
> dune-pdelab mailing list
> dune-pdelab at dune-project.org
> http://lists.dune-project.org/mailman/listinfo/dune-pdelab

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.dune-project.org/pipermail/dune-pdelab/attachments/20101201/fdf4cd5c/attachment.htm>