[Paraview] Icet test build fails with Paraview-3.12.0

Mon Jan 9 11:51:33 EST 2012

Hi,

Thanks - Am I corect in assuming hardware rendering?

I think I have finally fixed the problem - this is part of the patch from
Kitware
+  ELSE (DEFINED MPI_C_COMPILER)
+    # Older MPI module
+    MARK_AS_ADVANCED(CLEAR MPI_INCLUDE_PATH)
+    MARK_AS_ADVANCED(CLEAR MPI_LIBRARY)
+    MARK_AS_ADVANCED(CLEAR MPI_EXTRA_LIBRARY)
+
+    INCLUDE_DIRECTORIES(${MPI_INCLUDE_PATH})
+    SET(ICET_MPI_LIBRARIES ${MPI_LIBRARY} ${MPI_EXTRA_LIBRARY})
+  ENDIF(DEFINED MPI_C_COMPILER)

I had to chage the last line which is
SET(ICET_MPI_LIBRARIES ${MPI_LIBRARY} ${MPI_EXTRA_LIBRARY})

to the following
SET(ICET_MPI_LIBRARIES ${MPI_LIBRARY} ${MPI_C_LIBRARIES})

What I am finding is that, otherwise the mpicxx does get linked and then
when you do a gcc to compile it all breaks.  Replacing gcc with g++ works
OK as expected.

I am not sure if this is required specifically for my set up here or is it
a bug?

Regards,
Rakesh

On 9 January 2012 15:26, David E DeMarle <dave.demarle at kitware.com> wrote:

> ParaView depends on IceT for parallel rendering. If you don't have
> PARAVIEW_USE_PARALLEL on, or if you do by only intend to do parallel data
> processing (pvdataserver), then you don't need it, otherwise you do.
>
> David E DeMarle
> Kitware, Inc.
> R&D Engineer
> 21 Corporate Drive
> Clifton Park, NY 12065-8662
> Phone: 518-881-4909
>
>
>
> On Mon, Jan 9, 2012 at 10:21 AM, Rakesh Hammond <
> rakesh.hammond at googlemail.com> wrote:
>
>> Bruno,
>>
>> Thanks for your help - I got the latest CMakeLists.txt file for ICET and
>> tried to build again (after blowing away my build directory).  It breaks at
>> the same place as before.  On closer inspection, the link.txt file does
>> appear to have links to mpicxx file, removing that and running the same
>> line in the link.txt file builds the test fine.
>>
>> I am beginnning to wonder if some changes are required else where as
>> well?  Any pointeres anyone?
>>
>> On a different note, why does ParaView need IceT?
>>
>> I am tempted to turn it off if this action will have no adverse effects
>> on our requirement.
>>
>> Regards,
>> Rakesh
>>
>>  On 9 January 2012 08:55, Bruno Santos <bruno.santos at bluecape.com.pt>wrote:
>>
>>> Hi Rakesh,
>>>
>>> Good thing you asked!
>>>
>>> OK, so the bug tracker had the commit "number":
>>> 967348651061f039b3dbc7a71d70687d928977dd
>>> So I went to the repository for ParaView on Github:
>>> https://github.com/Kitware/ParaView/
>>> Clicked on one of the "commits" tab and replace "master" with the commit
>>> number: https://github.com/Kitware/ParaView/commits/master ->
>>> https://github.com/Kitware/ParaView/commits/967348651061f039b3dbc7a71d70687d928977dd
>>> So, since the problem was with IceT, I went to where it is in ParaView:
>>> https://github.com/Kitware/ParaView/tree/967348651061f039b3dbc7a71d70687d928977dd/Utilities
>>> OK, the respective folder only says "IceT - 916f22d", namely the commit
>>> number from where it comes. Went to the root folder of ParaView and looked
>>> at ".gitmodules":
>>> https://github.com/Kitware/ParaView/blob/967348651061f039b3dbc7a71d70687d928977dd/.gitmodules
>>>
>>> OK, IceT repo is here: git://paraview.org/IceT.git - so I jumped online
>>> to here: paraview.org/IceT.git -- automatic redirect by the website -->
>>> http://paraview.org/gitweb?p=IceT.git
>>> Clicked on one of the links that say "commit" and replace the number by
>>> "916f22d": paraview.org/gitweb?p=IceT.git;a=commit;h=916f22d
>>> Mmm, it's a merge. OK, then clicked on the "shortlog" link and this *jumped
>>> out* of the list:
>>> http://paraview.org/gitweb?p=IceT.git;a=commit;h=36fad20880740ecacedf9cf80f92a3255c19055b
>>> Clicked on the "diff" link and there we have it!
>>>
>>> Wait, I found another commit different from the one I indicated before.
>>> Apparently I didn't backtrack it properly.... Oh, both are from the same
>>> day! So both are needed! See the shortlog:
>>> http://paraview.org/gitweb?p=IceT.git;a=shortlog;h=e2231f1454c90d7d4fd16540d9468ffbdaaec3f7
>>>
>>> Best regards,
>>> Bruno
>>>
>>>
>>> On 07-01-2012 21:31, Rakesh Hammond wrote:
>>>
>>>  Bruno,
>>>
>>> Thanks - I think this is what I am after, I will have a play with this
>>> diff.
>>>
>>> For future reference, can you please tell me how did you find it?
>>>
>>> I think it would be nice to have a link from the mantis ticket to source
>>> code change !
>>>
>>> Regards,
>>> Rakesh
>>> On 7 January 2012 19:59, Bruno Santos <bruno.santos at bluecape.com.pt>wrote:
>>>
>>>> Hi Rakesh,
>>>>
>>>> If I backtracked it correctly, this should be the fix you are looking
>>>> for:
>>>>
>>>> http://paraview.org/gitweb?p=IceT.git;a=commitdiff;h=e2231f1454c90d7d4fd16540d9468ffbdaaec3f7
>>>>
>>>> Best regards,
>>>> Bruno
>>>>
>>>> Date: Sat, 7 Jan 2012 12:07:24 +0000
>>>> From: Rakesh Hammond <rakesh.hammond at googlemail.com>
>>>> Subject: [Paraview] Icet test build fails with Paraview-3.12.0
>>>>
>>>> Hi,
>>>>
>>>> I am trying to build the ParaView-3.12.0 (stable release on 08th of
>>>> November 2011) with an mpich library on Linux with gcc-4.5.2.
>>>>
>>>> The IceT tests unfortunately fails.  Having looked around the bug
>>>> tracker,
>>>> I found this ticket
>>>>
>>>> http://www.vtk.org/Bug/view.php?id=12716
>>>>
>>>> The problem I am having is same as what is described here, unfortunately
>>>> the fix was merged into the current development directory on the 11th of
>>>> November 2011.
>>>>
>>>> Is it possible to find out exactly what the fix was ? (for example
>>>> doing a
>>>> commitdif?). I am not familiar with git that much, but looking around I
>>>> was
>>>> unable to get the code difference.  If any one knows how to get the
>>>> changes, please send me the url link.
>>>>
>>>> Is there are any plans to release a patch for 3.12.0 any time soon?  At
>>>> work, we are not allowed to build from nightly builds, so I am a little
>>>> bit
>>>> stuck on how to progress further.  My options are either to wait for a
>>>> patch or make the necessary modifications for the stable 3.12.0 and
>>>> build.
>>>>
>>>> Thanks,
>>>> Rakesh
>>>>
>>>> _______________________________________________
>>>> Powered by www.kitware.com
>>>>
>>>> Visit other Kitware open-source projects at
>>>> http://www.kitware.com/opensource/opensource.html
>>>>
>>>> Please keep messages on-topic and check the ParaView Wiki at:
>>>> http://paraview.org/Wiki/ParaView
>>>>
>>>> Follow this link to subscribe/unsubscribe:
>>>> http://www.paraview.org/mailman/listinfo/paraview
>>>>
>>>
>>> Não foram detectados vírus nesta mensagem.
>>> Verificado por AVG - www.avg.com
>>> Versão: 2012.0.1901 / Base de dados de Vírus: 2109/4731 - Data de
>>> Lançamento: 01/08/12
>>>
>>>
>>>
>>> -- http://www.bluecape.com.pt
>>>
>>> blueCAPE Lda - CAE solutions
>>> Rua Fonte dos Corvos, n.29
>>> Casais da Serra
>>> 2665-305 Milharado
>>> PORTUGAL
>>> (+351) 219 755 345  Tel.(+351) 219 755 345  Fax(+351) 963 440 984  Mob.
>>>
>>>
>>
>> _______________________________________________
>> Powered by www.kitware.com
>>
>> Visit other Kitware open-source projects at
>> http://www.kitware.com/opensource/opensource.html
>>
>> Please keep messages on-topic and check the ParaView Wiki at:
>> http://paraview.org/Wiki/ParaView
>>
>> Follow this link to subscribe/unsubscribe:
>> http://www.paraview.org/mailman/listinfo/paraview
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.paraview.org/pipermail/paraview/attachments/20120109/200818be/attachment-0001.htm>