[vtkusers] vtkUnstructuredGrid::FindCell() fails to find pyramid cells on a Fujitsu FX10 (non-x86) platform?

[Andy Bauer]

Hi Takuya,

This has now been changed in the main repo.  Thanks for the help.

Andy

On Tue, Apr 17, 2012 at 7:09 AM, Andy Bauer <andy.bauer at kitware.com> wrote:

> Thanks for this fix.  I'll check it out.
>
> Andy
>
>
> On Tue, Apr 17, 2012 at 2:23 AM, Takuya OSHIMA <oshima at eng.niigata-u.ac.jp
> > wrote:
>
>> Hi Andy,
>>
>> Thank you for your reply. In fact I was using different versions of
>> vtkPyramid for the SPARC and the x86 platforms (the former from
>> exactly PV 3.14.1, the latter from a bit older git-master version
>> around Jan.-Feb. of this year) and had missed a recent commit
>> 6332bbee82445ed5a3c6cce81c0675f152ee8649 that was supposed to fix
>> gradient issue at apex of a pyramid. Obviously, the commit requires
>> another fix as attached. With the fix my program now works.
>>
>> I attached the same patch to the corresponding entry in the tracker:
>> http://vtk.org/Bug/view.php?id=12917
>>
>> Thanks,
>> Takuya
>>
>> Takuya OSHIMA, Ph.D.
>> Faculty of Engineering, Niigata University
>> 8050 Ikarashi-Ninocho, Nishi-ku, Niigata, 950-2181, JAPAN
>>
>>
>> From: Andy Bauer <andy.bauer at kitware.com>
>> Subject: Re: [vtkusers] vtkUnstructuredGrid::FindCell() fails to find
>> pyramid cells on a Fujitsu FX10 (non-x86) platform?
>> Date: Mon, 16 Apr 2012 09:30:32 -0400
>>
>> > Hi,
>> >
>> > I wonder if it's a tolerance issue in vtkPyramid::EvaluatePosition().
>>  The
>> > tolerance is set to 0.001.  You could try increasing that value as well
>> as
>> > increasing the 1.001 values and see if that solves your problem.
>> >
>> > Andy
>> >
>> > On Mon, Apr 16, 2012 at 8:43 AM, Takuya OSHIMA
>> > <oshima at eng.niigata-u.ac.jp>wrote:
>> >
>> >> Hi,
>> >>
>> >> We have built VTK as a part of ParaView 3.14.1 on a Fujitsu FX10
>> >> platform (SPARC64 IXfx). The built library mostly works, but for my
>> >> program vtkUnstructuredGrid::FindCell() fails to find pyramid cells in
>> >> my hex/pyramid/tetra-composite mesh. Contrarily, the same program
>> >> works as expected on a x86 platform (Mac OS X 64bit).
>> >>
>> >> I have searched through the FindCell() code for cell type specific
>> >> function calls and had a suspicion that vtkPyramid::EvaluatePosition()
>> >> (which FindCell() should internally be calling as a virtual function)
>> >> had an uninitialized variable or something but could not find anything
>> >> wrong. So far I am completely lost.
>> >>
>> >> Has anyone experienced (and solved) a similar problem? Any ideas or
>> >> thoughts?
>> >>
>> >> Platform: Fujitsu FX10 (SPARC64 IXfx)
>> >> VTK version: VTK included in ParaView 3.14.1
>> >> CMake version: 2.8.7
>> >> Compiler: Fujitsu's FCCpx (cross-compilation on an x86 Linux platform)
>> >> FCCpx_ENV (the compiler options): -Kfast -Xg -std=gnu++98
>> >>
>> >> Thanks,
>> >> Takuya
>> >>
>> >> Takuya OSHIMA, Ph.D.
>> >> Faculty of Engineering, Niigata University
>> >> 8050 Ikarashi-Ninocho, Nishi-ku, Niigata, 950-2181, JAPAN
>> >> _______________________________________________
>> >> 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 VTK FAQ at:
>> >> http://www.vtk.org/Wiki/VTK_FAQ
>> >>
>> >> Follow this link to subscribe/unsubscribe:
>> >> http://www.vtk.org/mailman/listinfo/vtkusers
>> >>
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.vtk.org/pipermail/vtkusers/attachments/20120418/126d5194/attachment.htm>