<br><font size=2 face="sans-serif">I'm using Paraview 2.6, and here are
the configuration options for compilers and mpi. If you want to see
more of my configuration, let me know. I spent quite awhile messing
around with these, especially to get distcc workking (our cluster contains
lots of 7 year old nodes... only together do they actually have notable
computing power, and compiling on one node would've taken forever). One
important thing was that this did NOT compile correctly with the new gcc
that we have (gcc-4.2 I believe) so we used the older one.</font>
<br>
<br><font size=2 face="sans-serif">//C compiler.</font>
<br><font size=2 face="sans-serif">CMAKE_C_COMPILER:FILEPATH=/usr/lib/distcc/bin/gcc-3.4</font>
<br><font size=2 face="sans-serif">[ which is linked to the normal distcc,
with the "gcc-3.4" binary in our PATH ]</font>
<br>
<br><font size=2 face="sans-serif">//CXX compiler.</font>
<br><font size=2 face="sans-serif">CMAKE_CXX_COMPILER:FILEPATH=/usr/lib/distcc/bin/g++-3.4</font>
<br><font size=2 face="sans-serif">[ which is linked to the normal distcc,
with the "g++-3.4" binary in our PATH ]</font>
<br>
<br><font size=2 face="sans-serif">//If a second mpi library is necessary,
specify it here.</font>
<br><font size=2 face="sans-serif">MPI_EXTRA_LIBRARY:FILEPATH=/usr/lib/libmpi++.so</font>
<br><font size=2 face="sans-serif">//Path to a file.</font>
<br><font size=2 face="sans-serif">MPI_INCLUDE_PATH:PATH=/usr/local/mpich/1.2.5.2/gnu/include</font>
<br><font size=2 face="sans-serif">//Path to a library.</font>
<br><font size=2 face="sans-serif">MPI_LIBRARY:FILEPATH=/usr/lib/libmpi.so</font>
<br><font size=2 face="sans-serif">[ CMake figured these out on its own.
We compiled and installed our own mpich, not (mpich-)mpd. ]</font>
<br>
<br><font size=2 face="sans-serif">Hope this helps,</font>
<br><font size=2 face="sans-serif">~Kyle</font>
<br><font size=2><tt><br>
> Message: 3<br>
> Date: Fri, 22 Jun 2007 12:31:01 -0400<br>
> From: "Robert Maynard" <thorin.hax@gmail.com><br>
> Subject: Re: [Paraview] Lam mpi + paraview<br>
> To: paraview@paraview.org<br>
> Message-ID:<br>
> <b45d6a1e0706220931n2ab240cdh1ccfa63a92ecb5e9@mail.gmail.com><br>
> Content-Type: text/plain; charset="iso-8859-1"<br>
> <br>
> My first attempt at getting MPI working with paraview was by using
the<br>
> default compiler and setting the MPI_INCLUDE_PATH and MPI_LIBRARY
paths. The<br>
> problem I was running into was that I was getting underfined references
when<br>
> linking the mpi libraries. The solution that Lam recommended was to
use<br>
> their compiler to solve the library problems. If somebody could send
me<br>
> their MPI_INCLUDE_PATH and MPI_LIBRARY configurations it would help
me out<br>
> alot.<br>
> <br>
> On 6/22/07, David E DeMarle <dave.demarle@kitware.com> wrote:<br>
> ><br>
> > That is odd.<br>
> ><br>
> > I have a feeling that lam's mpicc brings in or leaves out something<br>
> > important in the compilation and linking, especially when it
compiles the<br>
> > non parallel paraview app. When compiling against MPICH for example
we don't<br>
> > use MPICH's mpicc. But instead use normal cc and link in the
extra<br>
> > libraries. Instead of setting the compiler directly, have you
tried just<br>
> > turning on PARAVIEW_USE_MPI and then setting the MPI_INCLUDE_PATH
and<br>
> > MPI_LIBRARY instead?<br>
> ><br>
> > cheers<br>
> > Dave<br>
> ><br>
> ><br>
> ><br>
> > On 6/22/07, Robert Maynard <thorin.hax@gmail.com> wrote:<br>
> > ><br>
> > > the backtrace seems really short to me, I did compile paraview
with the<br>
> > > Debug flag.<br>
> > ><br>
> > > Program received signal SIGFPE, Arithmetic exception.<br>
> > > 0x03d79ac6 in ?? ()<br>
> > > (gdb) backtrace<br>
> > > #0 0x03d79ac6 in ?? ()<br>
> > > #1 0x00000000 in ?? ()<br>
> > > (gdb)<br>
> > ><br>
> > > On 6/22/07, David E DeMarle < dave.demarle@kitware.com>
wrote:<br>
> > > ><br>
> > > > I was thinking of a stack dump not a system call trace.
In gdb after<br>
> > > > the crash type backtrace and send that output to the
list.<br>
> > > ><br>
> > > > Thanks Robert.<br>
> > > > Dave<br>
> > > ><br>
> > > > On 6/22/07, Robert Maynard <thorin.hax@gmail.com>
wrote:<br>
> > > > ><br>
> > > > > Sure I have attached the strace output<br>
> > > > ><br>
> > > > ><br>
> > > > > On 6/22/07, David E DeMarle < dave.demarle@kitware.com>
wrote:<br>
> > > > > ><br>
> > > > > > Can you send a stack trace for the floating
point exception?<br>
> > > > > ><br>
> > > > > ><br>
> > > > > > On 6/22/07, Robert Maynard <thorin.hax@gmail.com
> wrote:<br>
> > > > > > ><br>
> > > > > > > Yes i understand I have to link to the
server in MPI now, but as<br>
> > > > > > > my first email stated, by I trying running
the client <br>
> now that I compiled<br>
> > > > > > > mpi support in, I get a runtime floating
point exception<br>
> before paraview<br>
> > > > > > > launches.<br>
> > > > > > ><br>
> > > > > > > On 6/22/07, David E DeMarle <dave.demarle@kitware.com
> wrote:<br>
> > > > > > > ><br>
> > > > > > > > In 3.0 the paraview executable
is not MPI parallel, only the<br>
> > > > > > > > server executables are.<br>
> > > > > > > > That is, you can no longer run
"mpirun -np N paraview" but<br>
> > > > > > > > instead have to do something like
"mpirun -np N <br>
> pvserver&; paraview" and<br>
> > > > > > > > then use File->Choose Server
to connect to the parallel server.<br>
> > > > > > > ><br>
> > > > > > > > cheers<br>
> > > > > > > > Dave DeMarle<br>
> > > > > > > ><br>
> > > > > > > > On 6/22/07, Robert Maynard <
thorin.hax@gmail.com> wrote:<br>
> > > > > > > ><br>
> > > > > > > > > I have compiled paraview with
mpi support using lam. I<br>
> > > > > > > > > changed the standard c and
c++ compilers for <br>
> paraview too mpicc and mpic++<br>
> > > > > > > > > so that all the<br>
> > > > > > > > > mpi references where aviable.
The build was correctly<br>
> > > > > > > > > compiled, the problem comes
when trying to launch <br>
> the standalone version of<br>
> > > > > > > > > paraview.<br>
> > > > > > > > ><br>
> > > > > > > > > When launching /bin/paraview
the error i get is "Floating<br>
> > > > > > > > > point exception", using
gdb to trace the error I get<br>
> > > > > > > > > New process 16350]<br>
> > > > > > > > ><br>
> > > > > > > > > Program received signal
SIGFPE, Arithmetic exception.<br>
> > > > > > > > > 0x05761ac6 in ?? ()<br>
> > > > > > > > ><br>
> > > > > > > > > Now if I try to run /bin/paraview
with mpirun -np 1, it<br>
> > > > > > > > > tells me that paraview did
not invoke MPI_INIT <br>
> before quitting. The problem<br>
> > > > > > > > > seems to only<br>
> > > > > > > > > happen with standalone version,
as /bin/pvserver runs<br>
> > > > > > > > > perfectly fine with mpirun<br>
> > > > > > > > ><br>
> > > > > > > > > _______________________________________________<br>
> > > > > > > > > ParaView mailing list<br>
> > > > > > > > > ParaView@paraview.org<br>
> > > > > > > > > http://www.paraview.org/mailman/listinfo/paraview<br>
> > > > > > > > ><br>
> > > > > > > > ><br>
> > > > > > > ><br>
> > > > > > ><br>
> > > > > > > _______________________________________________<br>
> > > > > > > ParaView mailing list<br>
> > > > > > > ParaView@paraview.org<br>
> > > > > > > http://www.paraview.org/mailman/listinfo/paraview<br>
> > > > > > ><br>
> > > > > > ><br>
> > > > > ><br>
> > > > ><br>
> > > > > _______________________________________________<br>
> > > > > ParaView mailing list<br>
> > > > > ParaView@paraview.org<br>
> > > > > http://www.paraview.org/mailman/listinfo/paraview<br>
> > > > ><br>
> > > > ><br>
> > > > ><br>
> > > ><br>
> > ><br>
> > > _______________________________________________<br>
> > > ParaView mailing list<br>
> > > ParaView@paraview.org<br>
> > > http://www.paraview.org/mailman/listinfo/paraview<br>
> > ><br>
> > ><br>
> ><br>
> -------------- next part --------------<br>
> An HTML attachment was scrubbed...<br>
> URL: http://public.kitware.<br>
> com/pipermail/paraview/attachments/20070622/4eed4dc3/attachment.html<br>
> <br>
> ------------------------------<br>
> <br>
> _______________________________________________<br>
> ParaView mailing list<br>
> ParaView@paraview.org<br>
> http://www.paraview.org/mailman/listinfo/paraview<br>
> <br>
> <br>
> End of ParaView Digest, Vol 38, Issue 52<br>
> ****************************************<br>
</tt></font>
<P>The server made the following annotations <br/>
---------------------------------------------------------------------------------<br/>
This message contains information that may be privileged or confidential and is the property of Beckman Coulter, Inc. It is intended only for the person to whom it is addressed. If you are not the intended recipient, you are not authorized to read, print, retain, copy, disseminate, distribute or use this message or any part thereof. If you receive this message in error, please notify the sender immediately and delete all copies of this message.
---------------------------------------------------------------------------------<br/>
</P>