<p>Hi Christophe,</p>
<p>On a slightly related topic: where do you use PETSc specifically inside Gmsh ? I haven't looked at the sources in a while but was curious. I will configure the latest source based on PETSc soon to understand the details. Any help/pointers will be much appreciated. </p>
<p>Thanks,<br>
Vijay</p>
<div class="gmail_quote">On Aug 23, 2011 7:08 AM, "Christophe Geuzaine" <<a href="mailto:cgeuzaine@ulg.ac.be">cgeuzaine@ulg.ac.be</a>> wrote:<br type="attribution">> <br>> On 23 Aug 2011, at 09:47, Mikhail Artemiev wrote:<br>
> <br>>> Hi all!<br>>> When I run the following easiest program:<br>>> <br>>> #include <mpi.h><br>>> #include "Gmsh.h"<br>>> int main(int argc, char **argv)<br>>> {<br>
>> MPI_Init(&argc, &argv);<br>>> GmshInitialize(argc, argv);<br>>> GmshFinalize();<br>>> MPI_Finalize();<br>>> return 0;<br>>> }<br>>> <br>>> I have the following error:<br>
>> <br>>> [artemiev-desktop:19676] *** An error occurred in MPI_Init<br>>> [artemiev-desktop:19676] *** on communicator MPI_COMM_WORLD<br>>> [artemiev-desktop:19676] *** MPI_ERR_OTHER: known error not in list<br>
>> [artemiev-desktop:19676] *** MPI_ERRORS_ARE_FATAL (your MPI job will now abort)<br>>> --------------------------------------------------------------------------<br>>> mpirun has exited due to process rank 1 with PID 19676 on<br>
>> node artemiev-desktop exiting without calling "finalize". This may<br>>> have caused other processes in the application to be<br>>> terminated by signals sent by mpirun (as reported here).<br>
>> --------------------------------------------------------------------------<br>>> --------------------------------------------------------------------------<br>>> Calling MPI_Init or MPI_Init_thread twice is erroneous.<br>
>> --------------------------------------------------------------------------<br>>> [artemiev-desktop:19674] 1 more process has sent help message help-mpi-errors.txt / mpi_errors_are_fatal<br>>> [artemiev-desktop:19674] Set MCA parameter "orte_base_help_aggregate" to 0 to see all help / error messages<br>
>> <br>>> I thought that "GmshInitialize" call "MPI_Init". Therefore I recompiled Gmsh with -DENABLE_MPI=0 option.<br>>> But the result was the same.<br>>> <br>>> Could you give a hint about how to use Gmsh API in parallel program?<br>
> <br>> Hi Mikhail - MPI_Init might also be called (indirectly) in GmshInitialize if Gmsh is compiled with PETSc support. Is this the case in your setup?<br>> <br>> <br>> <br>>> <br>>> Thanks<br>
>> Mikhail Artemiev<br>>> _______________________________________________<br>>> gmsh mailing list<br>>> <a href="mailto:gmsh@geuz.org">gmsh@geuz.org</a><br>>> <a href="http://www.geuz.org/mailman/listinfo/gmsh">http://www.geuz.org/mailman/listinfo/gmsh</a><br>
> <br>> -- <br>> Prof. Christophe Geuzaine<br>> University of Liege, Electrical Engineering and Computer Science <br>> <a href="http://www.montefiore.ulg.ac.be/~geuzaine">http://www.montefiore.ulg.ac.be/~geuzaine</a><br>
> <br>> <br>> <br>> <br>> _______________________________________________<br>> gmsh mailing list<br>> <a href="mailto:gmsh@geuz.org">gmsh@geuz.org</a><br>> <a href="http://www.geuz.org/mailman/listinfo/gmsh">http://www.geuz.org/mailman/listinfo/gmsh</a><br>
</div>