<div dir="ltr"><div class="gmail_default" style="font-family:verdana,sans-serif;font-size:small;color:#000000">I am pretty sure this is a bug in parmetis. A few years ago, I nailed down this "divide by zero" bug, and reported to them. They said they would take a look, but never did. This usually happens when the graph is relatively dense. </div><div class="gmail_default" style="font-family:verdana,sans-serif;font-size:small;color:#000000"><br></div><div class="gmail_default" style="font-family:verdana,sans-serif;font-size:small;color:#000000">You can try sequential Metis.</div><div class="gmail_default" style="font-family:verdana,sans-serif;font-size:small;color:#000000"><br></div><div class="gmail_default" style="font-family:verdana,sans-serif;font-size:small;color:#000000">Sherry</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Thu, Feb 14, 2019 at 10:38 PM Smith, Barry F. via petsc-users <<a href="mailto:petsc-users@mcs.anl.gov">petsc-users@mcs.anl.gov</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-style:solid;border-left-color:rgb(204,204,204);padding-left:1ex"><br>
<br>
> On Feb 15, 2019, at 12:30 AM, Marius Buerkle <<a href="mailto:mbuerkle@web.de" target="_blank">mbuerkle@web.de</a>> wrote:<br>
> <br>
> It works with all options for "-mat_superlu_dist_colperm" save parmetis. or do you mean to change some other option besides colperm?<br>
<br>
No, I didn't have any other suggestions. It could be that this option just more easily introduces zero pivots or you could report it to Sherry Li, the main SuperLU_DIST developer and see what she says.<br>
<br>
Barry<br>
<br>
> <br>
>> Gesendet: Freitag, 15. Februar 2019 um 04:16 Uhr<br>
>> Von: "Smith, Barry F." <<a href="mailto:bsmith@mcs.anl.gov" target="_blank">bsmith@mcs.anl.gov</a>><br>
>> An: "Marius Buerkle" <<a href="mailto:mbuerkle@web.de" target="_blank">mbuerkle@web.de</a>><br>
>> Cc: "PETSc users list" <<a href="mailto:petsc-users@mcs.anl.gov" target="_blank">petsc-users@mcs.anl.gov</a>><br>
>> Betreff: Re: [petsc-users] superlu_dist+parmetis<br>
>> <br>
>> <br>
>> Given the message the likely cause is SuperLU_DIST got a zero pivot on process 6. Presumably the colperm parmetis is one induced the zero pivot.<br>
>> <br>
>> Have you tried other superlu_dist options to find others that do not cause this?<br>
>> <br>
>> Barry<br>
>> <br>
>> <br>
>>> On Feb 14, 2019, at 6:44 PM, Marius Buerkle via petsc-users <<a href="mailto:petsc-users@mcs.anl.gov" target="_blank">petsc-users@mcs.anl.gov</a>> wrote:<br>
>>> <br>
>>> Dear PETSc team,<br>
>>> <br>
>>> I try to run superlu_dist+parmetis with " -mat_superlu_dist_colperm parmetis -mat_superlu_dist_parsymbfact" which gives me the following error<br>
>>> <br>
>>> [6]PETSC ERROR: Caught signal number 8 FPE: Floating Point Exception,probably divide by zero<br>
>>> [6]PETSC ERROR: Try option -start_in_debugger or -on_error_attach_debugger<br>
>>> [6]PETSC ERROR: or see <a href="http://www.mcs.anl.gov/petsc/documentation/faq.html#valgrind" rel="noreferrer" target="_blank">http://www.mcs.anl.gov/petsc/documentation/faq.html#valgrind</a><br>
>>> [6]PETSC ERROR: or try <a href="http://valgrind.org" rel="noreferrer" target="_blank">http://valgrind.org</a> on GNU/linux and Apple Mac OS X to find memory corruption errors<br>
>>> [651]PETSC ERROR: #1 User provided function() line 0 in unknown file<br>
>>> [653]PETSC ERROR: ------------------------------------------------------------------------<br>
>>> [653]PETSC ERROR: Caught signal number 15 Terminate: Some process (or the batch system) has told this process to end<br>
>>> [653]PETSC ERROR: Try option -start_in_debugger or -on_error_attach_debugger<br>
>>> [653]PETSC ERROR: or see <a href="http://www.mcs.anl.gov/petsc/documentation/faq.html#valgrind" rel="noreferrer" target="_blank">http://www.mcs.anl.gov/petsc/documentation/faq.html#valgrind</a><br>
>>> [653]PETSC ERROR: or try <a href="http://valgrind.org" rel="noreferrer" target="_blank">http://valgrind.org</a> on GNU/linux and Apple Mac OS X to find memory corruption errors<br>
>>> [653]PETSC ERROR: likely location of problem given in stack below<br>
>>> [653]PETSC ERROR: --------------------- Stack Frames ------------------------------------<br>
>>> [653]PETSC ERROR: Note: The EXACT line numbers in the stack are not available,<br>
>>> [653]PETSC ERROR: INSTEAD the line number of the start of the function<br>
>>> [653]PETSC ERROR: is given.<br>
>>> [653]PETSC ERROR: [653] SuperLU_DIST:pzgssvx line 465 /home/cdfmat_marius/prog/petsc/git/petsc/src/mat/impls/aij/mpi/superlu_dist/superlu_dist.c<br>
>>> [653]PETSC ERROR: [653] MatLUFactorNumeric_SuperLU_DIST line 314 /home/cdfmat_marius/prog/petsc/git/petsc/src/mat/impls/aij/mpi/superlu_dist/superlu_dist.c<br>
>>> [653]PETSC ERROR: [653] MatLUFactorNumeric line 3124 /home/cdfmat_marius/prog/petsc/git/petsc/src/mat/interface/matrix.c<br>
>>> [653]PETSC ERROR: --------------------- Error Message --------------------------------------------------------------<br>
>>> [653]PETSC ERROR: Signal received<br>
>>> [653]PETSC ERROR: See <a href="http://www.mcs.anl.gov/petsc/documentation/faq.html" rel="noreferrer" target="_blank">http://www.mcs.anl.gov/petsc/documentation/faq.html</a> for trouble shooting.<br>
>>> [653]PETSC ERROR: Petsc Development GIT revision: v3.10.3-980-g66b342c GIT Date: 2018-12-26 13:49:21 -0600<br>
>>> [653]PETSC ERROR: /home/cdfmat_marius/prog/transomat_latest_openmpi4.0/transomat on a named h023 by cdfmat_marius Wed Feb 13 23:58:21 2019<br>
>>> <br>
>>> Any idea?<br>
>>> <br>
>>> best,<br>
>>> marius<br>
>> <br>
>> <br>
<br>
</blockquote></div>