<div dir="ltr">I can reproduce the error on a linux machine with petsc-maint. It crashes at 2nd solve, on both processors:<div><br><div><div>Program received signal SIGSEGV, Segmentation fault.</div><div>0x00007f051dc835bd in pdgsequ (A=0x1563910, r=0x176dfe0, c=0x178f7f0, </div><div> rowcnd=0x7fffcb8dab30, colcnd=0x7fffcb8dab38, amax=0x7fffcb8dab40, </div><div> info=0x7fffcb8dab4c, grid=0x1563858)</div><div> at /sandbox/hzhang/petsc/arch-linux-gcc-gfortran/externalpackages/git.superlu_dist/SRC/pdgsequ.c:182</div><div>182 c[jcol] = SUPERLU_MAX( c[jcol], fabs(Aval[j]) * r[irow] );</div></div></div><div><br></div><div>The version of superlu_dist:</div><div><div>commit 0b5369f304507f1c7904a913f4c0c86777a60639</div><div>Author: Xiaoye Li <<a href="mailto:xsli@lbl.gov">xsli@lbl.gov</a>></div><div>Date: Thu May 26 11:33:19 2016 -0700</div><div><br></div><div> rename 'struct pair' to 'struct superlu_pair'.</div></div><div><br></div><div>Hong</div></div><div class="gmail_extra"><br><div class="gmail_quote">On Fri, Oct 21, 2016 at 5:36 AM, Anton Popov <span dir="ltr"><<a href="mailto:popov@uni-mainz.de" target="_blank">popov@uni-mainz.de</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div bgcolor="#FFFFFF" text="#000000"><span class="">
<br>
<div class="m_4838639753056563240moz-cite-prefix">On 10/19/2016 05:22 PM, Anton Popov
wrote:<br>
</div>
<blockquote type="cite">
<div class="gmail_default" style="font-family:comic sans ms,sans-serif;font-size:small">I looked at each
valgrind-complained item in your email dated Oct. 11. Those
reports are really superficial; I don't see anything wrong with
those lines (mostly uninitialized variables) singled out. I did
a few tests with the latest version in github, all went fine. </div>
<div class="gmail_default" style="font-family:comic sans ms,sans-serif;font-size:small"><br>
</div>
<div class="gmail_default" style="font-family:comic sans ms,sans-serif;font-size:small">Perhaps you can print your matrix
that caused problem, I can run it using your matrix.</div>
<div class="gmail_default" style="font-family:comic sans ms,sans-serif;font-size:small"><br>
</div>
<div class="gmail_default" style="font-family:comic sans ms,sans-serif;font-size:small">Sherry</div>
</blockquote></span>
Hi Sherry,<br>
<br>
I finally figured out a minimalistic setup (attached) that
reproduces the problem.<br>
<br>
I use petsc-maint:<span class=""><br>
<br>
git clone -b maint <a class="m_4838639753056563240moz-txt-link-freetext" href="https://bitbucket.org/petsc/petsc.git" target="_blank">https://bitbucket.org/petsc/<wbr>petsc.git</a><br>
<br></span>
and configure it in the debug mode without optimization using the
options:<br>
<br>
--download-superlu_dist=1 \<br>
--download-superlu_dist-<wbr>commit=origin/maint \<br>
<br>
Compile the test, assuming PETSC_DIR points to the described petsc
installation:<br>
<br>
make ex16<br>
<br>
Run with:<br>
<br>
mpirun -n 2 ./ex16 -f binaryoutput -pc_type lu
-pc_factor_mat_solver_package superlu_dist<br>
<br>
Matrix partitioning between the processors will be completely the
same as in our code (hard-coded).<br>
<br>
I factorize the same matrix twice with the same PC object.
Remarkably it runs fine for the first time, but fails for the
second.<br>
<br>
Thank you very much for looking into this problem.<br>
<br>
Cheers,<br>
Anton<br>
</div>
</blockquote></div><br></div>