On Sat, Nov 5, 2011 at 5:20 AM, Brad Aagaard <span dir="ltr"><<a href="mailto:baagaard@usgs.gov">baagaard@usgs.gov</a>></span> wrote:<br><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
Jed,<br>
<br>
I assume you mean src/ksp/ksp/examples/<u></u>tutorials/ex2<br>
I have not yet been able to duplicate the error with this example. Do you know of any other test cases that might be appropriate?<br>
<br>
My PyLith simulation with -malloc and -malloc_debug flags chokes on line 933 of src/mat/impls/baij/mpi/<u></u>baijjob.c with a EXEC_BAD_ACCESS.<div class="im"><br>
<br>
933 for (l=0; l<nzA; l++) cols[imark+l] = cstart + cworkA[l];<br>
<br></div>
When the memory error happens at this line, the debugger shows<br>
l = 1400100<br>
imark = 0<br>
cstart = 0<br>
cworkA[l] = 0<br>
cols[imark+l] = -17958193<br>
<br>
The address of the bad memory access matches the address of cols[imark+l].<br></blockquote><div><br></div><div>Brad, can you check C->rmap->bs when this happens? Since this is the only thing that changes, I am guessing that</div>
<div>it might be 0, and then you can get the wrong 'proc' which cascades. Can I run this here?</div><div><br></div><div> Thanks,</div><div><br></div><div> Matt</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
Brad<div class="im"><br>
<br>
<br>
On 11/4/11 8:39 PM, Jed Brown wrote:<br>
</div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="im">
On Fri, Nov 4, 2011 at 21:30, Brad Aagaard <<a href="mailto:baagaard@usgs.gov" target="_blank">baagaard@usgs.gov</a><br></div><div class="im">
<mailto:<a href="mailto:baagaard@usgs.gov" target="_blank">baagaard@usgs.gov</a>>> wrote:<br>
<br>
Barry,<br>
<br>
Sorry for the hg goof. Here is the changeset that I think is related<br>
to the problem.<br>
<br></div>
<a href="http://petsc.cs.iit.edu/petsc/__petsc-dev/rev/7782710e3f96" target="_blank">http://petsc.cs.iit.edu/petsc/<u></u>__petsc-dev/rev/7782710e3f96</a><div class="im"><br>
<<a href="http://petsc.cs.iit.edu/petsc/petsc-dev/rev/7782710e3f96" target="_blank">http://petsc.cs.iit.edu/<u></u>petsc/petsc-dev/rev/<u></u>7782710e3f96</a>><br>
<br>
<br>
If bs=1, then this doesn't change anything. As far as I know, it is not<br>
possible to create an SBAIJ(1) with map->bs != 1. Can you find a way for<br>
me to reproduce this? I see no problem with this, for example:<br>
<br>
mpiexec -n 4 ./ex2 -mat_type sbaij -pc_type asm<br>
</div></blockquote>
<br>
</blockquote></div><br><br clear="all"><div><br></div>-- <br>What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.<br>
-- Norbert Wiener<br>