On Tue, Aug 2, 2011 at 5:38 PM, Kuhlemann, Verena <span dir="ltr"><<a href="mailto:vkuhlem@emory.edu">vkuhlem@emory.edu</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;">
So, the first real error that I get is<br>
<br>
Invalid write of size 4<br>
==24909== at 0x9B3B9F: Moc_ComputePartitionParams__ (kwayrefine.c:215)<br>
==24909== by 0x9A1D28: Moc_Global_Partition__ (kmetis.c:231)<br>
==24909== by 0x9A1D00: Moc_Global_Partition__ (kmetis.c:228)<br>
==24909== by 0x9A1D00: Moc_Global_Partition__ (kmetis.c:228)<br>
==24909== by 0x9A1D00: Moc_Global_Partition__ (kmetis.c:228)<br>
==24909== by 0x9A1D00: Moc_Global_Partition__ (kmetis.c:228)<br>
==24909== by 0x9A1619: ParMETIS_V3_PartKway (kmetis.c:137)<br>
==24909== by 0x557AC8: MatPartitioningApply_Parmetis (pmetis.c:97)<br>
==24909== by 0x554058: MatPartitioningApply (partition.c:236)<br>
==24909== by 0x404DE2: main (runtests.c:99)<br>
<br>
Any ideas why?<br></blockquote><div><br></div><div>ParMetis is not robust to problems with the input (we are upgrading to the recent release</div><div>which is supposed to do more checking). I believe it can fail if you have empty partitions.</div>
<div>I would check the input.</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;">
I will try to use --download-mpich for valgrind debugging./<br>
<br>
Thanks,<br>
<br>
Verena<br>
________________________________________<br>
From: <a href="mailto:petsc-users-bounces@mcs.anl.gov">petsc-users-bounces@mcs.anl.gov</a> [<a href="mailto:petsc-users-bounces@mcs.anl.gov">petsc-users-bounces@mcs.anl.gov</a>] on behalf of Satish Balay [<a href="mailto:balay@mcs.anl.gov">balay@mcs.anl.gov</a>]<br>
Sent: Tuesday, August 02, 2011 1:35 PM<br>
To: PETSc users list<br>
Subject: Re: [petsc-users] valgrind<br>
<div><div></div><div class="h5"><br>
Also suggest using --download-mpich - for valgrind debugging. [it gives<br>
you a valgrind clean mpich]<br>
<br>
<br>
Satish<br>
<br>
On Tue, 2 Aug 2011, Matthew Knepley wrote:<br>
<br>
> On Tue, Aug 2, 2011 at 5:06 PM, Kuhlemann, Verena <<a href="mailto:vkuhlem@emory.edu">vkuhlem@emory.edu</a>> wrote:<br>
><br>
> > Hi,<br>
> ><br>
> > I am trying to find an error in my program with valgrind.<br>
> > The first message that I get is the following:<br>
> ><br>
><br>
> No, MPICH is doing something that valgrind does not understand with memory.<br>
> You can usually<br>
> ignore anything that goes back into MPI.<br>
><br>
> Matt<br>
><br>
><br>
> > Syscall param write(buf) points to uninitialised byte(s)<br>
> > ==22395== Syscall param write(buf) points to uninitialised byte(s)<br>
> > ==22393== at 0x73D92C0: __write_nocancel (in /lib64/<a href="http://libc-2.5.so" target="_blank">libc-2.5.so</a>)<br>
> > ==22393== by 0x7A76283: ibv_cmd_modify_qp (in<br>
> > /usr/lib64/libibverbs.so.1.0.0)<br>
> > ==22393== by 0x88B742A: ??? (in /usr/lib64/libmlx4-rdmav2.so)<br>
> > ==22393== by 0x7A7A363: ibv_modify_qp (in<br>
> > /usr/lib64/libibverbs.so.1.0.0)<br>
> > ==22393== by 0x67A3DDF: MPID_VIA_Init (viainit.c:1014)<br>
> > ==22393== by 0x67826BB: MPID_Init (mpid_init.c:66)<br>
> > ==22393== by 0x677438E: MPIR_Init (initutil.c:279)<br>
> > ==22393== by 0x432732: PetscInitialize (pinit.c:561)<br>
> > ==22395== at 0x73D92C0: __write_nocancel (in /lib64/<a href="http://libc-2.5.so" target="_blank">libc-2.5.so</a>)<br>
> > ==22395== by 0x7A76283: ibv_cmd_modify_qp (in<br>
> > /usr/lib64/libibverbs.so.1.0.0)<br>
> > ==22395== by 0x88B742A: ??? (in /usr/lib64/libmlx4-rdmav2.so)<br>
> > ==22395== by 0x7A7A363: ibv_modify_qp (in<br>
> > /usr/lib64/libibverbs.so.1.0.0)<br>
> > ==22395== by 0x67A3DDF: MPID_VIA_Init (viainit.c:1014)<br>
> > ==22395== by 0x67826BB: MPID_Init (mpid_init.c:66)<br>
> > ==22395== by 0x677438E: MPIR_Init (initutil.c:279)<br>
> > ==22395== by 0x432732: PetscInitialize (pinit.c:561)<br>
> > ==22395== by 0x4046FC: main (runtests.c:48)<br>
> > ==22395== Address 0x7feffef78 is on thread 1's stack<br>
> ><br>
> > I am not sure what this is suppose to tell me. Is the something wrong<br>
> > with my PetscInitialize?<br>
> ><br>
> > Thanks for the help,<br>
> ><br>
> > Verena<br>
> ><br>
> ><br>
> > ------------------------------<br>
> ><br>
> > This e-mail message (including any attachments) is for the sole use of<br>
> > the intended recipient(s) and may contain confidential and privileged<br>
> > information. If the reader of this message is not the intended<br>
> > recipient, you are hereby notified that any dissemination, distribution<br>
> > or copying of this message (including any attachments) is strictly<br>
> > prohibited.<br>
> ><br>
> > If you have received this message in error, please contact<br>
> > the sender by reply e-mail message and destroy all copies of the<br>
> > original message (including attachments).<br>
> ><br>
><br>
><br>
><br>
><br>
<br>
</div></div></blockquote></div><br><br clear="all"><br>-- <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>