On Tue, Jul 28, 2009 at 10:17 AM, Randall Mackie <span dir="ltr"><<a href="mailto:rlmackie862@gmail.com">rlmackie862@gmail.com</a>></span> wrote:<br><div class="gmail_quote"><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
I have run into a very difficult debugging problem. I have recently made some<br>
modifications to my PETSc code, to add some new features. When I compiled the<br>
code in debug mode (we are using the Intel compilers and mvapich on Infiniband),<br>
the code runs fine with any number of processes.<br>
<br>
When the code is compiled in optimize mode, it runs fine on, say, up to 32 processes,<br>
but not 64, bombing out someplace strange, with a Segmentation Violation.<br>
<br>
I've tried using Valgrind, but you can't use it with PETSc and my code compiled in<br>
Debug mode because the code finishes successfully, and the other problem I have with</blockquote><div><br>Sometimes valgrind will catch things even when code does not crash.<br> </div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<br>
Valgrind + mvapich is there are about a million messages spewed out, making it<br>
extremely difficult to see if there are really any issues in MY code. I've thought<br>
to have PETSc download and compile MPICH2, which I would hope would produce less<br>
output from Valgrind.</blockquote><div><br>In order to filter these out, you use a "suppressions file" for valgrind. The manual has a<br>good section on this and it should not be hard to wipre out most of them. Satish designed<br>
one for our unit tests.<br><br> Matt<br> </div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><br>
Anyone have any suggestions on how to debug this tricky situation? Any suggestions<br>
would be greatly appreciated.<br>
<br>
Randy<br>
</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>