<div>Hi all,</div>
<div> </div>
<div> I am preparing to do a fresh build and install with --enable-debug and run the same example (reported below) with </div>
<div>debugger (ugh!)... I also found the link below pointing to the similar issue as the one I reported, opened 3 weeks ago? </div>
<div>Any thoughts on the status or otherwise are appreciated as always...</div>
<div> </div>
<div><a href="http://trac.mcs.anl.gov/projects/mpich2/ticket/1147">http://trac.mcs.anl.gov/projects/mpich2/ticket/1147</a></div>
<div><br>--Sunil.</div>
<div><br> </div>
<div class="gmail_quote">On Fri, Jan 7, 2011 at 6:24 PM, Sunil Thomas <span dir="ltr"><<a href="mailto:sgthomas27@gmail.com">sgthomas27@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">
<div>Hi,</div>
<div> </div>
<div> I noticed that a simple test example code (just one C file) that used to run in parallel with an older installation of MPICH2 </div>
<div>(1.2.7 if I recall correctly) no longer runs with the latest stable release version (1.3.1). The following is the error I am getting:</div>
<div> </div>
<div><em><font color="#ff0000">$>mpiexec -f machinefile -n 2 ex5_hypre<br>APPLICATION TERMINATED WITH THE EXIT STRING: Hangup (signal 1)</font></em></div>
<div> </div>
<div> When I switch to my actual application (which basically solves a sparse system of linear equations) and try to run that </div>
<div>on 2 processors, it crashes due to a memory access fault error as follows:</div>
<div> </div>
<div><em><font color="#ff0000">unknown location(0): fatal error in ....: memory access violation at address: 0xfffffffe03bce414: no mapping at fault address</font></em></div>
<div> </div>
<div> It seems like the two are related. Both run just fine on 1 processor. Has anyone seen these issues and know a fix? Many </div>
<div>thanks!</div>
<div> </div><font color="#888888">
<div> </div>
<div>--Sunil.</div></font></blockquote></div><br>