To enable the debug mode you should configure Open MPI with --enable-debug. After compilation just run the test with gdb, and export a backtrace (gdb command bt or where).
George. On Oct 14, 2013, at 11:44 , Syed Ahsan Ali <ahsansha...@gmail.com> wrote: > George Can you please tell me how to run test in debug mode? > > On Fri, Oct 4, 2013 at 3:42 AM, George Bosilca <bosi...@icl.utk.edu> wrote: >> The output seems normal, there is no indication of anything weird happening. >> Also I tried to reproduce the error with other icc versions to no avail. >> >> Syed, >> >> Can you run the test in debug mode inside gdb? When the segfault happen >> please get the stack trace (gdb command where or bt) and send it to me. >> >> Thanks, >> George. >> >> >> On Oct 3, 2013, at 13:06 , Jeff Squyres (jsquyres) <jsquy...@cisco.com> >> wrote: >> >>> George -- >>> >>> See this thread on the users list: >>> >>> Thread starts here: >>> http://www.open-mpi.org/community/lists/users/2013/09/22682.php >>> Relevant messages start here: >>> http://www.open-mpi.org/community/lists/users/2013/09/22725.php >>> >>> Attached is a zip file from the user experiencing a failure in "make check" >>> in opal_datatype_test. I think he may have run "make -j X check" because >>> the output is a little jumbled. >>> >>> I'm unable to replicate the error on RHEL6/icc 14.0. >>> >>> Have you see this kind of failure before in opal_datatype_test / have any >>> idea how the error indicated in the output can happen? >>> >>> -- >>> Jeff Squyres >>> jsquy...@cisco.com >>> For corporate legal information go to: >>> http://www.cisco.com/web/about/doing_business/legal/cri/ >>> <testlog.zip>_______________________________________________ >>> devel mailing list >>> de...@open-mpi.org >>> http://www.open-mpi.org/mailman/listinfo.cgi/devel >> > > > > -- > Syed Ahsan Ali Bokhari > Electronic Engineer (EE) > > Research & Development Division > Pakistan Meteorological Department H-8/4, Islamabad. > Phone # off +92518358714 > Cell # +923155145014
