On Tue, 20 Aug 2013, John Peterson wrote: > It's not clear to me why running this example serially would > segfault while running it with mpirun would not.
I've seen MPI stacks which actually *required* the mpirun/mpiexec environment, to the point that "mpirun -np 1" was the only way to execute an MPI-compiled program on a single processor. I think even that case was nice enough to give an error message rather than a segfault, though. You're right that the latter is suspicious. --- Roy ------------------------------------------------------------------------------ Introducing Performance Central, a new site from SourceForge and AppDynamics. Performance Central is your source for news, insights, analysis and resources for efficient Application Performance Management. Visit us today! http://pubads.g.doubleclick.net/gampad/clk?id=48897511&iu=/4140/ostg.clktrk _______________________________________________ Libmesh-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/libmesh-users
