NetBSD port

[Satish Balay]

On Wed, 16 Dec 2009, Satish Balay wrote:

> There is a SEGV trapped by PETSc error handler. It doesn't know
> exactly where its hapenning. You'll have to run in a debugger to get
> the exact location of this error and the stack trace. [I suspect the
> segv is in OpenMPI code - but only a debugger can confirm/deny it]
> 
> Normally you could use -start_in_debugger with PETSc binary - assuming
> the remote nodes can communicate to your xserver on the desktop
> [directly or via ssh-portforwared-x11] to do this.

A couple more comments:

The SEGV error was with PISM code - however the PETSc example gave
malloc error - which doesn't really make any sense.

does this run consistantly give the same error?

Perhaps you can run the PETSc example with valgrind and see if it
detects any memory corruption. [looks like valgrind does run on
netbsd].

asterix:/home/balay/tmp/petsc-dist-test/src/snes/examples/tutorials>mpiexec -n 4 valgrind --tool=memcheck -q ./ex19 -dmmg_nlevels 4
lid velocity = 0.0016, prandtl # = 1, grashof # = 1
Number of Newton iterations = 2
lid velocity = 0.0016, prandtl # = 1, grashof # = 1
Number of Newton iterations = 2

You can also check if a slightly simpler example also crash on SGE -
i.e src/ksp/ksp/examples/tutorials/ex2.c

Satish