/*
Hi OpenMPI people,
This mail is itself the c code that will reproduce the
bug.
The basic idea is that in a sigle processor debian
machine, only TCP messaging, 1.1.2 (and previous
versions)
abort with this sample program during MPI_Barrier when
run with
three processes (more will fail too but
Hi OpenMPI people,
In the attached file is all the info (as simple as I
could reproduce it) of what I think is a bug.
The basic idea is that in a sigle processor debian
machine, only TCP messaging, 1.1.2 and previous abort
in the sample program during MPI_Barrier when run with
three processes (m
