On Jul 16, 2012, at 6:47 PM, Jed Brown wrote:
> On Mon, Jul 16, 2012 at 5:44 PM, Barry Smith <bsmith at mcs.anl.gov> wrote:
> I totally agree it is a hack and should never be used, BUT as far as I was
> aware in no place in PETSc did we do that? If that is it should be fixed. But
> the fix is likely not forcing mpiuni to have two communicators it is just to
> do things right.
>
> Do you not consider http://petsc.cs.iit.edu/petsc/petsc-dev/rev/169c19e9ea54
> to be a hack?
Jed,
Yes, that is absolutely a hack and does not belong there. But you are
totally miss understanding what I am saying: that hack is NEW. For 15 years
PETSc did NOT need a hack to work with MPIUNI (which has a single
communicator), thus I conclude that MPUNI is fine and something is wrong with
the PETSc thread comm stuff if it requires that hack. That is, why the heck
does petscthreadcomm depend on MPI_COMM_SELF != MPI_COMM_WORLD while for 20
years NOTHING ELSE IN PETSc (which is a dang lot more complicated than
petscthreadcomm) does not depend on MPI_COMM_SELF != MPI_COMM_WORLD???
In other words fix petscthreadcomm model; don't mess with a perfectly good
mpiuni.
Barry
