On Tue, Jan 22, 2013 at 4:48 PM, Jed Brown <jedbrown at mcs.anl.gov> wrote:
> > On Tue, Jan 22, 2013 at 4:39 PM, Karl Rupp <rupp at mcs.anl.gov> wrote: > >> Carefully fiddling with my local sources prior to the commit... I don't >> know of a way to test the backwards-macros other than inspecting the >> preprocessed code... >> >> Oh, and Barry's macro proposal was right (I forgot that this assumed an >> update of sources already) >> > > Use your Open MPI build and put > > #define OMPI_WANT_MPI_INTERFACE_WARNING > > before including mpi.h. They put in __attribute((deprecated)) so you'll > get warnings for all the old usage. Test that the new version still works > with MPI-1 by building with --with-mpi=0 (or download an illicit copy of > MPICH1). > This is pushed. I checked with MPIUNI and MPICH and it worked. I can't bring myself to build with OpenMPI. Matt -- What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead. -- Norbert Wiener -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20130124/88c161b7/attachment.html>
