On Feb 15, 2013, at 2:42 PM, Roy Stogner <royst...@ices.utexas.edu> wrote:
> > "trump --enable-petsc with --disable-mpi", why? > > PETSc can be built without MPI, and in fact an enable-petsc, > disable-mpi build is the first non-default configuration we're testing > in BuildBot. In case you want to --disable-mpi, of course?! I was just running a Cygwin configuration with $ ../configure --disable-mpi only to find LIBMESH_HAVE_MPI defined, courtesy of petsc.m4's MPI fallback stuff. In this case there is MPI on the system, What if I don't want MPI?? Should I need to remember $ ../configure --disable-mpi --disable-petsc? No one will remember that, they'll do what I did and compile, scratch their heads at the improper MPI symbols, and then (maybe) figure it out. At the very least, we should prevent LIBMESH_HAVE_MPI from getting defined when --disable-mpi is explicitly passed. Or at least prevent the MPI fallback stuff... Would that be acceptable? -Ben ------------------------------------------------------------------------------ The Go Parallel Website, sponsored by Intel - in partnership with Geeknet, is your hub for all things parallel software development, from weekly thought leadership blogs to news, videos, case studies, tutorials, tech docs, whitepapers, evaluation guides, and opinion stories. Check out the most recent posts - join the conversation now. http://goparallel.sourceforge.net/ _______________________________________________ Libmesh-devel mailing list Libmesh-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/libmesh-devel
