Hi, GROMACS dev here - I expect that will let that test pass. It just needs enough processes that can resolve MPI calls without deadlocking.
Mark On Thu, Jan 18, 2018 at 9:12 PM Nicholas Breen <nbr...@debian.org> wrote: > On Wed, Jan 17, 2018 at 11:22:27PM -0500, Aaron M. Ucko wrote: > > The only 3.0.x version available for x32 is 3.0.0-1, because newer > > versions build-depend on pmix, which in turn build-depends on pandoc, > > which is unavailable there. (I think the Haskell stack generally is.) > > With that version installed in my laptop's x32 chroot and everything > > else left at the versions available in unstable, I get *different* > > errors per the attached log, excerpted below. (Using unstable across > > the board yields the same errors as on the autobuilder.) > > Technically an improvement! The "not enough slots" error is a known > OpenMPI > bug-or-feature, depending on who you ask. As a workaround, could you > please > add this near the top of debian/rules? > > export OMPI_MCA_rmaps_base_oversubscribe=1 > > I'm not sure if this will let MdrunUtilityMpiUnitTests pass, but it should > at > least get further and to a more informative error. The "debian/rules > build-openmpi" target skips the very lengthy compilation of everything > else. > > A more powerful system shouldn't make any difference. > >
