OK. We probably need to move our PETSc pointer. From: Matthew Knepley <knepley@gmail.com> Date: Thursday, 7 March 2019 at 10:47 To: "Ham, David A" <david.ham@imperial.ac.uk> Cc: Patrick Farrell <patrick.farrell@maths.ox.ac.uk>, firedrake <firedrake@imperial.ac.uk> Subject: Re: [firedrake] firedrake install not working We fixed that. I think its in 3.10.4, if not we release at the end of the month. Matt On Thu, Mar 7, 2019 at 4:08 AM Ham, David A <david.ham@imperial.ac.uk<mailto:david.ham@imperial.ac.uk>> wrote: Note that the HDF5 in PETSc currently also has OpenMPI 4.0.0 issues. See: https://github.com/firedrakeproject/firedrake/issues/1380 The workaround is probably to use MPICH (which is better for other reasons too). On 07/03/2019, 08:44, "firedrake-bounces@imperial.ac.uk<mailto:firedrake-bounces@imperial.ac.uk> on behalf of Patrick Farrell" <firedrake-bounces@imperial.ac.uk<mailto:firedrake-bounces@imperial.ac.uk> on behalf of patrick.farrell@maths.ox.ac.uk<mailto:patrick.farrell@maths.ox.ac.uk>> wrote: On 06/03/2019 23:06, Lawrence Mitchell wrote: > It looks like libsupermesh (a new dependency) has some tests which use the c++ api to MPI. But that is deprecated for a long time and seemingly removed by openmpi 4 which you have? This should be fixed in libsupermesh PR #2: https://bitbucket.org/libsupermesh/libsupermesh/pull-requests/2/remove-depre... Patrick _______________________________________________ firedrake mailing list firedrake@imperial.ac.uk<mailto:firedrake@imperial.ac.uk> https://mailman.ic.ac.uk/mailman/listinfo/firedrake _______________________________________________ firedrake mailing list firedrake@imperial.ac.uk<mailto:firedrake@imperial.ac.uk> https://mailman.ic.ac.uk/mailman/listinfo/firedrake -- 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 https://www.cse.buffalo.edu/~knepley/<http://www.cse.buffalo.edu/~knepley/>