Open FlorianDeconinck opened 7 months ago
Note: MPICH 4.1 should be good with MAPL (and thus GEOS) now. At least in some testing.
But ftn
is its own beast and Cray MPI was weird in the old day...maybe it's better now?
That's my worry. Any Cray backed library comes with a risk of weird undocumented changes...
And since it's been a while we have look at this, we should start from scratch.
Presently CRAY compilers & MPICH cannot compile GEOS (BaseLibs and GEOS model).
SI team can spot us on this work but we need to drive it forward.
x-team: [SI] Matt T, Tom C. Parent: https://github.com/GEOS-ESM/SMT-Nebulae/issues/30