roystgnr
roystgnr
The solution vector is of type PARALLEL, and can only return components owned by the processor you're querying it from. Even current_local_solution is of type GHOSTED and will only return...
I forgot about that bit of cleverness in `index_range`; please ignore my suggested fix. John's almost certainly right about the real problem; the way we order DoFs in parallel depends...
The HERMITE elements are C1, but are limited to basically cartesian meshes (xi/eta/zeta need to be parallel to x/y/z, though you can stretch in any of those directions and/or do...
Actually, this ought to be mostly fixed by https://github.com/libMesh/libmesh/pull/2677 - the only remaining cases where we'd need serialization are where we're not just redistributing but we're on a mesh where...
I'd be quite happy with changing the default to ```TOLERANCE * TOLERANCE```. I don't recall putting much thought into the initial decision. And the main effect of a too-tight tolerance...
It looks as if somehow you had metaphysicl headers in include/*/ somewhere before running include/libmesh/rebuild_makefile.sh - could that possibly be the case? I just tried running that myself but it...
1.3.1 is here: https://github.com/libMesh/MetaPhysicL/tags - we moved the primary repository location a while back. I'll make a note of that at https://github.com/roystgnr/MetaPhysicL You're quite right that we should be supporting...
"Work in progress" means I can merge #2395 and undercut all your hard work without feeling *too* guilty?
Ouch, I barely changed browser tabs before github started yelling at us here about conflicting files. Sorry about the hassle. On the bright side, the expanded unit test coverage should...
@jwpeterson pointed out to me that our _impl.h convention (where most users *don't* need to include them because we instantiate the common case(s) manually) isn't intuitively obvious. Let's come up...