Arturo Vargas
Arturo Vargas
I ran the white-space remover in my editor and made more changes than expected. To simplify review I suggest hiding white-space changes.
> @artv3 did you force this merge b/c the CI was timing out? > > If you no longer need the branch, please delete it. Hi @rhornung67 , I used...
Might be still good to check for memory leaks.
> I did a little testing with Laghos. > > I changed the line: L2_FECollection L2FEC(order_e, dim/_, BasisType::Positive_/); > > to use the positive basis, as we do in non-AMR...
@rw-anderson when we do the coarsening would it still preserve bounds? In regards to the Remhos example.
> > @rw-anderson when we do the coarsening would it still preserve bounds? In regards to the Remhos example. > > No, simple L2 projection will definitely not be bounds...
~~Can we add method to return the element mass matrices (for the version of the solvers that assemble the matrices)? As well as re-assemble in case the mesh has moved?~~...
I suspect one of the issues is here: https://github.com/mfem/mfem/blob/11b3e0b4b5d21544c5c03f8d6f8f72d703676134/fem/bilininteg_hcurl.cpp#L1719-L1737 If we take a look at ope[9], we see that on the CPU the array gets overwritten with each `MFEM_FOR_EACH_THREAD` loop;...
Yes!
> @artv3, this is ready for you to take a look at and see if it would work for your purposes. Howdy! Just got back from my break. Will take...