Francesc Verdugo

Hi @JordiManyer, psparse! adds extra ghost ids if needed, it does not remove unnecessary ones. Even by modifying find_owner you will not fix the issue of having extra ghosts, unnecessary...

@JordiManyer even better: we can implement a new function `intersection_ghost` which is the counterpart of `union_ghost` (i.e. same function signature) and then add a new kw-arg in psparse! `combine_ghost=union_ghost` which...

See PR: https://github.com/fverdugo/PartitionedArrays.jl/pull/120

Hi @amartinhuertas ! I have been looking at the implementation of `PETScNonLinearSolver` and I believe that it is not completely inline with the spirit of the `NonlinearSolver` interface of Gridap....

Another minor fix. Add `,::Nothing` as last argument in https://github.com/gridap/GridapPETSc.jl/blob/3ba0de4ecd9c04ed1203f972b2c0409981aa7b5e/src/PETScNonlinearSolvers.jl#L179 The nonlinear solvers can be used either as ```julia cache = solve!(x,op) cache = solve!(x,op,cache) ``` or ```julia cache =...

We started to investigate the extension to quads and hex with @Kevin-Mattheus-Moerman preliminar code in branch: https://github.com/gridap/GridapMakie.jl/tree/quads_and_hexs For the record, I paste here the test we used: ```julia using Gridap...

@oriolcg the code below works. Two remarks: - Use interpolations in the reference space. We have encountered some other issues in the past when combining AgFEM + interpolations in the...

WIP in branch https://github.com/fverdugo/PartitionedArrays.jl/tree/benchmarks

Hi Carlos, I am refactoring the code with the main objective of improving the documentation so that others can contribute easier. See the work going on in branch https://github.com/fverdugo/PartitionedArrays.jl/tree/remove_abstractpdata (the...

Yes, I should have mentioned this, while closing the issue. It was in PR #134 as part of the implementation of smoothed aggregation AMG solvers.