[petsc-dev] DMDA_BOUNDARY_GHOSTED

[Mark F. Adams]

> 
> We have ghost cells in TS ex11, and I think we do them the correct way :) We have a function you register that fills them up, and we
> loop over them explicitly.
>  

Well damn Matt, this looks like the right way to do FV in PETSc, regardless of the nice BC abstraction, but am I going to get be seduced and crash into this island and turn into stone?  i.e., can stupid people deal with your Sections and Plexes and crap (Barry?).

I'm thinking of a much simpler problem (2D, 9-point stencil, constant coefficient Laplacian) and I want to add a convergence study wrapper around a code like this.  So have an outer loop that refines grids, gets errors, and prints convergence rates.  

Perhaps I could morph this example into a simpler example that shows how to do FV in the canonically blessed way in PETSc, abstracted BC functions, and a convergence test for FMG.  This might be a worthy addition.

I suppose I could keep TS and SNES even though I just plan on looking at a linear elliptic solve for now.

Any suggestions?

Mark