Julian Andrej

Try this. You're using the list for submeshes incorrectly. ``` Array fluid_attr(1); fluid_attr[0] = 1; Array solid_attr(1); solid_attr[0] = 2; ```

Now it looks like a Submesh bug to me. I'm trying to figure out how to solve that next week.

@ryanblack129 The combination of https://github.com/mfem/mfem/pull/4113 and https://github.com/mfem/mfem/pull/4151 will fix the issues mentioned here.

I don't have a preference

> When I submit PRs quite often I get relatively obscure errors due to the strict compilation flags, such as `-Wall -Wpedantic -Wshadow`. It is not really possible for me...

> @jandrej, can you elaborate on what the Newton mention is clarifying? Perhaps I am missing something, but the addition reads to me that `ImplicitSolve` will always implement a Newton...

As this is apparently more confusing than the previous docs, I'll close the PR.

go ahead

You have to set `KINSetNumMaxIters` to something larger as well (the default is 200, is that value kept?). You probably also want to allow KINSol to rebuild the Jacobian for...

Are you using the "performance" examples or is this the regular `examples/ex1p`?