Marc Fehling

I moved the `exchange_refinement_flags()` into an internal namespace, so we can also use it in `choose_p_over_h()`. I just moved it in the `include/distributed/tria.h` file. Let me know if you would...

> My understand is that p4est ensures that sibling cells are owned by the same process if they are _active_. In addition to the assertion that triggered in `choose_p_over_h`, I...

I would still like to come up with a minimal test that triggered the above assertion before this patch. With your comments, I have a good starting point to create...

This is the [freeplane](https://docs.freeplane.org/) mindmap from today's discussion (slightly adjusted): [SolutionTransfer.mm.zip](https://github.com/dealii/dealii/files/11931266/SolutionTransfer.mm.zip)

I like the idea to have one class that deals with all the ways to attach data for transfer/serialization. It is also easily extendible for users, especially when providing lambda...

I can reproduce the problem using `deal.ii` release 9.5.1 and the `aspect` main branch. The velocity field looks as follows for the uniformly and adaptively refined meshes. The velocity at...

Thank you for the quick response. I tried the following combinations: - `develop` branch, `Trilinos_ENABLE_COMPLEX_DOUBLE=ON`, `Trilinos_ENABLE_COMPLEX_FLOAT=ON` ``` KokkosKernels_INST_COMPLEX_DOUBLE ON KokkosKernels_INST_COMPLEX_FLOAT OFF KokkosKernels_INST_DOUBLE ON KokkosKernels_INST_FLOAT OFF Tpetra_INST_COMPLEX_DOUBLE ON Tpetra_INST_COMPLEX_FLOAT OFF Tpetra_INST_DOUBLE...

With `Trilinos_ENABLE_COMPLEX_DOUBLE=ON` and `Trilinos_ENABLE_COMPLEX_FLOAT=ON` I now get a warning: ``` CMake Warning at packages/tpetra/CMakeLists.txt:964 (MESSAGE): Trilinos_ENABLE_COMPLEX_FLOAT=ON, but Tpetra_INST_FLOAT=OFF. Disabling Scalar=std::complex in Tpetra by default. ``` This is helpful! Thank you!...

This should not influence the results of non-hp applications. We only had to bump the serialization version as we do not have a way to distinguish between hp and non-hp...

Until today, we implemented the new `types::fe_index` datatype in `DoFHandler` and all related accessor classes, or in other words, in all data structures within the `dofs/` folder. The question remains...