Ludwig

Sorry for the late reply, the queue time on the machine I have access to are a nightmare. I was able to test the MPI on CPU and can't reproduce...

It seem the issue was with kokkos, it compiles by now!

I revisited this and it looks like with intel25 and kokkos/4.6.01 even virtual functions are supported. So the only required change to work with SYCL is the change from `printf`...

To clarify the issue with the boundary conditions: In the default setup I can see two waves propagating through the phase diagram from both ends of the shock setup and...

Thanks a lot, @haykh! I did a quick check with the phase space plot I made last time and the weird waves are gone now!  So at least phenomenologically...

That sounds like a good option! I'll wait for your push and then re-run my tests to see if the magnetic field setup works the way it should.

On second thought: would it be possible to add a boundary condition `INIT` that uses `InitFields` in the same way that `ATMOSPHERE` used `DriveFields`? That would avoid duplicate code and...

The fixed BCs look good thanks a lot for implementing this! There was a bug in the Bfield setup, so please disregard the last plot for my previous test. I...

@vanthieg and I just sat down and tested the new BC. There is still the propagating wave and especially the dipole forming for the theta=90° setup is weird (output is...

As discussed on Slack the tests look good! This is the time evolution of the Bz component and temperature of a plasma drifting with gamma=10 (blue is the current entity...