The time-step task should ignore inhibited particles when computing the next time-step end.

Does what is says on the tin.

I am 99% sure this is correct but there might be corner cases I have not considered so an extra pair of eyes would be welcomed.

The only thing looks quite different is if we had a case when the cell only contained inhibited particles, when the ti min/max will now end up as zero. That is OK as these will be dealt with at the next rebuild, so I'm happy.

My analysis as well.

So we agree. I'll accept this as well.

merged

mentioned in commit 41561056

That actually looks like it was a bad idea.

Running mpirun -np 8 ../../swift_mpi --pin --cooling --star-formation --feedback --stars --cosmology --hydro --self-gravity -v 1 --threads=14 eagle_25.yml with debugging checks on crashes after 57 steps.

What would be the best strategy to revert such a merge commit?

I'll push the button above. Should be a clean revert.

mentioned in commit 15225fbe

mentioned in merge request !842 (merged)

mentioned in commit da42ff05

mentioned in merge request !843 (closed)