Skip to content

MPI communications for the SF split

Loic Hausammann requested to merge sf_split_mpi2 into master

It seems that not much had to be done in order to have the MPI version. I have updated the cell (un)packing functions for the sf_count with the gparts and also realized that I should move a bit the new particles in order to avoid a division by 0 (I am using the same code for the displacement than for the particle splitting).

To test it, I have run the SmallCosmoVolume with 4 ranks.

Merge request reports