add checks for valid tags when adding send/recv tasks, and used named constants for computing the MPI tag from the cell tag.

These exceed the range of int when processing EAGLE_50 on a single node with the new cell splitting value

Matthieu Schaller authored May 23, 2018 and Matthieu Schaller committed Jun 18, 2018

Extract the mesh gravity forces to a separate set of functions. Get the forces directly from the mesh to the gparts. Do not use the tree here.

Correctly use the global update counters in the cases where there are more than 'one int' worth of updates across all nodes. The limit per-node remains.

Add a new parameter to the gravity calculation to control the frequency of the tree rebuild using a fraction of the number of gpart updates.

Use the values gathered to compute the time-step based on the RMS peculiar velocity in cosmological runs.

Improve the way we define the time of the first snapshot and statistics in cosmology and non-cosmology modes.

Add the strict-prototype flag and correct all the prototypes as this creates issues with the Cray compiler.

Fix issue with attempt to dump partition when not using MPI

Can also change the number of queues to match the threads (expected behaviour), the number of tasks per cell and the size of MPI messages, these all could be useful when re-starting

Sugar of various kinds, new parameters to control restart file dumping, better adherence to the restart_ name space, dumps on exit for testing restarts, include in step properties

Make restart files dump after a number of hours from the start and synchronize dumps across MPI ranks

Moved restart_write into engine_step and trigger after drift_all

Tidies up the restart option

Add the possibility to read basic chemistry information from the parameter file and propagate this onto the particle at initialisation time. More uniform naming convention for the first initialisation of properties.