Gravity particles

Ok, that's another big one... This includes all the changes by @tt and @jregan regarding the use of an external potential.

The main changes are:

• Update to the unit with some more const statements.
• Addition of a set of physical constants and of an object to convert them in the chosen system of units.
• A new external gravity task and its dependencies.
• This task calls functions in a new file potential.h where users can define their external potentials.
• A new test case that just makes a bunch of particles orbit a point mass.

Can I suggest to keep this on the wiki ? https://gitlab.cosma.dur.ac.uk/swift/swiftsim/wikis/home

Added 1 commit:

• 2fc60ea7 - Pedro suggestions

Alright, I have applied these suggestions. Thanks !

Added 1 commit:

• c0aa3063 - Make the new example ready for the restored '-t' command-line parameter

I have also updated the new example in here such that everything works once !144 (merged) is merged into master.

Added 4 commits:

• c0aa3063...622ad2b6 - 3 commits from branch master
• 354bc99a - Merge branch 'master' into GravityParticles

When I run in MPI mode I get an error:

[00002.3] common_io.c:prepareXMFfile():363: Unable to open temporary file.

after large number of steps, but that seems random.

Ok, thanks. Was this with parallel or serial hdf5 ?

Serial.

The output when I run with -s and -g looks odd:

#   Step           Time      Time-step    Updates  g-Updates  Wall-clock time [ms]
       0   0.000000e+00   0.000000e+00          0          0                 0.000
       1   0.000000e+00   0.000000e+00    1024128    1024128              4554.576
       2   2.441406e-04   2.441406e-04          0          0               953.502
       3   2.441406e-04   0.000000e+00     821962     821962              3624.555
       4   4.882812e-04   2.441406e-04          0          0               373.328
       5   4.882812e-04   0.000000e+00    1024128    1024128              4505.958
       6   7.324219e-04   2.441406e-04          0          0               845.805
       7   7.324219e-04   0.000000e+00     825626     825626              3725.634
       8   9.765625e-04   2.441406e-04          0          0               375.137
       9   9.765625e-04   0.000000e+00    1024102    1024102              4669.597

what are these steps with a zero time step and why are particles apparently updated every other step only (this is the SodShock with a point massed added).

Sorry, don't worry about the output file problem, I think I was running the non-MPI swift in MPI mode. The real MPI version has much bigger problems,

Added 2 commits:

• 55720e1d - Deal with empty spaces.
• f3c535b4 - Get rid of the parts if we don't have any

Yes, I was trying to reproduce your crash but couldn't. On the way I fixed the other MPI issues though. All related to space having no (g)parts. That should be fixed now.

Let me now have a look at the Sod shock with -g.

Added 1 commit:

• 1d1e86c2 - Misplaced if-statement leading to 0-sized time-step for gparts with counterparts

Added 1 commit:

• c5a5c663 - Code formatting

Ok, fixed it. There was a misplaced if-statement in dokick that would lead to a mnimal time-step size of 0 for the gpart counterparts.

Thanks that has fixed these issues.