1. 24 Feb, 2016 2 commits
  2. 23 Feb, 2016 5 commits
    • Peter W. Draper's avatar
      Merge branch 'integer_timeline' into 'master' · 62ceefee
      Peter W. Draper authored
      Integer timeline
      
      This modified the way we do multiple time-steps. The time-steps are now placed on an integer timeline and can take values from 0 to 2^29 where 0 corresponds to the starting point of the simulation and 2^29 is the end point of the simulation. This effectively limits the number of time-steps that can be performed and sets an absolute minimal time-step size. If need be we can later on extend the number of time-steps allowed by switching to a long long but I have not seen this being necessary in any EAGLE or other simulations we ran.
      However, it prevents the assignment of 0-sized time-steps that occur in the current version of the code when using a final time which is not of the form 2^n (n positive or negative integer). This hence fixes issue #87.
      Quite a lot of changes scattered across the code but the only non-trivial ones are in runner_dokick() and engine_init().
      
      
      While working on it I have corrected the count of number of particles updated in a given time-step. This is now correct. Fixes issue #89.
      
      Finally, I have also added a message at start-up displaying which version of SPH we run.
      
      
      
      See merge request !94
      62ceefee
    • Matthieu Schaller's avatar
    • Matthieu Schaller's avatar
      Fixed segfault in MPI mode. · dee26e62
      Matthieu Schaller authored
      dee26e62
    • Matthieu Schaller's avatar
      Less verbose engine_init() in MPI mode · 4511d85d
      Matthieu Schaller authored
      4511d85d
    • Matthieu Schaller's avatar
  3. 22 Feb, 2016 2 commits
  4. 21 Feb, 2016 6 commits
  5. 20 Feb, 2016 6 commits
  6. 19 Feb, 2016 6 commits
  7. 18 Feb, 2016 6 commits
  8. 17 Feb, 2016 6 commits
    • Peter W. Draper's avatar
      Merge branch 'smoothing_length_derivative' into 'master' · 3b2b2be6
      Peter W. Draper authored
      Improved multi-timestep SPH
      
      This brings a range of improvement, mostly minor and of code clarity nature to the code. 
      It solves:
      
      - Issue #84. Particles are now <128 Bytes in size.
      - Issue #85. Smoothing-lengths now evolve more accurately and we implement a time-step criterion more conservative than default Gadget-2 but that is part of EAGLE.
      - Issue #88. We now have a 'minimal' hydro implementation that can be used both as a template and an example. The doxygen documentation is based on that version. The documentation is now very extensive. This model is a poor SPH implementation and should only be used to understand the structure of the code and not be seriously run.
      - Issue #95. The engine_collect_kick() function now correctly skips empty cells.
      - Issue #96. Conserved quantities are now written to a file. We need gravity before this really get useful.
      - The description of the SPH flavour in the HDF5 files is now moved to the hydro_io.h files for consistency.
      
      
      Most of it are straight-forward changes. Note that this does not fix the issue we have in MPI mode.
      
      See merge request !90
      3b2b2be6
    • Matthieu Schaller's avatar
      Compliation problem in DEFAULT_SPH · 5c966fdf
      Matthieu Schaller authored
      5c966fdf
    • Peter W. Draper's avatar
      Remove unused variable · 5a3ca6a9
      Peter W. Draper authored
      5a3ca6a9
    • Matthieu Schaller's avatar
      45e04036
    • Peter W. Draper's avatar
      Merge branch 'more_policies' into 'master' · 2b8930c4
      Peter W. Draper authored
      Engine policies for gravity and improved task mask construction
      
      This is a straightforward change to engine_step() and engine_init_particles() that should allow other users to add or remove tasks without having to comment-out portions of the code. 
      
      The right list of policies simply has to be passed to the engine when constructed. I have added three policies (hydro, external gravity and self gravity) to be future proof.
      
      I have also resurrected the two functions that print the content of a mask. These had disappeared following a merge. Finally, some more robust bit-mask checking tests are done when testing policies. 
      
      That should allow @jregan and @tt to add their external gravity tasks to the mask on line 1901 (in engine_step()) of engine.c without having to comment-out other parts of the code.
      
      See merge request !91
      2b8930c4
    • Peter W. Draper's avatar
      fe44d13b
  9. 15 Feb, 2016 1 commit