Commits · cf17832d56159f9c4c80df642b00f9e4cc415744 · SWIFT / SWIFTsim · GitLab

Snippets Groups Projects

Nov 23, 2017
- Removed more ddebugging code. Restored the gravity-pair construction loop based on the MAC. · cf17832d
  Matthieu Schaller authored 7 years ago
  
  cf17832d
- Removed debugging code. · 9e4c7055
  Matthieu Schaller authored 7 years ago
  
  9e4c7055
- Fixed a moronic bug. Unskipping MPI sends and recvs for gravity should not be... · 8e6278cb
  Matthieu Schaller authored 7 years ago
  8e6278cb
Nov 05, 2017
- Correct types for error messages. · 42249f8e
  Matthieu Schaller authored 7 years ago
  
  42249f8e
Oct 29, 2017
- Only vectorise subset if Gadget SPH scheme used. · dd133666
  James Willis authored 7 years ago
  
  dd133666
Oct 26, 2017
- Call vectorised version of doself_subset_density when vectorisation enabled. · 21fe8595
  James Willis authored 7 years ago
  
  21fe8595
Oct 09, 2017
- Code formatting. · 3f976040
  Matthieu Schaller authored 7 years ago
  
  3f976040
- First version of the exchange of multipoles via proxies. · 95ec482d
  Matthieu Schaller authored 7 years ago
  
  95ec482d
Oct 06, 2017
- Created a branching function for DOPAIR2 to only call the vectorised function... · d959e046
  James Willis authored 7 years ago
  d959e046
- Construct a list of local top-level cells that are then used in the unskipping · f6afa976
  Matthieu Schaller authored 7 years ago
  
  f6afa976
Oct 05, 2017
- Unskip the send/recv of gparts and multipoles. · fedf9b45
  Matthieu Schaller authored 7 years ago
  
  fedf9b45
Oct 01, 2017
- All gpart now find all of their interactions following correct boundary... · 6ea5c6b0
  Matthieu Schaller authored 7 years ago
  6ea5c6b0
- Do not escape the cells with no gparticles in the long-range gravity task. · c94fd6d6
  Matthieu Schaller authored 7 years ago
  
  c94fd6d6
Sep 29, 2017
- Added the send/recv tasks for the gpart and multipoles · cf7019f8
  Matthieu Schaller authored 7 years ago
  
  cf7019f8
Sep 28, 2017
- All-reduce all the top-level multipoles after a rebuild() to make sure we have... · cf71c7aa
  Matthieu Schaller authored 7 years ago
  cf71c7aa
Sep 27, 2017
- Also send dx_max_part and dx_max_gpart alongside ti_end_min at the end of a... · a8584ba9
  Matthieu Schaller authored 7 years ago
  a8584ba9
Sep 22, 2017
- Call vectorised dopair2_force_vec. · 83d2a266
  James Willis authored 7 years ago
  
  83d2a266
Sep 12, 2017
- Force compiler to report shadowed variables. Removed all shadowed variables.... · 9f88b224
  Matthieu Schaller authored 7 years ago
  9f88b224
Sep 11, 2017
- Do the PAIR interactions in the frame of cj, not in the frame of ci. · a98b6b56
  Matthieu Schaller authored 7 years ago
  
  a98b6b56
Sep 08, 2017
- Removed debugging line that skipped the ghosts · 8087179a
  Matthieu Schaller authored 7 years ago
  
  8087179a
- Everything is now consistent with the naive algorithm up to the ghosts · 6b55c516
  Matthieu Schaller authored 7 years ago
  
  6b55c516
Sep 07, 2017
- Removed debugging line that skipped the ghosts · fe3ff0aa
  Matthieu Schaller authored 7 years ago
  
  fe3ff0aa
- Everything is now consistent with the naive algorithm up to the ghosts · 2c5a5c9e
  Matthieu Schaller authored 7 years ago
  
  2c5a5c9e
Aug 28, 2017
- Do not do the L2L or L2P if a field tensor has received 0 contributions. · 982c7184
  Matthieu Schaller authored 7 years ago
  
  982c7184
Aug 24, 2017
- Restored periodic BC behaviour for the tree walk. · ae03b870
  Matthieu Schaller authored 7 years ago
  
  ae03b870
Aug 21, 2017
- Removed the (unused) sub-self and sub-pair gravity tasks · e08c2b7a
  Matthieu Schaller authored 7 years ago
  
  e08c2b7a
Aug 14, 2017
- Added a check to verify that a field tensor has been initialised before use. · e0a98486
  Matthieu Schaller authored 7 years ago
  
  e0a98486
- When unskipping gravity tasks, do a full tree walk to set the correct drift flags. · fa6bd92d
  Matthieu Schaller authored 7 years ago
  
  fa6bd92d
Aug 03, 2017
- use the new barrier. · a473067b
  Pedro Gonnet authored 7 years ago
  
  a473067b
Jul 30, 2017
- use the new barrier. · cd6bd663
  Pedro Gonnet authored 7 years ago
  
  cd6bd663
Jul 27, 2017
- allocate a separate sort array for each dimensions. · 9b2d165f
  Pedro Gonnet authored 7 years ago
  
  9b2d165f
- remove tests for undrifted particles in send/recv as not all sent/recvd... · d6361097
  Pedro Gonnet authored 7 years ago
  d6361097
Jul 25, 2017
- we only need ti_sorted in debug mode. · a15ce3e7
  Pedro Gonnet authored 7 years ago
  
  a15ce3e7
Jul 10, 2017
- No need to specify the cell_split_size if we use the default value. · c812010e
  Matthieu Schaller authored 7 years ago
  
  c812010e
- Add check to make sure that foreign cells do not have xparts · e5d24a2e
  Peter W. Draper authored 7 years ago
  e5d24a2e
Jul 07, 2017
- Code formatting · 90728e66
  Matthieu Schaller authored 7 years ago
  
  90728e66
- Re-instate the clearing of the sorted flag when receiving particles. · 7dd3298d
  Matthieu Schaller authored 7 years ago
  
  7dd3298d
Jul 06, 2017
- formatting. · 81265160
  Pedro Gonnet authored 7 years ago
  
  81265160
- don't re-sort existing dimensions if no cleanup required. · ca250f50
  Pedro Gonnet authored 7 years ago
  
  ca250f50
Jun 30, 2017
- Not need to clear sort flag on reception fo particles before density. This... · 51ea5f44
  Matthieu Schaller authored 7 years ago
  51ea5f44