Commits · f3fe550ac98601c9d6e7f5d291aad31a37a0ec44 · tsoc / openmm

29 Oct, 2013 1 commit
- Created ThreadPool to simplify parallel code · f3fe550a
  peastman authored Oct 29, 2013
  
  f3fe550a
24 Oct, 2013 3 commits
- Minor optimizations to CPU platform · a5bcc4be
  peastman authored Oct 24, 2013
  
  a5bcc4be
- Minor change to the neighbor list padding distance · 062a2db0
  peastman authored Oct 23, 2013
  
  062a2db0
- Do a more complex test to decide when to rebuild the neighbor list · 8d9950da
  peastman authored Oct 23, 2013
  
  8d9950da
23 Oct, 2013 3 commits
- Minor optimizations to neighbor list construction · 6ad86974
  peastman authored Oct 23, 2013
  
  6ad86974
- Parallelized the processing of exclusions for Ewald · d6b2eb75
  peastman authored Oct 23, 2013
  
  d6b2eb75
- Optimized algorithm for building neighbor list. · f301042f
  peastman authored Oct 23, 2013
  
  f301042f
22 Oct, 2013 1 commit
- Restructured nonbonded calculation to allow more efficient vectorization · 56902b45
  peastman authored Oct 22, 2013
  
  56902b45
21 Oct, 2013 1 commit
- Created vector classes to simplify SSE code · 6eacad45
  peastman authored Oct 21, 2013
  
  6eacad45
17 Oct, 2013 1 commit
- Fixed segfaults on some computers · 8f168506
  peastman authored Oct 17, 2013
  
  8f168506
15 Oct, 2013 4 commits
- Prevent segfaults when a simulation blows up · d389e504
  peastman authored Oct 15, 2013
  
  d389e504
- Bug fix · 2943d628
  peastman authored Oct 15, 2013
  
  2943d628
- Bug fix, plus a minor optimization to reference platform · dd175b23
  peastman authored Oct 15, 2013
  
  dd175b23
- Assorted optimizations · 79011828
  peastman authored Oct 14, 2013
  
  79011828
14 Oct, 2013 2 commits
- Tabulate the Ewald scale factor · f1533707
  peastman authored Oct 14, 2013
  
  f1533707
- Fixed a race condition · 9491eb72
  peastman authored Oct 14, 2013
  
  9491eb72
12 Oct, 2013 1 commit
- Do not rebuild the neighbor list every time step · 18f1cd3e
  peastman authored Oct 11, 2013
  
  18f1cd3e
11 Oct, 2013 4 commits
- Direct space nonbonded is multithreaded · b278bb6d
  peastman authored Oct 11, 2013
  
  b278bb6d
- Use optimized PME implementation when available · 3c435fb8
  peastman authored Oct 11, 2013
  
  3c435fb8
- PME works correctly in CPU platform · e22b8f53
  peastman authored Oct 11, 2013
  
  e22b8f53
- Continuing to optimize CPU nonbonded routines · 6b10b909
  peastman authored Oct 11, 2013
  
  6b10b909
10 Oct, 2013 4 commits
- Began creating optimized implementation of direct space nonbonded calculation · 9bc194b9
  peastman authored Oct 10, 2013
  
  9bc194b9
- Separated hardware query functions into a separate file. CPU platform checks... · 50d4f9d0
  peastman authored Oct 10, 2013
```
Separated hardware query functions into a separate file.  CPU platform checks for the correct hardware.
```
  50d4f9d0
- Neighborlist construction is multithreaded. · f4a1847a
  peastman authored Oct 10, 2013
  
  f4a1847a
- More optimizations to neighborlist construction, including SSE intrinsics · 3b3fd578
  peastman authored Oct 09, 2013
  
  3b3fd578
09 Oct, 2013 2 commits
- Optimizations to neighbor list construction · d4d9ce1f
  peastman authored Oct 09, 2013
  
  d4d9ce1f
- Early beginnings of creating an optimized CPU platform · d4a343e2
  peastman authored Oct 09, 2013
  
  d4a343e2