Commits · 0b35240df6e0f4e4178fe628bfe7c42a56587864 · tsoc / openmm

21 Dec, 2013 1 commit
- Parallelized summing the forces from different threads · 8bdf0829
  peastman authored Dec 20, 2013
  
  8bdf0829
19 Dec, 2013 3 commits
- Parallelized LangevinIntegrator for CPU platform · ddf1216d
  peastman authored Dec 19, 2013
  
  ddf1216d
- Fixed bug in detecting if AVX is supported · 6cde1860
  peastman authored Dec 19, 2013
  
  6cde1860
- Automatically select whether to use the SSE or AVX version of CpuNonbondedForce · d4af8747
  peastman authored Dec 18, 2013
  
  d4af8747
10 Dec, 2013 1 commit
- Minor optimizations to CPU platform · 9ec4f597
  peastman authored Dec 10, 2013
  
  9ec4f597
04 Dec, 2013 1 commit
- Enforce memory alignment to improve performance of vector operations. Also... · b1be68d8
  peastman authored Dec 04, 2013
```
Enforce memory alignment to improve performance of vector operations.  Also fixed bugs in an earlier optimization.
```
  b1be68d8
25 Nov, 2013 1 commit
- Fixed bugs in computing PME energy with CPU platform · 0ee8c785
  peastman authored Nov 25, 2013
  
  0ee8c785
05 Nov, 2013 1 commit
- Minor optimizations to CPU version of GBSAOBCForce · 775866b6
  peastman authored Nov 04, 2013
  
  775866b6
04 Nov, 2013 1 commit
- Created optimized CPU version of GBSAOBCForce · c97ca97b
  peastman authored Nov 04, 2013
  
  c97ca97b
01 Nov, 2013 1 commit
- Centralized the forces, positions, and thread pool so they can be shared between kernels · be61ee5b
  peastman authored Nov 01, 2013
  
  be61ee5b
30 Oct, 2013 1 commit
- Minor optimizations to CPU platform: avoid applying periodic boundary conditions unnecessarily · 62c64e3c
  peastman authored Oct 30, 2013
  
  62c64e3c
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
22 Oct, 2013 1 commit
- Restructured nonbonded calculation to allow more efficient vectorization · 56902b45
  peastman authored Oct 22, 2013
  
  56902b45
12 Oct, 2013 1 commit
- Do not rebuild the neighbor list every time step · 18f1cd3e
  peastman authored Oct 11, 2013
  
  18f1cd3e
11 Oct, 2013 3 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
10 Oct, 2013 1 commit
- Began creating optimized implementation of direct space nonbonded calculation · 9bc194b9
  peastman authored Oct 10, 2013
  
  9bc194b9
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