Commits · b8bae04c2a115dc5ffd90c70263d34e48034a787 · tsoc / openmm

11 Sep, 2014 2 commits
- Reversed a change that didn't help · 3e47d824
  peastman authored Sep 11, 2014
  
  3e47d824
- Switched to a slightly faster (and more accurate) approximation for erfc() · 4b5b08af
  peastman authored Sep 11, 2014
  
  4b5b08af
09 Sep, 2014 2 commits
- Cleaned up obsolete code · 68984320
  peastman authored Sep 09, 2014
  
  68984320
- Fixed bug in CustomManyParticleForce that caused crash · 904a9c10
  peastman authored Sep 08, 2014
  
  904a9c10
08 Sep, 2014 1 commit
- Rewrote findBlocksWithInteractions() in a way that is both faster and simpler. · bfeb8ac7
  peastman authored Sep 08, 2014
  
  bfeb8ac7
04 Sep, 2014 1 commit
- Bug fixes to OpenCL version of CustomManyParticleForce · 9f77efc2
  Peter Eastman authored Sep 04, 2014
  
  9f77efc2
03 Sep, 2014 1 commit
- Bug fix · 9cd18aeb
  peastman authored Sep 03, 2014
  
  9cd18aeb
30 Aug, 2014 1 commit
- Bug fixes to CUDA version of CustomManyParticleForce · b11c8061
  peastman authored Aug 29, 2014
  
  b11c8061
29 Aug, 2014 3 commits
- Bug fixes to CustomManyParticleForce · 3134fbb9
  peastman authored Aug 29, 2014
  
  3134fbb9
- Implemented UniqueCentralParticle mode in CUDA version of CustomManyParticleForce · c2d461f5
  peastman authored Aug 29, 2014
  
  c2d461f5
- Minor optimization to CUDA version of CustomManyParticleForce · a306aa58
  peastman authored Aug 29, 2014
  
  a306aa58
22 Aug, 2014 1 commit
- Optimizations to CUDA version of CustomManyParticleForce · 150d943a
  peastman authored Aug 22, 2014
  
  150d943a
21 Aug, 2014 1 commit
- CUDA version of CustomManyParticleForce uses neighbor list · bddaf4e7
  peastman authored Aug 21, 2014
  
  bddaf4e7
20 Aug, 2014 1 commit
- Wrote code to build neighbor list for CUDA version of CustomManyParticleForce · e3b631f6
  peastman authored Aug 20, 2014
  
  e3b631f6
19 Aug, 2014 3 commits
- Implemented exclusions for CUDA version of CustomManyParticleForce · 786e1ee7
  peastman authored Aug 19, 2014
  
  786e1ee7
- Implemented type filters for CUDA version of CustomManyParticleForce · 65b56ee1
  peastman authored Aug 19, 2014
  
  65b56ee1
- Continuing to work on CUDA implementation of CustomManyParticleForce · 785592f4
  peastman authored Aug 19, 2014
  
  785592f4
18 Aug, 2014 2 commits
- Beginning of CUDA implementation of CustomManyParticleForce · 890963b9
  peastman authored Aug 18, 2014
  
  890963b9
- avoid syntax error if tabulated function min is negative · e30242b6
  Mike Garrahan authored Aug 18, 2014
  
  e30242b6
12 Aug, 2014 2 commits
- Fixed uninitialized memory error when using multiple GPUs · 2279888f
  peastman authored Aug 12, 2014
  
  2279888f
- Avoid integer overflow when dealing with very large systems · edd91641
  peastman authored Aug 12, 2014
  
  edd91641
30 Jul, 2014 1 commit
- Fix crash with empty interaction groups · 92d5b696
  peastman authored Jul 30, 2014
  
  92d5b696
15 Jul, 2014 1 commit
- Fixed bug when running on a CPU device · c9c5997b
  peastman authored Jul 15, 2014
  
  c9c5997b
27 Jun, 2014 1 commit
- Fixed a rare memory corruption error when roundoff causes tile indices to be wrong · ece75910
  peastman authored Jun 27, 2014
  
  ece75910
06 May, 2014 1 commit
- Fixed bug using multiple devices with PME · f1991b3c
  peastman authored May 06, 2014
  
  f1991b3c
24 Apr, 2014 1 commit
- Created CUDA and OpenCL implementations of LocalCoordinatesSite · c88213f8
  peastman authored Apr 24, 2014
  
  c88213f8
17 Mar, 2014 1 commit
- Began creating CPU implementation of CustomNonbondedForce · 7ab954bb
  peastman authored Mar 17, 2014
  
  7ab954bb
31 Jan, 2014 1 commit
- Renamed methods for dealing with TabulatedFunctions to avoid problems in... · 75a04e7f
  peastman authored Jan 31, 2014
```
Renamed methods for dealing with TabulatedFunctions to avoid problems in wrapper APIs.  Also fixed a few bugs.
```
  75a04e7f
30 Jan, 2014 1 commit
- Created OpenCL and CUDA implementations of Continuous3DFunction · 3279c73a
  peastman authored Jan 30, 2014
  
  3279c73a
29 Jan, 2014 1 commit
- Created OpenCL and CUDA implementations of Continuous2DFunction · 9434c6e8
  peastman authored Jan 29, 2014
  
  9434c6e8
27 Jan, 2014 2 commits
- CPU, CUDA, and OpenCL platforms can now be built statically · 8285d658
  Yutong Zhao authored Jan 27, 2014
  
  8285d658
- Tabulated function parameters are hardcoded in the kernel instead of being... · 0a1a011d
  peastman authored Jan 27, 2014
```
Tabulated function parameters are hardcoded in the kernel instead of being stored in an array.  This makes the code simpler and may help performance slightly.
```
  0a1a011d
25 Jan, 2014 1 commit
- Created Discrete2DFunction and Discrete3DFunction · a773952e
  peastman authored Jan 24, 2014
  
  a773952e
24 Jan, 2014 2 commits
- Created Discrete1DFunction · 56e36449
  peastman authored Jan 24, 2014
  
  56e36449
- Converted CUDA and OpenCL platforms to process tabulated functions with the TabulatedFunction API · 81f26683
  peastman authored Jan 23, 2014
  
  81f26683
11 Jan, 2014 1 commit
- Same bug fix as the previous commit, but for CUDA · 1e857c31
  peastman authored Jan 10, 2014
  
  1e857c31
07 Jan, 2014 1 commit
- Very minor optimization · 06db7ac4
  peastman authored Jan 07, 2014
  
  06db7ac4
06 Jan, 2014 1 commit
- Improved logic for selecting a platform to better deal with exceptions in context initialization · d91e3f19
  peastman authored Jan 06, 2014
  
  d91e3f19
21 Dec, 2013 1 commit
- Fixed quotes in command line when using CudaHostCompiler option · cc11a449
  peastman authored Dec 20, 2013
  
  cc11a449
13 Dec, 2013 1 commit
- Added option for setting CUDA host compiler · b989fd9b
  peastman authored Dec 13, 2013
  
  b989fd9b