|
|
0425dd037e
|
Removing nall from kernel calls. Changing ellipsoid neighbor calls to use a correct block size.
|
2011-05-24 20:15:04 -04:00 |
|
|
|
e27ca98bb8
|
Untested update of entire GPU library to use multiple threads per atom.
|
2011-04-14 17:25:55 -04:00 |
|
|
|
2b37bf3d41
|
Some block sizes are now selected based on arch.
|
2011-04-13 16:32:43 -04:00 |
|
|
|
5ff6092ea1
|
Updating all preprocessor defines to be set by a kernel. Adding a threads
per atom option to fix gpu.
|
2011-04-13 13:57:56 -04:00 |
|
|
|
86f7f6cd42
|
Adding better error output to LAMMPS routines.
|
2011-04-03 11:51:26 -04:00 |
|
|
|
c1c9fd47de
|
Hybrid working, but not fully tested.
|
2011-01-27 17:52:08 -05:00 |
|
|
|
9d654ca91c
|
Adding support for non-arithmetic mixing to charmm and updating docs and
restart2data
|
2011-01-06 18:02:09 -05:00 |
|
|
|
7c5cef048f
|
Updating host-write buffer in CHARMM to zero all memory to avoid valgrind errors.
|
2011-01-05 19:02:39 -05:00 |
|
|
|
c1e53cf140
|
Final working charmm commit.
|
2010-12-30 19:52:32 -05:00 |
|
|
|
56b96c1722
|
Replacing with another test for faster type data in charmm.
|
2010-12-30 16:01:05 -05:00 |
|
|
|
b85a2a1cf7
|
Adding a test kernel for faster shared types in charmm.
|
2010-12-30 14:00:35 -05:00 |
|
|
|
bc5876cf5a
|
Bug fixes to generate first working charmm/long/gpu
|
2010-12-29 17:36:03 -05:00 |
|
|
|
6f4d1de96c
|
Initial commit of charmm/long/gpu style w/out testing.
|
2010-12-29 15:18:53 -05:00 |
|
