| commit | 1d8fe71690b08dcc000dcd4088f6edc7f63a937e | |
| author | Szilárd Páll <pszilard@kth.se> | |
| Sat, 8 Nov 2014 02:05:31 +0000 (8 03:05 +0100) | ||
| committer | Gerrit Code Review <gerrit@gerrit.gromacs.org> | |
| Sat, 29 Nov 2014 17:24:51 +0000 (29 18:24 +0100) | ||
| tree | 8388b564315c6c21b6a121f097a41cb886c698cf | treesnapshot (tar.gz zip) |
| parent | 477759badb466447e994ab0c72af9e2cd18eb6ca | commitdiff |
Add kernel and compiler suport for CC 3.7 devices
On compute capability 3.7 NVIDIA GPUs we can make use of the increased
register size by running 128 threads/block with keeping the minimum
number of blocks per multiprocessor at 16.
Change-Id: I84ec179a409668fe44fb9183cf3485c21bd53254
On compute capability 3.7 NVIDIA GPUs we can make use of the increased
register size by running 128 threads/block with keeping the minimum
number of blocks per multiprocessor at 16.
Change-Id: I84ec179a409668fe44fb9183cf3485c21bd53254
| cmake/gmxManageNvccConfig.cmake | diffblobblamehistory | |
| src/gromacs/mdlib/nbnxn_cuda/nbnxn_cuda.cu | diffblobblamehistory |