| commit | fa92dbed34ec7b916faaa2dd6f17c1b08a8ede0a | |
| author | Aleksei Iupinov <a.yupinov@gmail.com> | |
| Thu, 15 Feb 2018 12:27:48 +0000 (15 13:27 +0100) | ||
| committer | Aleksei Iupinov <a.yupinov@gmail.com> | |
| Fri, 16 Feb 2018 09:53:30 +0000 (16 10:53 +0100) | ||
| tree | 2b6f73e5ec2fbecd80689f314c5d12f6afde416c | treesnapshot (tar.gz zip) |
| parent | 03cb7abaceeb2f5227d996e0846a17ea5ead1a29 | commitdiff |
Fix PME for large systems with Fermi GPUs
PME spread/gather CUDA kernel scheduling did not account for
compute capability limitations. Realistically this has only
caused it to fail on CC 2.x with input systems larger than
2^18 ~= 262k atoms. This is now fixed for all CUDA architectures.
Fixes #2409
Change-Id: I59295b5d53a341d08a221aebb52e1db9f1e80107
PME spread/gather CUDA kernel scheduling did not account for
compute capability limitations. Realistically this has only
caused it to fail on CC 2.x with input systems larger than
2^18 ~= 262k atoms. This is now fixed for all CUDA architectures.
Fixes #2409
Change-Id: I59295b5d53a341d08a221aebb52e1db9f1e80107