PME gather CUDA kernel failing on Fermi - Redmine #2409
For inputs larger than ~350k:
Error while launching kernel pme_gather_kernel: invalid configuration argument
Reproduced on C2070 and GTX 580.
(from redmine: issue id 2409, created on 2018-02-14 by pszilard, closed on 2018-02-16)
- Changesets:
- Revision fa92dbed by Aleksei Iupinov on 2018-02-16T09:53:30Z:
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