Commit Graph

6 Commits

Author SHA1 Message Date
Robert Maynard
718caaaeac CudaAllocator allows managed memory to be explicitly disabled 2018-12-28 11:30:29 -05:00
Sujin Philip
1212081de1 Support deferred freeing for CUDA memory
Calls to 'cudaFree' block execution on all cuda devices. Reduce the number of
times this happens by having a deferred free mechanism that frees a pool
of pointers together when a threshold is reached.

Especially helpful during virtual object transfers that requires a few small
allocations and frees.
2018-08-16 12:05:36 -04:00
luz.paz
80b11afa24 Misc. typos
Found via `codespell -q 3` via downstream VTK
2018-01-30 06:51:47 -05:00
Kenneth Moreland
c3a3184d51 Update copyright for Sandia
Sandia National Laboratories recently changed management from the
Sandia Corporation to the National Technology & Engineering Solutions
of Sandia, LLC (NTESS). The copyright statements need to be updated
accordingly.
2017-09-20 15:33:44 -06:00
Allison Vacanti
0a828189ad Reuse user-provided cuda device pointers when possible. 2017-08-03 17:21:13 -04:00
David C. Lonie
bd042ec567 Add CudaAllocator to encapsulate runtime managed memory logic.
Unified memory is now used when we detect that the hardware supports it.
2017-07-31 09:08:27 -04:00