1. 19 Jan, 2021 4 commits
  2. 04 Jan, 2021 1 commit
  3. 22 Dec, 2020 1 commit
  4. 21 Dec, 2020 1 commit
  5. 18 Dec, 2020 2 commits
  6. 14 Dec, 2020 6 commits
  7. 20 Nov, 2020 1 commit
  8. 02 Oct, 2020 1 commit
  9. 11 Mar, 2020 1 commit
  10. 09 Mar, 2020 2 commits
  11. 25 Feb, 2020 1 commit
  12. 05 Feb, 2020 1 commit
    • Bram Veenboer's avatar
      Add set_grid and get_grid methods to Proxy · c4f2cb92
      Bram Veenboer authored
      These methods are intended for cases when the Proxy takes ownership over
      the grid, for instance in the CUDA Generic proxy, where a copy of the
      grid is kept in GPU memory. Only upon set_grid and get_grid, the host memory
      is synchronized with the GPU memory. This approach has a number
      consequences for the other CUDA proxies as well, hence a lot of changes
      are made throughout the code.
      c4f2cb92
  13. 31 Jan, 2020 4 commits
  14. 14 Jan, 2020 2 commits
  15. 28 Nov, 2019 1 commit
  16. 15 Nov, 2019 1 commit
  17. 01 Nov, 2019 5 commits
  18. 24 Oct, 2019 1 commit
  19. 16 Sep, 2019 1 commit
  20. 12 Sep, 2019 1 commit
  21. 10 Sep, 2019 1 commit
  22. 09 Sep, 2019 1 commit
    • Bram Veenboer's avatar
      Update to memory management in InstanceCUDA · 7fbf6631
      Bram Veenboer authored
      Rename methods and functions to make a distinction between memory
      allocation, memory registration (page-locking) and retrieving of
      pre-allocated memory. All cu::HostMemory objects not wrapped in a
      unique_ptr are contained in the h_misc_ vector, which uses a unique_ptr
      to make sure that the memory is eventually released.
      7fbf6631