Mentions légales du service

Skip to content
Snippets Groups Projects

Allocate panel memory manually if it is the read node

Closed LEANDRO NESI Lucas requested to merge lleandro/cppdiodon:allocate_panels into master

This will not let StarPU allocate this particular handle because we know that this data will never be in GPU; a simple malloc can be used instead of the CUDA_HOST_MALLOC, which is slow. Also, the free method submits a callback for the handle and then frees the memory and asks for a starpu invalidate.

Merge request reports

Loading
Loading

Activity

Filter activity
  • Approvals
  • Assignees & reviewers
  • Comments (from bots)
  • Comments (from users)
  • Commits & branches
  • Edits
  • Labels
  • Lock status
  • Mentions
  • Merge request status
  • Tracking
Please register or sign in to reply
Loading