Skip to content
Snippets Groups Projects

Misc. GPU improvements

Merged rarbore2 requested to merge edge_gpu into main
  • GPU backend fixes.
    • Emit float infinity properly.
    • Emit thread_block_tile variables at start of function.
  • RT backend fixes.
    • Lower intrinsics.
    • Emit float infinity properly.
  • Add max / min to monoid reduction cleaning (utilities for working w/ smallest/largest values of a datatype).
  • Don't outline scalar constants (TODO: inline these interprocedurally later).
  • Optimize edge detection schedules, especially GPU.
    • Emit a 2-level max reduction tree.

Merge request reports

Pipeline #201805 passed

Pipeline passed for ca9b2bf9 on edge_gpu

Merged by rarbore2rarbore2 4 weeks ago (Feb 21, 2025 3:12am UTC)

Loading

Pipeline #201806 passed

Pipeline passed for 2ce19c22 on main

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