Skip to content

[cuda.compute]: Disable sass checks for float16 merge sort#8053

Open
NaderAlAwar wants to merge 1 commit intoNVIDIA:mainfrom
NaderAlAwar:merge-sort-ldl-stl-disable
Open

[cuda.compute]: Disable sass checks for float16 merge sort#8053
NaderAlAwar wants to merge 1 commit intoNVIDIA:mainfrom
NaderAlAwar:merge-sort-ldl-stl-disable

Conversation

@NaderAlAwar
Copy link
Contributor

Description

#6642 made some kernel changes which lead to LDL and STL instructions in some of the specific cases we test in cuda.compute. This is not a JIT compilation issue since the C++/nvcc path also generates those instructions (a quick benchmark shows that this did not lead to a performance regression). I plan on doing a more thorough investigation of these issues in #7978.

Checklist

  • New or existing tests cover these changes.
  • The documentation is up to date with these changes.

@github-actions
Copy link
Contributor

🥳 CI Workflow Results

🟩 Finished in 1h 33m: Pass: 100%/56 | Total: 16h 55m | Max: 52m 57s

See results here.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

Status: In Review

Development

Successfully merging this pull request may close these issues.

2 participants