Skip to content

Conversation

@tdrwenski
Copy link

This PR tweaks the test cases in #529 and targets that branch (@adrienbernede)

  • Allow caliper to benchmark script to handle npasses (of the entire suite) as well as reps (of a single test in a row) when computing the average time per run. npasses currently has a bug so I am not using this setting at the moment (Incorrect complexity when npasses > 1 #563)
  • Add reps and passes to extra info (visible with tooltip on dashboard)
  • Exclude basic kernel test cases that vary ~20%
  • Note: GPU tests cases may still need work as those are currently commented out.

See my fork's dashboard here: https://tdrwenski.github.io/RAJAPerf/dev/bench/

@tdrwenski tdrwenski force-pushed the caliper-ci-gh-benchmark branch from 5a8ad06 to 8141412 Compare December 4, 2025 17:40
@tdrwenski
Copy link
Author

tdrwenski commented Dec 5, 2025

@adrienbernede -- I have rebased on your branch, retested with the Tuo test cases we previously had to comment out and got that working. I think this is ready to merge to your branch if you want to have a look!

@adrienbernede adrienbernede merged commit 5eb1c52 into llnl:woptim/caliper-ci Dec 15, 2025
@tdrwenski tdrwenski deleted the caliper-ci-gh-benchmark branch December 15, 2025 21:47
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants