🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/157214

• 📄 Preview Python docs built from this PR
• 📄 Preview C++ docs built from this PR
• ❓ Need help or want to give feedback on the CI? Visit the bot commands wiki or our office hours

Note: Links to docs will display an error until the docs builds have been completed.

⏳ 5 Pending, 2 Unrelated Failures

As of commit 2fedd18 with merge base bccb847 ():

This comment was automatically generated by Dr. CI and updates every 15 minutes.

@pytorchbot rebase

@pytorchbot started a rebase job onto refs/remotes/origin/viable/strict. Check the current status here

Successfully rebased gh/jansel/528/orig onto refs/remotes/origin/viable/strict, please pull locally before adding more changes (for example, via ghstack checkout https://github.com/pytorch/pytorch/pull/157214)

@pytorchbot merge --force " pr time benchmark job succeeded only effected test"

Merge started

Your change will be merged immediately since you used the force (-f) flag, bypassing any CI checks (ETA: 1-5 minutes). Please use -f as last resort and instead consider -i/--ignore-current to continue the merge ignoring current failures. This will allow currently pending tests to finish and report signal before the merge.

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

seems like it mainly failing due to win on symint_sum
addressed in this PR

I think maybe we need to split each benchmark into its own test this way only symint_sum would fail and others would succeed. @huydhn what do you think of this idea.

I think that only makes sense if:

1. you want to move symint_sum to a separate workflow that doesn't block merge
2. and symint_sum is the only one causing troubles

Generally, I would prefer to have a way for oncall to quickly disable the failing benchmark, so that it can be dealt with later without affecting others:

1. In the short term, maybe having a JSON file to include something like { "symint_sum": "skip" }
2. If more benchmarks are coming in the future, building a similar mechanism to disable a benchmark on CI via an issue like DISABLED test_non_contiguous_input_mm_plus_mm (__main__.TestMaxAutotune) #126867 is probably a good idea