🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/169485

• 📄 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

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

❌ 1 New Failure, 4 Unrelated Failures

As of commit e285381 with merge base afdff7f ():

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

[edited] The compile-time measurements are fairly flakey from day to day, so it's kind of impossible to say whether this would really affect compile time. For example, see these day-over-day comparisons the last few days:
https://hud.pytorch.org/benchmark/compilers?dashboard=torchinductor&startTime=Thu%2C%2020%20Nov%202025%2020%3A25%3A27%20GMT&stopTime=Thu%2C%2004%20Dec%202025%2020%3A25%3A27%20GMT&granularity=hour&mode=training&dtype=amp&deviceName=cuda%20(h100)&lBranch=main&lCommit=518c2b1b3dab9a2ef2849e04b3bc2f20c1c41db9&rBranch=main&rCommit=d5038950bacfe36bbf24a47a455fe76901deb8e8
https://hud.pytorch.org/benchmark/compilers?dashboard=torchinductor&startTime=Thu%2C%2020%20Nov%202025%2020%3A25%3A27%20GMT&stopTime=Thu%2C%2004%20Dec%202025%2020%3A25%3A27%20GMT&granularity=hour&mode=training&dtype=amp&deviceName=cuda%20(h100)&lBranch=main&lCommit=0c281dd78773b2bc17c58ead0e4cd4ac46e775c5&rBranch=main&rCommit=518c2b1b3dab9a2ef2849e04b3bc2f20c1c41db9
https://hud.pytorch.org/benchmark/compilers?dashboard=torchinductor&startTime=Thu%2C%2020%20Nov%202025%2020%3A25%3A27%20GMT&stopTime=Thu%2C%2004%20Dec%202025%2020%3A25%3A27%20GMT&granularity=hour&mode=training&dtype=amp&deviceName=cuda%20(h100)&lBranch=main&lCommit=320de0c6b0a3e7c6d2693ea5c28d5d0156ba7991&rBranch=main&rCommit=0c281dd78773b2bc17c58ead0e4cd4ac46e775c5

AFAIK, we've known about this compile-time flakiness since at least April, e.g., #152566

FYI @ezyang since you were asking about this feature for internal use case. See my comments above about compile-time measurements. I kinda doubt this feature would regress compile-time, but seems like we don't really have a way to measure.

@pytorchbot merge -i

Merge started

Your change will be merged while ignoring the following 5 checks: pull / linux-jammy-py3.10-clang12 / test (dynamo_wrapped, 1, 3, lf.linux.2xlarge), pull / linux-jammy-py3.10-gcc11 / test (docs_test, 1, 1, lf.linux.2xlarge), pull / linux-jammy-py3.13-clang12 / test (dynamo_wrapped, 1, 3, lf.linux.2xlarge), trunk / macos-py3-arm64 / test (default, 1, 3, macos-m1-stable), inductor-A100-perf-nightly / cuda12.8-py3.10-gcc11-sm80 / test (inductor_torchbench_perf, 3, 6, linux.aws.a100)

Learn more about merging in the wiki.

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