[FlexAttention] only calculate grads for buffers that require_grad #137451

drisspg · 2024-10-07T23:09:09Z

Stack from ghstack (oldest at bottom):

cc @Chillee @yanboliang @BoyuanFeng

[ghstack-poisoned]

pytorch-bot · 2024-10-07T23:09:12Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/137451

📄 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.

✅ No Failures

As of commit 4d89d3c with merge base 3346159 ():
💚 Looks good so far! There are no failures yet. 💚

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

…ire_grad" cc Chillee yanboliang BoyuanFeng [ghstack-poisoned]

drisspg · 2024-10-08T02:09:11Z

@pytorchbot merge

pytorchmergebot · 2024-10-08T02:10:53Z

Merge started

Your change will be merged once all checks pass (ETA 0-4 Hours).

Learn more about merging in the wiki.

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

Advanced Debugging

Check the merge workflow status
here

malfet · 2024-10-08T16:58:55Z

@pytorchbot revert -m "Need to revert it in order to be able to backout #136910" -c weird

pytorchmergebot · 2024-10-08T17:00:50Z

@pytorchbot successfully started a revert job. Check the current status here.
Questions? Feedback? Please reach out to the PyTorch DevX Team

…_grad (#137451)" This reverts commit 9f9d252. Reverted #137451 on behalf of https://github.com/malfet due to Need to revert it in order to be able to backout #136910 ([comment](#137451 (comment)))

pytorchmergebot · 2024-10-08T17:01:02Z

@drisspg your PR has been successfully reverted.

malfet · 2024-10-08T17:32:05Z

@pytorchbot merge -f "Looks like it could be relanded safely"

pytorchmergebot · 2024-10-08T17:33:47Z

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

Advanced Debugging

Check the merge workflow status
here

pytorchmergebot · 2024-10-08T17:33:57Z

Merge failed

Reason: Command git -C /home/runner/work/pytorch/pytorch cherry-pick -x f34daf2e863f5d9c65c1bc24d3d9d9f9475d4def returned non-zero exit code 1

Auto-merging torch/_higher_order_ops/flex_attention.py
CONFLICT (content): Merge conflict in torch/_higher_order_ops/flex_attention.py
error: could not apply f34daf2e86... [FlexAttention] only calculate grads for buffers that require_grad
hint: After resolving the conflicts, mark them with
hint: "git add/rm <pathspec>", then run
hint: "git cherry-pick --continue".
hint: You can instead skip this commit with "git cherry-pick --skip".
hint: To abort and get back to the state before "git cherry-pick",
hint: run "git cherry-pick --abort".
hint: Disable this message with "git config advice.mergeConflict false"

Details for Dev Infra team

Raised by workflow job

Part 2 of #136910 With some updates from: #137451 And updates to lib name + lazy registration [ghstack-poisoned]

Part 2 of #136910 With some updates from: #137451 And updates to lib name + lazy registration ghstack-source-id: d43ca91 Pull Request resolved: #137526

Part 2 of #136910 With some updates from: #137451 And updates to lib name + lazy registration cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy yf225 chenyang78 kadeng muchulee8 ColinPeppler amjames desertfire chauhang [ghstack-poisoned]

Part 2 of #136910 With some updates from: #137451 And updates to lib name + lazy registration ghstack-source-id: 4bd0169 Pull Request resolved: #137526