Skip to content

Conversation

@yf225
Copy link
Contributor

@yf225 yf225 commented Oct 17, 2024

After #137821, we will no longer be able to call the Compiled Autograd state getter under Dynamo tracing. One solution is to cache the "Compiled Autograd enabled" state outside of compile for FSDP2, and just read from the cache when we need the check. This is implemented by this PR.

Fixes #138177.

Stack from ghstack (oldest at bottom):

cc @XilunWu @H-Huang @awgu @kwen2501 @wanchaol @fegin @fduwjj @wz337 @wconstab @d4l3k @c-p-i-o @zhaojuanmao @mrshenli @rohan-varma @chauhang @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @chenyang78 @kadeng @amjames @rec @xmfan

yf225 added a commit that referenced this pull request Oct 17, 2024
@pytorch-bot
Copy link

pytorch-bot bot commented Oct 17, 2024

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/138187

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

✅ You can merge normally! (3 Unrelated Failures)

As of commit 0ab027e with merge base 795255a (image):

FLAKY - The following jobs failed but were likely due to flakiness present on trunk:

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

@pytorch-bot pytorch-bot bot added ciflow/inductor module: compiled autograd compiled_autograd module: dynamo oncall: distributed Add this issue/PR to distributed oncall triage queue release notes: distributed (fsdp) release notes category labels Oct 17, 2024
@yf225 yf225 changed the title [Traceable FSDP2] Add _compiled_autograd_enabled global state variable [Traceable FSDP2] Add _compiled_autograd_enabled global state variable Oct 17, 2024
@yf225 yf225 requested review from awgu and xmfan October 17, 2024 08:20
…state variable"

After #137821, we will no longer be able to call the Compiled Autograd state getter under Dynamo tracing. One solution is to cache the "Compiled Autograd enabled" state outside of compile for FSDP2, and just read from the cache when we need the check. This is implemented by this PR.

Fixes #138177.




cc XilunWu H-Huang awgu kwen2501 wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx chenyang78 kadeng chauhang amjames rec xmfan

[ghstack-poisoned]
…state variable"

After #137821, we will no longer be able to call the Compiled Autograd state getter under Dynamo tracing. One solution is to cache the "Compiled Autograd enabled" state outside of compile for FSDP2, and just read from the cache when we need the check. This is implemented by this PR.

Fixes #138177.




cc XilunWu H-Huang awgu kwen2501 wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx chenyang78 kadeng chauhang amjames rec xmfan

[ghstack-poisoned]
…state variable"

After #137821, we will no longer be able to call the Compiled Autograd state getter under Dynamo tracing. One solution is to cache the "Compiled Autograd enabled" state outside of compile for FSDP2, and just read from the cache when we need the check. This is implemented by this PR.

Fixes #138177.




cc XilunWu H-Huang awgu kwen2501 wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx chenyang78 kadeng chauhang amjames rec xmfan

[ghstack-poisoned]
yf225 added a commit that referenced this pull request Oct 17, 2024
…state variable"

After #137821, we will no longer be able to call the Compiled Autograd state getter under Dynamo tracing. One solution is to cache the "Compiled Autograd enabled" state outside of compile for FSDP2, and just read from the cache when we need the check. This is implemented by this PR.

Fixes #138177.




cc XilunWu H-Huang awgu kwen2501 wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx chenyang78 kadeng chauhang amjames rec xmfan

[ghstack-poisoned]
…state variable"

After #137821, we will no longer be able to call the Compiled Autograd state getter under Dynamo tracing. One solution is to cache the "Compiled Autograd enabled" state outside of compile for FSDP2, and just read from the cache when we need the check. This is implemented by this PR.

Fixes #138177.




cc XilunWu H-Huang awgu kwen2501 wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx chenyang78 kadeng chauhang amjames rec xmfan

[ghstack-poisoned]
…state variable"

After #137821, we will no longer be able to call the Compiled Autograd state getter under Dynamo tracing. One solution is to cache the "Compiled Autograd enabled" state outside of compile for FSDP2, and just read from the cache when we need the check. This is implemented by this PR.

Fixes #138177.




cc XilunWu H-Huang awgu kwen2501 wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx chenyang78 kadeng chauhang amjames rec xmfan

[ghstack-poisoned]
@yf225 yf225 requested a review from xmfan October 18, 2024 21:25
@yf225 yf225 added topic: not user facing topic category and removed release notes: distributed (fsdp) release notes category labels Oct 18, 2024
@pytorchmergebot
Copy link
Collaborator

@yf225 your PR has been successfully reverted.

pytorchmergebot added a commit that referenced this pull request Oct 19, 2024
…te variable (#138187)"

This reverts commit 0c913b3.

Reverted #138187 on behalf of https://github.com/yf225 due to linux-focal-rocm6.2-py3.10 / test (distributed, 1, 3, linux.rocm.gpu) test_compiled_autograd_ctx failed ([comment](#138187 (comment)))
…state variable"

After #137821, we will no longer be able to call the Compiled Autograd state getter under Dynamo tracing. One solution is to cache the "Compiled Autograd enabled" state outside of compile for FSDP2, and just read from the cache when we need the check. This is implemented by this PR.

Fixes #138177.




cc XilunWu H-Huang awgu kwen2501 wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o zhaojuanmao mrshenli rohan-varma chauhang voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx chenyang78 kadeng amjames rec xmfan

[ghstack-poisoned]
@yf225
Copy link
Contributor Author

yf225 commented Oct 19, 2024

@pytorchbot merge

@yf225 yf225 added ciflow/periodic Trigger jobs ran periodically on master (periodic.yml) on the PR ciflow/rocm Trigger "default" config CI on ROCm and removed Merged Reverted labels Oct 19, 2024
…state variable"

After #137821, we will no longer be able to call the Compiled Autograd state getter under Dynamo tracing. One solution is to cache the "Compiled Autograd enabled" state outside of compile for FSDP2, and just read from the cache when we need the check. This is implemented by this PR.

Fixes #138177.




cc XilunWu H-Huang awgu kwen2501 wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o zhaojuanmao mrshenli rohan-varma chauhang voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx chenyang78 kadeng amjames rec xmfan

[ghstack-poisoned]
…state variable"

After #137821, we will no longer be able to call the Compiled Autograd state getter under Dynamo tracing. One solution is to cache the "Compiled Autograd enabled" state outside of compile for FSDP2, and just read from the cache when we need the check. This is implemented by this PR.

Fixes #138177.




cc XilunWu H-Huang awgu kwen2501 wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o zhaojuanmao mrshenli rohan-varma chauhang voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx chenyang78 kadeng amjames rec xmfan

[ghstack-poisoned]
yf225 added a commit that referenced this pull request Oct 19, 2024
@pytorchmergebot
Copy link
Collaborator

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

@yf225
Copy link
Contributor Author

yf225 commented Oct 19, 2024

@pytorchbot merge

@pytorchmergebot
Copy link
Collaborator

The merge job was canceled or timed out. This most often happen if two merge requests were issued for the same PR, or if merge job was waiting for more than 6 hours for tests to finish. In later case, please do not hesitate to reissue the merge command
For more information see pytorch-bot wiki.

@pytorchmergebot
Copy link
Collaborator

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

@pytorchmergebot
Copy link
Collaborator

Merge failed

Reason: 1 jobs have failed, first few of them are: inductor-periodic / cuda12.1-py3.10-gcc9-sm80 / test (inductor_torchbench_smoketest_perf, 1, 1, linux.gcp.a100)

Details for Dev Infra team Raised by workflow job

@yf225
Copy link
Contributor Author

yf225 commented Oct 19, 2024

@pytorchbot merge -f "unrelated failures"

@pytorchmergebot
Copy link
Collaborator

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

@github-actions github-actions bot deleted the gh/yf225/146/head branch November 19, 2024 02:08
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ciflow/inductor ciflow/periodic Trigger jobs ran periodically on master (periodic.yml) on the PR ciflow/rocm Trigger "default" config CI on ROCm ciflow/trunk Trigger trunk jobs on your pull request keep-going Don't stop on first failure, keep running tests until the end Merged module: compiled autograd compiled_autograd module: dynamo module: fsdp oncall: distributed Add this issue/PR to distributed oncall triage queue topic: not user facing topic category

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants