Skip buffer in dense update #148533

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Closed

zoranzhao wants to merge 1 commit into pytorch:main from zoranzhao:export-D69553929

Member

zoranzhao commented Mar 5, 2025

Summary:
as title.

PyTorch Module buffer will not be published in delta publishing. In Quinn's previous diff, constant type annotations have been introduced.

In addition to skip constant, we also need to skip buffer if it is not found in the user-provided delta weights list

Test Plan: https://docs.google.com/document/d/1wiqUo0PyZ4g6YJIJlL_LE084ZEuE74iu74gZjqGGjWY/edit?tab=t.0#heading=h.dby6cwiw1xrn

Differential Revision: D69553929

pytorch-bot bot commented Mar 5, 2025 •

edited

Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/148533

📄 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.

❌ 1 New Failure, 4 Unrelated Failures

As of commit bd38921 with merge base 65dbc3b ():

NEW FAILURE - The following job has failed:

pull / win-vs2022-cpu-py3 / build (gh)
C:\actions-runner\_work\pytorch\pytorch\aten\src\ATen/native/quantized/cpu/OnednnUtils.h(445): error C2065: 'vnni_available': undeclared identifier

BROKEN TRUNK - The following jobs failed but were present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

inductor / cuda12.4-py3.10-gcc9-sm86 / test (inductor_timm, 2, 2, linux.g5.4xlarge.nvidia.gpu) (gh) (trunk failure)
mobilenetv3_large_100
trunk / win-vs2022-cpu-py3 / build (gh) (trunk failure)
C:\actions-runner\_work\pytorch\pytorch\aten\src\ATen/native/quantized/cpu/OnednnUtils.h(445): error C2065: 'vnni_available': undeclared identifier
trunk / win-vs2022-cuda12.1-py3 / build (gh) (trunk failure)
C:\actions-runner\_work\pytorch\pytorch\aten\src\ATen/native/quantized/cpu/OnednnUtils.h(445): error C2065: 'vnni_available': undeclared identifier

UNSTABLE - The following job is marked as unstable, possibly due to flakiness on trunk:

trunk / libtorch-linux-focal-cuda12.4-py3.10-gcc9-debug / build (gh) (#148495)
undefined reference to std::__throw_bad_array_new_length()'`

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

pytorch-bot bot added the ciflow/inductor label

Contributor

facebook-github-bot commented Mar 5, 2025

This pull request was exported from Phabricator. Differential Revision: D69553929

facebook-github-bot added the fb-exported label

Contributor

facebook-github-bot commented Mar 5, 2025

This pull request was exported from Phabricator. Differential Revision: D69553929

zoranzhao force-pushed the export-D69553929 branch from 1c09477 to 4fdcf4d Compare

March 5, 2025 05:25

22quinn approved these changes

View reviewed changes

pytorch-bot bot added the ciflow/trunk label

Member Author

zoranzhao commented Mar 5, 2025

@pytorchbot label "topic: not user facing"

pytorch-bot bot added the topic: not user facing label

zoranzhao force-pushed the export-D69553929 branch from 4fdcf4d to 640ad27 Compare

March 5, 2025 06:54

zoranzhao added a commit to zoranzhao/pytorch that referenced this pull request


          Skip buffer in dense update (pytorch#148533)

640ad27

Summary:

as title.

PyTorch Module buffer will not be published in delta publishing.  In Quinn's previous diff, constant type annotations have been introduced.

In addition to skip constant, we also need to skip buffer if it is not found in the user-provided delta weights list

Test Plan: https://docs.google.com/document/d/1wiqUo0PyZ4g6YJIJlL_LE084ZEuE74iu74gZjqGGjWY/edit?tab=t.0#heading=h.dby6cwiw1xrn

Reviewed By: 22quinn

Differential Revision: D69553929

Contributor

facebook-github-bot commented Mar 5, 2025

This pull request was exported from Phabricator. Differential Revision: D69553929

jingsh approved these changes

View reviewed changes

zoranzhao force-pushed the export-D69553929 branch from 640ad27 to 157705f Compare

March 5, 2025 18:39

Contributor

facebook-github-bot commented Mar 5, 2025

This pull request was exported from Phabricator. Differential Revision: D69553929

zoranzhao force-pushed the export-D69553929 branch from 157705f to c84db7a Compare

March 5, 2025 18:44

zoranzhao force-pushed the export-D69553929 branch from c84db7a to 896abd9 Compare

March 5, 2025 18:47

Contributor

facebook-github-bot commented Mar 5, 2025

This pull request was exported from Phabricator. Differential Revision: D69553929

zoranzhao force-pushed the export-D69553929 branch from 896abd9 to f295549 Compare

March 5, 2025 18:52

zoranzhao force-pushed the export-D69553929 branch from f295549 to 0df6d23 Compare

March 5, 2025 22:41

linux-foundation-easycla bot commented Mar 5, 2025 •

edited

Loading

The committers listed above are authorized under a signed CLA.

✅ login: zoranzhao / name: Zhuoran Zhao (bd38921)

Contributor

facebook-github-bot commented Mar 5, 2025

This pull request was exported from Phabricator. Differential Revision: D69553929

zoranzhao force-pushed the export-D69553929 branch from 0df6d23 to be0a9c6 Compare

March 5, 2025 22:45

zoranzhao force-pushed the export-D69553929 branch from d5ac2f4 to 61ec87c Compare

March 5, 2025 23:17

zoranzhao force-pushed the export-D69553929 branch from 61ec87c to 6c1ff6b Compare

March 6, 2025 00:58

zoranzhao force-pushed the export-D69553929 branch from 6c1ff6b to a2f6516 Compare

March 6, 2025 00:59

Contributor

facebook-github-bot commented Mar 6, 2025

This pull request was exported from Phabricator. Differential Revision: D69553929

1 similar comment

Contributor

facebook-github-bot commented Mar 6, 2025

This pull request was exported from Phabricator. Differential Revision: D69553929

zoranzhao added a commit to zoranzhao/pytorch that referenced this pull request


          [AOTI] Skip buffer in dense update (pytorch#148533)

5d84204

Summary:
Pull Request resolved: pytorch#148533

as title.

PyTorch Module buffer will not be published in delta publishing.  In Quinn's previous diff, constant type annotations have been introduced.

In addition to skip constant, we also need to skip buffer if it is not found in the user-provided delta weights list

Test Plan: https://docs.google.com/document/d/1wiqUo0PyZ4g6YJIJlL_LE084ZEuE74iu74gZjqGGjWY/edit?tab=t.0#heading=h.dby6cwiw1xrn

Reviewed By: jingsh, 22quinn

Differential Revision: D69553929

zoranzhao force-pushed the export-D69553929 branch 2 times, most recently from 5d84204 to 2d1c30e Compare

March 6, 2025 04:24

Contributor

facebook-github-bot commented Mar 6, 2025

This pull request was exported from Phabricator. Differential Revision: D69553929

zoranzhao force-pushed the export-D69553929 branch from 2d1c30e to 2bcf342 Compare

March 6, 2025 06:13

Contributor

facebook-github-bot commented Mar 6, 2025

This pull request was exported from Phabricator. Differential Revision: D69553929

zoranzhao added a commit to zoranzhao/pytorch that referenced this pull request


          [AOTI] Skip buffer in dense update (pytorch#148533)

433a06a

Summary:
Pull Request resolved: pytorch#148533

as title.

PyTorch Module buffer will not be published in delta publishing.  In Quinn's previous diff, constant type annotations have been introduced.

In addition to skip constant, we also need to skip buffer if it is not found in the user-provided delta weights list

Test Plan: https://docs.google.com/document/d/1wiqUo0PyZ4g6YJIJlL_LE084ZEuE74iu74gZjqGGjWY/edit?tab=t.0#heading=h.dby6cwiw1xrn

Reviewed By: jingsh, 22quinn

Differential Revision: D69553929

zoranzhao force-pushed the export-D69553929 branch 2 times, most recently from 433a06a to 97689b1 Compare

March 6, 2025 19:41

zoranzhao added a commit to zoranzhao/pytorch that referenced this pull request


          [AOTI] Skip buffer in dense update (pytorch#148533)

97689b1

Summary:

as title.

PyTorch Module buffer will not be published in delta publishing.  In Quinn's previous diff, constant type annotations have been introduced.

In addition to skip constant, we also need to skip buffer if it is not found in the user-provided delta weights list

Test Plan: https://docs.google.com/document/d/1wiqUo0PyZ4g6YJIJlL_LE084ZEuE74iu74gZjqGGjWY/edit?tab=t.0#heading=h.dby6cwiw1xrn

Reviewed By: jingsh, 22quinn

Differential Revision: D69553929


          [AOTI] Skip buffer in dense update (pytorch#148533)

bd38921

Summary:
Pull Request resolved: pytorch#148533

as title.

PyTorch Module buffer will not be published in delta publishing.  In Quinn's previous diff, constant type annotations have been introduced.

In addition to skip constant, we also need to skip buffer if it is not found in the user-provided delta weights list

Test Plan: https://docs.google.com/document/d/1wiqUo0PyZ4g6YJIJlL_LE084ZEuE74iu74gZjqGGjWY/edit?tab=t.0#heading=h.dby6cwiw1xrn

Reviewed By: jingsh, 22quinn

Differential Revision: D69553929

Contributor

facebook-github-bot commented Mar 6, 2025

This pull request was exported from Phabricator. Differential Revision: D69553929

zoranzhao force-pushed the export-D69553929 branch from 97689b1 to bd38921 Compare

March 6, 2025 19:45

Contributor

facebook-github-bot commented Mar 7, 2025

@pytorchbot merge -i

(Initiating merge automatically since Phabricator Diff has merged, merging with -i because oss signals were bypassed internally)

pytorchmergebot added the merging label

Collaborator

pytorchmergebot commented Mar 7, 2025

Merge started

Your change will be merged while ignoring the following 5 checks: pull / win-vs2022-cpu-py3 / build, trunk / win-vs2022-cuda12.1-py3 / build, trunk / win-vs2022-cpu-py3 / build, trunk / libtorch-linux-focal-cuda12.4-py3.10-gcc9-debug / build, inductor / cuda12.4-py3.10-gcc9-sm86 / test (inductor_timm, 2, 2, linux.g5.4xlarge.nvidia.gpu)

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 added the Merged label

pytorchmergebot closed this in

dfb4094

pytorchmergebot removed the merging label

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ciflow/inductor ciflow/trunk fb-exported Merged topic: not user facing