Skip to content

Conversation

@GregoryComer
Copy link
Member

@GregoryComer GregoryComer commented Jan 26, 2025

Summary

Add support for delegating view_copy in the XNNPACK delegate via the XNN static_reshape operator. This includes support for up to one dynamic dimension. It also includes conditional support for NHWC, so long as batch and channel are fixed (if batch or channel is modified, the reshape has to be done in the native dim order).

Add support for delegating view_copy in the XNNPACK delegate via the XNN static_reshape operator. This includes support for up to one dynamic dimension.

Test plan

I've added test coverage for view_copy, including dynamic shape support and practitioner constraints, to backends/xnnpack/test/ops/test_view_copy.py.If also added additional tests to cover the new view logic in channels_last_tagged_reshape_pass.

@GregoryComer GregoryComer added the release notes: xnnpack Changes to the XNNPack backend delegate label Jan 26, 2025
@pytorch-bot
Copy link

pytorch-bot bot commented Jan 26, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/7959

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

✅ No Failures

As of commit 7ceda56 with merge base 8af8252 (image):
💚 Looks good so far! There are no failures yet. 💚

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

@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Jan 26, 2025
@GregoryComer GregoryComer requested a review from mcr229 January 26, 2025 07:43
@facebook-github-bot
Copy link
Contributor

@GregoryComer has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.

@github-actions
Copy link

Looks like this PR hasn't been updated in a while so we're going to go ahead and mark this as Stale.
Feel free to remove the Stale label if you feel this was a mistake.
If you are unable to remove the Stale label please contact a maintainer in order to do so.
If you want the bot to never mark this PR stale again, add the no-stale label.
Stale pull requests will automatically be closed after 30 days of inactivity.

@github-actions github-actions bot added the stale PRs inactive for over 60 days label Aug 30, 2025
@github-actions
Copy link

Looks like this PR hasn't been updated in a while so we're going to go ahead and mark this as Stale.
Feel free to remove the Stale label if you feel this was a mistake.
If you are unable to remove the Stale label please contact a maintainer in order to do so.
If you want the bot to never mark this PR stale again, add the no-stale label.
Stale pull requests will automatically be closed after 30 days of inactivity.

GregoryComer added a commit to GregoryComer/executorch that referenced this pull request Nov 15, 2025
Summary:
Add support for delegating view_copy in the XNNPACK delegate via the XNN static_reshape operator. This includes support for up to one dynamic dimension. It also includes conditional support for NHWC, so long as batch and channel are fixed (if batch or channel is modified, the reshape has to be done in the native dim order).


Test Plan: I've added test coverage for view_copy, including dynamic shape support and practitioner constraints, to backends/xnnpack/test/ops/test_view_copy.py.If also added additional tests to cover the new view logic in channels_last_tagged_reshape_pass.

Differential Revision: D68691788

Pulled By: GregoryComer
@meta-codesync
Copy link

meta-codesync bot commented Nov 15, 2025

@GregoryComer has exported this pull request. If you are a Meta employee, you can view the originating Diff in D68691788.

GregoryComer added a commit to GregoryComer/executorch that referenced this pull request Dec 4, 2025
Summary:
Add support for delegating view_copy in the XNNPACK delegate via the XNN static_reshape operator. This includes support for up to one dynamic dimension. It also includes conditional support for NHWC, so long as batch and channel are fixed (if batch or channel is modified, the reshape has to be done in the native dim order).

Pull Request resolved: pytorch#7959

Test Plan: I've added test coverage for view_copy, including dynamic shape support and practitioner constraints, to backends/xnnpack/test/ops/test_view_copy.py.If also added additional tests to cover the new view logic in channels_last_tagged_reshape_pass.

Differential Revision: D68691788

Pulled By: GregoryComer
GregoryComer added a commit to GregoryComer/executorch that referenced this pull request Dec 4, 2025
Summary:
Add support for delegating view_copy in the XNNPACK delegate via the XNN static_reshape operator. This includes support for up to one dynamic dimension. It also includes conditional support for NHWC, so long as batch and channel are fixed (if batch or channel is modified, the reshape has to be done in the native dim order).


Test Plan: I've added test coverage for view_copy, including dynamic shape support and practitioner constraints, to backends/xnnpack/test/ops/test_view_copy.py.If also added additional tests to cover the new view logic in channels_last_tagged_reshape_pass.

Differential Revision: D68691788

Pulled By: GregoryComer
GregoryComer added a commit to GregoryComer/executorch that referenced this pull request Dec 4, 2025
Summary:
Add support for delegating view_copy in the XNNPACK delegate via the XNN static_reshape operator. This includes support for up to one dynamic dimension. It also includes conditional support for NHWC, so long as batch and channel are fixed (if batch or channel is modified, the reshape has to be done in the native dim order).


Test Plan: I've added test coverage for view_copy, including dynamic shape support and practitioner constraints, to backends/xnnpack/test/ops/test_view_copy.py.If also added additional tests to cover the new view logic in channels_last_tagged_reshape_pass.

Differential Revision: D68691788

Pulled By: GregoryComer
GregoryComer added a commit to GregoryComer/executorch that referenced this pull request Dec 4, 2025
Summary:
Add support for delegating view_copy in the XNNPACK delegate via the XNN static_reshape operator. This includes support for up to one dynamic dimension. It also includes conditional support for NHWC, so long as batch and channel are fixed (if batch or channel is modified, the reshape has to be done in the native dim order).


Test Plan: I've added test coverage for view_copy, including dynamic shape support and practitioner constraints, to backends/xnnpack/test/ops/test_view_copy.py.If also added additional tests to cover the new view logic in channels_last_tagged_reshape_pass.

Differential Revision: D68691788

Pulled By: GregoryComer
GregoryComer added a commit to GregoryComer/executorch that referenced this pull request Dec 4, 2025
Summary:
Add support for delegating view_copy in the XNNPACK delegate via the XNN static_reshape operator. This includes support for up to one dynamic dimension. It also includes conditional support for NHWC, so long as batch and channel are fixed (if batch or channel is modified, the reshape has to be done in the native dim order).


Test Plan: I've added test coverage for view_copy, including dynamic shape support and practitioner constraints, to backends/xnnpack/test/ops/test_view_copy.py.If also added additional tests to cover the new view logic in channels_last_tagged_reshape_pass.

Differential Revision: D68691788

Pulled By: GregoryComer
Summary:
Add support for delegating view_copy in the XNNPACK delegate via the XNN static_reshape operator. This includes support for up to one dynamic dimension. It also includes conditional support for NHWC, so long as batch and channel are fixed (if batch or channel is modified, the reshape has to be done in the native dim order).

Test Plan: I've added test coverage for view_copy, including dynamic shape support and practitioner constraints, to backends/xnnpack/test/ops/test_view_copy.py.If also added additional tests to cover the new view logic in channels_last_tagged_reshape_pass.

Differential Revision: D68691788

Pulled By: GregoryComer
@GregoryComer GregoryComer marked this pull request as ready for review December 4, 2025 06:57
@meta-codesync
Copy link

meta-codesync bot commented Dec 4, 2025

@GregoryComer has imported this pull request. If you are a Meta employee, you can view this in D68691788.

@GregoryComer GregoryComer merged commit ac4b939 into pytorch:main Dec 4, 2025
143 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. fb-exported meta-exported release notes: xnnpack Changes to the XNNPack backend delegate stale PRs inactive for over 60 days

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants