Skip to content

Conversation

MartinPavella
Copy link
Collaborator

Summary

The conversion of aten.view_copy to NeutronIR may require the insertion of extra Transpose operations, which may not be supported. This PR makes sure the view_copy is only delegated if the extra operations are supported too.

Test plan

Unit tests provided.

MartinPavella and others added 15 commits October 14, 2025 09:39
… node format.

The pass `RemoveGetItemPass` replaces a `max_pool2d_with_indices` node with a `max_pool2d` node, that doesn't require a GetItem afterward. The new operator must, however, preserve the original node format. Therefore, a copy of the pass was created in `backends/nxp/_passes`, where it was modified. The new directory was created, because the pass doesn't follow the `NeutronEdgePass` interface.
Before, the format inference was done during conversion to NeutronIR (after partitioning), so the partitioner didn't yet know the formats. Now, the partitioner has the format data, which can be used to accurately select nodes for delegation.
…der`, to delegate format related transpositions of input/output tensors.
…utputs to Neutron when possible.

Due to the different tensor formats used by Executorch and Neutron, the inputs/outputs often have to be transposed. This used to be done exclusively by the runtime. Now, the transpositions are done by Neutron when possible.
Before, this change was "hidden", and it was only done to some inputs/outputs.
…upported permutations.

The conversion of `aten.view_copy` to NeutronIR may sometimes require the insertion of extra `Transpose` operations, which aren't always supported. This change makes sure it doesn't happen.
@pytorch-bot
Copy link

pytorch-bot bot commented Oct 20, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/15270

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

❗ 1 Active SEVs

There are 1 currently active SEVs. If your PR is affected, please view them below:

✅ No Failures

As of commit bb49f1a with merge base 57a7903 (image):
💚 Looks good so far! There are no failures yet. 💚

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

@meta-cla meta-cla 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 Oct 20, 2025
@github-actions
Copy link

This PR needs a release notes: label

If your change should be included in the release notes (i.e. would users of this library care about this change?), please use a label starting with release notes:. This helps us keep track and include your important work in the next release notes.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "release notes: none"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

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.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants