NXP backend: added support for `aten.conv_transpose2d` using new MLIR flow by novak-vaclav · Pull Request #20807 · pytorch/executorch

novak-vaclav · 2026-07-09T17:03:04Z

Summary

Added support for aten.conv_transpose2d using new Neutron flow.

Test plan

tests can be manually run using pytest -c /dev/null backends/nxp/tests/

cc @robert-kalmar @JakeStevens @digantdesai @rascani @MartinPavella

pytorch-bot · 2026-07-09T17:03:08Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/20807

📄 Preview Python docs built from this PR

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

⏳ No Failures, 1 Pending

As of commit 857cb68 with merge base 43bdb7b ():
💚 Looks good so far! There are no failures yet. 💚

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

linux-foundation-easycla · 2026-07-09T17:03:12Z

❌ - login: @novak-vaclav / name: Vaclav Novak. The commit (857cb68) is not authorized under a signed CLA. Please click here to be authorized. For further assistance with EasyCLA, please visit our EasyCLA portal and chat with our support bot.

Copilot

Pull request overview

This PR extends the NXP backend’s new Neutron MLIR flow to support aten.conv_transpose2d (via aten.convolution with transposed=True), and updates the NXP test suite to validate delegation + numerical correctness across a broader set of transpose-conv shapes/parameters.

Changes:

Added a reusable Conv2dTransposedModule test model and expanded transpose-conv test coverage (delegated and non-delegated cases).
Refactored convolution delegation checks in ConvolutionConverter (regular + transposed) to use shared parameter extraction and updated Neutron constraints.
Adjusted ConvParameters to carry FX nodes for input/weight/bias and clarified handling of output_padding.

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 2 comments.

File	Description
backends/nxp/tests/models.py	Adds a `ConvTranspose2d`-based module used by backend tests.
backends/nxp/tests/ir/converter/node_converter/test_conv_converter.py	Reworks and significantly expands transpose-conv delegation/correctness tests.
backends/nxp/backend/ir/converter/node_converters/shared/conv_utils.py	Extends `ConvParameters` to include input/weight/bias FX nodes and updates `out_padding` typing/comment.
backends/nxp/backend/ir/converter/node_converters/ops_converters/convolution_converter.py	Updates delegation support checks and conversion flow to use the new `ConvParameters` structure and Neutron constraints.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

novak-vaclav · 2026-07-09T17:21:59Z

Fixed issues found by Copilot

Copilot

Pull request overview

Copilot reviewed 4 out of 4 changed files in this pull request and generated 3 comments.

Copilot

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

Copilot

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

roman-janik-nxp

Conv converter support checks are now more readable. Kudos for extensive tests. Please rabase and fix the found issues. Also 2 tests still fail on filename too long error.

Copilot

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

Copilot

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

Copilot

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

novak-vaclav · 2026-07-15T06:52:42Z

Fixed issues found by @roman-janik-nxp
Removed constraints for padding and kernel_size in CONV2D, as discussed with @robert-kalmar and the Neutron team
Refactored test_conv_converter so the tests use use_qat fixture, resulting in cleaner code (the fixture was not originally used because I needed to selectively mark either PTQ/QTA tests as xfail)

novak-vaclav · 2026-07-15T08:38:18Z

And shortened the names of the tests so pytest does not fail.

roman-janik-nxp

Good job!

Copilot

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

novak-vaclav · 2026-07-16T13:52:44Z

Thank you for the approval.
I rebased onto main. Once tests pass, I will merge.

novak-vaclav · 2026-07-16T17:05:00Z

Internal tests are passing, Github CI failures do not seem to be related to this code.
Merging.

novak-vaclav requested review from MartinPavella, irtrukhina, jirioc and roman-janik-nxp July 9, 2026 17:03

novak-vaclav requested a review from robert-kalmar as a code owner July 9, 2026 17:03

Copilot AI review requested due to automatic review settings July 9, 2026 17:03

novak-vaclav added module: nxp Issues related to NXP Neutron NPU delegation and code under backends/nxp/ release notes: nxp Changes to the NXP Neutron backend delegate labels Jul 9, 2026

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 Jul 9, 2026

Copilot started reviewing on behalf of novak-vaclav July 9, 2026 17:03 View session

Copilot AI reviewed Jul 9, 2026

View reviewed changes

Comment thread backends/nxp/backend/ir/converter/node_converters/ops_converters/convolution_converter.py Outdated

Comment thread backends/nxp/backend/ir/converter/node_converters/shared/conv_utils.py Outdated

novak-vaclav force-pushed the feature/EIEX-894-add-support-for-conv2d-transposed-using-MLIR-flow branch from d0fd211 to 83446f8 Compare July 9, 2026 17:20

Copilot AI review requested due to automatic review settings July 9, 2026 17:20

Copilot started reviewing on behalf of novak-vaclav July 9, 2026 17:21 View session

Copilot AI reviewed Jul 9, 2026

View reviewed changes

Comment thread backends/nxp/backend/ir/converter/node_converters/ops_converters/convolution_converter.py

Comment thread backends/nxp/backend/ir/converter/node_converters/ops_converters/convolution_converter.py Outdated

Comment thread backends/nxp/tests/models.py

Copilot AI review requested due to automatic review settings July 13, 2026 10:13

novak-vaclav force-pushed the feature/EIEX-894-add-support-for-conv2d-transposed-using-MLIR-flow branch from 83446f8 to b6f0fff Compare July 13, 2026 10:14

Copilot AI reviewed Jul 13, 2026

novak-vaclav force-pushed the feature/EIEX-894-add-support-for-conv2d-transposed-using-MLIR-flow branch from b6f0fff to 95f5378 Compare July 13, 2026 15:39

Copilot AI review requested due to automatic review settings July 13, 2026 15:39

Copilot AI reviewed Jul 13, 2026

roman-janik-nxp reviewed Jul 14, 2026

View reviewed changes

Comment thread backends/nxp/backend/ir/converter/node_converters/ops_converters/convolution_converter.py Outdated

Comment thread backends/nxp/tests/ir/converter/node_converter/test_conv_converter.py Outdated

novak-vaclav force-pushed the feature/EIEX-894-add-support-for-conv2d-transposed-using-MLIR-flow branch from 95f5378 to 509b206 Compare July 14, 2026 17:55

Copilot AI review requested due to automatic review settings July 14, 2026 17:55

Copilot AI reviewed Jul 14, 2026

novak-vaclav force-pushed the feature/EIEX-894-add-support-for-conv2d-transposed-using-MLIR-flow branch from 509b206 to d0197a9 Compare July 14, 2026 18:06

Copilot AI review requested due to automatic review settings July 15, 2026 06:44

novak-vaclav force-pushed the feature/EIEX-894-add-support-for-conv2d-transposed-using-MLIR-flow branch from d0197a9 to 360a4e7 Compare July 15, 2026 06:44

Copilot AI reviewed Jul 15, 2026

Copilot AI review requested due to automatic review settings July 15, 2026 06:50

novak-vaclav force-pushed the feature/EIEX-894-add-support-for-conv2d-transposed-using-MLIR-flow branch from 360a4e7 to 6423ced Compare July 15, 2026 06:50

Copilot AI reviewed Jul 15, 2026

novak-vaclav requested a review from roman-janik-nxp July 15, 2026 07:00

roman-janik-nxp approved these changes Jul 16, 2026

View reviewed changes

feat: added support for aten.conv_transpose2d using new MLIR flow

857cb68

Copilot AI review requested due to automatic review settings July 16, 2026 13:51

novak-vaclav force-pushed the feature/EIEX-894-add-support-for-conv2d-transposed-using-MLIR-flow branch from 6423ced to 857cb68 Compare July 16, 2026 13:51

Copilot AI reviewed Jul 16, 2026

novak-vaclav merged commit ab4271f into pytorch:main Jul 16, 2026
296 of 298 checks passed

novak-vaclav deleted the feature/EIEX-894-add-support-for-conv2d-transposed-using-MLIR-flow branch July 16, 2026 17:06

Uh oh!

Conversation

novak-vaclav commented Jul 9, 2026

Summary

Test plan

Uh oh!

pytorch-bot Bot commented Jul 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/20807

⏳ No Failures, 1 Pending

Uh oh!

linux-foundation-easycla Bot commented Jul 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

Uh oh!

novak-vaclav commented Jul 9, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Uh oh!

roman-janik-nxp left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Uh oh!

novak-vaclav commented Jul 15, 2026

Uh oh!

novak-vaclav commented Jul 15, 2026

Uh oh!

roman-janik-nxp left a comment

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Uh oh!

novak-vaclav commented Jul 16, 2026

Uh oh!

novak-vaclav commented Jul 16, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

pytorch-bot Bot commented Jul 9, 2026 •

edited

Loading

linux-foundation-easycla Bot commented Jul 9, 2026 •

edited

Loading