Cortex-M: avoid redundant convolution and pooling work - #22859
Merged
Merged
Conversation
Contributor
Author
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/22859
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit f420caa with merge base 026ca3f ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
meta-cla
Bot
added
the
CLA Signed
This was referenced Sep 15, 2026
This PR needs a
|
rascani
marked this pull request as ready for review
September 16, 2026 17:29
rascani
added a commit
that referenced
this pull request
Sep 16, 2026
For MVE convolutions with one input channel and more than eight output channels, emit regular-convolution weights at export. CMSIS-NN already repacks these depthwise weights and runs regular convolution on each inference, so this avoids repeated conversion and its temporary storage. Preserve depthwise dispatch for scalar/DSP and smaller channel counts. Also omit empty average-pooling pads. Update DS-CNN's convolution and padding expectations and cover the MVE dispatch threshold and pooling cases. Authored with AI assistance from Codex. ghstack-source-id: 5910700 ghstack-comment-id: 5689270859 Pull-Request: #22859
JakeStevens
approved these changes
Sep 16, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
For MVE convolutions with one input channel and more than eight output channels, emit regular-convolution weights at export. CMSIS-NN already repacks these depthwise weights and runs regular convolution on each inference, so this avoids repeated conversion and its temporary storage. Preserve depthwise dispatch for scalar/DSP and smaller channel counts. Also omit empty average-pooling pads.
Update DS-CNN's convolution and padding expectations and cover the MVE dispatch threshold and pooling cases.
Authored with AI assistance from Codex.