[ET-VK][ez] Fix registration for convolution operator#13253
Merged
Conversation
Pull Request resolved: #13226 ## Context Previously, I updated the implementation of `aten.cat.default` in D76305343 (#11508) since the original implementation had a bug. The new implementation only supported up to 3 input tensors, but several models require the need for up to 6 input tensors. This diff updates the capabilities of the `concat` op so that any arbitrary number of input tensors may be accepted. ## Changes * Update implementation of the concat shader to be able to be called repeatedly, allowing support for any number of input tensors. Differential Revision: [D79893084](https://our.internmc.facebook.com/intern/diff/D79893084/) ghstack-source-id: 301766080
Pull Request resolved: #13227 ## Context Update the registration of the convolution operator to indicate that the weight tensor is prepacked and should not undergo normal texture limits checking. The current registration may cause valid convolution operators to not be partitioned since the export logic will think the weight tensor is non representable using channels packed textures. An example weight size would be something like [256, 256, 1, 1] which would result in a texture with extents [1, 1, 16384] which may exceed texture limits on some machines. Differential Revision: [D79893086](https://our.internmc.facebook.com/intern/diff/D79893086/) ghstack-source-id: 301766081
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/13253
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New Failure, 6 Pending, 1 Unrelated FailureAs of commit 72c6367 with merge base 0f70a5d ( NEW FAILURE - The following job has failed:
BROKEN TRUNK - The following job failed but were present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
meta-cla
bot
added
the
CLA Signed
SS-JIA
approved these changes
Aug 9, 2025
agrima1304
pushed a commit
to agrima1304/executorch
that referenced
this pull request
Aug 26, 2025
## Context Update the registration of the convolution operator to indicate that the weight tensor is prepacked and should not undergo normal texture limits checking. The current registration may cause valid convolution operators to not be partitioned since the export logic will think the weight tensor is non representable using channels packed textures. An example weight size would be something like [256, 256, 1, 1] which would result in a texture with extents [1, 1, 16384] which may exceed texture limits on some machines. Differential Revision: [D79893086](https://our.internmc.facebook.com/intern/diff/D79893086/)
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.
This PR was created by the merge bot to help merge the original PR into the main branch.
ghstack PR number: #13227 by @SS-JIA
^ Please use this as the source of truth for the PR details, comments, and reviews
ghstack PR base: https://github.com/pytorch/executorch/tree/gh/SS-JIA/274/base
ghstack PR head: https://github.com/pytorch/executorch/tree/gh/SS-JIA/274/head
Merge bot PR base: https://github.com/pytorch/executorch/tree/gh/SS-JIA/273/orig
Merge bot PR head: https://github.com/pytorch/executorch/tree/gh/SS-JIA/274/orig
@diff-train-skip-merge