Skip to content

[ET-VK] Migrate ops to use DynamicDispatchNode #11353

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
merged 2 commits into from
Jun 4, 2025
Merged

[ET-VK] Migrate ops to use DynamicDispatchNode #11353

merged 2 commits into from
Jun 4, 2025

Conversation

pytorchbot
Copy link
Collaborator

This PR was created by the merge bot to help merge the original PR into the main branch.
ghstack PR number: #11312 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/235/base
ghstack PR head: https://github.com/pytorch/executorch/tree/gh/SS-JIA/235/head
Merge bot PR base: https://github.com/pytorch/executorch/tree/main
Merge bot PR head: https://github.com/pytorch/executorch/tree/gh/SS-JIA/235/orig
@diff-train-skip-merge

@SS-JIA
[ET-VK] Migrate ops to use DynamicDispatchNode
bc935a8 
Pull Request resolved: #11312

## Changes

* Migrate operators that are used in the llama model to use `DynamicDispatchNode` instead of `DispatchNode`

## Motivation

`DynamicDispatchNode` is a subclass of `DispatchNode` that allows dynamic selection of compute shaders, global and local work group sizing whenever the command buffer is encoded. This is critical for ensuring optimum performance when input shapes are dynamic, since it allows operators to select the best compute shader for the input conditions and also to adjust global work group sizing to launch the minimum number of work groups necessary.

Without this change, performance of llama 3.2 1B with dynamic shapes enabled is terrible (< 1 tok/s) because global work group sizing is determined based on maximum tensor sizes, which is based on the maximum sequence length. In practice, the sequence length dimension of tensors (even during the prefill phase) will not approach the maximum. This results in a lot of inactive threads launched during compute shader dispatches.
ghstack-source-id: 288057588

Differential Revision: [D75878398](https://our.internmc.facebook.com/intern/diff/D75878398/)
@pytorchbot pytorchbot requested a review from SS-JIA as a code owner June 4, 2025 05:36
@pytorch-bot PyTorch Bot
Copy link

pytorch-bot bot commented Jun 4, 2025 •
edited
Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/11353

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, 69 Pending

As of commit e666210 with merge base f50bfa9 (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 Jun 4, 2025
@github-actions GitHub Actions
Copy link

github-actions bot commented Jun 4, 2025

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.

@SS-JIA SS-JIA merged commit af0a246 into main Jun 4, 2025
95 checks passed
@SS-JIA SS-JIA deleted the gh/SS-JIA/235/orig branch June 4, 2025 05:51
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@SS-JIA SS-JIA SS-JIA approved these changes

Assignees
No one assigned
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
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@pytorchbot @SS-JIA @facebook-github-bot