[WebGPU] Fix UI jank caused by compositor starvation in batched dispatch#19418
Open
ssam18 wants to merge 2 commits intoapache:mainfrom
Open
[WebGPU] Fix UI jank caused by compositor starvation in batched dispatch#19418ssam18 wants to merge 2 commits intoapache:mainfrom
ssam18 wants to merge 2 commits intoapache:mainfrom
Conversation
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Code Review
This pull request introduces a maxDispatchesPerFlush property to the WebGPUContext to enable periodic command flushing during dispatches. A review comment points out a logic discrepancy where setting the value to 0 results in unlimited batching rather than the intended per-dispatch flushing, and provides a suggestion to correct the conditional check.
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Batching all dispatches into a single command buffer (introduced in #18871) blocks the browser compositor from interleaving its rendering work, causing visible frame drops and laggy scrolling during LLM inference. This adds a
maxDispatchesPerFlushfield (default 32) toWebGPUContextthat flushes the encoder periodically, giving the compositor regular breathing room while preserving most of the batching throughput gains. Setting it to 0 restores the old per-dispatch behavior. Fixes #19342