WebGPURenderer: Compute Shaders - Extend max workgroups capabilities#28846
Merged
sunag merged 2 commits intoJul 11, 2024
Conversation
📦 Bundle sizeFull ESM build, minified and gzipped.
🌳 Bundle size after tree-shakingMinimal build including a renderer, camera, empty scene, and dependencies.
|
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description
The current implementation of dispatchWorkgroups is limited to one dimension, which equals
device.limits.maxComputeWorkgroupsPerDimensionand is usually around 64k (65535).With the current implementation, the computeNode is initialized with a WorkgroupSizeX of 64, resulting in
65535 * 64 = 4,194,240.This PR increases this limit by automatically dispatching the workgroup in the second dimension, raising the default capabilities to
65535^2 * 64 = 274,869,518,400, which I believe should handle every case. 😄In the future, we could extend this by providing manual control of the X, Y, and Z dimensions for specific use cases, such as physics simulations. However, this should suffice for now, as using 3 dimensions or 1 dimension seems to yield the same performance output with most GPUs.
This contribution is funded by Utsubo