ggml webgpu: fix workgroup dispatch limit for large batch sizes#19965
Merged
reeselevine merged 5 commits intoggml-org:masterfrom Mar 3, 2026
Merged
Conversation
WebGPU limits workgroup sizes to 65535 per dimension. Large MUL_MAT operations with batch sizes exceedeing this limi would fail. * add compute_2d_workgroups() helper to split total workgroup ID across X/Y dimensions * update mul_mat_reg_tile.wgsl to reconstruct linear workgroup ID from 2D dispatch * update mul_mat_subgroup_matrix.wgsl to reconstruct linear workgroup ID from 2D dispatch * update mul_mat.wgsl to compute global index from 2D workgroup coordinates * refactor all three mul_mat dispatch paths to use the shared helper
2D workgroup dispatch can over-dispatch when total workgroups don't divide evenly into the 65535 per-dimension limit. Extra workgroups would compute invalid batch indices, causing memory corruption. * add batch_idx bound check to mul_mat_reg_tile.wgsl and mul_mat_subgroup_matrix.wgsl to prevent over-dispatched workgroups from accessing invalid memory * fixes test failures with large batch sizes (eg., bs=[128, 1024])
github-actions
bot
added
the
ggml
reeselevine
reviewed
Feb 28, 2026
Collaborator
reeselevine
left a comment
reeselevine
left a comment
There was a problem hiding this comment.
Glad this turned out to be a simple fix! We should soon implement a way to avoid launching too many workgroups in these cases too, since Safari does seem to limit the allowed number more aggressively than Chrome.
reeselevine
approved these changes
Mar 3, 2026
Collaborator
reeselevine
left a comment
reeselevine
left a comment
There was a problem hiding this comment.
The failing WebGPU CI is unrelated to this change, it's an issue introduced in #19772, which @nikhilJain17 is working on fixing separately.
ArberSephirotheca
pushed a commit
to ArberSephirotheca/llama.cpp
that referenced
this pull request
Mar 3, 2026
…-org#19965) * ggml-webgpu: fix workgroup dispatch limit for large batch sizes WebGPU limits workgroup sizes to 65535 per dimension. Large MUL_MAT operations with batch sizes exceedeing this limi would fail. * add compute_2d_workgroups() helper to split total workgroup ID across X/Y dimensions * update mul_mat_reg_tile.wgsl to reconstruct linear workgroup ID from 2D dispatch * update mul_mat_subgroup_matrix.wgsl to reconstruct linear workgroup ID from 2D dispatch * update mul_mat.wgsl to compute global index from 2D workgroup coordinates * refactor all three mul_mat dispatch paths to use the shared helper * ggml-webgpu: add bounds checking for over-dispatched workgroups 2D workgroup dispatch can over-dispatch when total workgroups don't divide evenly into the 65535 per-dimension limit. Extra workgroups would compute invalid batch indices, causing memory corruption. * add batch_idx bound check to mul_mat_reg_tile.wgsl and mul_mat_subgroup_matrix.wgsl to prevent over-dispatched workgroups from accessing invalid memory * fixes test failures with large batch sizes (eg., bs=[128, 1024]) * ggml-webgpu: add back TODO for spliting large sizes into batches * Optimize 2d workgroup provisioning * Set some parameters that increase speed --------- Co-authored-by: Reese Levine <reeselevine1@gmail.com>
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.
WebGPU limits workgroup counts to 65535 per dimension. MUL_MAT operations with batch sizes exceeding this limit would fail or corrupt memory.
This PR implements 2D workgroup dispatch to handle arbitrary batch sizes:
compute_2d_workgroups()helper to split workgroups across X/Y dimensions when exceeding the 65535 limit