ggml-webgpu: fix workgroup dispatching for several ops#23750
Merged
Conversation
github-actions
Bot
added
ggml
CISC
approved these changes
May 27, 2026
reeselevine
approved these changes
May 27, 2026
Contributor
reeselevine
left a comment
reeselevine
left a comment
There was a problem hiding this comment.
CI failure was fixed in #23733
gabe-l-hart
added a commit
to gabe-l-hart/llama.cpp
that referenced
this pull request
May 27, 2026
* origin/master: hexagon: add support for Q4_1 in MUL_MAT and MUL_MAT_ID (ggml-org#23647) ggml-webgpu: Fix how to dispatch WG to some ops (ggml-org#23750) vulkan: Switch MUL_MAT_VEC to 4 K per iteration for F16/32 (ggml-org#22887) vulkan: use GL_NV_cooperative_matrix_decode_vector for faster matmul (ggml-org#23541) vulkan: add REPEAT op support for f16 to f16. (ggml-org#23298) ci : move ARM jobs to self-hosted + disable kleidiai mac release (ggml-org#23780) vendor : update cpp-httplib to 0.46.0 (ggml-org#23650) pyproject : add conversion folder and update dependencies (ggml-org#23746) CUDA: restrict PDL to CTK >= 12.3 due to MSVC issues (ggml-org#23742) ci : bump cuda release to 13.3 (ggml-org#23749) common : fix env names to all have LLAMA_ARG_ prefix (ggml-org#23778) ci : fix windows ccaches (ggml-org#23777) ci : remove wasm test (ggml-org#23733) vulkan: avoid preferring transfer queue on AMD UMA devices (ggml-org#22455) ci : add ccache to server builds + fix undefined sanitizer build (ggml-org#23763) docs : fix duplicated "the" in granitevision and model-conversion docs (ggml-org#23767) convert: add MiniCPM5 tokenizer support (ggml-org#23384) server : fix the log message when using SSL (ggml-org#23393)
adrianhoehne
pushed a commit
to adrianhoehne/llama.cpp
that referenced
this pull request
May 28, 2026
fewtarius
pushed a commit
to fewtarius/llama.cpp
that referenced
this pull request
May 30, 2026
turbo-tan
pushed a commit
to turbo-tan/llama.cpp-tq3
that referenced
this pull request
Jun 2, 2026
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
3 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.
Overview
This PR fixes how workgroups are dispatched for several ops.
cpy: Dispatching workgroups on a single dimension can be insufficient for the specified tensor size, so I changed it to use two dimensions. This fixes the bug described in the additional information section.mul_mat_id_gather: A single dimension is sufficient for dispatching workgroups.argsort,conv2d,im2col,upscale: Usecompute_2d_workgroupsto avoid dispatching extra workgroups.I confirmed that
test-backend-ops testfor these ops passes.Additional information
test-backend-ops perf -b WebGPU -o CPYfailed with the following error atCPY(type_src=f32,type_dst=f32,ne_src=[786432,256,1,1],permute_src=[1,0,2,3],permute_dst=[0,0,0,0],_src_transpose=0).This suggests that we need to dispatch more workgroups for
cpy, although I'm not sure which models requires this.Requirements