ggml webgpu: minor set rows optimization#16810
Merged
CISC merged 9 commits intoggml-org:masterfrom Nov 5, 2025
Merged
Conversation
* updated optimization, fixed errors * non vectorized version now dispatches one thread per element * Simplify * Change logic for set_rows pipelines --------- Co-authored-by: Neha Abbas <nehaabbas@macbookpro.lan> Co-authored-by: Neha Abbas <nehaabbas@ReeseLevines-MacBook-Pro.local> Co-authored-by: Reese Levine <reeselevine1@gmail.com>
github-actions
bot
added
the
ggml
slaren
reviewed
Nov 1, 2025
| std::vector<ggml_tensor *> expert_views(n_expert_used); | ||
| for (int64_t i = 0; i < n_expert_used; ++i) { | ||
| expert_views[i] = ggml_view_2d(ctx, weighted, n_embd, n_tokens, weighted->nb[2], i * weighted->nb[1]); | ||
| expert_views[i] = ggml_view_2d(ctx, weighted, n_embd, n_tokens, weighted->nb[1], i * weighted->nb[1]); |
Member
There was a problem hiding this comment.
I don't think this change is correct.
Contributor
Author
There was a problem hiding this comment.
This update was testing some changes to the addition kernels in response to the discussion in #16857. But, it looks like the CUDA CI is failing with this change too, so if it's confirmed that the nb[2] is correct here I'll need to do a little more debugging to understand why the WebGPU add op is failing as currently written. I'll mark this PR as a draft for now to avoid it accidentally being merged.
Contributor
Author
There was a problem hiding this comment.
Quick update: Realized this due to the non-contiguity in the view tensors, which isn't supported yet in the kernels. I disabled support for non-contiguous tensors here and added a note so it can be added in the future.
This reverts commit ed710b3.
…or future support
Contributor
Author
CISC
approved these changes
Nov 5, 2025
gabe-l-hart
added a commit
to gabe-l-hart/llama.cpp
that referenced
this pull request
Nov 5, 2025
* origin/master: (21 commits) vulkan: Fix GGML_VULKAN_CHECK_RESULTS to better handle fusion (ggml-org#16919) examples(gguf): GGUF example outputs (ggml-org#17025) mtmd: allow QwenVL to process larger image by default (ggml-org#17020) server : do not default to multiple slots with speculative decoding (ggml-org#17017) mtmd: improve struct initialization (ggml-org#16981) docs: Clarify the endpoint that webui uses (ggml-org#17001) model : add openPangu-Embedded (ggml-org#16941) ggml webgpu: minor set rows optimization (ggml-org#16810) sync : ggml ggml : fix conv2d_dw SVE path (ggml/1380) CUDA: update ops.md (ggml-org#17005) opencl: update doc (ggml-org#17011) refactor: replace sprintf with snprintf for safer string handling in dump functions (ggml-org#16913) vulkan: remove the need for the dryrun (ggml-org#16826) server : do context shift only while generating (ggml-org#17000) readme : update hot topics (ggml-org#17002) ggml-cpu : bicubic interpolation (ggml-org#16891) ci : apply model label to models (ggml-org#16994) chore : fix models indent after refactor (ggml-org#16992) Fix garbled output with REPACK at high thread counts (ggml-org#16956) ...
Anico2
added a commit
to Anico2/llama.cpp
that referenced
this pull request
Jan 15, 2026
* Add buffer label and enable dawn-specific toggles to turn off some checks * Minor set_rows optimization (ggml-org#4) * updated optimization, fixed errors * non vectorized version now dispatches one thread per element * Simplify * Change logic for set_rows pipelines --------- Co-authored-by: Neha Abbas <nehaabbas@macbookpro.lan> Co-authored-by: Neha Abbas <nehaabbas@ReeseLevines-MacBook-Pro.local> Co-authored-by: Reese Levine <reeselevine1@gmail.com> * Comment on dawn toggles * Remove some comments * Implement overlap binary operators * Revert "Implement overlap binary operators" This reverts commit ed710b3. * Disable support for non-contiguous binary_op tensors and leave note for future support --------- Co-authored-by: neha-ha <137219201+neha-ha@users.noreply.github.com> Co-authored-by: Neha Abbas <nehaabbas@macbookpro.lan> Co-authored-by: Neha Abbas <nehaabbas@ReeseLevines-MacBook-Pro.local>
blime4
referenced
this pull request
in blime4/llama.cpp
Feb 5, 2026
* Add buffer label and enable dawn-specific toggles to turn off some checks * Minor set_rows optimization (#4) * updated optimization, fixed errors * non vectorized version now dispatches one thread per element * Simplify * Change logic for set_rows pipelines --------- Co-authored-by: Neha Abbas <nehaabbas@macbookpro.lan> Co-authored-by: Neha Abbas <nehaabbas@ReeseLevines-MacBook-Pro.local> Co-authored-by: Reese Levine <reeselevine1@gmail.com> * Comment on dawn toggles * Remove some comments * Implement overlap binary operators * Revert "Implement overlap binary operators" This reverts commit ed710b36f51ab3f53fa13db15c1685dc8678a32a. * Disable support for non-contiguous binary_op tensors and leave note for future support --------- Co-authored-by: neha-ha <137219201+neha-ha@users.noreply.github.com> Co-authored-by: Neha Abbas <nehaabbas@macbookpro.lan> Co-authored-by: Neha Abbas <nehaabbas@ReeseLevines-MacBook-Pro.local>
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
4 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.
SET_ROWSby having multiple threads work on each row, as well as vectorizationBetter matrix multiplication coming soon!