Repository navigation
Fix: Handle unaligned writes in ggml_backend_webgpu_buffer_set_tensor - #29471
Merged
Merged
Conversation
github-actions
Bot
added
ggml
yomaytk
approved these changes
Sep 28, 2026
yomaytk
left a comment
yomaytk
left a comment
Member
There was a problem hiding this comment.
The change looks good to me, thanks!
But the formatting CI fails, so could you format the target file? We recommend to use clang-format of version 22 or later.
Member
|
With the newly added tests, there is a new error for webgpu: |
Member
Ok, I'll look into it tomorrow. |
pierreguillot
pushed a commit
to Ircam-Partiels/llama.cpp
that referenced
this pull request
Oct 1, 2026
…sor (ggml-org#29471) * Fix: Handle unaligned writes in ggml_backend_webgpu_buffer_set_tensor * Clang formatting
1 task
frostyautumnleaf
pushed a commit
to frostyautumnleaf/llama.cpp
that referenced
this pull request
Oct 5, 2026
…sor (ggml-org#29471) * Fix: Handle unaligned writes in ggml_backend_webgpu_buffer_set_tensor * Clang formatting
edwardyoon
pushed a commit
to edwardyoon/focus-llama
that referenced
this pull request
Oct 7, 2026
…sor (ggml-org#29471) * Fix: Handle unaligned writes in ggml_backend_webgpu_buffer_set_tensor * Clang formatting (cherry picked from commit f00a64c)
pratiknarola-t
added a commit
to tetherto/qvac-fabric-llm.cpp
that referenced
this pull request
Oct 8, 2026
…nsor (upstream ggml-org#29471) Port of ggml-org/llama.cpp f00a64c.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
CI tests are failing consistently with an unaligned memory write for webgpu. Example here: https://github.com/ggml-org/llama.cpp/actions/runs/36205666525/job/108318129946?pr=27851.
This fix mirrors the tail section of ggml_backend_webgpu_buffer_set_tensor, which already checks if length % 4 != 0, with an analogous head section which checks if offset % 4 != 0.
Additional information
Requirements