ggml-cpu: fix data race for debug asserts#20148
Merged
JohannesGaessler merged 1 commit intoggml-org:masterfrom Mar 6, 2026
Merged
ggml-cpu: fix data race for debug asserts#20148JohannesGaessler merged 1 commit intoggml-org:masterfrom
JohannesGaessler merged 1 commit intoggml-org:masterfrom
Conversation
github-actions
Bot
added
the
ggml
ggerganov
approved these changes
Mar 6, 2026
bartowski1182
pushed a commit
to bartowski1182/llama.cpp
that referenced
this pull request
Mar 10, 2026
Ethan-a2
pushed a commit
to Ethan-a2/llama.cpp
that referenced
this pull request
Mar 20, 2026
Seunghhon
pushed a commit
to Seunghhon/llama.cpp
that referenced
this pull request
Apr 26, 2026
rsenthilkumar6
pushed a commit
to rsenthilkumar6/llama.cpp
that referenced
this pull request
May 1, 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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
While debugging the CI for #19802 I noticed that for >= 2 threads there seems to be a data race in the asserts that are added in debug builds to the CPU backend. In debug builds some ops check whether the output tensor is NaN but this wasn't correctly adapted in some places to consider the
i2andi3indices of the actual computation. So one thread can end up checking another thread's data, leading to a data race. There is no effect on release builds.