Fix division by zero bug #349
Merged
Conversation
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
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Outdated suggestions cannot be applied.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.
The bug was in the calculation of number of work items to use when computing FA on the CPU. In my case (maximum of 32 threads) it triggered with the GLM-4 model that has an unusually small number of KV heads (just 2). But I guess it can also trigger with a larger number of threads for more common numbers of KV heads.
Fixed by just using
max(1, nk). This will result in a far from optimal number of compute chunks, but at least it works.I'm working on a better strategy for dividing the work between the threads on this branch, but not quite ready for a PR yet.