CUDA: fix misaligned synchronization in FA #13469
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
This was referenced May 12, 2025
github-actions
bot
added
Nvidia GPU
CISC
approved these changes
May 12, 2025
Nexesenex
added a commit
to Nexesenex/croco.cpp
that referenced
this pull request
May 14, 2025
This reverts commit 95e1888.
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Outdated suggestions cannot be applied.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.
See discussion following #13430 (comment) .
I added a
__syncthreadsinstruction in #13438 to fix a race condition. However, this instruction is for some kernel configuration only executed for part of the warps so the points at which warps wait for each other become misaligned. On a sufficiently long context this then causes incorrect results.