Skip to content

Conversation

@STREM-hexagon
Copy link

Fixes a race condition discovered by the compute sanitizer. The writer threads might enter the next loop iteration before all reader threads have read their values. Adding __syncthreads() to avoid this.

@yuanknv
Copy link

yuanknv commented Aug 13, 2025

Thanks for catching and fixing this, it looks good to me. We will include this fix in the next patch for release-3.2.

@yuanknv yuanknv self-requested a review August 13, 2025 23:23
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants