Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix the task count check in TrafficController #11783

Merged
merged 6 commits into from
Dec 5, 2024

Conversation

jihoonson
Copy link
Collaborator

A follow-up to #11730. If that is the only task at the moment, the TrafficController should allow the task to process regardless of how much host memory it uses. However, this check was missing in the loop below, which would have all subsequent tasks hung if a task that uses a large memory is submitted while there are some tasks running.

        while (!throttle.canAccept(task)) { // <= task count check is missing
          condition.await()
        }

This PR fixes this bug. Additionally, it improves the TrafficController to use the condition variable instead of polling as suggested in #11730 (comment). Finally, the terminology in ThrottlingExecutorSuite has been fixed as suggested in #11730 (comment).

@jihoonson
Copy link
Collaborator Author

build

1 similar comment
@jihoonson
Copy link
Collaborator Author

build

@jihoonson jihoonson force-pushed the fix-traffic-controller branch from c64ee28 to b4b56a0 Compare November 27, 2024 18:34
@jihoonson
Copy link
Collaborator Author

build

@jihoonson
Copy link
Collaborator Author

build

revans2
revans2 previously approved these changes Dec 2, 2024
@jihoonson
Copy link
Collaborator Author

build

@jihoonson
Copy link
Collaborator Author

@gerashegalov @revans2 would you please have another look?

@jihoonson jihoonson merged commit f3ac8be into NVIDIA:branch-25.02 Dec 5, 2024
49 checks passed
@sameerz sameerz added the bug Something isn't working label Dec 11, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants