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

Fixing after 588 hang #1048

Closed
wants to merge 23 commits into from

Conversation

Projects
None yet
3 participants
@sithhell
Copy link
Member

commented Jan 17, 2014

This sequence of commits tries to get of the hangs in the after_588 regression test that occur from time to time.

Trying to fix the hang in the after_588 test:
Currently, the incref/decref requests have a cascade of different
locks which might lead to HPX threads getting suspended while other locks are held.
Those have been mostly "fixed", but now a "negative entry in reference count table"
exception occurs.
@hkaiser

This comment has been minimized.

Copy link
Member

commented on src/runtime/agas/addressing_service.cpp in 3cc95d0 Jan 15, 2014

This change looks valid

@hkaiser

This comment has been minimized.

Copy link
Member

commented on src/runtime/agas/incref_requests.cpp in 3cc95d0 Jan 15, 2014

Could you explain this code, please (perhaps add a comment)?

This comment has been minimized.

Copy link
Member Author

replied Jan 15, 2014

TBH, I have no clue why this was needed. The reason I added it was because I ran into the assert in the else branch. This might have been due to the race condition you described below.

@hkaiser

This comment has been minimized.

Copy link
Member

commented on src/runtime/naming/name.cpp in 3cc95d0 Jan 15, 2014

Moving this unlock will create a possible race conditions as the newly assigned credit will potentially get split before the incref handling is notified of the pending incref request. I ran into this one before...

@hkaiser

This comment has been minimized.

Copy link
Member

commented Jan 17, 2014

Please remove the special spinlock for gid_types (as discussed) and perhaps collapse all commits into a single one.

hkaiser added a commit that referenced this pull request Jan 17, 2014

@hkaiser hkaiser closed this Jan 17, 2014

@sithhell sithhell deleted the fixing_after_588_hang branch Jan 29, 2014

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
You can’t perform that action at this time.