Closed Bug 1655636 Opened 3 years ago Closed 3 years ago

avoid reinitializing the timers in HTTP3


(Core :: Networking: HTTP, enhancement, P2)




82 Branch
Tracking Status
firefox82 --- fixed


(Reporter: CuveeHsu, Assigned: jhsu)



(Whiteboard: [necko-triaged])


(1 file)

No description provided.
Summary: Reduce redundant timers for ProcessOutputAndEvents → Reduce redundant timers in HTTP3

Will this be fixed by bug 1646024?

Thanks for referring to this.
I've been patched bug 1646024 and I still saw several SetupTimer from Read or Write before the previous one is fired.

I had a patch and will post it later. It will ease the implementation of the timer probe in bug 1655566

We might overwrite the timer accenditally. Raise the severity.

Severity: S4 → S2

SetupTimer came from [Read|Write]SegmentsAgain and could be set before the previous one fire.
Therefore, it will accenditally remove the previous one and set the new timeout by current timer implementation.

If we set a long timeout, the previous timer could be wrongly removed.

Attachment #9166731 - Attachment description: Bug 1655636 - Reduce redundant timers for ProcessOutputAndEvents, r=dragana → Bug 1655636 - avoid reinitializing the timers for ProcessOutputAndEvents, r=dragana
Summary: Reduce redundant timers in HTTP3 → avoid reinitializing the timers in HTTP3
Assignee: CuveeHsu → nobody

There's a r+ patch which didn't land and no activity in this bug for 2 weeks.
:CuveeHsu, could you have a look please?
For more information, please visit auto_nag documentation.

Flags: needinfo?(CuveeHsu)
Pushed by
avoid reinitializing the timers for ProcessOutputAndEvents, r=dragana,necko-reviewers
Closed: 3 years ago
Resolution: --- → FIXED
Target Milestone: --- → 82 Branch
Assignee: nobody → jhsu

Thanks, Dragana.

Flags: needinfo?(CuveeHsu)
You need to log in before you can comment on or make changes to this bug.