A bunch of leak-like bugs have been reported recently, many of which look bad.
I figure it's useful to have a meta-bug because (a) some or all of them may have the same root cause (b) it'll help communicate the extent of the problem(s).
I've CC'd everyone who was CC'd to any of the dependent bugs. Please add any other leak-like bugs I might have missed.
Blocking on this to get it onto the radar of drivers (plus its blocking a bunch of blockers).
(In reply to comment #1)
> Blocking on this to get it onto the radar of drivers (plus its blocking a bunch
> of blockers).
Sigh. We need better tools for coordinating this kind of thing. I agree that release drivers should know that we still want to try to crank down memory usage a bit.
But I believe we normally don't block on meta bugs, and personally I think it's weird and confusing to 'block' on bugs that don't have a clear 'fixed' state.
No strong opinion on it. But I do think this is important.
http://hg.mozilla.org/mozilla-central/rev/0a2e06927c31 just landed for bug 630947. If this helps with any of the other bugs it'd be great to know.
Might note 620218 dealing with large memory issue also.
(In reply to comment #5)
> Might note 620218 dealing with large memory issue also.
Thanks for the pointer. That bug doesn't look leak-related, AFAICT, and so not relevant for this bug.
*** Bug 632807 has been marked as a duplicate of this bug. ***
Bug 624186 has a very simple test case which the submitter says leaks 100% of the time.
Now that Firefox 4 RC 1 has gone out, I'm closing this bug and copying all the unresolved bugs over to bug 640452, which is about leaks and quasi-leaks in Firefox 5. Please CC yourself to that bug if you're interested.
Marking as Verified Fixed as per comment 9.
This bug is being decommissioned in favour of the a new collection of
per-version leak tracking bugs. All its dependencies have been removed.
Bug 659855 is the replacement bug.