- Landing of all Cycle Collector related patches by peterv.
- Impact of new hashtable implementation by Graydon.
- Discuss next steps to profile and eliminate CC issues (i.e., dedicate an entire day, on a conf call, with a profiler, to isolate CC issues).
peterv: Landed a patch for addreff and release. Landed Graydon's new hashtable implementation. New patch to remove all debugging code is being reviewed.
Get a status on the following:
- Igor landed a patch that regressed the leaks a bit. See bug 375808
- bug 375063 (Leaking nsJSContexts and nsXBLDocGlobalObjects at shutdown): This is the bug that caused Igor to file bug 375270. Will probably want to fix bug 375270 instead of landing the fix for bug 375063.
- bug 375075 (Stop leaking the cycle collector on shutdown)
- These are a bunch of safety checks that need to be removed in order to improve the performance: bug 373693 (Cycle collector makes AddRef about 6x slower) and bug 373694 (Cycle collector makes Release about 11x slower), not done yet.
- bug 377606 New hashtable implementation bug.
Status of Pending CC Code in Bugzilla
- Add cycle collection to RDF datasources: awaiting sr
- Support cycle collection of refcounted non-xpcom objects (aka, the XBL patch)
- Make cycle collection suspect all native wrapper roots: backed out.
There will be a meeting on Monday, April 23rd, starting at 9:30am with dbaron, jonas, peterv (and anyone else I can find) in Building S at Mozilla to focus on CC for a day to get these regressions taken care of. If other's want to participate, please send email to email@example.com and I'll open up any communication channel necessary for you to participate.