CycleCollector/CCStatus-2007-05-11: Difference between revisions
Jump to navigation
Jump to search
| Line 15: | Line 15: | ||
* [https://bugzilla.mozilla.org/show_bug.cgi?id=368774 Support cycle collection of refcounted non-xpcom objects (aka, the XBL patch)]: in dbaron's review queue | * [https://bugzilla.mozilla.org/show_bug.cgi?id=368774 Support cycle collection of refcounted non-xpcom objects (aka, the XBL patch)]: in dbaron's review queue | ||
* [https://bugzilla.mozilla.org/show_bug.cgi?id=377884 Switch nsXPConnect::Traverse to use tracing] | * [https://bugzilla.mozilla.org/show_bug.cgi?id=377884 Switch nsXPConnect::Traverse to use tracing] | ||
* [https://bugzilla.mozilla.org/show_bug.cgi?id=379718 using trace API for reference counts] | |||
== Performance requirements == | == Performance requirements == | ||
Revision as of 18:54, 11 May 2007
« previous meeting | index | next meeting »
Agenda
- status of patches in progress
- performance requirements
- next steps on performance and on memory leaks
Status of Pending CC Code in Bugzilla
Last week's next steps: Wait for patches (first 4 of 5) to land, then test more:
- Jonas will write a test case to generate cycles via document. STATUS: Patch landed and there has been a significan improvement in performance. Now we're seeing a large majority of time spent in JS GC
- Support cycle collection of refcounted non-xpcom objects (aka, the XBL patch): in dbaron's review queue
- Switch nsXPConnect::Traverse to use tracing
- using trace API for reference counts
Performance requirements
Criteria for Performance Satisfactory:
- Double the current GC time.
Next steps on leaks
- Two patches dbaron's review queue.
- After the two patches, we need to have specific test cases and look for leaks.