[webkit-reviews] review requested: [Bug 29932] [CHROMIUM] Memory leak in V8EventListenerList : [Attachment 40388] Proposed fix for leak

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Wed Sep 30 16:09:56 PDT 2009


Adam Barth <abarth at webkit.org> has reassigned Stephen White
<senorblanco at chromium.org>'s request for review:
Bug 29932: [CHROMIUM] Memory leak in V8EventListenerList
https://bugs.webkit.org/show_bug.cgi?id=29932

Attachment 40388: Proposed fix for leak
https://bugs.webkit.org/attachment.cgi?id=40388&action=review

------- Additional Comments from Adam Barth <abarth at webkit.org>
I'm not sure this is right.  From the code, it looks like these vectors get
deleted when they're empty.  If they're still around, then they must not be
empty.	Are we leaking the objects contained in the vector?  Those objects seem
to be refcounted, but I don't see that we're grabbing a reference to them....

In any case, I'm probably not the right person to review this change.  We need
an expert on this code.

In general, I'd like to see this code become easier to understand.  Ideally
we'd use some of the OwnPtr / RefPtr self-documenting classes to explain the
expected ownership pattern.  A comment explaining the ownership pattern would
be useful.


More information about the webkit-reviews mailing list