[Webkit-unassigned] [Bug 9795] REGRESSION: Crash in [WebHTMLView(WebPrivate) _updateMouseoverWithEvent:]

[bugzilla-daemon at opendarwin.org]

http://bugzilla.opendarwin.org/show_bug.cgi?id=9795


ddkilzer at kilzer.net changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |NeedsRadar, Regression
           Priority|P2                          |P1
            Summary|Crash in                    |REGRESSION: Crash in
                   |[WebHTMLView(WebPrivate)    |[WebHTMLView(WebPrivate)
                   |_updateMouseoverWithEvent:] |_updateMouseoverWithEvent:]




------- Comment #5 from ddkilzer at kilzer.net  2006-07-11 04:10 PDT -------
(In reply to comment #4)
> I don't have any idea what "objc_msgSend_rtp" means, though.  Looks like
> a Very Bad Thing(tm) is happening with an internal ObjC method.

Doing some Google research, this points to an object either being
over-released:

http://www.cocoabuilder.com/archive/message/cocoa/2005/6/10/138462

Or being under-retained:

http://www.cocoabuilder.com/archive/message/cocoa/2005/12/24/153164

I'm not sure if this helps or not.  I guess I should review where _frame is
being retained and released to find out.  (This only happens on reloads, so it
may be some kind of race condition.)  Should I run WebKit+Safari with
MallocScribble on until it crashes again?

Since I'm seeing this bug intermittently (although I can't reproduce it on
demand), I'm marking this as a regression and a P1 with the NeedsRadar keyword.


-- 
Configure bugmail: http://bugzilla.opendarwin.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.