Bug 12264

Summary: REGRESSION (r18843): Webkit crashes on 43Folders.com home page
Product: WebKit Reporter: Patricia Warwick <pwarwick>
Component: New BugsAssignee: Nobody <webkit-unassigned>
Status: RESOLVED FIXED    
Severity: Normal CC: aroben, daniele.metilli, ddkilzer, mitz
Priority: P1 Keywords: HasReduction, Regression
Version: 420+   
Hardware: Mac   
OS: OS X 10.4   
Attachments:
Description Flags
Crash log
none
Candidate patch aroben: review+

Patricia Warwick
Reported 2007-01-14 06:09:21 PST
When I load www.43folders.com it causes Webkit to crash towards the end of writing the page.
Attachments
Crash log (21.17 KB, text/plain)
2007-01-14 06:09 PST, Patricia Warwick
no flags
Candidate patch (13.18 KB, patch)
2007-01-14 07:53 PST, mitz
aroben: review+
Patricia Warwick
Comment 1 2007-01-14 06:09:58 PST
Created attachment 12425 [details] Crash log
mitz
Comment 2 2007-01-14 07:03:32 PST
Reduction: <iframe><script></script></iframe>
mitz
Comment 3 2007-01-14 07:53:22 PST
Created attachment 12427 [details] Candidate patch This patch fixes the crash and changes WebKit's behavior in the included test to match Firefox and Opera. I DID NOT run most of the layout tests with this patch.
David Kilzer (:ddkilzer)
Comment 4 2007-01-14 08:50:10 PST
(In reply to comment #2) > Reduction: > <iframe><script></script></iframe> In the patch, the above looks like this (on purpose?): <iframe><script></iframe>PASS</script></iframe>
mitz
Comment 5 2007-01-14 08:55:41 PST
(In reply to comment #4) > (In reply to comment #2) > > Reduction: > > <iframe><script></script></iframe> > > In the patch, the above looks like this (on purpose?): > > <iframe><script></iframe>PASS</script></iframe> > Yes, the latter both crashes in TOT and demonstrates the incorrect parsing prior to r18843 (and the correct parsing when the patch is applied).
Adam Roben (:aroben)
Comment 6 2007-01-14 12:35:44 PST
Comment on attachment 12427 [details] Candidate patch r=me, although it would be great to add some ASSERTs where n is used just to make sure this doesn't happen again in the future.
mitz
Comment 7 2007-01-14 15:08:04 PST
*** Bug 12271 has been marked as a duplicate of this bug. ***
David Kilzer (:ddkilzer)
Comment 8 2007-01-14 16:51:55 PST
*** Bug 12274 has been marked as a duplicate of this bug. ***
David Kilzer (:ddkilzer)
Comment 9 2007-01-14 16:58:12 PST
This was causing too many crashes on public sites to hold up. I'll trust Adam or Mitz will go back to add the ASSERT statements. Committed revision 18848.
David Kilzer (:ddkilzer)
Comment 10 2007-01-14 16:59:28 PST
*** Bug 12266 has been marked as a duplicate of this bug. ***
Note You need to log in before you can comment on or make changes to this bug.