1. Open Safari (with nightly r24524) in Windows XP (rest of Safari used was 3.02 from the last official beta). 2. Go to URL http://www.elfa.se/se/ (Scandinavian electronics component vendor) 3. Safari immediately crashes, wanting to send bug report to Microsoft. This does not happen with latest official beta. Notes: Windows crash reports says the error occurs in safari.exe AppVer: 3.522.13.1 Offset 000607ad. From the detailed info: Exception code 0xc0000005, Flags: 0, Record: 0.
Does not crash with a local debug build of WebKit r24513 with Safari 3.0 (522.12) on Mac OS X 10.4.10 (8R218).
(In reply to comment #1) > Does not crash with a local debug build of WebKit r24513 with Safari 3.0 > (522.12) on Mac OS X 10.4.10 (8R218). Also works with a local debug build of WebKit r24534. Adding PlatformOnly keyword.
Thanks again for the bug report, Greger! Could you post a Dr. Watson crash log as well? Instructions on how to do that are here: http://webkit.org/quality/crashlogs.html
The site uses frames so I bet it's bug 14705 or at least related.
Created attachment 15651 [details] Dr. Watson log file (Windows XP) I used the default Dr. Watson settings: Crash Dump Type: Mini, Dump All Threads Content, Create Crash Dump File, Number of Instructions: 10, Number of Errors To Save: 10.
(In reply to comment #5) > Created an attachment (id=15651) [edit] > Dr. Watson log file (Windows XP) > > I used the default Dr. Watson settings: Crash Dump Type: Mini, Dump All Threads > Content, Create Crash Dump File, Number of Instructions: 10, Number of Errors > To Save: 10. Thanks, Greger! Can you please also attach your user.dmp file (as described on <http://webkit.org/quality/crashlogs.html>)?
Created attachment 15652 [details] Dr. Watson Dump File (Windows XP)
(In reply to comment #7) > Created an attachment (id=15652) [edit] > Dr. Watson Dump File (Windows XP) > Heh, you're fast (or I'm trigger-happy) :-)
(In reply to comment #8) > (In reply to comment #7) > > Created an attachment (id=15652) [edit] > > Dr. Watson Dump File (Windows XP) > > > > Heh, you're fast (or I'm trigger-happy) :-) > Yeah, well, I actually didn't read your comment before uploading the second file, so I guess you qualify as a fast shooter :-) I hope the logs help though!
I don't know if this is interesting, but I checked some older nightlies as well and this bug was introduced between r24441 and r24501 (which would give you about 60 check-ins to cover, right?).
(In reply to comment #10) > I don't know if this is interesting, but I checked some older nightlies as well > and this bug was introduced between r24441 and r24501 (which would give you > about 60 check-ins to cover, right?). That's very helpful! I've noted it in the bug title.
Most likely a dupe of bug 14705, resulting from <http://trac.webkit.org/projects/webkit/changeset/24490>.
(In reply to comment #12) > Most likely a dupe of bug 14705, resulting from > <http://trac.webkit.org/projects/webkit/changeset/24490>. > I saw that a preliminary fix for bug 14705 has been included in r24749, and I can confirm that this build works (for me) for the site in question.
*** This bug has been marked as a duplicate of 14705 ***