Bug 23292

Summary: WebKit r39852 crashes on launch
Product: WebKit Reporter: Craig W. Cadwallader <bugzilla>
Component: JavaScriptCoreAssignee: Gavin Barraclough <barraclough>
Status: RESOLVED FIXED    
Severity: Critical CC: barraclough, barton.tyner, shawnzone, webkit
Priority: P2    
Version: 528+ (Nightly build)   
Hardware: Mac   
OS: OS X 10.5   
Attachments:
Description Flags
crash report text
none
crash report for r39851
none
The patch
none
The other patch
oliver: review+
crash report for r39872 none

Description Craig W. Cadwallader 2009-01-13 07:39:47 PST
WebKit nightly r39852 crashes every time attempted to launch
Comment 1 Craig W. Cadwallader 2009-01-13 07:41:28 PST
Created attachment 26668 [details]
crash report text
Comment 2 Mark Rowe (bdash) 2009-01-13 12:26:51 PST
*** Bug 23299 has been marked as a duplicate of this bug. ***
Comment 3 Craig W. Cadwallader 2009-01-13 12:40:46 PST
Created attachment 26679 [details]
crash report for r39851

crash on launch r39851
Comment 4 Craig W. Cadwallader 2009-01-13 12:43:00 PST
meant this comment - r39839 does NOT crash on launch - with Comment #3 but got cut off when attaching report - sorry:
Comment 5 Mark Rowe (bdash) 2009-01-13 13:38:38 PST
*** Bug 23301 has been marked as a duplicate of this bug. ***
Comment 6 Shawn 2009-01-13 14:52:48 PST
(In reply to comment #4)
> meant this comment - r39839 does NOT crash on launch - with Comment #3 but got
> cut off when attaching report - sorry:
> 

Reference is to r39851  and r39872,  DO Crash upon launch. Reproducible. Sent crash log, I'm new to Bugzilla, not Webkit nightlies used as the primary browser.  r39839 does NOT crash upon launch.  I reverted.
Comment 7 Gavin Barraclough 2009-01-13 16:11:48 PST
Created attachment 26686 [details]
The patch

Yep, I seem to have broken the interpreter pretty nastily in r39851.
Comment 8 Gavin Barraclough 2009-01-13 16:13:31 PST
Created attachment 26687 [details]
The other patch

Gah, - platform.h change.
Comment 9 Darin Adler 2009-01-13 16:15:13 PST
Comment on attachment 26687 [details]
The other patch

r=me

Why not & rather than &&?
Comment 10 Gavin Barraclough 2009-01-13 16:16:19 PST
Sending        JavaScriptCore/ChangeLog
Sending        JavaScriptCore/runtime/JSImmediate.h
Transmitting file data ..
Committed revision 39879.
Comment 11 Craig W. Cadwallader 2009-01-13 17:56:03 PST
Created attachment 26701 [details]
crash report for r39872
Comment 12 Craig W. Cadwallader 2009-01-13 17:57:56 PST
crash on launch continues with r39872 (see attached crash report)

Should this be entered as new bug report?
Comment 13 Gavin Barraclough 2009-01-13 18:36:19 PST
Craig,

This should be fixed in revision r39879.  If you see crashes on r39879 or above, please reopen this bug and check it's assigned to me, & I'll fix.  But fingers crossed, I think all platforms should now be good.

Many thanks,
Gavin.
Comment 14 Shawn 2009-01-13 21:44:07 PST
(In reply to comment #13)
> Craig,
> 
> This should be fixed in revision r39879.  If you see crashes on r39879 or
> above, please reopen this bug and check it's assigned to me, & I'll fix.  But
> fingers crossed, I think all platforms should now be good.
> 
> Many thanks,
> Gavin.
> 

Resolved
Comment 15 Craig W. Cadwallader 2009-01-13 22:48:24 PST
(In reply to comment #13)
> Craig,
> 
> This should be fixed in revision r39879.  If you see crashes on r39879 or
> above, please reopen this bug and check it's assigned to me, & I'll fix.  But
> fingers crossed, I think all platforms should now be good.
> 
> Many thanks,
> Gavin.
> 
Gavin,

r39879 wasn't posted to the nightlies list but r39881 and r39882 are and the problem is resolved. 

Thanks!

Craig

Comment 16 Gavin Barraclough 2009-01-14 15:08:28 PST
*** Bug 23293 has been marked as a duplicate of this bug. ***