[10.8] crash in libunwind

RESOLVED WORKSFORME

Status

()

Core
General
--
critical
RESOLVED WORKSFORME
5 years ago
5 years ago

People

(Reporter: Usul, Unassigned)

Tracking

({crash, regression, topcrash})

15 Branch
x86_64
Mac OS X
crash, regression, topcrash
Points:
---

Firefox Tracking Flags

(Not tracked)

Details

(Whiteboard: [gs][tbird topcrash][startupcrash], crash signature, URL)

(Reporter)

Description

5 years ago
This bug was filed from the Socorro interface and is 
report bp-793d6f79-f291-45c3-ac75-4cc0a2120910 .
============================================================= 

0 	libunwind.dylib 	libunwind.dylib@0x1610 	
1 	libunwind.dylib 	libunwind.dylib@0x158e 	
2 	libunwind.dylib 	libunwind.dylib@0x150d 	
3 	libobjc.A.dylib 	libobjc.A.dylib@0xb82e 	
4 	CoreFoundation 	CoreFoundation@0x12796f 	
5 	AppKit 	AppKit@0x19f62f 	
6 	AppKit 	AppKit@0xa2c5b5 	
7 	AppKit 	AppKit@0x19f5e1 	
8 	AppKit 	AppKit@0x19f62f 	
9 	AppKit 	AppKit@0x19f5e1 	
10 	CoreFoundation 	CoreFoundation@0x43e0f 	
11 	CoreFoundation 	CoreFoundation@0x4426f 	
12 	libssl.0.9.8.dylib 	libssl.0.9.8.dylib@0x37fff 	
13 	CoreFoundation 	CoreFoundation@0x43e0f 	
14 	CoreFoundation 	CoreFoundation@0x4426f 	
15 	libssl.0.9.8.dylib 	libssl.0.9.8.dylib@0x37fff 	
16 	AppKit 	AppKit@0x158ce7 	
17 	CoreFoundation 	CoreFoundation@0x44251 	
18 	libmozglue.dylib 	je_calloc 	_string.h:80
19 		@0x3 	
20 	ColorSync 	ColorSync@0xa603a 	
21 	libCSync.A.dylib 	libCSync.A.dylib@0xb04a 	
22 	ColorSync 	ColorSync@0xa6055 	
23 	libsystem_c.dylib 	libsystem_c.dylib@0x19865 	
24 	libsystem_c.dylib 	libsystem_c.dylib@0x1a174 	
25 	ColorSync 	ColorSync@0x19b35 	
26 	CoreFoundation 	CoreFoundation@0xfa66 	
27 	libmozglue.dylib 	je_free 	memory/jemalloc/jemalloc.c:1674
(Reporter)

Updated

5 years ago
Blocks: 790234
These have a strong correlation with "Ubiquity":
100% (191/191) vs.  24% (199/840) Ubiquity

http://en.wikipedia.org/wiki/Ubiquity_%28Firefox%29

Updated

5 years ago
Keywords: regression
Hardware: All → x86_64
Summary: crash in libunwind → [10.8] crash in libunwind

Comment 2

5 years ago
me,too.
My Clash report here.
https://crash-stats.mozilla.com/report/index/bp-e853a4c9-2f3d-4508-b3cb-a2f8e2120912

Comment 3

5 years ago
Tb 15.0.0 on OS X 10.8.1 Everything's gonna be alright.
Tb 15.0.1 on OS X 10.8.1 Always Crash.

Updated

5 years ago
Whiteboard: [gs

Comment 4

5 years ago
no seamonkey crashes with libunwind, fwiw
Keywords: topcrash
Whiteboard: [gs → [gs][tbird topcrash][startupcrash]
(In reply to Youhei Tooyama from comment #3)
> Tb 15.0.0 on OS X 10.8.1 Everything's gonna be alright.
> Tb 15.0.1 on OS X 10.8.1 Always Crash.

Youhei, are you using fresh installs, or is 15.0.1 updated from 15.0? I would be very interested to know what happens with fresh installs, if you've just been updating from 15.0.
why doesn't nsLookAndFeel.mm use try-catch block?

Comment 7

5 years ago
(In reply to Mark Banner (:standard8) from comment #5)
> (In reply to Youhei Tooyama from comment #3)
> > Tb 15.0.0 on OS X 10.8.1 Everything's gonna be alright.
> > Tb 15.0.1 on OS X 10.8.1 Always Crash.
> 
> Youhei, are you using fresh installs, or is 15.0.1 updated from 15.0? I
> would be very interested to know what happens with fresh installs, if you've
> just been updating from 15.0.

Yes,
1.Install Tb 15.
2.Update Tb 15 to 15.0.1 ->boot Crash Reporter

and 
1.Install Tb 15.
2.Override Tb 15.0.1 ->boot Crash Reporter
Youhei,

Would you be able to give this build a try? It is based on 15.0.1 with a couple of patches removed to take it back towards 15.0 a little bit.

http://ftp.mozilla.org/pub/mozilla.org/thunderbird/try-builds/bugzilla@standard8.plus.com-e5945169b9f9/try-comm-central-macosx64/thunderbird-15.0.1.en-US.mac.dmg

Comment 9

5 years ago
Youhei, do you find this to be fixed in version 16 ?
Flags: needinfo?(mozilla.dancho)

Comment 10

5 years ago
Sorry.
Not reproduce the problem now.

Not reproduced:
Thunderbird 15.0.1 + OS X 10.8.2
Thunderbird 16 +  OS X 10.8.2

Thanks.
Flags: needinfo?(mozilla.dancho)
Yeah, unfortunately we don't know what fixed it, but it definitely seems to have gone away now. Resolving as WFM, we can always reopen/file a bug if it happens again.
Status: NEW → RESOLVED
Last Resolved: 5 years ago
Resolution: --- → WORKSFORME
You need to log in before you can comment on or make changes to this bug.