544640 - Firefox 3.6 Crash Report [@ XPCWrappedNative::GetNewOrUsed(XPCCallContext&, nsISupports*, XPCWrappedNativeScope*, XPCNativeInterface*, nsWrapperCache*, int, XPCWrappedNative**) ]

Status: RESOLVED WORKSFORME

(Core :: XPConnect, defect)

Description

[Carsten Book [:Tomcat]]

Currently top #14 in the Top Crash List +190 positions up 

http://crash-stats.mozilla.com/report/list?range_value=2&range_unit=weeks&signature=XPCWrappedNative%3A%3AGetNewOrUsed%28XPCCallContext%26%2C%20nsISupports*%2C%20XPCWrappedNativeScope*%2C%20XPCNativeInterface*%2C%20nsWrapperCache*%2C%20int%2C%20XPCWrappedNative**%29&version=Firefox%3A3.6

seems in some cases a start up crash - on all platforms

Comment 1

[chris hofmann]

this might be a variation of the yslow crash in bug 542203.  a lot of the stack looks the same

its hard to understand why it should be increasing so quickly on 3.6.

Frame  	Module  	Signature [Expand]  	Source
0 	xul.dll 	XPCWrappedNative::GetNewOrUsed 	js/src/xpconnect/src/xpcwrappednative.cpp:340
1 	xul.dll 	XPCWrappedNative::GetNewOrUsed 	js/src/xpconnect/src/xpcwrappednative.cpp:454
2 	xul.dll 	XPCConvert::NativeInterface2JSObject 	js/src/xpconnect/src/xpcconvert.cpp:1199
3 	xul.dll 	XPCConvert::NativeData2JS 	js/src/xpconnect/src/xpcconvert.cpp:471
4 	xul.dll 	XPCConvert::NativeData2JS 	js/src/xpconnect/src/xpcprivate.h:2974
5 	xul.dll 	XPCWrappedNative::CallMethod 	js/src/xpconnect/src/xpcwrappednative.cpp:2809
6 	xul.dll 	XPC_WN_GetterSetter 	js/src/xpconnect/src/xpcwrappednativejsops.cpp:1784
7 	js3250.dll 	js_Invoke 	js/src/jsinterp.cpp:1360
8 	js3250.dll 	js_InternalInvoke 	js/src/jsinterp.cpp:1423
9 	js3250.dll 	js_GetPropertyHelper 	js/src/jsobj.cpp:4277
10 	js3250.dll 	js_Interpret 	js/src/jsops.cpp:1520
11 	js3250.dll 	js_Invoke 	js/src/jsinterp.cpp:1368
12 	xul.dll 	nsXPCWrappedJSClass::CallMethod 	js/src/xpconnect/src/xpcwrappedjsclass.cpp:1696
13 	xul.dll 	nsXPCWrappedJS::CallMethod 	js/src/xpconnect/src/xpcwrappedjs.cpp:570
14 	xul.dll 	PrepareAndDispatch 	xpcom/reflect/xptcall/src/md/win32/xptcstubs.cpp:114
15 	xul.dll 	SharedStub 	xpcom/reflect/xptcall/src/md/win32/xptcstubs.cpp:141
16 	xul.dll 	nsObserverService::NotifyObservers 	xpcom/ds/nsObserverService.cpp:182
17 	xul.dll 	nsHttpHandler::NotifyObservers 	netwerk/protocol/http/src/nsHttpHandler.cpp:498
18 	xul.dll 	nsHttpChannel::ProcessResponse 	netwerk/protocol/http/src/nsHttpChannel.cpp:986
19 	xul.dll 	nsHttpChannel::OnStartRequest 	netwerk/protocol/http/src/nsHttpChannel.cpp:5157
20 	xul.dll 	nsInputStreamPump::OnStateStart 	netwerk/base/src/nsInputStreamPump.cpp:439
21 	xul.dll 	nsInputStreamPump::OnInputStreamReady 	netwerk/base/src/nsInputStreamPump.cpp:395
22 	xul.dll 	nsOutputStreamReadyEvent::Run 	xpcom/io/nsStreamUtils.cpp:191
23 	xul.dll 	nsThread::ProcessNextEvent 	xpcom/threads/nsThread.cpp:527
24 	xul.dll 	nsBaseAppShell::Run 	widget/src/xpwidgets/nsBaseAppShell.cpp:170
25 	xul.dll 	nsAppStartup::Run 	toolkit/components/startup/src/nsAppStartup.cpp:182
26 	nspr4.dll 	PR_GetEnv 	
27 	firefox.exe 	wmain 	toolkit/xre/nsWindowsWMain.cpp:120
28 	firefox.exe 	__tmainCRTStartup 	obj-firefox/memory/jemalloc/crtsrc/crtexe.c:591
29 	kernel32.dll 	BaseThreadInitThunk 	
30 	ntdll.dll 	__RtlUserThreadStart 	
31 	ntdll.dll 	_RtlUserThreadStart

Comment 2

[Steven Michaud [:smichaud] (Retired)]

> this might be a variation of the yslow crash in bug 542203.  a lot
> of the stack looks the same

It's not particularly strongly associated with YSlow or Firebug,
according to today's interesting-addons files at
http://people.mozilla.com/crash_analysis/.

It's more strongly associated with the Yandex Bar
(https://addons.mozilla.org/addon/3495).

Maybe this bug's and bug 542203's crashes are caused by a single,
underlying Firefox bug.

Comment 3

[chris hofmann]

I commented over in bug 542203 that there is an interesting pattern of both of these bugs crashing in much higher volume at about the 10 minute mark after start up.

https://bugzilla.mozilla.org/show_bug.cgi?id=542203#c55

Comment 4

[Mike Beltzner [:beltzner, not reading bugmail]]

This dropped back down to #92 - might have been a data poop, or more worryingly, related to third-party integration code :(

Comment 5

[chris hofmann]

Attachment: daily crash rate histogram

yeah, could have been data poop on 1/24-1/29 that have a big tug on the ranking.  lots of things going on during that time.  post 3.6 release but before the updated yslow got out there I think, plus we hit snags in the socorro upgrade IIRC.

fact remains that we are running about 100 crashes higher recently than before those few days of spike, and if you combine with daily crashes from Bug 542120 it would move higher in the ranking to a point that it needs to remain on the crash kill radar and we need to actively be looking for some way to reduce.

Comment 6

[Joe Drew (not getting mail)]

This should almost certainly not be in Core :: General.

Comment 7

[Martijn Wargers (dead)]

I crashed 2 times with this stack on:
http://expansive-derivation.ossreleasefeed.com/2010/07/25-mostly-developer-chrome-extensions-you-need/
or:
http://expansive-derivation.ossreleasefeed.com/2010/07/getting-started-with-css3-pie/

The top of the stack might be more useful than normally:
http://crash-stats.mozilla.com/report/index/ba81c5d5-8eee-43e3-9d41-b365d2100910
0  	js3250.dll  	JSObject::init  	 js/src/jsobj.h:277
1 	js3250.dll 	JS_NewSystemObject 	js/src/jsdbgapi.cpp:1865
2 	xul.dll 	XPCWrappedNative::GetNewOrUsed 	js/src/xpconnect/src/xpcwrappednative.cpp:571
3 	xul.dll 	XPCConvert::NativeInterface2JSObject 	js/src/xpconnect/src/xpcconvert.cpp:1199
4 	xul.dll 	XPCConvert::NativeData2JS 	js/src/xpconnect/src/xpcconvert.cpp:471
etc...

Comment 8

[Wayne Mery (:wsmwk)]

no crashes after firefox 4.0b4

Comment 9

[Sheila Mooney]

Since there are none of these on any recent release, I am going to resolve as works for me. If it's pops up again, likely a new bug or we can reopen.