763142 - plugin-container crashes when closing Firefox that is built with --disable-crashreporter if Flash 11.3 is loaded.

Status: RESOLVED DUPLICATE of bug 747683

(Core Graveyard :: Plug-ins, defect)

Description

[Tetsuro Kato [:tete]]

User Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:13.0) Gecko/20100101 Firefox/13.0
Build ID: 20120601045813

Steps to reproduce:

I built Firefox 13.0 by specifying --disable-crashreporter in my mozconfig file. And I updated Flash Player to version 11.3.300.257.



Actual results:

When closing the build after loading the following URL, plugin-container.exe frequently crashes.
http://get.adobe.com/jp/flashplayer/completion/aih/?exitcode=0

I also built Firefox 13.0 without specifying --disable-crashreporter in my mozconfig. When using this build, the probability that a plugin-container.exe crashes greatly reduces.

I compared the probabilities that plugin-container.exe crashes after opening the above URL with each build:
With --disable-crashreporter: 9/10
Without --disable-crashreporter: 1/10



Expected results:

It's preferable that plugin-container doesn't crash.

Comment 1

[xunxun]

I can reproduce the issue on my firefox 13 and my aurora edition with --disable-crashreporter

Comment 2

[xunxun]

If I change dom.ipc.plugins.enabled to false,
firefox NO crash.

It should be plugin-container.exe itself problem.

Comment 3

[xunxun]

When I use VS debug attach plugin-container.exe

after closing, it will crash on

crtexe.c  572 line

__except ( _XcptFilter(GetExceptionCode(), GetExceptionInformation()) )




Stack is :

 	msvcr100.dll!___crtExitProcess()  + 0x17 byte	
 	msvcr100.dll!__cinit()  + 0xb689 byte	
 	msvcr100.dll!_exit()  + 0x11 byte	
>	plugin-container.exe!__tmainCRTStartup()  line 572	     C

Comment 4

[Scoobidiver (away)]

It might be related to bug 749928.

Comment 5

[Tetsuro Kato [:tete]]

I can also reproduce the problem with VC8 x86 trunk build.
I hear that the problem is reproduced with VC10 x86 trunk build, but it isn't reproduced with VC10 x64 trunk build.

Comment 6

[xunxun]

Don't know if it's the Adobe's side bug:

https://bugbase.adobe.com/index.cfm?event=bug&id=3176400

Comment 7

[Tetsuro Kato [:tete]]

Debugging information from WinDbg.

Access violation (c0000005)
ntdll_77430000!RtlpWaitOnCriticalSection+0xbd:
77468dc9 ff4014          inc     dword ptr [eax+14h]  ds:002b:00000014=????????

Stack text:
ntdll_77430000!RtlpWaitOnCriticalSection+0xbd
ntdll_77430000!RtlEnterCriticalSection+0x150
npswf32!native_ShockwaveFlash_TCallLabel+0x147054
npswf32!native_ShockwaveFlash_TCallLabel+0x1476ec
npswf32!native_ShockwaveFlash_TCallLabel+0x145f7a
npswf32!native_ShockwaveFlash_TCallLabel+0x143397
npswf32!native_ShockwaveFlash_TCallLabel+0x1433b9
npswf32!native_ShockwaveFlash_TCallLabel+0x14343b
npswf32!native_ShockwaveFlash_TCallLabel+0x132fdb
npswf32!native_ShockwaveFlash_TCallLabel+0x128f85
npswf32!native_ShockwaveFlash_TCallLabel+0x132c47
npswf32!BrokerMainW+0x2928f5
npswf32!BrokerMainW+0x1fb281
npswf32!BrokerMainW+0x1fb4b1
npswf32!BrokerMainW+0x1fb51c
ntdll_77430000!LdrpCallInitRoutine+0x14
ntdll_77430000!LdrpUnloadDll+0x375
ntdll_77430000!LdrUnloadDll+0x4a
KERNELBASE!FreeLibrary+0x15
nspr4!PR_UnloadLibrary+0x62 [nsprpub\pr\src\linking\prlink.c @ 999]
xul!mozilla::plugins::PluginModuleChild::~PluginModuleChild+0x22 [dom\plugins\ipc\pluginmodulechild.cpp @ 117]
xul!mozilla::plugins::PluginProcessChild::~PluginProcessChild+0x11 [firefoxobjdebug\dist\include\mozilla\plugins\pluginprocesschild.h @ 26]
xul!mozilla::plugins::PluginProcessChild::`scalar deleting destructor'+0x8
xul!XRE_InitChildProcess+0x275 [toolkit\xre\nsembedfunctions.cpp @ 487]
plugin_container!NS_internal_main+0x3d [ipc\app\mozillaruntimemain.cpp @ 48]
plugin_container!wmain+0xf0 [toolkit\xre\nswindowswmain.cpp @ 102]
plugin_container!__tmainCRTStartup+0x10f [f:\dd\vctools\crt_bld\self_x86\crt\src\crtexe.c @ 594]
KERNEL32!BaseThreadInitThunk+0xe
ntdll_77430000!__RtlUserThreadStart+0x70
ntdll_77430000!_RtlUserThreadStart+0x1b

Comment 8

[Tetsuro Kato [:tete]]

The problem seems to have been fixed by fixing Bug 747683.

Comment 9

[xunxun]

(In reply to Tetsuro Kato (Tete) from comment #8)
> The problem seems to have been fixed by fixing Bug 747683.

Maybe can mark it with FIXED if you make sure?

Comment 10

[Tetsuro Kato [:tete]]

I have confirmed the problem is fixed in trunk by fixing Bug 747683. So I change the status.