Last Comment Bug 560037 - Firefox 3.6.4 Crash Report [@ mozilla::plugins::PPluginInstanceChild::CallPStreamNotifyConstructor(mozilla::plugins::PStreamNotifyChild*, nsCString const&, nsCString const&, bool const&, nsCString const&, bool const&, short*) ]
: Firefox 3.6.4 Crash Report [@ mozilla::plugins::PPluginInstanceChild::CallPSt...
Status: NEW
[oopp-watchlist]
: crash
Product: Core
Classification: Components
Component: Plug-ins (show other bugs)
: Trunk
: x86 Windows XP
: -- normal with 5 votes (vote)
: ---
Assigned To: Nobody; OK to take it and work on it
:
Mentors:
: 570950 (view as bug list)
Depends on:
Blocks:
  Show dependency treegraph
 
Reported: 2010-04-17 10:55 PDT by chris hofmann
Modified: 2015-10-13 07:26 PDT (History)
8 users (show)
See Also:
Crash Signature:
QA Whiteboard:
Iteration: ---
Points: ---
Has Regression Range: ---
Has STR: ---


Attachments
windebug (windbg) logfile of firefox.exe - ffx4beta8pre (83.40 KB, text/plain)
2010-10-08 05:18 PDT, abittner
no flags Details

Description chris hofmann 2010-04-17 10:55:24 PDT
possible new crash, or maybe just a new signature on an old crash.  #12 in very early 3.6.4 data.  not much to go on yet other than watch for more data.

stack looks like

http://crash-stats.mozilla.com/report/index/4ea5a66c-e506-4c64-9c14-0067b2100416

0  	xul.dll  	mozilla::plugins::PPluginInstanceChild::CallPStreamNotifyConstructor  	 obj-firefox/ipc/ipdl/PPluginInstanceChild.cpp:532
1 	xul.dll 	mozilla::plugins::child::_geturlnotify 	dom/plugins/PluginModuleChild.cpp:806
2 	npctrl.dll 	npctrl.dll@0xca7fd 	
3 	npctrl.dll 	npctrl.dll@0x89af0 	
4 	npctrl.dll 	npctrl.dll@0x6bf62 	
5 	npctrl.dll 	npctrl.dll@0x4453c 	
6 	npctrl.dll 	npctrl.dll@0x138bd 	
7 	npctrl.dll 	npctrl.dll@0x13ba3 	
8 	npctrl.dll 	npctrl.dll@0x13d4d 	
9 	npctrl.dll 	npctrl.dll@0x13de4 	
10 	npctrl.dll 	npctrl.dll@0x13e38 	
11 	npctrl.dll 	npctrl.dll@0x13f4b 	
12 	npctrl.dll 	npctrl.dll@0x4709 	
13 	user32.dll 	InternalCallWinProc 	
14 	user32.dll 	UserCallWinProcCheckWow 	
15 	user32.dll 	DispatchMessageWorker 	
16 	user32.dll 	DispatchMessageW 	
17 	xul.dll 	base::MessagePumpForUI::ProcessMessageHelper 	ipc/chromium/src/base/message_pump_win.cc:361
18 	xul.dll 	base::MessagePumpForUI::ProcessNextWindowsMessage 	ipc/chromium/src/base/message_pump_win.cc:336
19 	xul.dll 	base::MessagePumpForUI::DoRunLoop 	ipc/chromium/src/base/message_pump_win.cc:205
20 	xul.dll 	base::MessagePumpWin::RunWithDispatcher 	ipc/chromium/src/base/message_pump_win.cc:52
21 	xul.dll 	base::MessagePumpWin::Run 	ipc/chromium/src/base/message_pump_win.h:78
22 	xul.dll 	MessageLoop::RunInternal 	ipc/chromium/src/base/message_loop.cc:216
23 	xul.dll 	MessageLoop::RunHandler 	
24 	xul.dll 	MessageLoop::Run 	ipc/chromium/src/base/message_loop.cc:173
25 	xul.dll 	base::Thread::ThreadMain 	ipc/chromium/src/base/thread.cc:165
26 	xul.dll 	`anonymous namespace'::ThreadFunc 	ipc/chromium/src/base/platform_thread_win.cc:26
27 	kernel32.dll 	kernel32.dll@0x13676 	
28 	ntdll.dll 	ntdll.dll@0x39d71 	
29 	ntdll.dll 	ntdll.dll@0x39d44

more at

http://crash-stats.mozilla.com/report/list?range_value=2&range_unit=weeks&signature=mozilla%3A%3Aplugins%3A%3APPluginInstanceChild%3A%3ACallPStreamNotifyConstructor%28mozilla%3A%3Aplugins%3A%3APStreamNotifyChild*%2C%20nsCString%20const%26%2C%20nsCString%20const%26%2C%20bool%20const%26%2C%20nsCString%20const%26%2C%20bool%20const%26%2C%20short*%29&version=Firefox%3A3.6.4
Comment 1 Benjamin Smedberg AWAY UNTIL 2-AUG-2016 [:bsmedberg] 2010-04-17 11:28:17 PDT
Both silverlight calling NPN_GetURLNotify. The stack implies that the associated plugin instance (NPP) has already been destroyed.

It's an oopsie, if that wasn't clear.
Comment 2 abittner 2010-06-09 05:04:59 PDT
i get constant and immediate crashes on new micrsofot wei-share site, running on ffx 3.7a5nightly running in a tab.

site: http://www.weishare.net/

currently running:
Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.3a5pre) Gecko/20100608 Minefield/3.7a5pre ( .NET CLR 3.5.30729; .NET4.0E)


sample crash:
http://crash-stats.mozilla.com/report/index/bp-7afda4e8-113e-4433-9e1f-7a2b12100609
http://crash-stats.mozilla.com/report/index/bp-b453eb3d-304a-4b56-8097-31a6a2100609
http://crash-stats.mozilla.com/report/index/bp-c2dc1379-89d1-4185-a462-0d6f52100609
http://crash-stats.mozilla.com/report/index/bp-830139ea-8f3c-460b-a029-a68832100609
http://crash-stats.mozilla.com/report/index/bp-e0a480ac-ad2f-41f6-9e73-a88d42100609
Comment 3 timeless 2010-06-09 08:39:20 PDT
*** Bug 570950 has been marked as a duplicate of this bug. ***
Comment 4 abittner 2010-06-18 10:31:54 PDT
still crashes on me with current 3.7a5pre build
http://crash-stats.mozilla.com/report/index/59cea5cc-98a4-4f50-99d6-326a22100618

every hundredth time of reloading the weishare page it actually works (silverlight starts loading up and working), it even stays for a while with reloading the tab from there on, and eventually fails like after the third or fourth reload once it was working.

maybe its some race condition when the browser is under some load (for example i was crashing/submitbugreporter/reloading the weishare tab quite quickly and eventually it did work after a few tens of reload attempts) it seems to succeed to make silverlight work.

is there actually some trace plugin/addon for the mozilla apps where i could figure out what parts and components of the mozilla apps eat up cpu cycles.

with the recent 3.7a5pre builds i am having slowdowns when firefox runs for many days with many windows and many tabs opened and being under "heavy" (actually normal for me)  use. surfing many sites, closing, opening many tabs, having a number of addons, etc.

this slowdown is especially recognizable after a few days when single clicks or scrolling inside tabs or even only navigating back and forth inside a tab (history) takes like 10-20seconds each and everything is painfully slow. javascript, history, ram-usage or such stuff comes to my mind.

this has never been so bad ever since firefox arrived at the 3.5 or 3.6builds, but now with 3.7ax its seems to get worse again.... :(

so after a few days i currently need to quit and restart the firefox app all the way as its barely usable any more :(

any hints? profiling apps for the non-developer folks for us, or any possibility to measure and log/record cpu-cycles inside the app and its various parts/threads/functions, and any way to submit these profiling results for review?


thanks.
Comment 5 abittner 2010-06-29 07:47:18 PDT
firefox 4.0beta1 build1
Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:2.0b1) Gecko/20100628 Firefox/4.0b1 ( .NET CLR 3.5.30729; .NET4.0E)

seems to be unaffected of this bug. tried with weishare. no crash of the plugin there any more.
Comment 6 abittner 2010-07-13 15:02:06 PDT
the crashing on the weishare site has returned with:

Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:2.0b1) Gecko/20100630 Firefox/4.0b1 ( .NET CLR 3.5.30729; .NET4.0E)

running on winxp/sp3/latestpatches

about:buildconfig
Source

Built from http://hg.mozilla.org/mozilla-central/rev/65c30e4ee631
Build platform
target
i686-pc-mingw32
Build tools
Compiler 	Version 	Compiler flags
cl 	14.00.50727.762 	-TC -nologo -W3 -Gy -Fdgenerated.pdb -DNDEBUG -DTRIMMED -Zi -Zi -UDEBUG -DNDEBUG -GL -wd4624 -wd4952 -O1
cl 	14.00.50727.762 	-GR- -TP -nologo -Zc:wchar_t- -W3 -Gy -Fdgenerated.pdb -wd4800 -DNDEBUG -DTRIMMED -Zi -Zi -UDEBUG -DNDEBUG -GL -wd4624 -wd4952 -O1
Configure arguments

--enable-application=browser --enable-update-channel=beta --enable-update-packaging --enable-jemalloc --enable-tests --enable-official-branding

-----------------

:( too bad.

http://crash-stats.mozilla.com/report/index/5d932fbb-43f3-46fa-bf47-9fade2100713
http://crash-stats.mozilla.com/report/index/b8e172bf-f943-42ab-9e88-f8d392100713

sample crashes.

regards.
Comment 7 XtC4UaLL [:xtc4uall] 2010-09-06 16:22:08 PDT
On 1.9.2 Branch this is now low Volume (40-50 Crashes/Week):
with e.g. NPSWF32.dll 10.1.82.7 (besides the Silverlight ones):

Frame  	Module  	Signature  	Source
0 	xul.dll 	mozilla::plugins::PPluginInstanceChild::CallPStreamNotifyConstructor(mozilla::plugins::PStreamNotifyChild*,nsCString const&,nsCString const&,bool const&,nsCString const&,bool const&,short*) 	obj-firefox/ipc/ipdl/PPluginInstanceChild.cpp:532
1 	xul.dll 	mozilla::plugins::child::_geturlnotify 	dom/plugins/PluginModuleChild.cpp:831
2 	NPSWF32.dll 	NPN_GetURLNotify 	F_1687171941____________________________________________________________:198
3 	NPSWF32.dll 	F_884039186___________________________ 	F_36257560_______________________________________________________________________:319
4 	NPSWF32.dll 	F_1119657076____________________________________________________________________________________________________________________________________________ 	F1438600658____________________________________:20048
5 	NPSWF32.dll 	F_590358957_______________ 	F1196556351________________________________________:43
6 	NPSWF32.dll 	F473088221____________________________________________ 	F1441475132___________________________________________________________________:5202
7 		@0xfffffffe 	
8 	NPSWF32.dll 	F1718314630________________________ 	F1438600658____________________________________:11671 

On Trunk/2.0 (30 Crashes/Week, most of those in Silverlight), but here another Flash one:
bp-546b54a7-6437-48e3-9d07-7e5892100902
Frame  	Module  	Signature  	Source
0 	xul.dll 	mozilla::plugins::PPluginInstanceChild::CallPStreamNotifyConstructor(mozilla::plugins::PStreamNotifyChild*,nsCString const&,nsCString const&,bool const&,nsCString const&,bool const&,short*) 	obj-firefox/ipc/ipdl/PPluginInstanceChild.cpp:712
1 	xul.dll 	mozilla::plugins::child::_geturlnotify 	dom/plugins/PluginModuleChild.cpp:903
2 	NPSWF32.dll 	NPN_GetURLNotify 	F_1687171941____________________________________________________________:198
3 	NPSWF32.dll 	F_884039186___________________________ 	F_36257560_______________________________________________________________________:319
4 	NPSWF32.dll 	F_1119657076____________________________________________________________________________________________________________________________________________ 	F1438600658____________________________________:20048
5 	NPSWF32.dll 	F723062336________________________ 	F_2142896994_______________________________________:77
6 	NPSWF32.dll 	F_1662446635__________________________ 	F_1624107517_____________________________________________________:270
7 	NPSWF32.dll 	F473088221____________________________________________ 	F1441475132___________________________________________________________________:5185
8 	NPSWF32.dll 	F1949876734____________________________________________________ 	F_1624107517_____________________________________________________:185
9 	NPSWF32.dll 	F1949876734____________________________________________________ 	F_1624107517_____________________________________________________:185
10 	NPSWF32.dll 	F1949876734____________________________________________________ 	F_1624107517_____________________________________________________:185
11 	NPSWF32.dll 	F_1447065881___________________________________ 	F_99398417_________________________________________:4044
12 	NPSWF32.dll 	F_215975579_______________________________________________________________________________________________ 	F1438600658____________________________________:23838
Comment 9 abittner 2010-10-07 08:18:47 PDT
still crashes with ffx4 beta7pre and beta8pre as well.

example for beta8pre:
http://crash-stats.mozilla.com/report/index/8f379cd8-a64e-4822-b2d5-52ad82101007

most of times right after a complete fresh start of firefox.exe process from scratch and then navigating to the www.weishare.net site the first attempt to view that website works normally for the microsoft silverlight plugin. but as soon as ctrl+r, refresh, f5, all following attempts to view and refresh this site again bring up the plugincrash.

it takes ages again while still running this instance of firefox to be able to succeed in getting the weishare website to run again normally with the silverlight plugin.


btw, i just went to firefox4 beta8pre a little while ago via the about box for the first time, so i guess today was the switchover to the beta8pre nightlies, and this build
Build identifier: Mozilla/5.0 (Windows NT 5.1; rv:2.0b8pre) Gecko/20101007 Firefox/4.0b8pre
about:buildconfig
Source Built from http://hg.mozilla.org/mozilla-central/rev/5bb4f093a50b

gave me weishare crashes as well, but the very first one of the silverlight plugin crashes crashes at weishare was a different stack signature:

see it here
http://crash-stats.mozilla.com/report/index/bp-e3618d3c-16eb-4c2d-823a-f74932101007

there are only two other of this stacksignature:
http://crash-stats.mozilla.com/report/list?product=Firefox&query_search=signature&query_type=exact&query=mozilla%3A%3Aplugins%3A%3APPluginInstanceChild%3A%3ADeallocSubtree%28%29&date=10%2F07%2F2010%2008%3A12%3A27&range_value=30&range_unit=days&hang_type=any&process_type=any&plugin_field=&plugin_query_type=&plugin_query=&do_query=1&admin=&signature=mozilla%3A%3Aplugins%3A%3APPluginInstanceChild%3A%3ADeallocSubtree%28%29



the signature is:
mozilla::plugins::PPluginInstanceChild::DeallocSubtree() 


in contrast to the one that we are discussing here:
mozilla::plugins::PPluginInstanceChild::CallPStreamNotifyConstructor(mozilla::plugins::PStreamNotifyChild*, nsCString const&, nsCString const&, bool const&, nsCString const&, bool const&, short*) 


maybe this is a possibility to track down this bug or those other crash stackreports are related and can give further hints to this behaviour.
Comment 10 abittner 2010-10-08 05:18:23 PDT
Created attachment 481804 [details]
windebug (windbg) logfile of firefox.exe - ffx4beta8pre

was actually trying to windebug (windbg) firefox for a completely different bug, it was stopping in the debugger because of this.

attaching log.

firefox4 beta8pre

Note You need to log in before you can comment on or make changes to this bug.