642600 - Firefox Crash [@ hang | __semwait_signal ]

Status: RESOLVED WONTFIX

(Core Graveyard :: Plug-ins, defect, P3)

Description

[Marcia Knous [:marcia]]

Seen while running Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:2.0b13pre) Gecko/20110317 Firefox/4.0b13pre. Shockwave Flash Version: 10.2.152.33. http://tinyurl.com/4euo8k8 to crashes in the last two weeks - fairly low volume.

STR:
1. Loaded March Madness http://www.ncaa.com/mmod/player/console.html
2. Game was playing and then it crashed

https://crash-stats.mozilla.com/report/index/bp-22931692-aa59-4877-961b-47cb22110317

Frame 	Module 	Signature [Expand] 	Source
0 	libSystem.B.dylib 	__semwait_signal 	
1 	libSystem.B.dylib 	pthread_cond_timedwait$UNIX2003 	
2 	AVA_HD_VP3.dylib 	AVA_HD_VP3.dylib@0x2511f 	
3 	AVA_HD_VP3.dylib 	AVA_HD_VP3.dylib@0x793b 	
4 	AVA_HD_VP3.dylib 	AVA_HD_VP3.dylib@0x7b28 	
5 	AVA_HD_VP3.dylib 	AVA_HD_VP3.dylib@0x71cd 	
6 	AVA_HD_VP3.dylib 	AVA_HD_VP3.dylib@0x77a2 	
7 	AppleVA 	VP3DecodeFrame 	
8 	VideoToolbox 	VP3VideoDecoder_DecodeFrame 	
9 	VideoToolbox 	vtDecompressionDuctDecodeSingleFrame 	
10 	VideoToolbox 	VTDecompressionSessionDecodeFrame 	
11 	VideoDecodeAcceleration 	VDADecoderDecode 	
12 	FlashPlayer-10.6 	F_536693969_________________________ 	
13 	FlashPlayer-10.6 	F_537007577_______________________ 	
14 	FlashPlayer-10.6 	F_173265829____________________________ 	
15 	FlashPlayer-10.6 	F_1121905458_____________________________________ 	
16 	FlashPlayer-10.6 	F_455237104__________________________________ 	
17 	FlashPlayer-10.6 	F1723263386______________________ 	
18 	FlashPlayer-10.6 	F1286739385________________ 	
19 	FlashPlayer-10.6 	F_1743300739_______ 	
20 	FlashPlayer-10.6 	F1868032823____________ 	
21 	FlashPlayer-10.6 	F944075138________ 	
22 	FlashPlayer-10.6 	F1533022358_____________ 	
23 	CoreFoundation 	__CFRunLoopDoSources0 	
24 	CoreFoundation 	__CFRunLoopRun 	
25 	CoreFoundation 	CFRunLoopRunSpecific 	
26 	CoreFoundation 	CFRunLoopRunInMode 	
27 	HIToolbox 	RunCurrentEventLoopInMode 	
28 	HIToolbox 	ReceiveNextEventCommon 	
29 	HIToolbox 	BlockUntilNextEventMatchingListInMode 	
30 	AppKit 	_DPSNextEvent 	
31 	AppKit 	-[NSApplication nextEventMatchingMask:untilDate:inMode:dequeue:] 	
32 	AppKit 	-[NSApplication run] 	
33 	XUL 	base::MessagePumpNSApplication::DoRun 	ipc/chromium/src/base/message_pump_mac.mm:677
34 	XUL 	base::MessagePumpCFRunLoopBase::Run 	ipc/chromium/src/base/message_pump_mac.mm:213
35 	XUL 	MessageLoop::Run 	ipc/chromium/src/base/message_loop.cc:219
36 	XUL 	XRE_InitChildProcess 	toolkit/xre/nsEmbedFunctions.cpp:515
37 	plugin-container 	main 	ipc/app/MozillaRuntimeMain.cpp:80
38 	plugin-container 	plugin-container@0xec5 	
39 		@0x7

Comment 1

[Bob Clary [:bc] (inactive)]

I hit this today in Nightly on Mac OS X 10.5. The browser consumed 100% cpu and was unresponsive. I did a sample and found multiple threads blocked on __semwait_signal. This is also the subject of a support article https://support.mozilla.org/en-US/questions/897129

There are 41 instances of this in the last week in socorro for Firefox 7, 8, 9.

Comment 2

[Bob Clary [:bc] (inactive)]

PS. I forgot to mention the browser recovered after several minutes...

Comment 3

[Steven Michaud [:smichaud] (Retired)]

It's normal for a thread to be waiting on __semwait_signal (or select... or kevent).  You'll see many such threads if you run a homemade build of Firefox (without symbols stripped) in gdb, do ctrl-c to break into gdb, then enter "thread apply all bt".  This doesn't mean Firefox is hung.  It's the threads that *aren't* waiting on __semwait_signal and friends that may tell you what's causing the hang (if there is one).

So this bug is technically invalid -- it's potentially many different bugs with many different causes.

Comment 4

[Steven Michaud [:smichaud] (Retired)]

By the way, the stack in comment #0 (bp-22931692-aa59-4877-961b-47cb22110317) has *no* threads waiting on anything but a "normal" target.  And a lot of the threads are Flash threads.  So the comment #0 stack is an example of a Flash hang.  Which we can't do anything about, since we don't have Flash source-code, or even non-obfuscated Flash symbols.

Comment 5

[Bob Clary [:bc] (inactive)]

I'm sorry I didn't save the inspection report when I hit it. We *can* work with Adobe to help alleviate the symptoms. Having the browser hit 100% CPU and beachball for 5 or more minutes isn't an experience that most people would tolerate.

Comment 6

[Virtual_ManPL [:Virtual] 🇵🇱 - (please needinfo? me - so I will see your comment/reply/question/etc.)]

I'm marking this bug as WORKSFORME as bug crashlog signature didn't appear from a long time (over half year) in Firefox (except some obsolete Fx <49).

Comment 7

[Virtual_ManPL [:Virtual] 🇵🇱 - (please needinfo? me - so I will see your comment/reply/question/etc.)]

Ups, my bad and due to ( bug #1348631 ) looks like there are sill crashes, so reopening.

Comment 8

[Sylvestre Ledru [:Sylvestre]]

Closing because no crashes reported for 12 weeks.