568827 - Firefox Crash [@ nsXPCWrappedJSClass::DelegatedQueryInterface(nsXPCWrappedJS*, nsID const&, void**) ]

Status: RESOLVED INCOMPLETE

Description

[chris hofmann]

volume increase on this signature on the first day of the lastest 3.6.4 builds (20100523185824) getting a bit more exposure.   we should keep an eye on this as the the number of users on the new builds grow to see if it tapers off.

checking --- nsXPCWrappedJSClass::DelegatedQueryInterface.nsXPCWrappedJS...nsID.const...void... 20100527-crashdata.csv
found in: 3.6.3 3.6.4 3.0.8 3.0.19 3.5.9 3.0.10 3.0.1
release total-crashes
              nsXPCWrappedJSClass::DelegatedQueryInterface.nsXPCWrappedJS...nsID.const...void... crashes
                         pct.
all     367855  20      5.43692e-05
3.6.3   255738  8       3.1282e-05
3.6.4   23009   5       0.000217306
3.0.8   503     2       0.00397614
3.0.19  11571   2       0.000172846
3.5.9   32443   1       3.08233e-05
3.0.10  972     1       0.00102881
3.0.1   1600    1       0.000625

os breakdown
nsXPCWrappedJSClass::DelegatedQueryInterface.nsXPCWrappedJS...nsID.const...void...Total 20
Win5.1  0.75
Win6.0  0.05
Win6.1  0.20

stack looks like

http://crash-stats.mozilla.com/report/index/95faa625-d4d9-4169-9fcd-550c02100527

Frame  	Module  	Signature [Expand]  	Source
0 	xul.dll 	nsXPCWrappedJSClass::DelegatedQueryInterface 	js/src/xpconnect/src/xpcwrappedjsclass.cpp:709
1 	xul.dll 	nsXPCWrappedJS::QueryInterface 	js/src/xpconnect/src/xpcwrappedjs.cpp:185
2 	xul.dll 	nsCOMPtr_base::assign_from_qi_with_error 	obj-firefox/xpcom/build/nsCOMPtr.cpp:105
3 	xul.dll 	NS_GetWeakReference 	obj-firefox/xpcom/build/nsWeakReference.cpp:108
4 		@0x81579b 	
5 	xul.dll 	nsObserverList::AddObserver 	xpcom/ds/nsObserverList.cpp:57
6 	xul.dll 	nsObserverService::AddObserver 	xpcom/ds/nsObserverService.cpp:135
7 	xul.dll 	NS_InvokeByIndex_P 	xpcom/reflect/xptcall/src/md/win32/xptcinvoke.cpp:102
8 	xul.dll 	XPCWrappedNative::CallMethod 	js/src/xpconnect/src/xpcwrappednative.cpp:2722
9 	xul.dll 	XPC_WN_CallMethod 	js/src/xpconnect/src/xpcwrappednativejsops.cpp:1740
10 	js3250.dll 	js_Invoke 	js/src/jsinterp.cpp:1360
11 	js3250.dll 	js_Interpret 	js/src/jsops.cpp:2240
12 	js3250.dll 	js_Invoke 	js/src/jsinterp.cpp:1368
13 	js3250.dll 	array_extra 	js/src/jsarray.cpp:3197
14 	js3250.dll 	array_forEach 	js/src/jsarray.cpp:3253
15 	js3250.dll 	js_Interpret 	js/src/jsops.cpp:2208
16 	js3250.dll 	js_Execute 	js/src/jsinterp.cpp:1601
17 	js3250.dll 	JS_EvaluateUCScriptForPrincipals 	js/src/jsapi.cpp:5070
18 	js3250.dll 	JS_EvaluateScriptForPrincipals 	js/src/jsapi.cpp:5034
19 	js3250.dll 	js3250.dll@0x7e1f 	
20 	xul.dll 	NS_InvokeByIndex_P 	xpcom/reflect/xptcall/src/md/win32/xptcinvoke.cpp:102
21 	xul.dll 	XPCWrappedNative::CallMethod 	js/src/xpconnect/src/xpcwrappednative.cpp:2722
22 	xul.dll 	XPC_WN_CallMethod 	js/src/xpconnect/src/xpcwrappednativejsops.cpp:1740
23 	js3250.dll 	js_Invoke 	js/src/jsinterp.cpp:1360
24 	js3250.dll 	js_Interpret 	js/src/jsops.cpp:2240
25 	js3250.dll 	js_Invoke 	js/src/jsinterp.cpp:1368
26 	js3250.dll 	js_InternalInvoke 	js/src/jsinterp.cpp:1423
27 	js3250.dll 	js_GetPropertyHelper 	js/src/jsobj.cpp:4267
28 	js3250.dll 	js_Interpret 	js/src/jsops.cpp:1520
29 	js3250.dll 	js_Invoke 	js/src/jsinterp.cpp:1368
30 	xul.dll 	nsXPCWrappedJSClass::CallMethod 	js/src/xpconnect/src/xpcwrappedjsclass.cpp:1696
31 	xul.dll 	nsXPCWrappedJS::CallMethod 	js/src/xpconnect/src/xpcwrappedjs.cpp:570
32 	xul.dll 	PrepareAndDispatch 	xpcom/reflect/xptcall/src/md/win32/xptcstubs.cpp:114
33 	xul.dll 	SharedStub 	xpcom/reflect/xptcall/src/md/win32/xptcstubs.cpp:141
34 	xul.dll 	nsCreateInstanceByContractID::operator 	obj-firefox/xpcom/build/nsComponentManagerUtils.cpp:210
35 	xul.dll 	nsCOMPtr_base::assign_from_helper 	obj-firefox/xpcom/build/nsCOMPtr.cpp:150
36 	xul.dll 	nsCOMPtr<nsISupports>::operator= 	obj-firefox/dist/include/nsCOMPtr.h:1040
37 	xul.dll 	nsAppStartupNotifier::Observe 	
38 		@0x135827f

more at

http://crash-stats.mozilla.com/report/list?product=Firefox&query_search=signature&query_type=contains&query=nsXPCWrappedJSClass%3A%3ADelegatedQueryInterface%28nsXPCWrappedJS*%2C%20nsID%20const%26%2C%20void**%29&date=05%2F28%2F2010%2007%3A22%3A08&range_value=1&range_unit=weeks&hang_type=any&process_type=any&plugin_field=&plugin_query_type=&plugin_query=&do_query=1&signature=nsXPCWrappedJSClass%3A%3ADelegatedQueryInterface%28nsXPCWrappedJS*%2C%20nsID%20const%26%2C%20void**%29

not much in the urls or comments to go on yet.

Comment 1

[chris hofmann]

appears to be associated with some kind of repeated slowdown.  here are all the comments received for this signature in May with reference to version and build ids.

   1 3.0.1 2008070208 this computer is way to slow
   1 3.0.4 2008102920 this ish ant workin

  15 3.6 20100115144158 \N
   1 3.6.3 20100401080539 Getting totally fed up with FF crashing.
   1 3.6.3 20100401080539 WHY ALWAYS HAVE THE PROBLEMS AUTO SHUTDOWN
  66 3.6.3 20100401080539 \N
   1 3.6.3 20100401080539 das ist langsam nicht mehr normal, dass der mozilla firefox laufend abstürzt, | heute schon 15 mal

gtranslate: this is no more slowly normal for the mozilla firefox crashes constantly, | 15 times today

   1 3.6.4 20100503122926 \N
   5 3.6.4 20100523185824 \N
   1 3.7a3 20100315090913 \N

Comment 2

[Daniel Veditz [:dveditz]]

Not enough info to block 1.9.2.x at this point, but keep an eye on it.

Comment 3

[Wayne Mery (:wsmwk)]

~60% are startup
relatively rare
~75% are XP

of recent crashes, some are about:sessionrestore
like bp-860f8bb6-b9a6-49e1-bd85-b98002121209

Comment 4

[Anthony Hughes (:ashughes) [QA]]

This is spiking with Firefox 31.0b9. Previously this averaged about 0.35 crashes per 1MM ADU, with Beta 9 it spiked to 98 C/1MM ADU and is now at 68 C/1MM ADU after a couple days. In pure volume this is not high enough to be considered a topcrash but the spike concerns me.

Comment 5

[Sylvestre Ledru [:sylvestre]]

I removed the rare word from the whiteboard. 
Do we know if it impacts 32?

Comment 6

[Anthony Hughes (:ashughes) [QA]]

I don't see any crashes for Firefox 33 or 32. Curiously though I don't see any crashes for 31.0b9, 31.0b99, or 31.0 either. I'm wondering if this is one of those random build time crashes?

Comment 7

[Release mgmt bot [:sylvestre / :calixte]]

Crash volume for signature 'nsXPCWrappedJSClass::DelegatedQueryInterface':
 - nightly (version 50): 0 crash from 2016-06-06.
 - aurora  (version 49): 1 crash from 2016-06-07.
 - beta    (version 48): 4 crashes from 2016-06-06.
 - release (version 47): 24 crashes from 2016-05-31.
 - esr     (version 45): 0 crash from 2016-04-07.

Crash volume on the last weeks:
             Week N-1   Week N-2   Week N-3   Week N-4   Week N-5   Week N-6   Week N-7
 - nightly          0          0          0          0          0          0          0
 - aurora           0          0          0          0          0          1          0
 - beta             1          0          2          0          0          0          1
 - release          2          9          2          4          1          3          3
 - esr              0          0          0          0          0          0          0

Affected platforms: Windows, Mac OS X

Comment 8

[Wayne Mery (:wsmwk)]

Only two 47.0.1 crashes in one week - both startup - seems not worth keeping a bug open for.

(In reply to Anthony Hughes (:ashughes) [GFX][QA][Mentor] from comment #6)
> I'm wondering if this is one of those random build time crashes?

What do you mean by "random build time"?

Comment 9

[Anthony Hughes (:ashughes) [QA]]

(In reply to Wayne Mery (:wsmwk, NI for questions) from comment #8)
> What do you mean by "random build time"?

We used to (maybe still do) get crashes due to a bug in AMD CPUs that manifests randomly at build time because we use PGO on Windows. Simply retriggering the build usually resolves the issue but it's not something we know about until users start hitting the crash with the broken build. We chemspilled Firefox 19 for one of these and it's the reason we started doing RCs in addition to Beta.

All that said, this bug really predates all of that so it's probably irrelevant.