564625 - Crash [@ libclient.dylib@0x8b560 ] [@ libclient.dylib@0xa31cb ]

Status: RESOLVED INCOMPLETE

(Plugins Graveyard :: Java (Java Embedding Plugin), defect)

Description

[chris hofmann]

some interesting comments from one or two users.  maybe related tickled by the new updates that go pushed the the plugin page?


20100506-crashdata.csv:libclient.dylib@0x8b560 3.6.4 Mac OS X 10.5.8 9L31a
https://www.mozilla.com/en-US/plugincheck/
Checking Add-ons.
http://crash-stats.mozilla.com/report/index/4aac1b36-afab-4c2b-9c72-cd9432100506

20100506-crashdata.csv:libclient.dylib@0x8b560 3.6.4 Mac OS X 10.5.8 9L31a
https://www.mozilla.com/en-US/plugincheck/
Keeps crashing on your Add-on Check Up page.
http://crash-stats.mozilla.com/report/index/1b89319a-04c7-4437-8737-041b92100506


and some other place where this crash is seen.

20100503-crashdata.csv:libclient.dylib@0x8b560 3.5.9 Mac OS X 10.5.8 9L31a
http://chicagoist.com/
\N
http://crash-stats.mozilla.com/report/index/ed6b9ab5-acbb-4f36-9234-2e8752100503

20100503-crashdata.csv:libclient.dylib@0x8b560 3.5.9 Mac OS X 10.5.8 9L31a
http://www.google.com/search?q=ork+poster+chicago&ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a
\N
http://crash-stats.mozilla.com/report/index/46e30242-cca0-4f0a-b607-6ebd32100503

20100504-crashdata.csv:libclient.dylib@0x8b560 3.6.3 Mac OS X 10.5.8 9L31a
http://feku-francemetrologie.com/
\N
http://crash-stats.mozilla.com/report/index/e96f9d4a-6e1d-4d31-943f-8da172100504

20100504-crashdata.csv:libclient.dylib@0x8b560 3.6.3 Mac OS X 10.5.8 9L31a
http://www.hairscene.net/female-page/?album=2&gallery=6
\N
http://crash-stats.mozilla.com/report/index/785d45d6-87eb-4e3b-a676-9f5af2100504

Comment 1

[chris hofmann]

stacks look like

Frame  	Module  	Signature [Expand]  	Source
0 	libclient.dylib 	libclient.dylib@0x8b560 	
1 	JavaEmbeddingPlugin 	+[AppletView isJava142:] 	AppletView.m:1735
2 	JavaEmbeddingPlugin 	+[AppletView maybeCreateJavaVM:extraOptsNum:vmPtr:envPtr:initAWT:] 	AppletView.m:2298
3 	JavaEmbeddingPlugin 	JEPCreateJavaVM 	Controller.m:6724
4 	MRJPlugin 	MRJSession::open 	MRJSession.cpp:913
5 	MRJPlugin 	MRJPlugin::StartupJVM 	MRJPlugin.cpp:669
6 	MRJPlugin 	MRJPlugin::CreateSecureEnv 	MRJPlugin.cpp:725
7 	XUL 	nsNPAPIPlugin::CreateSecureEnv 	modules/plugin/base/src/nsNPAPIPlugin.cpp:718
8 	XUL 	CreateProxyJNI 	modules/oji/src/ProxyJNI.cpp:1732
9 	XUL 	XUL@0xb5cc84 	
10 	XUL 	create_java_vm_impl 	modules/oji/src/lcglue.cpp:342
11 	XUL 	jsj_ConnectToJavaVM 	js/src/liveconnect/jsj.c:474
12 	XUL 	JSJ_AttachCurrentThreadToJava 	js/src/liveconnect/jsj.c:698
13 	XUL 	map_js_context_to_jsj_thread_impl 	modules/oji/src/lcglue.cpp:180
14 	XUL 	jsj_EnterJava 	js/src/liveconnect/jsj_utils.c:465
15 	XUL 	JavaPackage_resolve 	js/src/liveconnect/jsj_JavaPackage.c:199
16 	libmozjs.dylib 	js_LookupPropertyWithFlags 	js/src/jsobj.cpp:3854
17 	libmozjs.dylib 	js_GetPropertyHelper 	js/src/jsobj.cpp:4191
18 	libmozjs.dylib 	js_Interpret 	js/src/jsops.cpp:1518
19 	libmozjs.dylib 	js_Invoke 	js/src/jsinterp.cpp:1368
20 	XUL 	nsXPCWrappedJSClass::CallMethod 	js/src/xpconnect/src/xpcwrappedjsclass.cpp:1696
21 	XUL 	PrepareAndDispatch 	xpcom/reflect/xptcall/src/md/unix/xptcstubs_ppc_rhapsody.cpp:182
22 	XUL 	XUL@0xc07870 	
23 	XUL 	nsEventListenerManager::HandleEventSubType 	content/events/src/nsEventListenerManager.cpp:1041
24 	XUL 	nsEventListenerManager::HandleEvent 	content/events/src/nsEventListenerManager.cpp:1147
25 	XUL 	nsEventTargetChainItem::HandleEvent 	content/events/src/nsEventDispatcher.cpp:246
26 	XUL 	nsEventTargetChainItem::HandleEventTargetChain 	content/events/src/nsEventDispatcher.cpp:310
27 	XUL 	nsEventDispatcher::Dispatch 	content/events/src/nsEventDispatcher.cpp:573
28 	XUL 	nsEventDispatcher::DispatchDOMEvent 	content/events/src/nsEventDispatcher.cpp:636
29 	XUL 	nsEventListenerManager::DispatchEvent 	content/events/src/nsEventListenerManager.cpp:1274
30 	XUL 	nsDOMEventRTTearoff::DispatchEvent 	content/base/src/nsGenericElement.cpp:1672
31 	XUL 	nsIDOMEventTarget_DispatchEvent 	dom_quickstubs.cpp:5271
32 	libmozjs.dylib 	js_Interpret 	js/src/jsops.cpp:2208
33 	libmozjs.dylib 	js_Execute 	js/src/jsinterp.cpp:1601
34 	libmozjs.dylib 	JS_EvaluateUCScriptForPrincipals 	js/src/jsapi.cpp:5070
35 	XUL 	nsJSContext::EvaluateString 	dom/base/nsJSEnvironment.cpp:1748
36 	XUL 	nsScriptLoader::EvaluateScript 	content/base/src/nsScriptLoader.cpp:711
37 	XUL 	nsScriptLoader::ProcessRequest 	content/base/src/nsScriptLoader.cpp:625
38 	XUL 	nsScriptLoader::ProcessScriptElement 	content/base/src/nsScriptLoader.cpp:577
39 	XUL 	nsScriptElement::MaybeProcessScript 	content/base/src/nsScriptElement.cpp:193
40 	XUL 	nsHTMLScriptElement::MaybeProcessScript 	content/html/content/src/nsHTMLScriptElement.cpp:564
41 	XUL 	HTMLContentSink::ProcessSCRIPTEndTag 	content/html/document/src/nsHTMLContentSink.cpp:3112
42 	XUL 	SinkContext::CloseContainer 	content/html/document/src/nsHTMLContentSink.cpp:1014
43 	XUL 	HTMLContentSink::CloseContainer 	content/html/document/src/nsHTMLContentSink.cpp:2392
44 	XUL 	CNavDTD::CloseContainer 	parser/htmlparser/src/CNavDTD.cpp:2762
45 	XUL 	CNavDTD::HandleEndToken 	parser/htmlparser/src/CNavDTD.cpp:1641
46 	XUL 	CNavDTD::HandleToken 	parser/htmlparser/src/CNavDTD.cpp:721
47 	XUL 	CNavDTD::BuildModel 	parser/htmlparser/src/CNavDTD.cpp:304
48 	XUL 	nsParser::BuildModel 	parser/htmlparser/src/nsParser.cpp:2456
49 	XUL 	nsParser::ResumeParse 	parser/htmlparser/src/nsParser.cpp:2337
50 	XUL 	nsParser::ContinueInterruptedParsing 	parser/htmlparser/src/nsParser.cpp:1833
51 	XUL 	nsRunnableMethod<nsContentSink, void>::Run 	nsThreadUtils.h:282
52 	XUL 	nsThread::ProcessNextEvent 	xpcom/threads/nsThread.cpp:527
53 	XUL 	NS_ProcessPendingEvents_P 	nsThreadUtils.cpp:200
54 	XUL 	nsBaseAppShell::NativeEventCallback 	widget/src/xpwidgets/nsBaseAppShell.cpp:125
55 	XUL 	nsAppShell::ProcessGeckoEvents 	widget/src/cocoa/nsAppShell.mm:506
56 	CoreFoundation 	CFRunLoopRunSpecific

Comment 2

[Steven Michaud [:smichaud] (Retired)]

That crashes happen at libclient.dylib@0x8b560 tells me nothing about
them.  It's only the second element in these stacks that might be
diagnostic:

+[AppletView isJava142:]

So I've emailed Aravind to ask him to change Socorro to treat all
stacks whose top element begins with 'libclient.dylib@' as having
compound signatures (as is already done for stacks whose top element
is 'objc_msgSend').  (I also asked him to make similar changes for
'libawt.jnilib@' and 'libobjc.A.dylib@').

Once that's done, I'll be able to search Socorro for '+[AppletView
isJava142:]' -- which is needed to make any further progress on this
bug.

Comment 3

[Steven Michaud [:smichaud] (Retired)]

Of the 6 crashes at libclient.dylib@0x8b560, only two happened on
Firefox 3.6.4 (two others happened on FF 3.6.3, and two more on FF
3.5.9):

http://crash-stats.mozilla.com/query/query?product=Firefox&version=ALL%3AALL&platform=mac&range_value=1&range_unit=weeks&date=05%2F08%2F2010+14%3A51%3A56&query_search=signature&query_type=contains&query=libclient.dylib%400x8b560&build_id=&process_type=all&do_query=1

(FF 3.6.4 and 3.5.10 bundle JEP 0.9.7.3 -- a new version.  Previous FF
versions bundle JEP 0.9.7.2.  But these crashes happen on both
versions, so these crashes aren't tied to the new version of the JEP.)

Comment 4

[chris hofmann]

Attachment: other mac java skiplist candidates

we get topmost source file in the .csv files now so I search for anything with /Applet View/ and got the attachement.   We don't get very good line number info yet so its not a real good list for isolating stacks for this bug, but the attachement might suggest other places where we need to apply skiplists.

Comment 5

[Steven Michaud [:smichaud] (Retired)]

Looking through your list, the following additions to the skip list
(to prefixSignatureRegEx) strike me as most likely to be useful
(besides the ones I've already suggested):

AppKit@
CoreFoundation@
Foundation@
_objc_error
libSystem.B.dylib@
libhotspot.dylib@
libzip.jnilib@

In my experience, stacks that start with JVM_* are almost always
useless -- because all of the symbols in the top few levels are either
absent or spurious.

Comment 6

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

We shouldn't just be SkipListing wholesale the OS libraries that aren't resymbolized, though; we should rather try to get symbols for them.

I've just this week gotten a new set of symbols from a 10.4.11 user whose crashes had unresymbolized OS library frames, and I'm in the process of downloading 10.5 updates to take a fresh 10.5 partition all the way to current to pick up symbols for updates released before I took over the Mac OS X symbols.

I agree that the Java libs tend to end up without symbols in stacks even when we have symbols for them (cl and I have both seen that happen in our Java crashes, and our module IDs matched what we had symbols for).

Comment 7

[Steven Michaud [:smichaud] (Retired)]

I'm not entirely sure what "skip list" means in this context.

I don't want the elements *skipped* that I listed in comment #5.
Rather (as I said in comment #2) I want them treated like
'objc_msgSend' currently is:  I want them them to be considered as
part of a compound signature.  For example

libclient.dylib@0x8b560 | +[AppletView isJava142:]

Comment 8

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

(In reply to comment #7)
> I don't want the elements *skipped* that I listed in comment #5.
> Rather (as I said in comment #2) I want them treated like
> 'objc_msgSend' currently is:  I want them them to be considered as
> part of a compound signature.  For example

Ah, OK.  Compounding/prepending is much better. :-)

I also see bug 565412 is covering a more limited set of libraries than comment 5, which I think is a very good thing; I left a more extensive comment about that in bug 565412 comment 5.

Comment 10

[Benjamin Smedberg]

Closing old bugs in the Plugins component. We aren't going to track issues in 3rd-party plugins in the Mozilla bug tracker. In addition, support for NPAPI plugins will be removed at the end of this year; for more details see the post at https://blog.mozilla.org/futurereleases/2015/10/08/npapi-plugins-in-firefox/

If there is a serious bug in Firefox, it needs to be filed in the "Core" product, "Plug-Ins" component.