879854 - [debug][spew] WARNING: NS_ENSURE_TRUE(AttrToDataProp(attr, prop)) failed: file /content/html/content/src/nsDOMStringMap.cpp, line 166

Status: RESOLVED WONTFIX

(Firefox OS Graveyard :: General, defect)

Description

[Mike Habicher [:mikeh] (high bugzilla latency)]

Observed on an Inari DEBUG build:
- gecko: b2g18:09dc1ae3b1b5
- gaia: v1.0.1:37b6527c3f73a497f75d057e90386f77ff5a552b

This line appears over and over again in the logcat output:

06-05 13:03:52.990   111   111 I Gecko   : [Parent 111] WARNING: NS_ENSURE_TRUE(AttrToDataProp(attr, prop)) failed: file /home/mikeh/dev/mozilla/m-c/b2g18/content/html/content/src/nsDOMStringMap.cpp, line 166

Comment 1

[Mike Habicher [:mikeh] (high bugzilla latency)]

See also bug 879856.

Comment 2

[Mike Habicher [:mikeh] (high bugzilla latency)]

gdb bt:

Breakpoint 1, nsDOMStringMap::RemovePropInternal (this=0x48b39cc0, aAttr=0x48c59cc0) at /home/mikeh/dev/mozilla/m-c/b2g18/content/html/content/src/nsDOMStringMap.cpp:166
166	  NS_ENSURE_TRUE(AttrToDataProp(attr, prop), NS_OK);
(gdb) n
165	  nsAutoString prop;
(gdb) bt
#0  nsDOMStringMap::RemovePropInternal (this=0x48b39cc0, aAttr=0x48c59cc0) at /home/mikeh/dev/mozilla/m-c/b2g18/content/html/content/src/nsDOMStringMap.cpp:165
#1  0x4113b858 in nsDOMStringMapRemoveProp::Run (this=0x487affa0) at /home/mikeh/dev/mozilla/m-c/b2g18/content/html/content/src/nsDOMStringMap.cpp:63
#2  0x40fdaa6e in nsContentUtils::AddScriptRunner (aRunnable=0x487affa0) at /home/mikeh/dev/mozilla/m-c/b2g18/content/base/src/nsContentUtils.cpp:5046
#3  0x4113aa1c in nsDOMStringMap::RemoveProp (this=<value optimized out>, aAttr=0x48c59cc0) at /home/mikeh/dev/mozilla/m-c/b2g18/content/html/content/src/nsDOMStringMap.cpp:158
#4  0x410e815c in nsGenericHTMLElement::UnsetAttr (this=0x49856ba0, aNameSpaceID=0, aAttribute=0x48c59cc0, aNotify=true) at /home/mikeh/dev/mozilla/m-c/b2g18/content/html/content/src/nsGenericHTMLElement.cpp:2035
#5  0x4113b108 in nsDOMStringMap::RemoveDataAttr (this=<value optimized out>, aProp=<value optimized out>) at /home/mikeh/dev/mozilla/m-c/b2g18/content/html/content/src/nsDOMStringMap.cpp:147
#6  0x4121f42e in nsDOMStringMapSH::DelProperty (this=0x465bbf80, wrapper=<value optimized out>, cx=<value optimized out>, obj=<value optimized out>, id=..., vp=0xbecf5c10, _retval=0xbecf5b07)
    at /home/mikeh/dev/mozilla/m-c/b2g18/dom/base/nsDOMClassInfo.cpp:8719
#7  0x41571d8e in XPC_WN_Helper_DelProperty (cx=0x48b01d10, obj=..., id=..., vp=...) at /home/mikeh/dev/mozilla/m-c/b2g18/js/xpconnect/src/XPCWrappedNativeJSOps.cpp:911
#8  0x41d54454 in js::CallJSPropertyOp (cx=0x48b01d10, op=0x41571d0d <XPC_WN_Helper_DelProperty>, receiver=..., id=..., vp=...) at /home/mikeh/dev/mozilla/m-c/b2g18/js/src/jscntxtinlines.h:437
#9  0x41d5dc8e in js::baseops::DeleteGeneric (cx=0x48b01d10, obj=..., id=..., rval=..., strict=1) at /home/mikeh/dev/mozilla/m-c/b2g18/js/src/jsobj.cpp:4777
#10 0x41d5df30 in js::baseops::DeleteProperty (cx=0x48b01d10, obj=..., name=<value optimized out>, rval=..., strict=1) at /home/mikeh/dev/mozilla/m-c/b2g18/js/src/jsobj.cpp:4809
#11 0x41d250fa in JSObject::deleteProperty (cx=0x48b01d10, obj=..., name=..., rval=..., strict=true) at /home/mikeh/dev/mozilla/m-c/b2g18/js/src/jsobjinlines.h:194
#12 0x41fcde64 in js::mjit::stubs::DelProp<1> (f=..., name_=<value optimized out>) at /home/mikeh/dev/mozilla/m-c/b2g18/js/src/methodjit/StubCalls.cpp:1444
#13 0x41eed2c6 in JaegerStubVeneer () from /home/mikeh/dev/mozilla/btg024/objdir-gecko-b2g18-debug/dist/bin/libxul.so
#14 0x41eed2c6 in JaegerStubVeneer () from /home/mikeh/dev/mozilla/btg024/objdir-gecko-b2g18-debug/dist/bin/libxul.so
Backtrace stopped: previous frame identical to this frame (corrupt stack?)

Comment 3

[Lukas Blakk [:lsblakk] use ?needinfo]

User impact?

Comment 4

[Mike Habicher [:mikeh] (high bugzilla latency)]

(In reply to [e-mail] [:lsblakk] from comment #3)
>
> User impact?

This line basically appears constantly and endlessly in the logcat, which could mean something is spinning on the failure, which would result in poor battery life.

Comment 5

[bhavana bajaj [:bajaj]]

Needsinfo Mike here to help with some numbers how  battery life may really be impacted ? As is we may not be able to block on the issue and need some data driven evidence to make a call.

Mike, is it possible to do a test run for a good/bad build to start with or may be check some battery related automated test we may already have ?

Comment 6

[Wayne Chang [:wchang]]

Triage agrees to minus but track.

Comment 7

[Mike Habicher [:mikeh] (high bugzilla latency)]

This spew is still happening. Looks like the error is printed over << 100 TIMES PER SECOND >>.

Evidently this is not really causing anything to fail. I'm tempted to just change the NS_ENSURE_TRUE() call to an if (! /* true */) return;, but we should probably figure out who is wailing on this function. Looking at the call stack above, we seem to be deleting a JS property. Hopefully figuring out which one is easy to locate.

Comment 8

[Mike Habicher [:mikeh] (high bugzilla latency)]

Attachment: Find out which attribute is causing the spew

06-26 15:35:10.130   474   474 I Gecko   : nsDOMStringMap::RemovePropInternal: AttrToDataProp() failed, attr='hidden'

That doesn't help narrow the field much, does it?

Comment 9

[Mike Habicher [:mikeh] (high bugzilla latency)]

Looks like the status bar is the culprit:

(gdb) call DumpJSStack()
0 sb_updateSignal() ["app://system.gaiamobile.org/js/statusbar.js":485]
    this = [object Object]
1 sb_handleEvent() ["app://system.gaiamobile.org/js/statusbar.js":272]
    this = [object Object]

Over several iterations on this breakpoint, the callstack looks the same.

frame 1 context:
      case 'callschanged':
>       this.update.signal.call(this);
        break;

frame 0 context:
      flightModeIcon.hidden = true;
>     icon.hidden = false;

alive, are you the right person to look into this?

Comment 10

[(inactive after 6/18) Alive Kuo [:alive]]

Not sure what to do, so shall we use removeAttribute('hidden') ?

Comment 11

[Mike Habicher [:mikeh] (high bugzilla latency)]

What would that do?

I would assume we want the icon hidden for a reason--I suppose the question is why is the icon being hidden 100 times/second?

Comment 12

[(inactive after 6/18) Alive Kuo [:alive]]

(In reply to Mike Habicher [:mikeh] from comment #11)
> What would that do?
> 
> I would assume we want the icon hidden for a reason--I suppose the question
> is why is the icon being hidden 100 times/second?

OK, I know what's the problem now, I'll investigate it soon.

Comment 13

[Vlad Filippov (:vladikoff)]

Can still reproduce this with a flashed build from `master` on Sept 14th, 2013.
Rooted device: ZTE Open (full_inari).

Comment 14

[BMO Automation]

Firefox OS is not being worked on