782096 - java.lang.NullPointerException: at android.os.Parcel.readException(Parcel.java) on Galaxy Tab 7.7" (GT-P6800) with Honeycomb

Status: RESOLVED FIXED

(Firefox for Android Graveyard :: Keyboards and IME, defect)

Description

[Scoobidiver (away)]

There are 65 crashes in 15.0 Beta, including bp-c2304ebc-8734-4ed8-8c28-6f76f2120811.

java.lang.NullPointerException
	at android.os.Parcel.readException(Parcel.java:1327)
	at android.os.Parcel.readException(Parcel.java:1275)
	at com.android.internal.view.IInputMethodManager$Stub$Proxy.hideSoftInput(IInputMethodManager.java:606)
	at android.view.inputmethod.InputMethodManager.hideSoftInputFromWindow(InputMethodManager.java:857)
	at android.view.inputmethod.InputMethodManager.hideSoftInputFromWindow(InputMethodManager.java:829)
	at org.mozilla.gecko.AwesomeBar$6.onFocusChange(AwesomeBar.java:241)
	at android.view.View.onFocusChanged(View.java:3419)
	at android.widget.TextView.onFocusChanged(TextView.java:7614)
	at android.widget.EditText.onFocusChanged(EditText.java:136)
	at android.view.View.unFocus(View.java:3347)
	at android.view.ViewGroup.unFocus(ViewGroup.java:630)
	at android.view.ViewGroup.requestChildFocus(ViewGroup.java:512)
	at android.view.ViewGroup.requestChildFocus(ViewGroup.java:518)
	at android.view.ViewGroup.requestChildFocus(ViewGroup.java:518)
	at android.view.ViewGroup.requestChildFocus(ViewGroup.java:518)
	at android.view.ViewGroup.requestChildFocus(ViewGroup.java:518)
	at android.view.View.handleFocusGainInternal(View.java:3239)
	at android.view.View.requestFocus(View.java:4460)
	at android.view.View.requestFocus(View.java:4410)
	at android.view.View.requestFocus(View.java:4388)
	at android.widget.TabHost.onTouchModeChanged(TabHost.java:204)
	at android.view.ViewTreeObserver.dispatchOnTouchModeChanged(ViewTreeObserver.java:606)
	at android.view.ViewRoot.ensureTouchModeLocally(ViewRoot.java:2331)
	at android.view.ViewRoot.ensureTouchMode(ViewRoot.java:2315)
	at android.view.View.requestFocusFromTouch(View.java:4485)
	at org.mozilla.gecko.AwesomeBar.onKeyDown(AwesomeBar.java:471)
	at android.view.KeyEvent.dispatch(KeyEvent.java:2540)
	at android.app.Activity.dispatchKeyEvent(Activity.java:2280)
	at com.android.internal.policy.impl.PhoneWindow$DecorView.dispatchKeyEvent(PhoneWindow.java:1819)
	at android.view.ViewRoot.deliverKeyEventPostIme(ViewRoot.java:2963)
	at android.view.ViewRoot.handleMessage(ViewRoot.java:2234)
	at android.os.Handler.dispatchMessage(Handler.java:99)
	at android.os.Looper.loop(Looper.java:132)
	at android.app.ActivityThread.main(ActivityThread.java:4123)
	at java.lang.reflect.Method.invokeNative(Native Method)
	at java.lang.reflect.Method.invoke(Method.java:491)
	at com.android.internal.os.ZygoteInit$MethodAndArgsCaller.run(ZygoteInit.java:849)
	at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:607)
	at dalvik.system.NativeStart.main(Native Method)

More reports at:
https://crash-stats.mozilla.com/report/list?signature=java.lang.NullPointerException%3A+at+android.os.Parcel.readException%28Parcel.java%29

Comment 1

[Chris Peterson [:cpeterson]]

Curiously, these hideSoftInputFromWindow() crashes are all Galaxy Tab users running Android 3.2 (Honeycomb). 

The NPE is [re]thrown by Parcel.readException(), which means the original NPE was thrown from InputMethodManagerService's hideSoftInput(). I wonder if there was a bug in Honeycomb's InputMethodManagerService that was fixed in ICS? Our client should not be able to trigger NPEs in the service!

Comment 2

[Chris Peterson [:cpeterson]]

Attachment: catch-NPE.patch

Catch unexpected NullPointerException from InputMethodManagerService.

Comment 3

[Chris Peterson [:cpeterson]]

(In reply to Chris Peterson (:cpeterson) from comment #2)
> Created attachment 656966 [details] [diff] [review]
> catch-NPE.patch
> 
> Catch unexpected NullPointerException from InputMethodManagerService.

snorp, have you had a chance to review my fix for this InputMethodManagerService NullPointerException we discussed in last week's Bug Deep Dive meeting?

Comment 4

[James Willcox (:snorp) ([e-mail]) (he/him)]

Comment on attachment 656966 [details] [diff] [review]
catch-NPE.patch

Review of attachment 656966 [details] [diff] [review]:
-----------------------------------------------------------------

Gross, but looks good.

Comment 5

[Chris Peterson [:cpeterson]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/41cf3c361d9d

Comment 6

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/41cf3c361d9d

Comment 7

[Chris Peterson [:cpeterson]]

Comment on attachment 656966 [details] [diff] [review]
catch-NPE.patch

[Approval Request Comment]
Bug caused by (feature/regressing bug #): N/A
User impact if declined: Intermittent crashes for some Honeycomb tablet users. I suspect this is a bug in Honeycomb that was fixed in ICS+.
Testing completed (on m-c, etc.): Tested locally, BUT this is an intermittent bug I have not repro'd myself.
Risk to taking this patch (and alternatives if risky): Low risk. The code change simply catches a NullPointerException from Google code instead of crashing.
String or UUID changes made by this patch: N/A

Comment 8

[Lukas Blakk [:lsblakk] use ?needinfo]

Comment on attachment 656966 [details] [diff] [review]
catch-NPE.patch

mobile-only, low risk, still early in beta - we can take this to branches.

Comment 9

[Chris Peterson [:cpeterson]]

https://hg.mozilla.org/releases/mozilla-aurora/rev/61502f07b33c

Comment 10

[Chris Peterson [:cpeterson]]

https://hg.mozilla.org/releases/mozilla-beta/rev/e34089cb1120

Comment 11

[Kevin Brosnan [Ex-Mozilla]]

This is still crashing at about the same rate as mentioned in comment 0.

Comment 12

[Mark Finkle (:mfinkle) (use needinfo?)]

Let's open a new bug for this. The patch was landed and pushed all the way to beta.