1158309 - Exception handling message "Gecko:Ready"

Status: RESOLVED FIXED

(Firefox for Android Graveyard :: General, defect)

Description

[Geoff Brown [:gbrown]]

Some failed test logs contain

12:02:08     INFO -  04-24 11:54:49.333 E/GeckoApp(  787): Exception handling message "Gecko:Ready":
12:02:08     INFO -  04-24 11:54:49.333 E/GeckoApp(  787): java.lang.NullPointerException
12:02:08     INFO -  04-24 11:54:49.333 E/GeckoApp(  787): 	at org.mozilla.gecko.GeckoApp.geckoConnected(GeckoApp.java:2387)
12:02:08     INFO -  04-24 11:54:49.333 E/GeckoApp(  787): 	at org.mozilla.gecko.GeckoApp.handleMessage(GeckoApp.java:667)
12:02:08     INFO -  04-24 11:54:49.333 E/GeckoApp(  787): 	at org.mozilla.gecko.BrowserApp.handleMessage(BrowserApp.java:1822)
12:02:08     INFO -  04-24 11:54:49.333 E/GeckoApp(  787): 	at org.mozilla.gecko.EventDispatcher.dispatchEvent(EventDispatcher.java:208)
12:02:08     INFO -  04-24 11:54:49.333 E/GeckoApp(  787): 	at org.mozilla.gecko.EventDispatcher.dispatchEvent(EventDispatcher.java:178)
12:02:08     INFO -  04-24 11:54:49.333 E/GeckoApp(  787): 	at org.mozilla.gecko.GeckoAppShell.handleGeckoMessage(GeckoAppShell.java:2253)
12:02:08     INFO -  04-24 11:54:49.333 E/GeckoApp(  787): 	at org.mozilla.gecko.mozglue.GeckoLoader.nativeRun(Native Method)
12:02:08     INFO -  04-24 11:54:49.333 E/GeckoApp(  787): 	at org.mozilla.gecko.GeckoAppShell.runGecko(GeckoAppShell.java:382)
12:02:08     INFO -  04-24 11:54:49.333 E/GeckoApp(  787): 	at org.mozilla.gecko.GeckoThread.run(GeckoThread.java:187)

For examples, see https://bugzilla.mozilla.org/show_bug.cgi?id=1054292#c1803 and https://bugzilla.mozilla.org/show_bug.cgi?id=1054292#c1804.

Comment 1

[Geoff Brown [:gbrown]]

:jchen -- I'm pretty sure this is related to bug 1150693; can you have a look?

Comment 2

[(inactive) Jim Chen [:jchen] [:darchons]]

Attachment: Initialize LayerView sooner (v1)

This should fix the race condition. It's kind of gross that we need a focus hack to get tests to pass, but it was coincidence that we didn't need the hack before.

Comment 3

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

Comment on attachment 8599486 [details] [diff] [review]
Initialize LayerView sooner (v1)

Review of attachment 8599486 [details] [diff] [review]:
-----------------------------------------------------------------

::: mobile/android/base/GeckoApp.java
@@ +1414,5 @@
>                  ((SurfaceView)mCameraView).getHolder().setType(SurfaceHolder.SURFACE_TYPE_PUSH_BUFFERS);
>              }
>          }
>  
> +        // XXX our editor tests require the GeckoView to have focus to pass, so we have to

I don't think the XXX adds anything :)

Comment 4

[Pulsebot]

https://hg.mozilla.org/integration/fx-team/rev/7552ffb7bd0c

Comment 5

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/7552ffb7bd0c

Comment 7

[Kevin Brosnan [Ex-Mozilla]]

[Tracking Requested - why for this release]: This is a bad bug. We should consider Aurora and maybe even 38.0.5. See bug 1129074 for how it manifests and https://input.mozilla.org/en-US/?q=white+or+blank&product=Firefox+for+Android&date_start=2015-02-04&selected=90d for user feedback that may be relevant.

Comment 8

[Sylvestre Ledru [:Sylvestre]]

Jim, can you provide an uplift request to mozilla release? (for 38.0.5beta)

Comment 9

[(inactive) Jim Chen [:jchen] [:darchons]]

We'd have to uplift both bug 1150693 and this bug. It's kind of risky...

Comment 10

[Sylvestre Ledru [:Sylvestre]]

OK. Thanks.
Let it ride the train then!

Comment 11

[(inactive) Jim Chen [:jchen] [:darchons]]

Comment on attachment 8599486 [details] [diff] [review]
Initialize LayerView sooner (v1)

Approval Request Comment

[Feature/regressing bug #]: N/A

[User impact if declined]: This bug is a regression from bug 1150693. It should be uplifted together with bug 1150693.

[Describe test coverage new/current, TreeHerder]: m-c; locally

[Risks and why]: Same risk as bug 1150693

[String/UUID change made/needed]: None

Comment 12

[Ryan VanderMeulen [:RyanVM]]

Comment on attachment 8599486 [details] [diff] [review]
Initialize LayerView sooner (v1)

This landed on m-c when it was still Gecko 40, so it doesn't need to land on Aurora.

Comment 13

[Liz Henry (:lizzard)]

I'd like this to ride along with 40 unless it is definitely needed in beta along with bug 1150693.

Comment 14

[Liz Henry (:lizzard)]

Wontfix for 39 as discussed in bug 1150693.