862913 - [Lockscreen] lockscreen shown without background image for a while when turning the screen back on

Status: RESOLVED WONTFIX

(Firefox OS Graveyard :: Gaia::System::Lockscreen, defect)

Description

[leo.bugzilla.gecko]

1. Title : sometimes clock is showed without background image when turning the screen back on
2. Precondition : leave the device untim screen turned off
3. Tester's Action : After for a while(10mins or few hours after), turn on the screen using power button or home button. 
4. Detailed Symptom (ENG.) : we can see the lockscreen but background image which is same with wallpaper for a short time. After a second, the background image is shown. 
5. Expected : should be appeared the lockscreen with background image.
6. Reproducibility: Y
   1)Frequency Rate : rarely

Comment 1

[Justin Lebar (not reading bugmail)]

If we end up doing bug 862970, I think that will solve this problem.

Comment 2

[Jinho Hwang [:jeffhwang]]

Michael made a patch for this bug to remove calling setVisible() when shell.js got the sizemodechange event. And when I applied the patch, I couldn't reproduce this.
Also, it can fix bug 825174 too. Mike could you please attach the patch here.

Comment 3

[Justin Lebar (not reading bugmail)]

Please note that this approach will cause other problems, until Gaia is modified so that it setVisible(false)'s all frames other than the system app when the screen is turned off.

Comment 4

[Jinho Hwang [:jeffhwang]]

Yes, You're right Justin. I'm also concern about that.
Could you give me some more details on your expectation about the other problems that this approach will cause? 
when I applied the patch and tested, there was one problem(rarely the screen turning back on 2 seconds after I press the power button) but it was device specific only happened on leo and they solved. 
After this I could't find any others.
I'd like to hear your opinions on this.

Comment 5

[Justin Lebar (not reading bugmail)]

If you don't setVisible(false) all other frames, the frames won't know that they're no longer visible, and they won't send an event down to Gaia or the app informing Gaia/the app that the frame is no longer visible.  Missing this event would be a web-facing regression.  It's been known to cause all sorts of problems, like high energy consumption when playing music.

Comment 6

[leo.bugzilla.gecko]

Attachment: prevent to do setVisible(false), if there is a LockScreen enabled.

Comment 7

[Jinho Hwang [:jeffhwang]]

Thanks Justin for the information.
I think there is a way to solve this bug without any other side-effect like high energy consumption you mentioned on comment #5. 
We can just prevent to do setVisible(false) only there is lockscreen enabled. Otherwise, we do not change anything  on sizemodechange event.
It works because.. since the lockscreen enabled, current application running on fg wil get the setVisible(false) when the power key pressed to turn off the screen by window manager.

So, I think it can be a workaround for this, Until we get the bug 862970 fixed.
I attached the tiny patch for this here ... just in case..

Comment 8

[Sotaro Ikeda [:sotaro]]

It seems that all ui is under 'screen'
https://github.com/mozilla-b2g/gaia/blob/master/apps/system/index.html#L297

Comment 9

[Sotaro Ikeda [:sotaro]]

sorry, ignore comment #8

Comment 10

[Sotaro Ikeda [:sotaro]]

(In reply to JinhoHwang from comment #2)
> Also, it can fix bug 825174 too. Mike could you please attach the patch here.

I applied attachment 740707 [details] [diff] [review] to buri device. I still see some bug symptoms of bug 825174. When power on/off button pushed quickly twice, white flash was displayed relatively long time(about 1sec).

Comment 11

[Sotaro Ikeda [:sotaro]]

(In reply to Sotaro Ikeda [:sotaro] from comment #10)
> I applied attachment 740707 [details] [diff] [review] to buri device. I
> still see some bug symptoms of bug 825174. When power on/off button pushed
> quickly twice, white flash was displayed relatively long time(about 1sec).

I happen on buri device. As in Bug 825174 comment #33, it seems different problem.

Comment 12

[Jinho Hwang [:jeffhwang]]

(In reply to Sotaro Ikeda [:sotaro] from comment #10)

> I applied attachment 740707 [details] [diff] [review] to buri device. I
> still see some bug symptoms of bug 825174. 

the attachment 740707 [details] [diff] [review] is not the one mentioned on comment #2.
it only works for this lockscreen bug(when there is a lockscreen enabled) 

if you just remove this line(http://dxr.mozilla.org/mozilla-central/b2g/chrome/content/shell.js#l427) in shell.js, you won't be able to reproduce the Bug 825174 on your device.

> When power on/off button pushed
> quickly twice, white flash was displayed relatively long time(about 1sec).

I will check about it. Thanks for the report.

Comment 13

[Sotaro Ikeda [:sotaro]]

(In reply to JinhoHwang from comment #12)
> (In reply to Sotaro Ikeda [:sotaro] from comment #10)
>
> > When power on/off button pushed
> > quickly twice, white flash was displayed relatively long time(about 1sec).
> 
> I will check about it. Thanks for the report.

Sorry, It happens only on mozilla's default system app. It do not happen on partner's built ROM. See Bug 825174 comment #36.

Comment 14

[Jinho Hwang [:jeffhwang]]

I just attached a patch attachment 749180 [details] [diff] [review] for this bug and Bug 825174 to prevent flashing cached resources & updating imagelockingstate which is the main cause of this bug when turning off the screen. 
pls, see https://bugzilla.mozilla.org/show_bug.cgi?id=825174#c40 too.

Comment 15

[BMO Automation]

Firefox OS is not being worked on