817888 - Call waiting should turn the screen back on

Status: VERIFIED FIXED

(Firefox OS Graveyard :: Gaia::System, defect, P1)

Description

[Michael Schwartz [:m4]]

Connect a MO/MT call, wait for screen to turn off, initiate a MT call (with CW enabled).  Expected: screen turns on; Observed: it does not.

In the logs, I can see that the update in call state is served to Gaia and if the screen is manually turned on, Gaia is displaying the CW call correctly.

Comment 1

[Faramarz Rashed]

this seems like it's necessary for certification; mvines?

Comment 2

[Michael Vines [:m1] [:evilmachines]]

Yeah, this is definitely something that I don't think we could ship without a fix for.

Comment 3

[Damon Sicore (:damons)]

Who owns this?

Comment 4

[Dylan Oliver [:doliver]]

Assigned to Etienne and cc: Francisco in case he's a better match for it.

Comment 5

[Dylan Oliver [:doliver]]

Moving into C2 due to 802677 blocker.

Comment 6

[Tim Guan-tin Chien [:timdream] (please needinfo)]

This should be implemented in a way that when the second call incoming, the dialer should request another screen wake lock.

I think I can come up with a fix tomorrow Taipei time... will steal if I can verify my fix works before anyone else has work on it.

Comment 7

[Etienne Segonzac (:etienne)]

(In reply to Tim Guan-tin Chien [:timdream] (MoCo-TPE) from comment #6)
> This should be implemented in a way that when the second call incoming, the
> dialer should request another screen wake lock.
> 
> I think I can come up with a fix tomorrow Taipei time... will steal if I can
> verify my fix works before anyone else has work on it.

Hey Tim, I have a fix almost ready, I'll ping you for review in a few.
Hope we're not conflicting.

Comment 8

[Etienne Segonzac (:etienne)]

Attachment: Patch proposal

Comment 9

[Tim Guan-tin Chien [:timdream] (please needinfo)]

Comment on attachment 688811 [details] [diff] [review]
Patch proposal

r=me.

I am not sure screen_manager.js change is necessary (supposedly, when you get a screen lock, screen_manager.js:L82 should turn on the screen for you). That could be a bug. If that's the case, turning the screen on at L184 is ok.

Your fix is much more complete than what I had in mind!

Comment 10

[Chris Jones [:cjones] inactive; ni?/f?/r? if you need me]

Is this ready to land?

Comment 11

[Etienne Segonzac (:etienne)]

https://github.com/mozilla-b2g/gaia/commit/e82cd0503ddce8c3845b1da57da723d15227b144

Comment 12

[Etienne Segonzac (:etienne)]

(In reply to Tim Guan-tin Chien [:timdream] (MoCo-TPE) from comment #9)
> Comment on attachment 688811 [details] [diff] [review]
> Patch proposal
> 
> r=me.
> 
> I am not sure screen_manager.js change is necessary (supposedly, when you
> get a screen lock, screen_manager.js:L82 should turn on the screen for you).
> That could be a bug. If that's the case, turning the screen on at L184 is ok.

Opening a followup for this, since indeed the screen wasn't turned back on simply by requesting a lock.

Comment 13

[Tim Guan-tin Chien [:timdream] (please needinfo)]

I should share the blame; I should have find out this during review :'(

I did check the |screenLock| variable and found one on L207, but now I see it's totally different scope.

Comment 14

[Etienne Segonzac (:etienne)]

And the effect was really sneaky: we got the multiple call but the UI was weird.
Spent some time in CSS before finding it :)

Comment 15

[mlevin]

Verified fixed on Unagi build 20130103070201.