994408 - [B2G][Browser]Dragging youtube slider can cause video to skip to end, any subsequent slider adjustments auto snap to end of video

Status: RESOLVED WORKSFORME

(Core :: Audio/Video, defect)

Description

[Ross Kuhlman ([e-mail])]

Attachment: YoutubeSkip.txt

The youtube video slider does not function properly.  After dragging the slider to skip ahead in the video, the slider will automatically jump to the end of the video. After this occurs, the slider will repeatedly snap to the end of the video, and the user will have to drag the slider multiple times in order to resume playing.

Repro Steps:
1) Update a Buri to BuildID: 20140409130909
2) Launch Browser app.
3) Watch a youtube video.
4) Drag the youtube slider forward until it is past the buffered area.

Actual
The slider jumps ahead to the end of the video, and sticks there.

Expected:
The slider jumps to the point indicated by the user, and resumes playing from there.

Master M- Environmental Variables:
Device: Buri Master M-C MOZ
BuildID: 20140409130909
Gaia: 9d0b1bdf746823a94b13e6574c1d8304dc584763
Gecko: 5a5ed08df529
Version: 31.0a1
Firmware Version: v1.2-device.cfg

Notes:

Repro frequency: 60%
See attached: logcat

Comment 1

[Jason Smith [:jsmith]]

Does this happen on 1.4?

Comment 2

[Ross Kuhlman ([e-mail])]

I am unable to reproduce the issue on 1.4

v1.4 Environmental Variables:
Device: Buri v1.4 MOZ
BuildID: 20140409000202
Gaia: 26983f356ecb1bcf30e862d334b5de790071803e
Gecko: e450e07e3a58
Version: 30.0a2
Firmware Version: v1.2-device.cfg

Comment 3

[Jason Smith [:jsmith]]

I'm wondering if this is a fast seek regression.

Comment 4

[Jason Smith [:jsmith]]

(In reply to Jason Smith [:jsmith] from comment #3)
> I'm wondering if this is a fast seek regression.

Chris - What do you think?

Comment 5

[Chris Pearce [:cpearce (Not reading bugmail)]]

Possible. Can we get a regression range search?

Comment 6

[Jason Smith [:jsmith]]

(In reply to Chris Pearce (:cpearce) from comment #5)
> Possible. Can we get a regression range search?

Yup - I'll get someone assigned to do that.

Comment 7

[Matthew Vaughan]

MOZILLA INBOUND:

- Last Working -
Device: Buri Master (1.5) MOZ RIL
BuildID: 20140331200633
Gaia: eee8caa81a368f0feace718201ed15a423812c18
Gecko: d313210d2619
Version: 31.0a1
Firmware Version: v1.2-device.cfg

- First Broken -
Device: Buri Master (1.5) MOZ RIL
BuildID: 20140331204432
Gaia: eee8caa81a368f0feace718201ed15a423812c18
Gecko: 35180f110e44
Version: 31.0a1
Firmware Version: v1.2-device.cfg


**This looks to be a gecko issue since the gaia is the same for both builds.**

Push log: http://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=d313210d2619&tochange=35180f110e44

Comment 8

[Jason Smith [:jsmith]]

This is either caused by bug 631058 or bug 778077.

Comment 9

[Jason Smith [:jsmith]]

(In reply to Jason Smith [:jsmith] from comment #8)
> This is either caused by bug 631058 or bug 778077.

Chris - Do you know which bug caused this regression?

Comment 10

[Chris Pearce [:cpearce (Not reading bugmail)]]

I don't know. I can't reproduce this bug in my Otoro.

But I can reproduce a bug where if I seek to the end of a m.youtube.com video (into an unbuffered range) and then try to re-play the video the browser locks up. Sometimes I get audio playback, but I don't see the video frame update, and the UI is unresponsive. And I see lots of this in adb logcat:

E/libgenlock(  104): perform_lock_unlock_operation: GENLOCK_IOC_LOCK failed (lockType0x2, err=Connection timed out fd=246)
E/Adreno200-EGLSUB(  104): LockImage() genlock_lock_buffer GENLOCK_READ_LOCK failed

Comment 11

[bhavana bajaj [:bajaj]]

(In reply to Chris Pearce (:cpearce) from comment #10)
> I don't know. I can't reproduce this bug in my Otoro.
> 
> But I can reproduce a bug where if I seek to the end of a m.youtube.com
> video (into an unbuffered range) and then try to re-play the video the
> browser locks up. Sometimes I get audio playback, but I don't see the video
> frame update, and the UI is unresponsive. And I see lots of this in adb
> logcat:
> 
> E/libgenlock(  104): perform_lock_unlock_operation: GENLOCK_IOC_LOCK failed
> (lockType0x2, err=Connection timed out fd=246)
> E/Adreno200-EGLSUB(  104): LockImage() genlock_lock_buffer GENLOCK_READ_LOCK
> failed

Chris, any way you can try it on a Buri which was reported here ? That may eliminate the device dependency here.

Comment 12

[Chris Pearce [:cpearce (Not reading bugmail)]]

No. We don't have one of those in the Auckland office.

Comment 13

[Jason Smith [:jsmith]]

Eric - Can someone on your team take this since you guys have access to production devices?

Comment 14

[Eric Chou [:ericchou] [:echou]]

Sure. Assign to myself first and I'll delegate to Bruce or Blake after discussing with them.

Comment 15

[Eric Chou [:ericchou] [:echou]]

Bruce will work on this.

Comment 16

[Bruce Sun [:brsun]]

I've tried the latest master, I haven't seen this issue happens yet:
 - Gaia      265e1ac4ee71ad6190335c974bfce33f783edfce
 - Gecko     https://hg.mozilla.org/mozilla-central/rev/ea25086073c8
 - BuildID   20140427160202
 - Version   31.0a1
 - Firmware Version: v1.2-device.cfg

I can reproduce this issue on master build with BuildID:20140409130909. But I only encountered maybe 3~5 times this afternoon. The reproduce rate is not so high on my environment.

Hi rkuhlman,
Do you have any specific youtube sources by which we can more easily reproduce this issue?
Would you mind helping me try the latest master to see if we can still reproduce this issue on that?
Thanks for your help in advance.

Comment 17

[Ross Kuhlman ([e-mail])]

I had the most success when watching videos that are 5+ minutes in length, no specific sources. I am so far unable to reproduce the issue on todays master build.

Master M-C Environmental Variables:
Device: Buri Master M-C MOZ
BuildID: 20140430040206
Gaia: badc73ee7f108fa631150bded0cc57e92aad810e
Gecko: e19812f56952
Version: 32.0a1
Firmware Version: v1.2-device.cfg