985060 - [B2G][Everything.me] Searching in e.me then swiping to the right to the homescreen while the background image is loading results in icons looking sliced or broken

Status: RESOLVED WORKSFORME

(Core :: Graphics: Layers, defect)

Description

[Rachel Pribble [:rpribble]]

Attachment: BrokenIcons.ogg

Description:
Performing a search in the e.me bar, then swiping from right to left to view the homescreen while the background image is still loading on the e.me can result in the left row of homescreen items looking sliced or broken.

Repro Steps:
1) Updated Buri to BuildID: 20140318000203
2) Tap the e.me search bar
3) Type a few letters (4 used in test)
4) During or immediately after the loading symbol, before the background search image has finished loading, swipe from right to left to move to the homescreen
5) Observe the left row of homescreen icons appear broken

Actual:
Icons appear broken.

Expected:
No icons appear broken.

Environmental Variables:
Device: Buri v1.4 MOZ ril
BuildID: 20140318000203
Gaia: c03a6af9028c4b74a84b5a98085bbb0c07261175
Gecko: 3776f72f1967
Version: 30.0a2
Firmware Version: v1.2-device.cfg

Notes:

Repro frequency: 2/10, 20%
See attached: Video

Comment 1

[Rachel Pribble [:rpribble]]

This issue does not occur on the Buri v1.3 MOZ ril.

Environmental Variables:
Build ID: 20140318004002
Gecko: https://hg.mozilla.org/releases/mozilla-b2g28_v1_3/rev/cf2042938526
Gaia: 2ea2aab306bd1c941719160cdcb49ee9d755dc17
Platform Version: 28.0
Firmware Version: V1.2-device.cfg

Comment 2

[Jason Smith [:jsmith]]

Does this reproduce with tiling disabled?

Comment 3

[Milan Sreckovic [:milan] (needinfo for best results)]

Perhaps a duplicate of bug 984490?

Comment 4

[No-Jun Park [:njpark]]

It happens also with tiling disabled on 1.4

Comment 5

[Milan Sreckovic [:milan] (needinfo for best results)]

Hmm.  I was just going to say that it looks remarkably like bug 983883, bug 984531, bug 984618, but those are currently being treated as tiling only.

Comment 6

[Jason Smith [:jsmith]]

(In reply to Milan Sreckovic [:milan] from comment #5)
> Hmm.  I was just going to say that it looks remarkably like bug 983883, bug
> 984531, bug 984618, but those are currently being treated as tiling only.

Looks like we need a window then.

Comment 7

[KTucker [:KTucker][Inactive 3/4/2016]]

The following regression window was found using b2g-inbound builds:

- Last Working -
1.4 Environmental Variables:
Device: buri 1.4 MOZ
BuildID: 20140307085255
Gaia: 5de5cab90e1a83d81c0fc506912696d85b1865fe
Gecko: 6175ca5a9da4
Version: 30.0a1
Firmware Version: v1.2-device.cfg

- First Broken -
1.4 Environmental Variables:
Device: buri 1.4 MOZ
BuildID: 20140307091726
Gaia: 5de5cab90e1a83d81c0fc506912696d85b1865fe
Gecko: 45253e02bde3
Version: 30.0a1
Firmware Version: v1.2-device.cfg

Gecko Pushlog: http://hg.mozilla.org/integration/b2g-inbound/pushloghtml?fromchange=6175ca5a9da4&tochange=45253e02bde3

Comment 8

[Jason Smith [:jsmith]]

Can you include the gecko & gaia swap?

Assuming this is gecko related, then this was caused by bug 977757.

Comment 9

[KTucker [:KTucker][Inactive 3/4/2016]]

This is a Gecko issue. The Gaia is the same for both builds.

Comment 10

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Seems likely to be layers-related, given that will-change is changing our layerization decisions, though it's possible it could be closer to FrameLayerBuilder.

Comment 11

[Chris Lord [:cwiiis]]

A bit of a long shot, but is there any fixed position content involved here? I'm wondering if this is another manifestation of bug 874950. Even if it isn't, it looks like a similar issue - where content is incorrectly being occluded...

Comment 12

[Chris Lord [:cwiiis]]

(In reply to Chris Lord [:cwiiis] from comment #11)
> A bit of a long shot, but is there any fixed position content involved here?
> I'm wondering if this is another manifestation of bug 874950. Even if it
> isn't, it looks like a similar issue - where content is incorrectly being
> occluded...

In fact, thinking about this, bug 874950 only fixes this for layers with displayports, but I suppose the same thing could happen for async-transformed layers too.

Comment 13

[Chris Lord [:cwiiis]]

I'm on PTO for two weeks and haven't had the time to look at this yet. needinfo'ing myself in case this isn't picked up again before then.

Comment 14

[Kai-Chih Hu [:khu]]

Milan, because Chris is on PTO for 2 weeks and this bug is a 1.4+ bug, could we find another one to help? Thanks!

Comment 15

[Andrew Overholt [:overholt]]

Attachment: frame in video

The video's great but I thought a frame of it showing the problem may be helpful.

Comment 16

[Milan Sreckovic [:milan] (needinfo for best results)]

(In reply to Kevin Hu [:khu] from comment #14)
> Milan, because Chris is on PTO for 2 weeks and this bug is a 1.4+ bug, could
> we find another one to help? Thanks!

If something opens up, but I wouldn't count on it.

Comment 17

[Kai-Chih Hu [:khu]]

This bug has no update for a while. Not sure if we have any action plan or not. :)

Comment 18

[Milan Sreckovic [:milan] (needinfo for best results)]

We wait for Chris to come back and we look at this in priority order.

Comment 19

[Milan Sreckovic [:milan] (needinfo for best results)]

(In reply to Milan Sreckovic [:milan] from comment #18)
> We wait for Chris to come back and we look at this in priority order.

Basically, as suggested by comment 16.

Comment 20

[Kai-Chih Hu [:khu]]

Thanks, Milan.

Comment 21

[Milan Sreckovic [:milan] (needinfo for best results)]

Chris, can you take a look at this now you're back?

Comment 22

[Chris Lord [:cwiiis]]

I can't reproduce this using the instructions in comment #0/trying to follow the video on latest master. Is this still an issue for the reporter?

Comment 23

[Harvey]

This issue does NOT reproduce in the latest Master build.

2.0 Environmental Variables:
Device: Buri 2.0
BuildID: 20140430115207
Gaia: e729838c39856046f5526468b18cfa507bd613a6
Gecko: 69ae2efbe943
Version: 30.0
Firmware Version: v1.2-device.cfg

Comment 24

[Jason Smith [:jsmith]]

(In reply to Harvey from comment #23)
> This issue does NOT reproduce in the latest Master build.
> 
> 2.0 Environmental Variables:
> Device: Buri 2.0
> BuildID: 20140430115207
> Gaia: e729838c39856046f5526468b18cfa507bd613a6
> Gecko: 69ae2efbe943
> Version: 30.0
> Firmware Version: v1.2-device.cfg

Can you check 1.4 as well?

Comment 25

[Harvey]

This issue does NOT reproduce on 1.4

Environmental Variables:
Device: Buri v1.4 
BuildID: 20140430000201
Gaia: 81e97c3ca58be0487292011bc59efa4cebab30be
Gecko: 123485e733d5
Version: 30.0
Firmware Version: v1.2-device.cfg