1148718 - [Nexus5][Video]The video preview is grey in comfirm page while selecting a video as attachment in message.

Status: VERIFIED FIXED

(Core :: Graphics, defect)

Description

[Alissa（leave from Mozilla）]

Attachment: logcat1.txt

[1.Description]:
[Nexus 5 v2.2&3.0][Video]Add a video as attachments in message, in the comfirm page, the first view of video cannot be displayed, the view is grey, sometimes, waiting some time or lock/unlock screen, the first view displayed, but if we lock and unlock screen agian, the view disappeared agian.
Found at:15:47
see attachments:logcat1.txt, video 1.3gp

[2.Testing Steps]: 
Precondition:You have taken a video from camera less than 295kb.
1.Launch message.
2.Tap New button.
3.Tap attachment button.
4.Tap video.
5.Select the video you taken from camera.

[3.Expected Result]: 
5.The fist view of view should be displayed in comfirm page.

[4.Actual Result]: 
5.In the video preview page, the video is grey, if we lock and unlock screen, the preview can be displayed in time.

[5.Reproduction build]: 
Device: Flame 2.2[Unaffected]
Build ID               20150327162502
Gaia Revision          473cd63f53c855299b719285d9b95e3f2910782f
Gaia Date              2015-03-27 20:14:43
Gecko Revision         https://hg.mozilla.org/releases/mozilla-b2g37_v2_2/rev/b358619def45
Gecko Version          37.0
Device Name            flame
Firmware(Release)      4.4.2
Firmware(Incremental)  eng.cltbld.20150327.194944
Firmware Date          Fri Mar 27 19:49:53 EDT 2015
Bootloader             L1TC000118D0

Device:Nexus5 3.0[Affected]
Build ID               20150327160203
Gaia Revision          9cc496cecc37d7a29f9279827cdf6e4891211f67
Gaia Date              2015-03-27 13:55:18
Gecko Revision         https://hg.mozilla.org/mozilla-central/rev/44e454b5e93b
Gecko Version          39.0a1
Device Name            hammerhead
Firmware(Release)      5.0
Firmware(Incremental)  eng.cltbld.20150327.192727
Firmware Date          Fri Mar 27 19:27:40 EDT 2015
Bootloader             HHZ12d

Devicce:Nexus5 2.2[Affected]
Build ID               20150327162502
Gaia Revision          473cd63f53c855299b719285d9b95e3f2910782f
Gaia Date              2015-03-27 20:14:43
Gecko Revision         https://hg.mozilla.org/releases/mozilla-b2g37_v2_2/rev/b358619def45
Gecko Version          37.0
Device Name            hammerhead
Firmware(Release)      5.0
Firmware(Incremental)  eng.cltbld.20150327.195950
Firmware Date          Fri Mar 27 20:00:03 EDT 2015
Bootloader             HHZ12d

[6.Reproduction Frequency]: 
Always Recurrence,5/5

[7.TCID]: 
Free Test

[8.Note]:
This issue doesn't exist on flame 2.2

Comment 1

[Alissa（leave from Mozilla）]

Attachment: 1.3gp

Comment 2

[Alissa（leave from Mozilla）]

[3.Expected Result]: 
5.The fist view of video should be displayed in comfirm page.

Comment 3

[Hermes Cheng (inactive after July 27, 2015)]

I prefer not a blocker. Jerry, do you know who can take a look? It might not be a gaia problem but a gecko one. I am not sure.

Comment 4

[Jerry Shih[:jerry] (UTC+8) (inactive)]

I think it's gecko issue. When I press the homescreen button to get the card-view snapshot, I can see the video content.

Comment 5

[Jerry Shih[:jerry] (UTC+8) (inactive)]

I try to dump the image content at ImageContainer::SetCurrentImage()[1] and the image data is correct. So we might have problem on b2g side. Investigating.

[1]
https://dxr.mozilla.org/mozilla-central/source/gfx/layers/ImageContainer.cpp?from=SetCurrentImage&case=true#202

Comment 6

[Jerry Shih[:jerry] (UTC+8) (inactive)]

Attachment: init TextureSource for current TextureHost when attach. v1

In this use case(see comment 0), we call UseTextureHost() before Attach(). Thus, we got failed in both PrepareTextureSource() and ImageLayerComposite::RenderLayer(). This patch just do PrepareTextureSource() again for the existed TextureHost.
Sotaro and Nical, what do you think about this patch? Should we handle other type of XXXComposite? I only can image that async video causes this problem.

Comment 7

[Jerry Shih[:jerry] (UTC+8) (inactive)]

2.2? for n5-l issue

Comment 8

[Jerry Shih[:jerry] (UTC+8) (inactive)]

try link for this patch:
https://treeherder.mozilla.org/#/jobs?repo=try&revision=b602a02d04a0

Comment 9

[Nicolas Silva [:nical]]

Comment on attachment 8587377 [details] [diff] [review]
init TextureSource for current TextureHost when attach. v1

Review of attachment 8587377 [details] [diff] [review]:
-----------------------------------------------------------------

The logic looks good to me, but I would prefer not to add a new virtual method just for this, see my comment below. Good catch, though.

::: gfx/layers/composite/CompositableHost.h
@@ +156,5 @@
>      SetLayer(aLayer);
>      mAttached = true;
>      mKeepAttached = aFlags & KEEP_ATTACHED;
> +
> +    UseCurrentTextureHost();

Please replace this by:

if (mFrontBuffer) {
  UseTextureHost(mFrontBuffer)
}

I'd rather not add a new method.

Comment 10

[Nicolas Silva [:nical]]

(In reply to Jerry Shih[:jerry] (UTC+8) from comment #6)
> Sotaro and Nical, what do you think about this patch? Should we handle other
> type of XXXComposite? I only can image that async video causes this problem.

I think that we don't need to handle this for the other Compositable classes. ImageHost gets the most complex scenarios when it comes to attaching/detaching to layers, due to async-video. Since canvas workers will go through ImageHost as well, and painted layers never use the ImageBridge we should be good with only ImageHost handling this case.

Comment 11

[Jerry Shih[:jerry] (UTC+8) (inactive)]

Attachment: init TextureSource for current TextureHost when attach. v2

Call use UseTextureHost() if we already have textureHost.

Comment 12

[Jerry Shih[:jerry] (UTC+8) (inactive)]

Please land attachment 8587490 [details] [diff] [review] to m-c.

try link:
https://treeherder.mozilla.org/#/jobs?repo=try&revision=8848864d01fb

this patch is similar to the previous one:
https://treeherder.mozilla.org/#/jobs?repo=try&revision=b602a02d04a0

Comment 13

[Jerry Shih[:jerry] (UTC+8) (inactive)]

I saw some tests failed. Checking for the new patch....

Comment 14

[Jerry Shih[:jerry] (UTC+8) (inactive)]

Please land attachment 8587490 [details] [diff] [review] to m-c. Try result seems good.
try link:
https://treeherder.mozilla.org/#/jobs?repo=try&revision=8848864d01fb

Comment 15

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/integration/b2g-inbound/rev/e4471cda8083

Comment 16

[Jerry Shih[:jerry] (UTC+8) (inactive)]

Comment on attachment 8587490 [details] [diff] [review]
init TextureSource for current TextureHost when attach. v2

[Approval Request Comment]
Bug caused by (feature/regressing bug #): none
User impact if declined:
We have timing issue at nexus5-l. The first frame of image will not show in comment 0 use case.
Testing completed:
Pass try on m-c. Test on 2.2 manually. I can see the first frame on both version.
Risk to taking this patch (and alternatives if risky):
Low. This patch just calls the texture source init function when it's attached to layer.
String or UUID changes made by this patch: none

Comment 17

[Phil Ringnalda (:philor)]

https://hg.mozilla.org/mozilla-central/rev/e4471cda8083

Comment 18

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/releases/mozilla-b2g37_v2_2/rev/9931ab333e5a

Comment 19

[Alissa（leave from Mozilla）]

This issue had been successfully verified on nexus5 2.2&3.0, there is no device of status-firefox40.
Reproducing rate: 0/5
Device: Nexus 5 2.2[Verified]
Build ID               20150407162504
Gaia Revision          ea735c21bfb0d78333213ff0376fce1eac89ead6
Gaia Date              2015-04-07 20:58:15
Gecko Revision         https://hg.mozilla.org/releases/mozilla-b2g37_v2_2/rev/3f86ddb7f719
Gecko Version          37.0
Device Name            hammerhead
Firmware(Release)      5.0
Firmware(Incremental)  eng.cltbld.20150407.195148
Firmware Date          Tue Apr  7 19:52:02 EDT 2015
Bootloader             HHZ12d

Device: Nexus5 3.0[Verified]
Build ID               20150407160201
Gaia Revision          84cbd4391fb7175d5380fa72c04d68873ce77e6d
Gaia Date              2015-04-07 17:33:14
Gecko Revision         https://hg.mozilla.org/mozilla-central/rev/078128c2600a
Gecko Version          40.0a1
Device Name            hammerhead
Firmware(Release)      5.0
Firmware(Incremental)  eng.cltbld.20150407.192944
Firmware Date          Tue Apr  7 19:30:00 EDT 2015
Bootloader             HHZ12d

Comment 20

[Hermes Cheng (inactive after July 27, 2015)]

Thanks, Alissa.