Closed Bug 975146 Opened 12 years ago Closed 12 years ago

crash in mozilla::layers::CompositorOGL::CopyToTarget(mozilla::gfx::DrawTarget*, mozilla::gfx::Matrix const&)

Categories

(Core :: Graphics: Layers, defect)

ARM
Gonk (Firefox OS)
defect
Not set
critical

Tracking

()

RESOLVED FIXED
mozilla30
blocking-b2g 1.4+
Tracking Status
firefox28 --- wontfix
firefox29 --- wontfix
firefox30 --- fixed
b2g-v1.4 --- fixed

People

(Reporter: bzumwalt, Assigned: sotaro)

Details

(Keywords: crash, regression, reproducible, Whiteboard: dogfood1.4)

Crash Data

Attachments

(1 file, 1 obsolete file)

This bug was filed from the Socorro interface and is report bp-396629c4-a300-447b-beac-39fb72140220. ============================================================= [@ mozilla::layers::CompositorOGL::CopyToTarget(mozilla::gfx::DrawTarget*, mozilla::gfx::Matrix const&)] Description: Crash occurred while pressing Home Button and Power Button at same time to take screenshot when Firefox logo appeared after restarting phone. Phone vibrated as though screenshot were taken. This was followed by crackling noises from phone speaker. During this time the phone screen remained on the circular Firefox logo, but shortly after returned to animated Firefox mascot screen before proceeding to Firefox logo then booting normally. No screenshot is present in gallery app. This process was repeated successfully several times Repro Steps: 1) Updated Buri to BuildID: 20140220040203 2) Long press power button and select restart 3) When second Firefox OS logo screen appears press power button and homescreen button at same time as if to take screenshot Actual: Taking screenshot on Firefox logo screen while booting results in crash. Expected: User is unable to attempt screenshot while phone is booting. Environmental Variables: Device: Buri v1.4 Master Mozilla RIL BuildID: 20140220040203 Gaia: 6e71ab4da1b08586ea0c758edb7aa199ee34cd2f Gecko: 660b62608951 Version: 30.0a1 Firmware Version: V1.2-device.cfg Notes: Repro frequency: 3/3, 100%
blocking-b2g: --- → 1.4?
Component: General → Graphics: Layers
Product: Firefox OS → Core
Regression Window: Last Working Environmental Variables: Device: Buri BuildID: 20140213165315 Gaia: 3c2e748e4d57551527366eba00428a0f5ccc4012 Gecko: 0f18070eb5d6 Version: 30.0a1 Base Image: V1.2-device.cfg First Broken Environmental Variables: Device: Buri BuildID: 20140213170315 Gaia: ade88673d00414c3177f7444543b2fa01324708e Gecko: 5f1d4098333f Version: 30.0a1 Base Image: V1.2-device.cfg Last Working Gaia/First Broken Gecko: Issue DOES reproduce Gaia: 3c2e748e4d57551527366eba00428a0f5ccc4012 Gecko: 5f1d4098333f First Broken Gaia/Last Working Gecko: Issue does NOT reproduce Gaia: ade88673d00414c3177f7444543b2fa01324708e Gecko: 0f18070eb5d6
QA Contact: jzimbrick
Assignee: nobody → dglastonbury
I debugged a little bit. The following function returned null pointer. Both rect.with and rect.height are -1. It comes from mWidgetSize.width, mWidgetSize.height. RefPtr<DataSourceSurface> source = Factory::CreateDataSourceSurface(rect.Size(), gfx::SurfaceFormat::B8G8R8A8);
(In reply to Sotaro Ikeda [:sotaro] from comment #3) > I debugged a little bit. The following function returned null pointer. Both > rect.with and rect.height are -1. It comes from mWidgetSize.width, > mWidgetSize.height. > > RefPtr<DataSourceSurface> source = > Factory::CreateDataSourceSurface(rect.Size(), > gfx::SurfaceFormat::B8G8R8A8); The above is about CompositorOGL::CopyToTarget().
From the crash log in Comment 0, the crash happened withing HwComposer rendering. If falls into incorrect rendering path. It should be rendered by OpenGL. Yeah, from it, it becomes clear that it it triggered by bug 946720. But it is not caused by bug 946720's fix. bug 946720 changed the synchronous image layer to use gralloc buffers. It promote to use hw composer more than before.
Dan, I am going to take a bug.
Assignee: dglastonbury → sotaro.ikeda.g
Attachment #8382476 - Flags: review?(bjacob)
Comment on attachment 8382476 [details] [diff] [review] patch - Do not use hw composer for screenshot Review of attachment 8382476 [details] [diff] [review]: ----------------------------------------------------------------- I feel that I am not quite competent to decide whether adding a DrawTarget member to LayerManagerComposite is the right thing to do. Forwarding the review request.
Attachment #8382476 - Flags: review?(bjacob) → review?(bas)
blocking-b2g: 1.4? → 1.4+
Bas, can you review the patch soon? This bug is b2g 1.4+ bug. Need to be fixed soon.
Flags: needinfo?(bas)
Comment on attachment 8382476 [details] [diff] [review] patch - Do not use hw composer for screenshot Review of attachment 8382476 [details] [diff] [review]: ----------------------------------------------------------------- It seems like a simple enough fix for now, it isn't too pretty but I don't think there's value in trying to come up with something more clever.
Attachment #8382476 - Flags: review?(bas) → review+
Flags: needinfo?(bas)
unbitrot. Carry "bas: review+".
Attachment #8382476 - Attachment is obsolete: true
Attachment #8385373 - Flags: review+
Status: NEW → RESOLVED
Closed: 12 years ago
Resolution: --- → FIXED
Target Milestone: --- → mozilla30
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: