1082222 - Firefox 33 black screen on all windows

Status: RESOLVED INCOMPLETE

(Core :: Graphics: Layers, defect)

Description

[Justin Wood (:Callek)]

So I have a win7 "Thinkpad Carbon X1" with the intel graphics driver.

OMTC is on.

Firefox 34, as of ~ the last week has been causing black-screen (black chrome and content) while all UX elements are still responsive (cursor changes, I can type in chatzilla etc)

My memory use (per windows task manager) is < 1GB, despite this being a 64bit win7 install.

I was asked to test build1 of 33.0 final and while that was also broken it was broken in different (worse) ways, so I can't definitively say if the backouts between it and build 2 solve this.

Today was the first day that I had 5 near-continuous cases of failure, at least 4 of them are on todays aurora. (note that aurora was uplifted to beta today, so this is code that corresponds to what will be beta1, hence blocking and tracking requests)

Comment 1

[Justin Wood (:Callek)]

ugh-oh just reproduced on firefox 33 build2:

Application Basics
------------------

Name: Firefox
Version: 33.0
User Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:33.0) Gecko/20100101 Firefox/33.0
Multiprocess Windows: 0/1

Graphics
--------

Adapter Description: Intel(R) HD Graphics 4000
Adapter Drivers: igdumdim64 igd10iumd64 igd10iumd64 igdumdim32 igd10iumd32 igd10iumd32
Adapter RAM: Unknown
Device ID: 0x0166
Direct2D Enabled: true
DirectWrite Enabled: true (6.2.9200.16571)
Driver Date: 11-15-2013
Driver Version: 9.18.10.3359
GPU #2 Active: false
GPU Accelerated Windows: 1/1 Direct3D 11 (OMTC)
Vendor ID: 0x8086
WebGL Renderer: Google Inc. -- ANGLE (Intel(R) HD Graphics 4000 Direct3D9Ex vs_3_0 ps_3_0)
windowLayerManagerRemote: true
AzureCanvasBackend: direct2d
AzureContentBackend: direct2d
AzureFallbackCanvasBackend: cairo
AzureSkiaAccelerated: 0

Comment 2

[Henrik Skupin [:whimboo][⌚️UTC+2]]

[Tracking Requested - why for this release]:
If it happens with 33 and 35 it's very likely that also 34 should be affected.

Comment 3

[Justin Wood (:Callek)]

brief update/status/information:

I never tried on 35, that was a mistaken flag set.

After my multiple issues on 34 today, I tried to downgrade to 33, which I got 32 by accident (stub installer)

Then I tried 33 build 1 per kairo's suggestion, and got severe corruption (e.g. windows taskbar was failing too, and windows desktop background was all-black)

then switch to 33 build 2, and was able to reproduce in < a minute. Chose to restart after I seen a new issue on 33 build 2 (flash of black, then all UI drawing seemed halted, as in no UX feedback, which was different than my experience before)

After restart, couldn't reproduce within 2 min on 33 build 2, so switched back to aurora for now, lets see what happens.

Comment 4

[Liz Henry (:lizzard) (relman/hg->git project)]

I don't think we should panic on this for 33. Let's wait and see what happens tomorrow and this week.

Comment 5

[Robert Kaiser]

(In reply to Justin Wood (:Callek) from comment #1)
> Adapter Description: Intel(R) HD Graphics 4000
> Driver Version: 9.18.10.3359

That at least does not show up as anything problematic we would blocklist from anything, from what I can see.

> GPU #2 Active: false

I somewhat wonder if the GPU #2 has anything to do with the issues.

Comment 6

[Sylvestre Ledru [:Sylvestre]]

Tracking in case this spikes with the release.

Comment 7

[Justin Wood (:Callek)]

I'm marking this WFM for now, no signs of any corruption since my system restart. Including on the aurora where I hit it multiple times in one day.

Comment 8

[Sylvestre Ledru [:Sylvestre]]

It seems to impact other people
http://www.nextinpact.com/news/90409-firefox-33-openh264-pour-videoconferences-et-chromecast-sur-version-mobile.htm#/comment/5194275 (French)

Comment 9

[Sylvestre Ledru [:Sylvestre]]

Duplicate of bug 1037828?

Comment 10

[Tracy Walker [:tracy]]

(In reply to Sylvestre Ledru [:sylvestre] from comment #9)
> Duplicate of bug 1037828?

I think it is.  Also, I would hazard to guess that, when callek restarted his system it applied the update to his video driver. Thus his WFM.

Comment 12

[GMA]

https://downloadcenter.intel.com/Detail_Desc.aspx?DwnldID=24329

Latest Intel drivers for Ivy Bridge HD4000/HD2500 is 10.18.10.3945.

Comment 13

[Justin Wood (:Callek)]

(In reply to [:tracy] Tracy Walker - QA Mentor from comment #10)
> (In reply to Sylvestre Ledru [:sylvestre] from comment #9)
> > Duplicate of bug 1037828?
> 
> I think it is.  Also, I would hazard to guess that, when callek restarted
> his system it applied the update to his video driver. Thus his WFM.

To be clear, my driver is still

Driver Date: 11-15-2013
Driver Version: 9.18.10.3359

And I had no windows updates pending when I restarted. So I am quite puzzled as well for why I hit it and am no longer hitting it. But it is certainly a dupe of Bug 1083071

Comment 14

[Jeff Muizelaar [:jrmuizel]]

Since this doesn't happen at start up it looks like it's not a dup of 1083071

Comment 15

[u279076]

Justin, have you seen this recently?

Comment 16

[Justin Wood (:Callek)]

I'm not primarily using windows anymore, but my Aurora Firefox on win8 (that fiance uses a lot) hasn't seen it, no.

Comment 17

[u279076]

(In reply to Justin Wood (:Callek) from comment #16)
> I'm not primarily using windows anymore, but my Aurora Firefox on win8 (that
> fiance uses a lot) hasn't seen it, no.

Thanks Justin. I am closing this as INCOMPLETE given this information. Please reopen the bug report if it happens again.