Created attachment 8511440 [details] Yahoo Black screen bug.docx User Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:33.0) Gecko/20100101 Firefox/33.0 Build ID: 20141011015303 Steps to reproduce: Open yahoo at https://www.yahoo.com/ and start scrolling down the screen, normal is to scan news items down the center of the page. Firefox 33.0 Windows 7, service pack 1, 64 bit, Dell E4310 laptop, 8 gig memory Actual results: Getting a black screen in the center only. Expected results: shown normal screen and news items
I tested in normal mode about 7:00 pm and get heavy black screen. In safe mode at 7:00 pm, no black screen but small glitches as I scroll down. The glitches I am seeing may be an artifact from web access. I'm on 10 meg DSL, but that does not necessarily mean fast load times. I retested in normal mode at 3:00 a.m. and there are no glitches and no issues. This suggests that something about high usage on the net is interfering with normal page rendering and that it is exacerbated by one or more extensions/addons. I did not see this until Firefox 33.0 and am now running Firefox 33.0.1. Would it be worth the effort of reverting to Firefox 32.0 to see if it eliminates the problem? There are a few more addons that are not shown, but they are all disabled. I put screen shots of my addons and extensions here: http://www.selectedplants.com/miscan/image001.png http://www.selectedplants.com/miscan/image003.png http://www.selectedplants.com/miscan/image005.png
Update that I backed down to firefox 32.0.3 and the problem totally disappears. I speculate that version 33 implements a new page rendering algorithm and breaks one of the plugins like silverlight or shockwave. I will stay on version 32 until you get this resolved. I don't like watching jittery videos and seeing black screen on yahoo.
Would you please check our 33.0.2 build http://ftp.mozilla.org/pub/mozilla.org/firefox/candidates/33.0.2-candidates/build2/win32/
Upped version to 33.0.2 and definitely see the problem. Un-checked hardware acceleration and glitches are gone. Turned hardware acceleration back on and turned off all add-ons/extensions and see the problem. At this point, it looks like some kind of interaction when hardware acceleration is turned on. I'm leaving 33.0.2 loaded and hardware acceleration un-checked. This appears to work fine for now. It is not a fix though. I found out that watching just about any video on Vimeo is an easy way to demonstrate the issue.
Nical, Bas, does it ring a bell to you? Darrel, could you attach the result of about:support? Thanks
(In reply to Sylvestre Ledru [:sylvestre] from comment #6) > Nical, Bas, does it ring a bell to you? > > Darrel, could you attach the result of about:support? Thanks Nope, I haven't seen blackness like this as localized in any other bugs yet, I don't think. About:support will be critical here.
So give a sigh of relief. It was an outdated driver. http://downloads.dell.com/video/VIDEO_DRVR_WIN_R296701.EXE I loaded the driver, turned on hardware acceleration and the glitches are now gone.
Darrel, you didn't get the value in about:support before updating? :/
(In reply to Darrel Jones from comment #8) > So give a sigh of relief. It was an outdated driver. > > http://downloads.dell.com/video/VIDEO_DRVR_WIN_R296701.EXE > > I loaded the driver, turned on hardware acceleration and the glitches are > now gone. Probably just another dupe of bug 1083071?
Here is the relevant section post-update: Adapter Description Intel(R) HD Graphics Adapter Drivers igdumd64 igd10umd64 igdumdx32 igd10umd32 Adapter RAM Unknown Device ID 0x0046 Direct2D Enabled true DirectWrite Enabled true (6.2.9200.16571) Driver Date 1-30-2013 Driver Version 184.108.40.20693 GPU #2 Active false GPU Accelerated Windows 2/2 Direct3D 11 (OMTC) Vendor ID 0x8086 WebGL Renderer Google Inc. -- ANGLE (Intel(R) HD Graphics Direct3D9Ex vs_3_0 ps_3_0) windowLayerManagerRemote true AzureCanvasBackend direct2d AzureContentBackend direct2d AzureFallbackCanvasBackend cairo AzureSkiaAccelerated 0
Milan/Jeff/Bas - Do we need to block another driver version to resolve this bug?
I believe we were already blocking the driver version that was causing problems, and the issue was showing up in the software mode. So, whatever would fix this, is probably not a blocklisting solution, and :nical has been tracking SW backend issues...
AFAIK, there is no current action to take with this bug. At the very least, this is wontfix for 34.
Haven't been hearing about this from user feedback during beta 35 and since comment 15 suggests there isn't any action to take I'm removing tracking on this.
Closing this bug since we've made no progress in several months. Please reopen this bug report if you have any new leads.