[i946gz] Xserver black screen on 3/28/09 update of Jaunty

[Craig McDaniel]

Binary package hint: xserver-xorg-core

Ubuntu Jaunty (development branch)
Release: 9.04
Package: xserver-xorg-core 2:1.6.0
xserver-xorg-video-intel 2:2.6.3

New bug: Xserver was working fine on Jaunty until apt-get update this morning brought in a new xserver-xorg-core. I run apt-update fairly often and this was working fine over the last week.

Machine is Asus P22 with Intel Corporation 82946GZ/GL Integrated Graphics Controller (rev 02). Using DVI-D output connected to Gateway FPD218SW 1680x1050 resolution monitor. (No VGA connector on card; only DVI-D output).

Boots normally until X starts. No output to monitor after that point. X thinks it's running OK because you can hear the Ubuntu sound prompt when the logon screen should be showing. I can logon "blind" to the black screen after hearing the Ubuntu sound and then Alt-F1, logon to the terminal screen blind and then "sudo shutdown -r now" to restart the machine.

Additional note: X server was not actually working perfectly even before this upgrade. It had this same black screen symptom when trying to come back from a suspend.

Xorg.0.log is attached.

[lspci]
00:00.0 Host bridge [0600]: Intel Corporation 82946GZ/PL/GL Memory Controller Hub [8086:2970] (rev 02)
     Subsystem: ASUSTeK Computer Inc. Device [1043:81ea]
00:02.0 VGA compatible controller [0300]: Intel Corporation 82946GZ/GL Integrated Graphics Controller [8086:2972] (rev 02)
     Subsystem: ASUSTeK Computer Inc. Device [1043:81ea]

[Vojtěch Trefný (vojtech.trefny)]

Same problem here, but my graphics card is Nvidia 8600GT (same problem with binary drivers even with opensource, now running at vesa driver...).

[Bryce Harrington (bryce)]

Which version of xorg-server (exactly) did you have installed when the problem began? Did the issue completely go away when you downgraded to the previous version? What version was that?

Fwiw, all of the recent changes to xorg-server have been things that could not cause a black screen issue, so I'm suspicious that what you're seeing may be unrelated, especially since you indicate you've seen this failure before.

Anyway, for how to create a full bug report about a black screen issue please see http://wiki.ubuntu.com/X/Troubleshooting which includes a guide on this specific issue that you should follow.

[Craig McDaniel (craigmcd)]

Thanks for the suggestions Bryce. My main problem is solved for now, but some mysteries remain:

First, I disabled gdm to diagnose. I downgraded to the previous versions of xserver-xorg-core and xserver-common using "dpkg -i /var/cache/apt/archive/<package>.deb" but continued to get a black screen when starting X manually.

Mystery #1: Finally I tried enabling gdm again and it worked! I switched back to the latest xserver-xorg-core and xserver-common and it still worked! Other than adding the additional xserver-xorg-video-intel-dbg package, I should be back to the same configuration as when the issue first occurred.

Mystery #2: I still don't understand why I can't start the X server manually. It always works when using "/etc/init.d/gdm start". When I kill gdm and try to start X using "/usr/X11R6/bin/X :0" or "startx" nothing but a black screen appears. I even tried duplicating the way gdm runs X, but still had no luck:

/usr/X11R6/bin/X :0 -br -audit 0 -auth /var/lib/gdm/:0.Xauth -nolisten tcp vt7

gdm is doing something to make X work that I can't reproduce. If I can ever figure out how to start X manually, it may help me resolve the issue with resuming from a suspend. I have attached various log files. The main difference between the Xorg.0.log files is that the one that broken one doesn't show this section:

(II) intel(0): EDID vendor "GWY", prod id 2186
(II) intel(0): Using EDID range info for horizontal sync
(II) intel(0): Using EDID range info for vertical refresh
(II) intel(0): Printing DDC gathered Modelines:
(II) intel(0): Modeline "1680x1050"x0.0 146.37 1680 1784 1960 2240 1050 1056 1086 1089 +hsync +vsync (65.3 kHz)

The broken one also includes this error message:

(EE) intel(0): First SDVOC output reported failure to sync

but that only shows up after doing a VT switch.

[Bryce Harrington (bryce)]

[This is an automatic notification.]

A new major version of the -intel driver is now available in Karmic.

This version includes a major reworking of the acceleration
architecture, which resolves a huge number of issues. We do not know
whether it resolves the issue you reported.

Would you mind testing Karmic Alpha-2 and seeing if it is still a
problem? CD ISO images are available here:

  http://cdimages.ubuntu.com/releases/karmic/

If the issue can still be reproduced on karmic, please report here with
your findings, and attach a fresh Xorg.0.log from your test, and we will
be able to forward the bug upstream.

Otherwise, if the bug no longer exists in Karmic, let us know that as
well.

In the off chance you encounter different bugs while attempting to test
Karmic, please report those as new bug reports.

[wouter bolsterlee (wbolster)]

Looks like this is the same as Debian#496838. Any clue what exactly causes/fixes this?

[Bryce Harrington (bryce)]

Wouter, actually I'm not confident it's the same bug; there's a variety of different things that can cause the exact symptoms reported for this bug, and Debian #496838 seems to be pretty specifically about use of gdmflexiserver. Can you explain further on why you think this may be related to that bug?

I'm also setting this back to Incomplete since it still needs re-tested against Karmic.

[Geir Ove Myhr (gomyhr)]

This bug report has been inactive for quite some time and in the meantime Ubuntu has moved along. I hope the reported problem has been resolved in the current version of Ubuntu. If it has not and there are people willing to follow up with information collecting and testing, it is best to file a new bug report with the latest Ubuntu development version (10.04 still works, but soon 10.10 is better), using the ubuntu-bug command if possible. Feel free to mention the bug number of this bug.

We are sorry that we are not always have the capacity to look at all reported bugs in a timely manner. Occasionally we have to purge old bug reports where no progress has been made, whether it is due to reporter not being able to respond or lack of manpower on the Ubuntu side, in order to keep our list of "active" bugs current. Sorry for the inconvenience this may cause.

Here are a few tips that will make it easier for triagers and developers to process your bug:
* Read https://wiki.ubuntu.com/X/Reporting
* Browse https://wiki.ubuntu.com/X and see if there is some relevant information for you problem
* Respond quickly to questions from triagers
* Try changing things to see if the behaviour changes, e.g.
  * A new mainline kernel from http://kernel.ubuntu.com/~kernel-ppa/mainline/?C=M;O=D (more information at https://wiki.ubuntu.com/KernelTeam/MainlineBuilds)
  * The xorg-edgers PPA: https://launchpad.net/~xorg-edgers/+archive/ppa
  * Various options in `/etc/X11/xorg.conf` (see `man intel`)
* Browse other bug reports to see if you find a similar one and provide the information that is being asked for there.