KMS: notebook boots but stays black

Bug #425165 reported by Wolfgang Kufner
78
This bug affects 12 people
Affects Status Importance Assigned to Milestone
linux (Fedora)
Fix Released
Medium
linux (Ubuntu)
Fix Released
Medium
Unassigned
Nominated for Karmic by Wolfgang Kufner

Bug Description

[Status]
Bug is fixed in Lucid.
Bug is not fixed in Karmic.

[Workaround 1 (preferred)]
As workaround disable kernel mode setting (KMS).
To disable KMS add the option nomodeset to the kernel boot options in the grub config (for GRUB 2: edit /etc/default/grub and add nomodeset to GRUB_CMDLINE_LINUX, then run sudo update-grub; for GRUB 1: edit /boot/grub/menu.lst and add nomodeset to the end of the line beginning with # kopt=, then run sudo update-grub).

[Workaround 2 (alternative)]
Use an external monitor:
Connect to the VGA port before booting. It will show you a desktop on the external display after booting, however Gnome may place the menu on the dysfunctional internal display (LVDS1). To get your menu to the external display switch off the internal display. To do so press Alt+F2 to get a "Run Application" window, type "gnome-terminal" in there and click Run. A terminal window will appear. Write "xrandr --output LVDS1 --off" (without the hyphens) in the terminal and press enter.

[Report]
The default boot process on the very recent acer extensa 5635z does not display anything after grub. The backlight is on but everything stays black.
The system is karmic with xorg-edgers. The karmic desktop live 64bit daily of September 5th shows the same behaviour.
Luckily an external display connencted before boot displays everything just fine. The internal display is again always black. So I did an ubuntu-bug linux just after boot up with the external monitor connected. Screenshot1.png shows that the monitors are in mirror mode at 1024x768.
At a previous boot I played around with the display settings a bit more (via the GUI). I could set the external monitor to its native resolution. That displayed fine. I could also set the internal monitor to its 1366x768 native resolution, but it stayed black. A screenshot captured all the contents of the two screens just fine. See Screenshot2.png.

Booting with a current live Fedora 12 alpha 1 shows the same general behaviour. Booting with external monitor there works also.
update: the bug is now fixed in fedora 20091111

Checked again with fully updated karmic as of 2009-09-13 and its new kernel 2.6.31-10-generic. Still black after boot with kernel modesetting.

Adding nomodeset to the boot parameters does, of course, circumvent this kernelmodesetting bug.

Please let me know what further information is needed to fix this bug.

ProblemType: Bug
Architecture: amd64
ArecordDevices:
 **** List of CAPTURE Hardware Devices ****
 card 0: Intel [HDA Intel], device 0: CONEXANT Analog [CONEXANT Analog]
   Subdevices: 1/1
   Subdevice #0: subdevice #0
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: anita 2993 F.... pulseaudio
CRDA: Error: [Errno 2] No such file or directory
Card0.Amixer.info:
 Card hw:0 'Intel'/'HDA Intel at 0xf0700000 irq 22'
   Mixer name : 'Conexant CX20561 (Hermosa)'
   Components : 'HDA:14f15051,10250253,00100000'
   Controls : 14
   Simple ctrls : 7
Date: Sun Sep 6 13:22:52 2009
DistroRelease: Ubuntu 9.10
HibernationDevice: RESUME=UUID=52e9071d-964b-4eea-97f4-f57aa45d2ad9
MachineType: Acer Extensa 5635Z
Package: linux-image-2.6.31-9-generic 2.6.31-9.29
PccardctlIdent:

PccardctlStatus:

ProcCmdLine: BOOT_IMAGE=/boot/vmlinuz-2.6.31-9-generic root=UUID=071a1c23-d094-4719-b16f-5f7c90414d9d ro quiet splash
ProcEnviron:
 PATH=(custom, user)
 LANG=en_US.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.31-9.29-generic
RelatedPackageVersions:
 linux-backports-modules-2.6.31-9-generic N/A
 linux-firmware 1.16
SourcePackage: linux
Uname: Linux 2.6.31-9-generic x86_64
dmi.bios.date: 05/07/2009
dmi.bios.vendor: Phoenix
dmi.bios.version: V0.3213
dmi.board.name: BA50-MV
dmi.board.vendor: Acer
dmi.board.version: Not Applicable
dmi.chassis.type: 10
dmi.chassis.vendor: Acer
dmi.chassis.version: N/A
dmi.modalias: dmi:bvnPhoenix:bvrV0.3213:bd05/07/2009:svnAcer:pnExtensa5635Z:pvrNotApplicable:rvnAcer:rnBA50-MV:rvrNotApplicable:cvnAcer:ct10:cvrN/A:
dmi.product.name: Extensa 5635Z
dmi.product.version: Not Applicable
dmi.sys.vendor: Acer

00:02.0 VGA compatible controller [0300]: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller [8086:2a42] (rev 07)
 Subsystem: Acer Incorporated [ALI] Device [1025:013c]

Revision history for this message
Wolfgang Kufner (wolfgangkufner) wrote :
Revision history for this message
Wolfgang Kufner (wolfgangkufner) wrote :
tags: added: karmic
description: updated
Revision history for this message
Wolfgang Kufner (wolfgangkufner) wrote :
description: updated
Revision history for this message
Ladislav Urbanek (ladislav-urbanek) wrote :

I can confirm this issue. I have the same laptop and the same problem in Karmic.
I tested openSUSE 11.1 and had no problem with this. Ubuntu 9.04 no problem too.

Changed in linux (Ubuntu):
status: New → Confirmed
description: updated
Revision history for this message
In , Wolfgang (wolfgang-redhat-bugs) wrote :

Booting with Test Day:2009-09-11 Intel live cd 64bit the inbuilt display always stays black after grub. Only the backlight goes on. After a while I can hear the melody played when it logs in automatically. But the display stays black. I shut it down blindly with power button + return. It shut down regularly.
Booting with an external display attached gives me a picture on the external only. I switched the internal off with xrandr. The backlight went off and I got the menubar and taskbar onto the external monitor.
I have repeated this (external monitor attached) again with drm.debug=1 and produced the attached logs.
A screenshot (Screenshot.png) taken shows the whole screen, including the part I never see.

also tried switching internal on with xrandr --output LVDS1 --auto. the backlight comes back on but no picture.

with nomodeset it works fine.
dpms: screensaver does not kick in but power management does and it recovers correctly.
enabling desktop effects fails with something like hardware not supported.
Xv works.
fastuserswitching works.
VT terminal switching works.
suspend resume works with external.

I have also tried this with current ubuntu karmic and filed a bug there:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/425165

00:02.0 VGA compatible controller [0300]: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller [8086:2a42] (rev 07) (prog-if 00 [VGA controller])
 Subsystem: Acer Incorporated [ALI] Device [1025:0253]

/etc/X11/xorg.conf does not exist.

Revision history for this message
In , Wolfgang (wolfgang-redhat-bugs) wrote :

Created attachment 360949
I can screenshot what I can not see

Revision history for this message
In , Wolfgang (wolfgang-redhat-bugs) wrote :

Created attachment 360950
/var/log/messages

Revision history for this message
In , Wolfgang (wolfgang-redhat-bugs) wrote :

Created attachment 360951
dmesg

Revision history for this message
In , Wolfgang (wolfgang-redhat-bugs) wrote :

Created attachment 360952
Xorg.0.log

Revision history for this message
In , Wolfgang (wolfgang-redhat-bugs) wrote :

Created attachment 360953
smoltprofile.txt

Revision history for this message
In , Wolfgang (wolfgang-redhat-bugs) wrote :

Created attachment 360954
xrandr.txt

Changed in linux (Ubuntu):
importance: Undecided → Medium
status: Confirmed → Triaged
Revision history for this message
In , Adam (adam-redhat-bugs) wrote :

Please do:

% sudo dd if=/dev/mem of=/tmp/rom bs=64k skip=12 count=1

and attach that, and /proc/acpi/dsdt, to this bug.

Revision history for this message
In , Wolfgang (wolfgang-redhat-bugs) wrote :

Created attachment 362172
sudo dd if=/dev/mem of=/tmp/rom bs=64k skip=12 count=1

Revision history for this message
In , Wolfgang (wolfgang-redhat-bugs) wrote :

Created attachment 362173
/proc/acpi/dsdt

Changed in linux (Fedora):
status: Unknown → Confirmed
Revision history for this message
Wolfgang Kufner (wolfgangkufner) wrote :

I just nominated this bug for karmic because I really hope that one of precious few notebooks currently widely available in german speaking parts of europe without having to pay windows tax will not greet the user with an unyielding black screen upon booting with the shiny new ubuntu karmic.

Revision history for this message
In , Matěj (matj-redhat-bugs) wrote :

Since this bugzilla report was filed, there have been several major updates in various components of the Xorg system, which may have resolved this issue. Users who have experienced this problem are encouraged to upgrade their system to the latest version of their packages (at least F12Beta, but even better if the very latest versions).

Please, if you experience this problem on the up-to-date system, let us now in the comment for this bug, or whether the upgraded system works for you.

If you won't be able to reply in one month, I will have to close this bug as INSUFFICIENT_DATA. Thank you.

[This is a bulk message for all open Fedora Rawhide Xorg-related bugs. I'm adding myself to the CC list for each bug, so I'll see any comments you make after this and do my best to make sure every issue gets proper attention.]

Revision history for this message
johnny (jrb81) wrote :

I have the same problem. I use the kernel parameter i915.modset=0 and thereby I can work. But the graphic performance is significantly worse than before the upgrade from Ubuntu 9.04 to 9.10.

Revision history for this message
johnny (jrb81) wrote :

P.S. I have an ASUS X5DIJ (K series) with an Intel GMA 4500M graphic chipset.

description: updated
Revision history for this message
In , Wolfgang (wolfgang-redhat-bugs) wrote :

verified fixed in nightly-composes desktop-x86_64-20091111.16.iso

thanks :-)

Revision history for this message
Wolfgang Kufner (wolfgangkufner) wrote :

the bug is now fixed in fedora:
verified fixed in fedora nightly-composes desktop-x86_64-20091111.16.iso

description: updated
Revision history for this message
In , Chris (chris-redhat-bugs) wrote :

Wolfgang, thank you very much for your efforts in finding and testing this bug! Setting it to CLOSED RAWHIDE due to your verification.

--
Fedora Bugzappers volunteer triage team
https://fedoraproject.org/wiki/BugZappers

Changed in linux (Fedora):
status: Confirmed → Fix Released
Revision history for this message
gunthers (hackmetz) wrote :

Same problem here - black screen with Ubuntu 9.10 Live and Archlinux.
Could we find out how the guys at Fedora fixed that bug? Wolfgang, maybe you could contact Adam Jackson who asked for additional info. He also maintains intel video drivers for Fedora.

Revision history for this message
Wolfgang Kufner (wolfgangkufner) wrote :

Fixed in lucid. I just checked with ubuntu lucid-desktop-amd64 20091116 (booting the iso from hd via grub2 loopback).

Changed in linux (Ubuntu):
status: Triaged → Fix Committed
Revision history for this message
gunthers (hackmetz) wrote :

fixed in 2.6.32-rc7
(also comes with vast improvement in X-snappiness)

Revision history for this message
Bo Zhang (edstudios) wrote :

I have the same laptop 5635Z and experiencing the same bug. I've found someone on the forum pointed out that this is due to incorrect detect of TV output hence the main display is turned off. Look at this:

http://<email address hidden>/msg44165.html

With his patch it would fix the problem, however TV function will be completely disabled.

Revision history for this message
Bo Zhang (edstudios) wrote :

For your verification, connect an external monitor via Laptop's VGA connection and turn on, it should display the booting process, then GDM without problem. AFAIK this is a show stopper for Karmic, most of the laptops with Intel GMA suffered this problem. It definitely have to be fixed in Karmic otherwise it is unusable. Disabling KMS using i915.modeset=0 results very low performance hence is not perferred.

Revision history for this message
johnny (jrb81) wrote :

Hello everybody,

someone knows about any Karmic progress? Another month passed and we've still worse than Windows performance since we're forced to boot with the i915.modest option ...
Anyway thanks for your hard work in general and advance!

Greets, jb

Revision history for this message
Nicolas Dumoulin (nicolas-dumoulin) wrote :

As jayarbie, I'm interested for having information about backporting the issue in karmic.
I've migrated the laptop of friends from Win to Kubuntu, and I don't think it's a good idea to put them on a dev version.
For information, the bug occurs on a dell inspiron 15z (I've searched for a long for this bug report ;-) ).

Thank you

Revision history for this message
David Perfors (dnperfors) wrote :

i am also interested in having the solution backported. I am not willing to run a development version on my laptop (not in this early stage of development) but still want to work with ubuntu if possible...

Thanks.

Revision history for this message
David Perfors (dnperfors) wrote :

After installing 9.10 on my acer extensa 5635z I was not able to get things working with the i915.modeset=0 option (unlike using the live cd which worked for me)
I decided that I would give 10.04 alpha1 a shot and there it was fixed for me. 8.04.3 didn't work either... (didn't try 9.04)

Revision history for this message
gunthers (hackmetz) wrote :

Not really a solution, but if you run Ubuntu 9.10 you might want to get 2.6.32 as workaround until this gets backported (if it will)
https://wiki.ubuntu.com/KernelTeam/MainlineBuilds?action=show&redirect=KernelMainlineBuilds

Revision history for this message
David Perfors (dnperfors) wrote :

running 10.04 alpha1 and did an update today... My screen was black again :( (kernel 2.6.32-8.12, kernel 2.6.32-7.10 does work) Strange because my last update was only a view days ago... and I think I was running that kernel (8.12) for some days now...

Revision history for this message
Wolfgang Kufner (wolfgangkufner) wrote :

I just checked with ubuntu lucid-desktop-amd64 20091219 (booting the iso from hd via grub2 loopback) on the acer extensa 5635z. The display with kernel mode setting was fine.
I plan to recheck with the next iso.

Revision history for this message
Wolfgang Kufner (wolfgangkufner) wrote :

I just checked with ubuntu lucid-desktop-amd64 20091221 (booting the iso from hd via grub2 loopback) on the acer extensa 5635z. The display with kernel mode setting was fine.
I also tried to switch off kernel mode setting with the "nomodeset" parameter. That also works.

@David Perfors
I notice that the newest kernel for lucid is 2.6.32-9.13. Does that kernel work for you?

Revision history for this message
Henning Moll (drscott) wrote :

Are there any differences between the two solutions 'i915.modeset=0' and 'nomodeset'? Which one to prefer?

(also having an Extensa 5635z)

Revision history for this message
Wolfgang Kufner (wolfgangkufner) wrote :

Personally I have always used "nomodeset" with my Extensa 5635z and it works fine 100%, but I guess they both do exactly the same thing.

https://wiki.ubuntu.com/KarmicKoala/ReleaseNotes speaks of both "nomodeset" and "i915.modeset=0".

And graphics on my 5635z with kernel mode setting switched off are NOT slower than on my otherwise very similar 5630z with KMS.

Revision history for this message
alpha-centauri (ubuntu-alexander-becker-privat) wrote :

Have same issue.

posted bug here

https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/502153

Would be nice if the fix could be included in 10.04

description: updated
description: updated
Revision history for this message
Jeremy Foshee (jeremyfoshee) wrote :

alpha-centauri,
      Have you encountered this issue in the current Alpha of Lucid? If so, please run apport-collect -p linux 425165 so that I can get some relevant current information for the team to work from.

Thanks in advance!

-JFo

Changed in linux (Ubuntu):
status: Fix Committed → Incomplete
Revision history for this message
Andrey Voronov (voronov84) wrote : apport information

AlsaVersion: Advanced Linux Sound Architecture Driver Version 1.0.21.
Architecture: amd64
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: andrey 2184 F.... pulseaudio
 /dev/snd/pcmC0D0p: andrey 2184 F...m pulseaudio
CRDA: Error: [Errno 2] Нет такого файла или каталога
Card0.Amixer.info:
 Card hw:0 'Intel'/'HDA Intel at 0xfe8f8000 irq 22'
   Mixer name : 'Realtek ALC883'
   Components : 'HDA:10ec0883,1043829f,00100002'
   Controls : 35
   Simple ctrls : 20
Card1.Amixer.info:
 Card hw:1 'SAA7134'/'saa7134[0] at 0xfebffc00 irq 18'
   Mixer name : 'SAA7134 Mixer'
   Components : ''
   Controls : 6
   Simple ctrls : 3
Card2.Amixer.info:
 Card hw:2 'HDMI'/'HDA ATI HDMI at 0xfe9fc000 irq 17'
   Mixer name : 'ATI R6xx HDMI'
   Components : 'HDA:1002aa01,00aa0100,00100100'
   Controls : 4
   Simple ctrls : 1
Card2.Amixer.values:
 Simple mixer control 'IEC958',0
   Capabilities: pswitch pswitch-joined penum
   Playback channels: Mono
   Mono: Playback [off]
DistroRelease: Ubuntu 10.04
Frequency: This has only happened once.
HibernationDevice: RESUME=UUID=065620fb-b452-46bd-aaac-56c3ab902835
InstallationMedia: Ubuntu 10.04 "Lucid Lynx" - Alpha amd64 (20100206)
IwConfig: Error: [Errno 2] Нет такого файла или каталога
MachineType: System manufacturer P5KC
Package: linux (not installed)
ProcCmdLine: BOOT_IMAGE=/vmlinuz-2.6.32-12-generic root=UUID=7b9f2bf0-926a-4056-a0fd-a6fc4122a2b7 ro quiet splash
ProcEnviron:
 PATH=(custom, user)
 LANG=ru_UA.utf8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.32-12.17-generic
Regression: Yes
RelatedPackageVersions: linux-firmware 1.29
Reproducible: No
RfKill: Error: [Errno 2] Нет такого файла или каталога
Tags: lucid regression-potential
TestedUpstream: Yes
Uname: Linux 2.6.32-12-generic x86_64
UserGroups: adm admin cdrom dialout lpadmin plugdev sambashare
dmi.bios.date: 10/30/2007
dmi.bios.vendor: American Megatrends Inc.
dmi.bios.version: 0804
dmi.board.asset.tag: To Be Filled By O.E.M.
dmi.board.name: P5KC
dmi.board.vendor: ASUSTeK Computer INC.
dmi.board.version: Rev 1.xx
dmi.chassis.asset.tag: Asset-1234567890
dmi.chassis.type: 3
dmi.chassis.vendor: Chassis Manufacture
dmi.chassis.version: Chassis Version
dmi.modalias: dmi:bvnAmericanMegatrendsInc.:bvr0804:bd10/30/2007:svnSystemmanufacturer:pnP5KC:pvrSystemVersion:rvnASUSTeKComputerINC.:rnP5KC:rvrRev1.xx:cvnChassisManufacture:ct3:cvrChassisVersion:
dmi.product.name: P5KC
dmi.product.version: System Version
dmi.sys.vendor: System manufacturer

tags: added: apport-collected
Revision history for this message
Andrey Voronov (voronov84) wrote : AlsaDevices.txt

apport information

Revision history for this message
Andrey Voronov (voronov84) wrote : AplayDevices.txt

apport information

Revision history for this message
Andrey Voronov (voronov84) wrote : ArecordDevices.txt

apport information

Revision history for this message
Andrey Voronov (voronov84) wrote : BootDmesg.txt

apport information

Revision history for this message
Andrey Voronov (voronov84) wrote : Card0.Amixer.values.txt

apport information

Revision history for this message
Andrey Voronov (voronov84) wrote : Card0.Codecs.codec.0.txt

apport information

Revision history for this message
Andrey Voronov (voronov84) wrote : Card1.Amixer.values.txt

apport information

Revision history for this message
Andrey Voronov (voronov84) wrote : Card2.Codecs.codec.0.txt

apport information

Revision history for this message
Andrey Voronov (voronov84) wrote : CurrentDmesg.txt

apport information

Revision history for this message
Andrey Voronov (voronov84) wrote : Lspci.txt

apport information

Revision history for this message
Andrey Voronov (voronov84) wrote : Lsusb.txt

apport information

Revision history for this message
Andrey Voronov (voronov84) wrote : PciMultimedia.txt

apport information

Revision history for this message
Andrey Voronov (voronov84) wrote : ProcCpuinfo.txt

apport information

Revision history for this message
Andrey Voronov (voronov84) wrote : ProcInterrupts.txt

apport information

Revision history for this message
Andrey Voronov (voronov84) wrote : ProcModules.txt

apport information

Revision history for this message
Andrey Voronov (voronov84) wrote : UdevDb.txt

apport information

Revision history for this message
Andrey Voronov (voronov84) wrote : UdevLog.txt

apport information

Revision history for this message
Andrey Voronov (voronov84) wrote : WifiSyslog.txt

apport information

Revision history for this message
Andrey Voronov (voronov84) wrote :

Oops i do not have this issue on laptop and desktop, sorry.

Revision history for this message
Jeremy Foshee (jeremyfoshee) wrote :

Changing back to Fix Committed as the code has been pulled for lucid.

-JFo

Changed in linux (Ubuntu):
status: Incomplete → Fix Released
Revision history for this message
Jeremy Foshee (jeremyfoshee) wrote :

rather Fix Released as it is available via alpha.

-JFo

Changed in linux (Fedora):
importance: Unknown → Medium
To post a comment you must log in.