[Solved] X won't start after latest update

Joekickass · 2013-03-20 08:52:28

Hi,

updated my Arch Linux system yesterday. After that X won't start anymore. Problem is most likely related to x-server-1.14.0-2, xf86-video-nouveau-1.0.6-2 and/or linux-3.8.3-2. Any help on how to debug X-server is appreciated!

Uname:

Linux toni-arch 3.8.3-2-ARCH #1 SMP PREEMPT Sun Mar 17 13:04:22 CET 2013 x86_64 GNU/Linux

lspci:

00:00.0 Host bridge [0600]: Intel Corporation 3rd Gen Core processor DRAM Controller [8086:0154] (rev 09)
00:01.0 PCI bridge [0604]: Intel Corporation Xeon E3-1200 v2/3rd Gen Core processor PCI Express Root Port [8086:0151] (rev 09)
00:01.1 PCI bridge [0604]: Intel Corporation Xeon E3-1200 v2/3rd Gen Core processor PCI Express Root Port [8086:0155] (rev 09)
00:01.2 PCI bridge [0604]: Intel Corporation Xeon E3-1200 v2/3rd Gen Core processor PCI Express Root Port [8086:0159] (rev 09)
00:02.0 VGA compatible controller [0300]: Intel Corporation 3rd Gen Core processor Graphics Controller [8086:0166] (rev 09)
00:14.0 USB controller [0c03]: Intel Corporation 7 Series/C210 Series Chipset Family USB xHCI Host Controller [8086:1e31] (rev 04)
00:16.0 Communication controller [0780]: Intel Corporation 7 Series/C210 Series Chipset Family MEI Controller #1 [8086:1e3a] (rev 04)
00:1a.0 USB controller [0c03]: Intel Corporation 7 Series/C210 Series Chipset Family USB Enhanced Host Controller #2 [8086:1e2d] (rev 04)
00:1b.0 Audio device [0403]: Intel Corporation 7 Series/C210 Series Chipset Family High Definition Audio Controller [8086:1e20] (rev 04)
00:1c.0 PCI bridge [0604]: Intel Corporation 7 Series/C210 Series Chipset Family PCI Express Root Port 1 [8086:1e10] (rev c4)
00:1c.1 PCI bridge [0604]: Intel Corporation 7 Series/C210 Series Chipset Family PCI Express Root Port 2 [8086:1e12] (rev c4)
00:1d.0 USB controller [0c03]: Intel Corporation 7 Series/C210 Series Chipset Family USB Enhanced Host Controller #1 [8086:1e26] (rev 04)
00:1f.0 ISA bridge [0601]: Intel Corporation HM77 Express Chipset LPC Controller [8086:1e57] (rev 04)
00:1f.2 SATA controller [0106]: Intel Corporation 7 Series Chipset Family 6-port SATA Controller [AHCI mode] [8086:1e03] (rev 04)
00:1f.3 SMBus [0c05]: Intel Corporation 7 Series/C210 Series Chipset Family SMBus Controller [8086:1e22] (rev 04)
01:00.0 VGA compatible controller [0300]: NVIDIA Corporation Device [10de:0fd5] (rev a1)
01:00.1 Audio device [0403]: NVIDIA Corporation GK107 HDMI Audio Controller [10de:0e1b] (rev a1)
03:00.0 Ethernet controller [0200]: Broadcom Corporation Device [14e4:16a3] (rev 10)
03:00.1 SD Host controller [0805]: Broadcom Corporation NetXtreme BCM57765 Memory Card Reader [14e4:16bc] (rev 10)
04:00.0 Network controller [0280]: Broadcom Corporation BCM4331 802.11a/b/g/n [14e4:4331] (rev 02)

pacman log:

[2013-03-19 08:28] Running 'pacman -Syu'
[2013-03-19 08:28] synchronizing package lists
[2013-03-19 08:28] starting full system upgrade
[2013-03-19 08:29] upgraded bash (4.2.042-3 -> 4.2.045-1)
[2013-03-19 08:29] upgraded acpid (2.0.18-1 -> 2.0.18-2)
[2013-03-19 08:29] upgraded brasero (3.6.1-1 -> 3.6.1-2)
[2013-03-19 08:29] upgraded polkit (0.109-1 -> 0.110-1)
[2013-03-19 08:29] warning: /etc/colord.conf installed as /etc/colord.conf.pacnew
[2013-03-19 08:29] upgraded colord (0.1.28-1 -> 0.1.30-2)
[2013-03-19 08:29] upgraded gdk-pixbuf2 (2.26.5-1 -> 2.26.5-2)
[2013-03-19 08:29] upgraded geoclue (0.12.0-5 -> 0.12.99-1)
[2013-03-19 08:29] upgraded gettext (0.18.2-1 -> 0.18.2.1-1)
[2013-03-19 08:29] upgraded gpgme (1.3.1-5 -> 1.3.2-1)
[2013-03-19 08:29] upgraded gtk-update-icon-cache (2.24.16-1 -> 2.24.17-1)
[2013-03-19 08:29] upgraded libxi (1.6.2-1 -> 1.7-1)
[2013-03-19 08:29] upgraded gtk2 (2.24.16-1 -> 2.24.17-1)
[2013-03-19 08:29] upgraded lftp (4.4.4-1 -> 4.4.5-1)
[2013-03-19 08:29] upgraded libassuan (2.0.3-1 -> 2.1.0-1)
[2013-03-19 08:29] upgraded libraw1394 (2.0.7-2 -> 2.1.0-1)
[2013-03-19 08:29] upgraded libdc1394 (2.1.3-2 -> 2.2.1-1)
[2013-03-19 08:29] upgraded libproxy (0.4.7-2 -> 0.4.11-1)
[2013-03-19 08:29] upgraded libwbclient (3.6.12-2 -> 3.6.13-1)
[2013-03-19 08:29] >>> Updating module dependencies. Please wait ...
[2013-03-19 08:29] >>> Generating initial ramdisk, using mkinitcpio.  Please wait...
[2013-03-19 08:29] ==> Building image from preset: 'default'
[2013-03-19 08:29]   -> -k /boot/vmlinuz-linux -c /etc/mkinitcpio.conf -g /boot/initramfs-linux.img
[2013-03-19 08:29] ==> Starting build: 3.8.3-2-ARCH
[2013-03-19 08:29]   -> Running build hook: [base]
[2013-03-19 08:29]   -> Running build hook: [udev]
[2013-03-19 08:29]   -> Running build hook: [autodetect]
[2013-03-19 08:29]   -> Running build hook: [block]
[2013-03-19 08:29]   -> Running build hook: [filesystems]
[2013-03-19 08:29] ==> WARNING: Hook 'usbinput' is deprecated. Replace it with 'keyboard' in your config
[2013-03-19 08:29]   -> Running build hook: [keyboard]
[2013-03-19 08:29]   -> Running build hook: [fsck]
[2013-03-19 08:29] ==> Generating module dependencies
[2013-03-19 08:29] ==> Creating gzip initcpio image: /boot/initramfs-linux.img
[2013-03-19 08:29] ==> Image generation successful
[2013-03-19 08:29] ==> Building image from preset: 'fallback'
[2013-03-19 08:29]   -> -k /boot/vmlinuz-linux -c /etc/mkinitcpio.conf -g /boot/initramfs-linux-fallback.img -S autodetect
[2013-03-19 08:29] ==> Starting build: 3.8.3-2-ARCH
[2013-03-19 08:29]   -> Running build hook: [base]
[2013-03-19 08:29]   -> Running build hook: [udev]
[2013-03-19 08:29]   -> Running build hook: [block]
[2013-03-19 08:29]   -> Running build hook: [filesystems]
[2013-03-19 08:29] ==> WARNING: Hook 'usbinput' is deprecated. Replace it with 'keyboard' in your config
[2013-03-19 08:29]   -> Running build hook: [keyboard]
[2013-03-19 08:29]   -> Running build hook: [fsck]
[2013-03-19 08:29] ==> Generating module dependencies
[2013-03-19 08:29] ==> Creating gzip initcpio image: /boot/initramfs-linux-fallback.img
[2013-03-19 08:29] ==> Image generation successful
[2013-03-19 08:29] upgraded linux (3.7.10-1 -> 3.8.3-2)
[2013-03-19 08:29] upgraded man-pages (3.48-1 -> 3.50-1)
[2013-03-19 08:29] upgraded mpfr (3.1.1.p2-2 -> 3.1.2-1)
[2013-03-19 08:29] upgraded perl (5.16.2-4 -> 5.16.3-1)
[2013-03-19 08:29] upgraded smbclient (3.6.12-2 -> 3.6.13-1)
[2013-03-19 08:29] upgraded samba (3.6.12-2 -> 3.6.13-1)
[2013-03-19 08:29] upgraded sg3_utils (1.33-1 -> 1.35-1)
[2013-03-19 08:29] upgraded sudo (1.8.6.p7-1 -> 1.8.6.p7-2)
[2013-03-19 08:29] upgraded texinfo (5.0-1 -> 5.1-1)
[2013-03-19 08:29] upgraded upower (0.9.19-1 -> 0.9.20-1)
[2013-03-19 08:29] upgraded util-linux (2.22.2-1 -> 2.22.2-2)
[2013-03-19 08:29] upgraded xf86-input-evdev (2.7.3-2 -> 2.7.3-3)
[2013-03-19 08:29] upgraded xf86-input-synaptics (1.6.3-1 -> 1.6.3-2)
[2013-03-19 08:29] upgraded xf86-video-intel (2.21.3-1 -> 2.21.4-1)
[2013-03-19 08:29] upgraded xf86-video-nouveau (1.0.6-1 -> 1.0.6-2)
[2013-03-19 08:29] upgraded xorg-server-common (1.13.3-1 -> 1.14.0-2)
[2013-03-19 08:29] upgraded xorg-server (1.13.3-1 -> 1.14.0-2)
[2013-03-19 08:29] upgraded xorg-server-devel (1.13.3-1 -> 1.14.0-2)

Xorg log:

...
[     9.102] (**) intel(G0): Triple buffering? enabled
[     9.102] (**) intel(G0): Framebuffer tiled
[     9.102] (**) intel(G0): Pixmaps tiled
[     9.102] (**) intel(G0): 3D buffers tiled
[     9.102] (**) intel(G0): SwapBuffers wait enabled
[     9.102] (==) intel(G0): video overlay key set to 0x101fe
[     9.126] (II) intel(G0): Output VGA1 has no monitor section
[     9.153] (II) intel(G0): EDID for output VGA1
[     9.153] (II) intel(G0): Output VGA1 disconnected
[     9.153] (WW) intel(G0): No outputs definitely connected, trying again...
[     9.153] (II) intel(G0): Output VGA1 disconnected
[     9.153] (WW) intel(G0): Unable to find connected outputs - setting 1024x768 initial framebuffer
[     9.153] (II) intel(G0): Using default gamma of (1.0, 1.0, 1.0) unless otherwise stated.
[     9.153] (II) intel(G0): Kernel page flipping support detected, enabling
[     9.153] (==) intel(G0): DPI set to (96, 96)
[     9.153] (II) Loading sub module "fb"
[     9.153] (II) LoadModule: "fb"
[     9.153] (II) Loading /usr/lib/xorg/modules/libfb.so
[     9.153] (II) Module fb: vendor="X.Org Foundation"
[     9.153] 	compiled for 1.14.0, module version = 1.0.0
[     9.153] 	ABI class: X.Org ANSI C Emulation, version 0.4
[     9.153] (II) Loading sub module "dri2"
[     9.153] (II) LoadModule: "dri2"
[     9.153] (II) Module "dri2" already built-in
[     9.153] (--) Depth 24 pixmap format is 32 bpp
[     9.153] (II) intel(G0): [DRI2] Setup complete
[     9.153] (II) intel(G0): [DRI2]   DRI driver: i965
[     9.153] (II) intel(G0): Allocated new frame buffer 2880x1800 stride 11776, tiled
[     9.153] (II) UXA(256): Driver registered support for the following operations:
[     9.153] (II)         solid
[     9.153] (II)         copy
[     9.153] (II)         composite (RENDER acceleration)
[     9.154] (II)         put_image
[     9.154] (II)         get_image
[     9.154] (==) intel(G0): Backing store disabled
[     9.154] (==) intel(G0): Silken mouse enabled
[     9.154] (II) intel(G0): Initializing HW Cursor
[     9.154] (II) intel(G0): RandR 1.2 enabled, ignore the following RandR disabled message.
[     9.154] (==) intel(G0): DPMS enabled
[     9.154] (==) intel(G0): Intel XvMC decoder enabled
[     9.154] (II) intel(G0): Set up textured video
[     9.154] (II) intel(G0): [XvMC] xvmc_vld driver initialized.
[     9.154] (II) intel(G0): direct rendering: DRI2 Enabled
[     9.154] (==) intel(G0): hotplug detection: "enabled"
[     9.154] 
Fatal server error:
[     9.154] AddScreen/ScreenInit failed for gpu driver 0 -1
[     9.154] 
[     9.154] (EE) 
Please consult the The X.Org Foundation support 
	 at http://wiki.x.org
 for help. 
[     9.154] (EE) Please also check the log file at "/var/log/Xorg.5.log" for additional information.
[     9.154] (EE) 
[     9.160] Server terminated with error (1). Closing log file.

List of /etc/X11 and /etc/X11/xorg.conf.d/ :

# ls -la /etc/X11
drwxr-xr-x  2 root root 4096 Nov 18 23:27 .bak
drwxr-xr-x  2 root root 4096 Aug  2  2011 twm
drwxr-xr-x  3 root root 4096 Oct 31 19:17 xinit
drwxr-xr-x  2 root root 4096 Mar 19 08:29 xorg.conf.d
# ls -la /etc/X11/xorg.conf.d
-rw-r--r-- 1 root root 1099 Aug 27  2012 10-evdev.conf
-rw-r--r-- 1 root root  548 Nov  5 20:25 10-mtrack.bak
-rw-r--r-- 1 root root  377 Aug 27  2012 10-quirks.conf
-rw-r--r-- 1 root root  542 Nov  5 20:21 10-synaptics.conf.pacsave
-rw-r--r-- 1 root root 1855 Feb  1 16:25 50-synaptics.conf

Dmesg doesn't really say anything useful.

I've encountered similar problems when starting X before, but never really found any good ways of getting more info on why X-server won't start. Any help is very appreciated!

/ Tomas

Last edited by Joekickass (2013-03-20 14:22:14)

Lone_Wolf · 2013-03-20 11:27:55

Troubleshooting X is often tough, but usually going through Xorg.0.log with fine-combed tooth gives some clues.

From what i can see you have 2 videocards, an integrated intel one and a nvidia one.
The partial xorg log you posted indicates X doesn't detect any connected outputs on the intel, but there's nothing about the nvidia card.
We'll need the full Xorg log.

Which drivers do you have installed (intel nouvea , nvidia ) ?

Problems with starting X can be caused by loading kernel modules with incorrect settings, often this is only visible in dmesg, please post that also.

P.S.
dmesg and Xorg logs tend to be very long, consider using a pastebin site.

Last edited by Lone_Wolf (2013-03-20 11:28:46)

Joekickass · 2013-03-20 12:18:08

Lone_Wolf wrote:

Troubleshooting X is often tough, but usually going through Xorg.0.log with fine-combed tooth gives some clues.

From what i can see you have 2 videocards, an integrated intel one and a nvidia one.
The partial xorg log you posted indicates X doesn't detect any connected outputs on the intel, but there's nothing about the nvidia card.
We'll need the full Xorg log.
Which drivers do you have installed (intel nouvea , nvidia ) ?
Problems with starting X can be caused by loading kernel modules with incorrect settings, often this is only visible in dmesg, please post that also.
P.S.
dmesg and Xorg logs tend to be very long, consider using a pastebin site.

Thanks for your reply! Yes, you are correct, I have a Nvidia Optimus setup with one intel and one nvidia card. I'm using nouveau for the nvidia card, which is also the one enabled from start. I have vga_switcheroo installed but never got it to work on this machine, so the Intel driver is loaded but never actually used. The hardware is relatively new so I kinda expect stuff to break sometimes. That's why I also ask on pointers on what to look for, so I can ask more specific questions in the forums...

I have added the logs you requested below:

Xorg:
http://pastebin.com/VeFg1RAD

dmesg:
http://pastebin.com/tKBVyseM

/ Tomas

Kilz · 2013-03-20 12:55:20

More background on this issue can be found Here and Here.

Joekickass · 2013-03-20 12:58:11

Yeah, I wrote some of those posts, but found no solution in neither of them. Thanks anyway!

/ Tomas

Lone_Wolf · 2013-03-20 13:52:17

Both dmesg and Xorg log point to problems with the intel card .
(Your monitor is connected to the nouveau card, so it seems you are lucky and have 2 cards that each have their own output).

There have been threads about 3.8.x and problems with intel cards (forum and arch-general or arch-dev mailing list i think), so i'd say we can try 2 things :

- remove xf86-video-intel , so X can only use nouveau .
- downgrade the kernel to the 3.7.x versions

You state that the system enables the nvidia card from the start.
If that happens during (u)efi setup, before booting an OS the first option seems the best choice.

fsckd · 2013-03-20 14:02:22

Don't cross-post. If you wanted a thread moved, you could have just asked us moderators. Closing the other thread since this one has activity.

Joekickass · 2013-03-20 14:10:20

fsckd wrote:

Don't cross-post. If you wanted a thread moved, you could have just asked us moderators. Closing the other thread since this one has activity.

Sorry about that. Now I know

Joekickass · 2013-03-20 14:21:38

Lone_Wolf wrote:

Both dmesg and Xorg log point to problems with the intel card .
(Your monitor is connected to the nouveau card, so it seems you are lucky and have 2 cards that each have their own output).
There have been threads about 3.8.x and problems with intel cards (forum and arch-general or arch-dev mailing list i think), so i'd say we can try 2 things :
- remove xf86-video-intel , so X can only use nouveau .
- downgrade the kernel to the 3.7.x versions
You state that the system enables the nvidia card from the start.
If that happens during (u)efi setup, before booting an OS the first option seems the best choice.

Ah, so even if the intel card is not used, it can still mess up the X-server startup?

... and you're right! Removing xf86-video-intel solved the problem:

sudo pacman -Rs xf86-video-intel

Much better than downgrading kernel and/or x-server which was my first thought!

However, now I would like to know how I can fix the Intel card. But I will put that in a separate thread since I've already been reprimanded by moderators

Thanks for your help! By the way, what line(s) tipped you off in Xorg.log?

/ Tomas

Lone_Wolf · 2013-03-20 22:04:14

[     9.153] (WW) intel(G0): No outputs definitely connected, trying again...
[     9.153] (II) intel(G0): Output VGA1 disconnected
[     9.153] (WW) intel(G0): Unable to find connected outputs - setting 1024x768 initial framebuffer

This seems a bit weird, i'm 99% sure i've seen xorg logs with multiple cards where cards without connected screens gave no problem.

from dmesg :

[    4.663960] i915 0000:00:02.0: setting latency timer to 64
[    4.704125] i915 0000:00:02.0: irq 45 for MSI/MSI-X
[    4.704131] [drm] Supports vblank timestamp caching Rev 1 (10.10.2010).
[    4.704132] [drm] Driver supports precise vblank timestamp query.
[    4.704138] i915 0000:00:02.0: Invalid ROM contents
[    4.704168] [drm] failed to find VBIOS tables
[    4.704220] vgaarb: device changed decodes: PCI:0000:00:02.0,olddecodes=io+mem,decodes=none:owns=io+mem
[    4.704222] vgaarb: transferring owner from PCI:0000:00:02.0 to PCI:0000:01:00.0
[    4.730994] [drm] failed to retrieve link info, disabling eDP
[    5.304151] [drm] GMBUS [i915 gmbus vga] timed out, falling back to bit banging on pin 2
[    5.317426] i915 0000:00:02.0: No connectors reported connected with modes
[    5.317429] [drm] Cannot find any crtc or sizes - going 1024x768

Invalid ROM contents and unable to find VBIOS tables normally happens with graphic cards that are to new and not yet supported by the kernel.
The i915 has been in use for years already, it doesn't make sense the kernel can't use it correctly.

In short, it was the combination of xorg and dmesg that convinced me the intel module has a problem on your system (and possibly on other intel hardware).

Joekickass · 2013-03-21 07:09:38

Lone_Wolf wrote:

[     9.153] (WW) intel(G0): No outputs definitely connected, trying again...
[     9.153] (II) intel(G0): Output VGA1 disconnected
[     9.153] (WW) intel(G0): Unable to find connected outputs - setting 1024x768 initial framebuffer
This seems a bit weird, i'm 99% sure i've seen xorg logs with multiple cards where cards without connected screens gave no problem.
from dmesg :
[    4.663960] i915 0000:00:02.0: setting latency timer to 64
[    4.704125] i915 0000:00:02.0: irq 45 for MSI/MSI-X
[    4.704131] [drm] Supports vblank timestamp caching Rev 1 (10.10.2010).
[    4.704132] [drm] Driver supports precise vblank timestamp query.
[    4.704138] i915 0000:00:02.0: Invalid ROM contents
[    4.704168] [drm] failed to find VBIOS tables
[    4.704220] vgaarb: device changed decodes: PCI:0000:00:02.0,olddecodes=io+mem,decodes=none:owns=io+mem
[    4.704222] vgaarb: transferring owner from PCI:0000:00:02.0 to PCI:0000:01:00.0
[    4.730994] [drm] failed to retrieve link info, disabling eDP
[    5.304151] [drm] GMBUS [i915 gmbus vga] timed out, falling back to bit banging on pin 2
[    5.317426] i915 0000:00:02.0: No connectors reported connected with modes
[    5.317429] [drm] Cannot find any crtc or sizes - going 1024x768
Invalid ROM contents and unable to find VBIOS tables normally happens with graphic cards that are to new and not yet supported by the kernel.
The i915 has been in use for years already, it doesn't make sense the kernel can't use it correctly.
In short, it was the combination of xorg and dmesg that convinced me the intel module has a problem on your system (and possibly on other intel hardware).

Very nice, thanks again!

Arch Linux

#1 2013-03-20 08:52:28

[Solved] X won't start after latest update

#2 2013-03-20 11:27:55

Re: [Solved] X won't start after latest update

#3 2013-03-20 12:18:08

Re: [Solved] X won't start after latest update

#4 2013-03-20 12:55:20

Re: [Solved] X won't start after latest update

#5 2013-03-20 12:58:11

Re: [Solved] X won't start after latest update

#6 2013-03-20 13:52:17

Re: [Solved] X won't start after latest update

#7 2013-03-20 14:02:22

Re: [Solved] X won't start after latest update

#8 2013-03-20 14:10:20

Re: [Solved] X won't start after latest update

#9 2013-03-20 14:21:38

Re: [Solved] X won't start after latest update

#10 2013-03-20 22:04:14

Re: [Solved] X won't start after latest update

#11 2013-03-21 07:09:38

Re: [Solved] X won't start after latest update

Board footer