Arch Linux

Please read this before reporting a bug:
https://wiki.archlinux.org/title/Bug_reporting_guidelines

Do NOT report bugs when a package is just outdated, or it is in the AUR. Use the 'flag out of date' link on the package page, or the Mailing List.

REPEAT: Do NOT report bugs for outdated packages!
Tasklist

FS#15475 - [xf86-video-intel] X fails to start after upgrade - Intel driver problem

Attached to Project: Arch Linux
Opened by robin wood (dninja) - Friday, 10 July 2009, 22:26 GMT
Last edited by Jan de Groot (JGC) - Thursday, 22 October 2009, 17:05 GMT
Task Type Bug Report
Category Packages: Extra
Status Closed
Assigned To Jan de Groot (JGC)
Architecture i686
Severity High
Priority Normal
Reported Version
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 0
Private No

Details

Description:
I've just done an upgrade and now X won't start. The error from the xorg log file is at the bottom

Additional info:

intel-dri 7.4.4-1
videoproto 2.2.2-2
xf86-video-intel 2.7.1-1
xorg-apps 7.4-2
xorg-font-utils 7.4-2
xorg-fonts-100dpi 1.0.1-2
xorg-fonts-75dpi 1.0.1-2
xorg-fonts-alias 1.0.1-2
xorg-fonts-encodings 1.0.2-3
xorg-fonts-misc 1.0.0-4
xorg-res-utils 1.0.3-3
xorg-server 1.6.1.901-3
xorg-server-utils 7.4-6
xorg-twm 1.0.4-3
xorg-utils 7.4-4
xorg-xauth 1.0.3-1
xorg-xinit 1.1.1-1
xorg-xkb-utils 7.4-2
xorg-xsm 1.0.1-2

This happened last time I did an upgrade so I rolled back any package associated with X or Intel and that fixed it.

Steps to reproduce:

Just starting x.

(**) Keyboard0: CustomKeycodes disabled
(II) XINPUT: Adding extended input device "Keyboard0" (type: KEYBOARD)

Backtrace:
0: /usr/bin/X(xorg_backtrace+0x3b) [0x813158b]
1: /usr/bin/X(xf86SigHandler+0x9e) [0x80bd1ce]
2: [0xb8069400]
3: /usr/lib/xorg/modules/drivers//intel_drv.so(intel_batch_flush+0xe0) [0xb7a896
20]
4: /usr/lib/xorg/modules/drivers//intel_drv.so [0xb7a952a6]
5: /usr/bin/X [0x8179113]
6: /usr/bin/X [0x81419fe]
7: /usr/bin/X(BlockHandler+0x58) [0x8090808]
8: /usr/bin/X(WaitForSomething+0x10c) [0x812f02c]
9: /usr/bin/X(Dispatch+0x80) [0x808c350]
10: /usr/bin/X(main+0x395) [0x8072005]
11: /lib/libc.so.6(__libc_start_main+0xe6) [0xb7c20a36]
12: /usr/bin/X [0x80714b1]

Fatal server error:
Caught signal 11. Server aborting


Please consult the The X.Org Foundation support
at http://wiki.x.org
for help.
Please also check the log file at "/var/log/Xorg.0.log" for additional information.

(II) UnloadModule: "mouse"
(II) UnloadModule: "kbd"
(II) AIGLX: Suspending AIGLX clients for VT switch
This task depends upon

Closed by  Jan de Groot (JGC)
Thursday, 22 October 2009, 17:05 GMT
Reason for closing:  Fixed
Additional comments about closing:  Should be fixed with 2.6.31 kernel.
Comment by robin wood (dninja) - Friday, 10 July 2009, 22:58 GMT
Some extra information, I just did some roll backs and found that it is one of these packages that broke the system

intel-dri-7.2-2*
xf86-input-evdev-2.1.2-1*
xf86-input-keyboard-1.3.2-1*
xf86-input-mouse-1.3.0-1*
xf86-video-intel-2.4.3-1*
xorg-server-utils-7.4-2*
xorg-server-1.5.3-4*
xorg-utils-7.4-2*

These are the versions it wants to upgrade them to:


Targets (8): intel-dri-7.4.4-1 xf86-input-evdev-2.2.2-1
xf86-input-keyboard-1.3.2-2 xf86-input-mouse-1.4.0-2
xf86-video-intel-2.7.1-1 xorg-server-utils-7.4-6
xorg-server-1.6.1.901-3 xorg-utils-7.4-4

This is the output from lspci

00:02.0 VGA compatible controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 03)
00:02.1 Display controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 03)

And the machine is a Toshiba Equium A200-1V0
Comment by Gladstone (gladstone) - Saturday, 11 July 2009, 21:17 GMT
I get the same problem with xf86-video-intel-2.7.99.901-1, which I've pinpointed down to Xorg not respecting my:

Option "AccelMethod" "exa"

setting in xorg.conf.

$ cat /var/log/Xorg.0.log.old | grep -i accel
(--) intel(0): Using UXA for acceleration
(II) composite (RENDER acceleration)
(WW) intel(0): Option "AccelMethod" is not used

This system does not work with UXA and it seems EXA isn't working anymore/has been removed already.
Comment by robin wood (dninja) - Saturday, 11 July 2009, 21:22 GMT
I don't have that line in my xorg.conf but I do see this in my log file

(WW) intel(0): libpciaccess reported 0 rom size, guessing 64kB
(**) intel(0): Using XAA for acceleration
(II) intel(0): 2 display pipes available.

Do you have a fix/work around?
Comment by Lysikov Stanislav (stave) - Sunday, 12 July 2009, 19:32 GMT
I have the some problem after upgrade. In log file:

(EE)Intel(1): No valid modes.
...
Fatal server error:
Caught signal 11. Server aborting
Comment by kang (kang) - Wednesday, 15 July 2009, 00:41 GMT
same problem, but xf86-video-intel-newest from AUR which has version xf86-video-intel-2.7.99.902-1 works fine again (extra has xf86-video-intel-2.7.99.901-1)
Comment by robin wood (dninja) - Wednesday, 15 July 2009, 22:37 GMT
That worked for me as well.

I assume for the long term we just keep an eye on the xf86-video-intel package in the normal repository and when that hits 2.7.99.901-1 we can go back to it. Is that correct?
Comment by Jan de Groot (JGC) - Saturday, 18 July 2009, 21:51 GMT
Please try with xorg-server, intel-dri and xf86-video-intel from testing.
Comment by mark (mmm) - Thursday, 30 July 2009, 06:36 GMT
testing works for me flawlessly
Comment by Roman Kyrylych (Romashka) - Thursday, 30 July 2009, 11:20 GMT
Anyone else can confirm that testing fixes the problem now?
(if noone replies in the following days, I suggest closing this as Fixed)
Comment by Gladstone (gladstone) - Thursday, 30 July 2009, 11:39 GMT
As of 2.7.99.902-1 (which is now in extra), everything works fine for me. Should be the same for the testing version.
Comment by Daniele C. (legolas558) - Tuesday, 18 August 2009, 17:49 GMT
  • Field changed: Percent Complete (100% → 0%)
I have the same problem:

(EE) intel(0): Output LVDS enabled but has no modes
(EE) intel(0): No valid modes.
(EE) Screen(s) found, but none have a usable configuration.

I am using:

extra/xorg-server 1.6.3-3
extra/xf86-video-intel 2.7.99.902-1

I am not using stock kernel but linus' git kernel 2.6.31-rc6-00099-gdf4ecf1
Comment by Jan de Groot (JGC) - Tuesday, 18 August 2009, 17:59 GMT
What happens when using stock kernel?
Comment by Daniele C. (legolas558) - Wednesday, 09 September 2009, 09:53 GMT
When using stock kernel the exact same thing happened.

Now I am using 2.6.31-rc9 (vanilla) and xf86-video-intel-2.8.1-1 and the bug is gone.

Can be closed.

Loading...