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Do NOT report bugs when a package is just outdated, or it is in the AUR. Use the 'flag out of date' link on the package page, or the Mailing List.

REPEAT: Do NOT report bugs for outdated packages!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#11995 - XOrg spontaneously crashes

Attached to Project: Arch Linux
Opened by name withheld (Gullible Jones) - Tuesday, 04 November 2008, 16:39 GMT
Last edited by Jan de Groot (JGC) - Friday, 10 April 2009, 21:09 GMT
Task Type Bug Report
Category Packages: Extra
Status Closed
Assigned To Jan de Groot (JGC)
Architecture All
Severity Low
Priority Normal
Reported Version None
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 1
Private No

Details

Description:

XOrg can spontaneously exit without any warning; if restarted, it will immediately exit again, which is highly annoying for those of us using display managers.

Additional info:
* Using XOrg and Intel driver from Extra, EXA enabled as per default, with XDM as the display manager


Steps to reproduce:

Unknown. Xorg.0.log is attached.
   Xorg.0.log.old (34.3 KiB)
This task depends upon

Closed by  Jan de Groot (JGC)
Friday, 10 April 2009, 21:09 GMT
Reason for closing:  Fixed
Additional comments about closing:  Assuming fixed in xorg-server 1.6.0, haven't seen this bug for a few months now.
Comment by Jan de Groot (JGC) - Friday, 07 November 2008, 09:28 GMT
Please try xorg-server and xf86-video-intel from testing (be sure to install xf86-input-evdev too or check documentation http://wiki.archlinux.org/index.php/Xorg_input_hotplugging). The crashes with 1.4.2 and xf86-video-intel 2.3.1 are known to me, I've had them on a daily base. With xorg-server and xf86-video-intel from testing I stil get these crashes sometimes, but far less often.
Comment by name withheld (Gullible Jones) - Friday, 07 November 2008, 16:59 GMT
FWIW, I switched to using XAA and that fixed this... I think we can write this down as more EXA bugginess.
Comment by Jan de Groot (JGC) - Friday, 07 November 2008, 21:11 GMT
Note that there's no interest in XAA upstream anymore, so any bugs you will find using XAA will get closed as wontfix.
Comment by name withheld (Gullible Jones) - Saturday, 08 November 2008, 04:48 GMT
Ugh, guess I'll try the Testing driver then...
Comment by name withheld (Gullible Jones) - Sunday, 09 November 2008, 01:37 GMT
Hey, just a thought - I had set INTEL_BATCH=1 for performance reasons, could that have been causing the crashes?
Comment by Tiago Pierezan Camargo (tcamargo) - Tuesday, 18 November 2008, 14:31 GMT
Same problem here. After a crash, X won't work anymore. A reboot is needed.

xorg-server-1.4.2-2
xf86-video-intel-2.3.2-1

It's hard to tell which package (or combination of packages) triggered the problem because I did a big update last Sunday.

I will try the packages from testing.
   Xorg.0.log.old (37.9 KiB)
Comment by name withheld (Gullible Jones) - Tuesday, 18 November 2008, 18:56 GMT
Update - the problem was INTEL_BATCH... If you have this enabled, turn it off, and your crashes should disappear.
Comment by Tiago Pierezan Camargo (tcamargo) - Tuesday, 18 November 2008, 19:56 GMT
I never set INTEL_BATCH. With the packages from testing I still don't get a single crash.
Comment by Tiago Pierezan Camargo (tcamargo) - Sunday, 23 November 2008, 09:19 GMT
No crash so far, but 2 problems:
1) I get "Failed to initialize TTM buffer manager. Falling back to classic." every time I try to run an opengl application. Google Earth performance is terrible. I only installed it today so I don't know if it ran faster before.
2) When I run xrandr some artifacts appear on screen for a few seconds. I also got this with other apps but most notably with xrandr.
Comment by Jan de Groot (JGC) - Sunday, 21 December 2008, 22:33 GMT
INTEL_BATCH isn't the real cause of this bug. This bug appears once per month on my office workstation with Q35 where I'm using a dualhead setup that exceeds the maximum screen size for DRI. This means that I always have software rendering without intel DRI.
Note that the logs we capture are not the ones that show the actual bugs. When this happens to me, the screen freezes for a few seconds, X crashes and gdm restarts X a few times. Because the hardware is in an impossible state, X will never come up and will give the error you find in the logs. The only solution is rebooting then.

Loading...