[i830] when memory runs low, xorg locks up and does not unlock when memory conditions improve

[themusicgod1]

Binary package hint: xorg

When I run low on memory, say, 20k or so remaining, xorg locks up (doesn't respond to input at all, incl mouse movements/capslock). OK that makes sense...but the problem is that once memory situation improves (say, by me sshing from another computer to kill the process that's taking up the memory, whether it be firefox, gnu electric, or whathaveyou), xorg remains locked up, even if reniced to a more realtime priority. It remains at 0% cpu usage, too, even while the system is idle and I am giving it input via mouse/touchscreen.

Unlike the other reports on the intel drivers, Xorg is not using a disproportionate amount of memory -- the amount of memory used is fine, it is just not adjusting to memory conditions on the system.

Ubuntu Karmic / evtouch touchscreen / intel drivers probably from below

(also apologies that the attachments that were supposed to go towards this bug report went to the last one -- had i known it was going to attach all that stuff I would have just used the launchpad interface :) )

ProblemType: Bug
Architecture: i386
Date: Mon Nov 9 17:04:51 2009
DistroRelease: Ubuntu 9.10
InstallationMedia: Ubuntu 9.10 "Karmic Koala" - Release i386 (20091028.5)
Lsusb: Bus 001 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
MachineType: Matsushita Electric Industrial Co.,Ltd. CF-28STJGZDM
Package: xorg 1:7.4+3ubuntu7
PccardctlIdent:
 Socket 0:
   no product info available
 Socket 1:
   no product info available
PccardctlStatus:
 Socket 0:
   no card
 Socket 1:
   no card
ProcCmdLine: root=/dev/sda3 ro quiet splash
ProcEnviron:
 LANG=en_CA.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.31-14.48-generic
RelatedPackageVersions:
 xserver-xorg 1:7.4+3ubuntu7
 libgl1-mesa-glx 7.6.0-1ubuntu4
 libdrm2 2.4.14-1ubuntu1
 xserver-xorg-video-intel 2:2.9.0-1ubuntu2
 xserver-xorg-video-ati 1:6.12.99+git20090929.7968e1fb-0ubuntu1
SourcePackage: xorg
Uname: Linux 2.6.31-14-generic i686
XsessionErrors:
 (gnome-settings-daemon:3711): GLib-CRITICAL **: g_propagate_error: assertion `src != NULL' failed
 (gnome-settings-daemon:3711): GLib-CRITICAL **: g_propagate_error: assertion `src != NULL' failed
 (nautilus:3744): Eel-CRITICAL **: eel_preferences_get_boolean: assertion `preferences_is_initialized ()' failed
 (polkit-gnome-authentication-agent-1:3749): GLib-CRITICAL **: g_once_init_leave: assertion `initialization_value != 0' failed
dmi.bios.date: 04/02/2003
dmi.bios.vendor: Phoenix Technologies K.K.
dmi.bios.version: V3.00L12
dmi.board.name: CF28-3
dmi.board.vendor: Matsushita Electric Industrial Co.,Ltd.
dmi.board.version: 001
dmi.chassis.asset.tag: No Asset Tag
dmi.chassis.type: 10
dmi.chassis.vendor: Matsushita Electric Industrial Co.,Ltd.
dmi.chassis.version: 001
dmi.modalias: dmi:bvnPhoenixTechnologiesK.K.:bvrV3.00L12:bd04/02/2003:svnMatsushitaElectricIndustrialCo.,Ltd.:pnCF-28STJGZDM:pvr003:rvnMatsushitaElectricIndustrialCo.,Ltd.:rnCF28-3:rvr001:cvnMatsushitaElectricIndustrialCo.,Ltd.:ct10:cvr001:
dmi.product.name: CF-28STJGZDM
dmi.product.version: 003
dmi.sys.vendor: Matsushita Electric Industrial Co.,Ltd.
fglrx: Not loaded
system:
 distro: Ubuntu
 architecture: i686kernel: 2.6.31-14-generic

[lspci]
00:02.0 VGA compatible controller [0300]: Intel Corporation 82830 CGC [Chipset Graphics Controller] [8086:3577] (rev 04)
     Subsystem: Matsushita Electric Industrial Co., Ltd. Device [10f7:8338]

[themusicgod1 (themusicgod1)]

I tested this on the karmic, lucid, and debian sid/experimental versions of xserver-xorg-video-intel and the problem remained. Installed debugging symbols for xorg, and brought it up in gdb(in a 'normal' situation, ie when gdm first loads, the backtrace had meaningful information, although not (yet) source lines), and when the freeze occurs, all I get is
"
Backtrace stopped: previous frame identical to this frame()
#2 0x00e9f07 in ?? () from lib.so.6
#1 0x00f07ae3 in ?? () from lib.so.6
#0 0x00d01422 in ?? __kernel_vsyscall ()
"

It should be noted that this same hardware appears to not exhibit this bug on debian stable where the i810 driver is used (it's an 830).

[Bryce Harrington (bryce)]

Have you had a chance to test this on lucid? Unfortunately I do not have specific info that it's fixed, however I did pull in several i8xx fixes from upstream, so am hopeful.

I'd suggest booting your kernel with i915.modeset=0 because it's been found that this chip doesn't work well with KMS (lp #542208). We'll be blacklisting kms for that chip to resolve that bug, but I'm curious about whether your issue will be remaining.

[themusicgod1 (themusicgod1)]

I did test it in lucid but not either of the betas, or with this modeset option. I'll see if I can try that later.

[Launchpad Janitor (janitor)]

[Expired for xserver-xorg-video-intel (Ubuntu) because there has been no activity for 60 days.]