low memory corruption and ext3 corruption starting googleearth ("radeon" Xorg driver)

[Björn Schließmann]

I'm experiencing ext3 root fs corruption after starting a user space application. I first noticed it when starting googleearth in Gnome from a terminal; the prompt returned but almost no commands were found and the root filesystem had automatically been remounted ro. I suspected a hardware problem, but after rebooting and fscking a smartctl long selftest went fine (also, my hd is under 200 service hours old).

I reproduced the problem once more, this time in a VT (Ctrl-Alt-Fn) while logged in to Gnome. Starting googleearth, nothing happened (seems to be by design, on my other machine where googleearth makes no other problems it's the same). So I tried "DISPLAY=:0 googleearth". The prompt did not return, but in dmesg there were no errors yet. So I switched to X (Alt-F7). I saw the Googleearth splash screen for a short time and it vanished again. Switching to VT1 again I now saw a few kernel ext3 warnings. Trying to issue commands, hell broke loose; messages about low memory corruption and many ext3 errors.

I'll attach "screenshots" of dmesg excerpts (sorry, couldn't get it out in text form -- no scp, no nc, no mount, nothing I'd have needed worked).

I'll do further tries if it helps, just drop a note.

Ah BTW, googleearth worked for me in the previous release.

ProblemType: Bug
Architecture: i386
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: chronoss 2270 F.... pulseaudio
CRDA:
 country DE:
  (2400 - 2483 @ 40), (N/A, 20)
  (5150 - 5350 @ 40), (N/A, 20), NO-OUTDOOR, DFS
  (5470 - 5725 @ 40), (N/A, 26), NO-OUTDOOR, DFS
Card0.Amixer.info:
 Card hw:0 'I82801DBICH4'/'Intel 82801DB-ICH4 with AD1981B at irq 5'
   Mixer name : 'Analog Devices AD1981B'
   Components : 'AC97a:41445374'
   Controls : 26
   Simple ctrls : 18
Date: Mon Feb 8 23:26:20 2010
DistroRelease: Ubuntu 9.10
HibernationDevice: RESUME=UUID=ba5e9423-6d83-46d6-9a22-0b33ef96ebe3
Lsusb:
 Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
 Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
 Bus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
 Bus 002 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
MachineType: IBM 2373A91
Package: linux-image-2.6.31-19-generic 2.6.31-19.56
ProcCmdLine: root=UUID=4d015652-6ae1-4e87-acc6-e8c43a344ba5 ro lapic rootflags=data=journal vga=791
ProcEnviron:
 PATH=(custom, no user)
 LANG=en_US.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.31-19.56-generic
RelatedPackageVersions:
 linux-backports-modules-2.6.31-19-generic 2.6.31-19.21
 linux-firmware 1.25
RfKill:
 0: phy0: Wireless LAN
  Soft blocked: no
  Hard blocked: no
SourcePackage: linux
Uname: Linux 2.6.31-19-generic i686
UserAsoundrc:
 # ALSA library configuration file

 # Include settings that are under the control of asoundconf(1).
 # (To disable these settings, comment out this line.)
 </home/chronoss/.asoundrc.asoundconf>
WpaSupplicantLog:

XsessionErrors:
 (gnome-settings-daemon:2288): GLib-CRITICAL **: g_propagate_error: assertion `src != NULL' failed
 (polkit-gnome-authentication-agent-1:2327): GLib-CRITICAL **: g_once_init_leave: assertion `initialization_value != 0' failed
 (nautilus:2323): Eel-CRITICAL **: eel_preferences_get_boolean: assertion `preferences_is_initialized ()' failed
 (gnome-panel:2322): Gdk-WARNING **: /build/buildd/gtk+2.0-2.18.3/gdk/x11/gdkdrawable-x11.c:952 drawable is not a pixmap or window
dmi.bios.date: 06/18/2007
dmi.bios.vendor: IBM
dmi.bios.version: 1RETDRWW (3.23 )
dmi.board.name: 2373A91
dmi.board.vendor: IBM
dmi.board.version: Not Available
dmi.chassis.asset.tag: No Asset Information
dmi.chassis.type: 10
dmi.chassis.vendor: IBM
dmi.chassis.version: Not Available
dmi.modalias: dmi:bvnIBM:bvr1RETDRWW(3.23):bd06/18/2007:svnIBM:pn2373A91:pvrThinkPadT41:rvnIBM:rn2373A91:rvrNotAvailable:cvnIBM:ct10:cvrNotAvailable:
dmi.product.name: 2373A91
dmi.product.version: ThinkPad T41
dmi.sys.vendor: IBM

[Björn Schließmann (b-schliessmann)]

Well, I did a few experiments. When trying with glxgears, the same "low memory corruption" happened, but glxgears didn't crash and my root filesystem didn't break. It seems the scope of this error depends on how intensely 3D acceleration is used. Please see attached file. "Corrupted low memory" starts about line 180, a traceback is at the bottom.

[Björn Schließmann (b-schliessmann)]

I tried again with kernel mode setting enabled, and the problem doesn't occur anymore.

But -- in the console appears, directly after starting glxgears or googleearth, the following. I'm not sure where it comes from, but "WARN_ONCE" sounds like kernel. There's no traceback though. Xorg.0.log shows nothing suspicious.

*********************************WARN_ONCE*********************************
File radeon_tcl.c function radeon_run_tcl_render line 499
Rendering was 96 commands larger than predicted size. We might overflow command buffer.
***************************************************************************

[Jeremy Foshee (jeremyfoshee)]

Hi Björn,

This bug was reported a while ago and there hasn't been any activity in it recently. We were wondering if this is still an issue? Can you try with the latest development release of Ubuntu? ISO CD images are available from http://cdimage.ubuntu.com/releases/ .

If it remains an issue, could you run the following command from a Terminal (Applications->Accessories->Terminal). It will automatically gather and attach updated debug information to this report.

apport-collect -p linux 519055

Also, if you could test the latest upstream kernel available that would be great. It will allow additional upstream developers to examine the issue. Refer to https://wiki.ubuntu.com/KernelMainlineBuilds . Once you've tested the upstream kernel, please remove the 'needs-upstream-testing' tag. This can be done by clicking on the yellow pencil icon next to the tag located at the bottom of the bug description and deleting the 'needs-upstream-testing' text. Please let us know your results.

Thanks in advance.

    [This is an automated message. Apologies if it has reached you inappropriately; please just reply to this message indicating so.]

[Björn Schließmann (b-schliessmann)]

With the lucid 2.6.32-21-generic kernel this is no more an issue (tested with kernel modesetting off; error doesn't appear with either EXA or XAA).

Would it still make sense to test the mainline kernel?

[Jeremy Foshee (jeremyfoshee)]

Björn,
    I think we are fine with this. I have marked the bug Fix released.

Thanks!

~JFo

[Björn Schließmann (b-schliessmann)]

Ok, and no worries! :)