[i965gm] GPU lockup (IPEHR: 0x08000000)

[PresuntoRJ]

Binary package hint: xserver-xorg-video-intel

Not sure if its a xorg problem, since it could also be in gdm or something else. I keep my system up-to-date while testing (just run an apt-get upgrade) and have been testing banshee when the screen went black ... I could not regain access to the system locally with Ctrl+Alt+Bckspc (set to kill X), or even Ctlr+Alt+F2... I had to ssh to this box, but killing gdm or xorg was not enough ... eventually I gave up and rebooted it, so I could post this bug report.

ProblemType: Crash
DistroRelease: Ubuntu 11.04
Package: xserver-xorg-video-intel 2:2.14.0-4ubuntu7
ProcVersionSignature: Ubuntu 2.6.38-8.42-generic-pae 2.6.38.2
Uname: Linux 2.6.38-8-generic-pae i686
Architecture: i386
Chipset: i965gm
CompizPlugins: No value set for `/apps/compiz-1/general/screen0/options/active_plugins'
CompositorRunning: compiz
DRM.card0.LVDS.1:
 status: connected
 enabled: enabled
 dpms: On
 modes: 1280x800
 edid-base64: AP///////wBMozM2AAAAAAAPAQOAIRV4Cof1lFdPjCcnUFQAAAABAQEBAQEBAQEBAQEBAQEB7hoAgFAgEDAQMBMAS88QAAAZAAAADwAAAAAAAAAAACOHAmQCAAAA/gBTQU1TVU5HCiAgICAgAAAA/gBMVE4xNTRYMy1MMDYKAHA=
DRM.card0.VGA.1:
 status: disconnected
 enabled: disabled
 dpms: Off
 modes:
 edid-base64:
Date: Wed Apr 20 13:53:50 2011
DistUpgraded: Log time: 2011-04-09 18:13:48.268164
DistroCodename: natty
DistroVariant: ubuntu
DkmsStatus: virtualbox-ose, 4.0.4, 2.6.38-8-generic, i686: installed
ExecutablePath: /usr/share/apport/apport-gpu-error-intel.py
GraphicsCard:
 Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (primary) [8086:2a02] (rev 0c) (prog-if 00 [VGA controller])
   Subsystem: Lenovo Device [17aa:383e]
   Subsystem: Lenovo Device [17aa:383e]
InstallationMedia: Ubuntu 10.10 "Maverick Meerkat" - Release i386 (20101007)
InterpreterPath: /usr/bin/python2.7
MachineType: LENOVO 0769AAP
ProcCmdline: /usr/bin/python /usr/share/apport/apport-gpu-error-intel.py
ProcEnviron:

ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-2.6.38-8-generic-pae root=UUID=4b092518-b996-429c-b41e-9bf39183163c ro quiet splash vt.handoff=7
ProcKernelCmdLine_: BOOT_IMAGE=/boot/vmlinuz-2.6.38-8-generic-pae root=UUID=4b092518-b996-429c-b41e-9bf39183163c ro quiet splash vt.handoff=7
RelatedPackageVersions:
 xserver-xorg 1:7.6+4ubuntu3
 libdrm2 2.4.23-1ubuntu6
 xserver-xorg-video-intel 2:2.14.0-4ubuntu7
Renderer: Unknown
SourcePackage: xserver-xorg-video-intel
Title: [i965gm] GPU lockup
UpgradeStatus: Upgraded to natty on 2011-04-10 (10 days ago)
UserGroups:

dmi.bios.date: 06/11/2008
dmi.bios.vendor: LENOVO
dmi.bios.version: 68ET37WW
dmi.board.name: IEL10
dmi.board.vendor: LENOVO
dmi.board.version: REFERENCE
dmi.chassis.asset.tag: No Asset Tag
dmi.chassis.type: 10
dmi.chassis.vendor: No Enclosure
dmi.chassis.version: N/A
dmi.modalias: dmi:bvnLENOVO:bvr68ET37WW:bd06/11/2008:svnLENOVO:pn0769AAP:pvr3000N200:rvnLENOVO:rnIEL10:rvrREFERENCE:cvnNoEnclosure:ct10:cvrN/A:
dmi.product.name: 0769AAP
dmi.product.version: 3000 N200
dmi.sys.vendor: LENOVO
version.compiz: compiz 1:0.9.4+bzr20110415-0ubuntu2
version.libdrm2: libdrm2 2.4.23-1ubuntu6
version.libgl1-mesa-dri: libgl1-mesa-dri 7.10.2-0ubuntu2
version.libgl1-mesa-dri-experimental: libgl1-mesa-dri-experimental N/A
version.libgl1-mesa-glx: libgl1-mesa-glx 7.10.2-0ubuntu2
version.xserver-xorg: xserver-xorg 1:7.6+4ubuntu3
version.xserver-xorg-video-ati: xserver-xorg-video-ati 1:6.14.0-0ubuntu4
version.xserver-xorg-video-intel: xserver-xorg-video-intel 2:2.14.0-4ubuntu7
version.xserver-xorg-video-nouveau: xserver-xorg-video-nouveau 1:0.0.16+git20110107+b795ca6e-0ubuntu7

[Bryce Harrington (bryce)]

Have you encountered other system freezes like this one since upgrading to natty? Did you experience freezes like this on maverick? Are you able to reproduce this freeze reliably?

[PresuntoRJ (fabio-tleitao)]

This has never happened in Maverick, or any other Ubuntu release before
that (since 8.04 on this same box), only in Natty, and at least twice (the
second time is the one I have filed the bug)

2011/4/20 Bryce Harrington <email address hidden>

> Have you encountered other system freezes like this one since upgrading
> to natty? Did you experience freezes like this on maverick? Are you
> able to reproduce this freeze reliably?
>
> ** Changed in: xserver-xorg-video-intel (Ubuntu)
> Status: New => Incomplete
>
> --
> You received this bug notification because you are a direct subscriber
> of the bug.
> https://bugs.launchpad.net/bugs/767425
>
> Title:
> [i965gm] GPU lockup (IPEHR: 0x08000000)
>
> To unsubscribe from this bug, go to:
>
> https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/767425/+subscribe
>

--
Fábio Leitão
..-. .- -... .. --- .-.. . .. - .- --- ...-.-

[Bryce Harrington (bryce)]

Presunto, alright I think what I'd like to have you do is to let it freeze a couple more times, so you get a feel for the frequency of the freezes (weekly, daily, etc.) and/or if any conditions or steps seem to tend to lead to the freeze.

After you've gotten enough experience with the freeze, then try installing a newer kernel from here:

  http://kernel.ubuntu.com/~kernel-ppa/mainline/drm-next/

Then run on that kernel for a sufficient period of time to convince you that the freeze is either still there, or is no longer present. If the freeze is still present, attach a fresh dmesg and /sys/kernel/debug/dri/0/i915_error_state from right after a freeze, and we can send the bug report upstream. If it is no longer present, perhaps there is a patch the kernel team can help isolate and backport to natty.

[PresuntoRJ (fabio-tleitao)]

I have just downloaded the newer kernel there (form the current folder) linux-image-2.6.38-996-generic_2.6.38-996.201103251543_i386.deb and its headers (all and i386)... lets see how the do. ;)

Is this some specially compiled debug version of the kernel? Or just with some newer patches?

[PresuntoRJ (fabio-tleitao)]

Just downloaded and installed the lastest Kernel available there as current (linux-image-2.6.38-996-generic_2.6.38-996.201103251543_i386.deb) and its headers (all and i386)... lets see how they do ;)

Is this a specially built kernel with debug enabled? or just some newer patches?

It seems to boot alright, but (as a notebook that I keep my hands on) the machine felt a little warmer ... I wanted to check the temperature (in case the kernel was forcing something odd with the CPUs) but I have found out I am unable to collect temperature with acpi tools now (it used to work just fine, even with landscape-sysinfo, and its also gone)

I can still see the speed step its keeping the processor though, via indicator-cpufreq, and it seemes quiet usual (the occasional oscillation to top speed, and then back to slowest or some middle clock)

The battery information seems coherent too, the detection of power source and the estimated percentage (currently 97%) to discharge when not connected... but it does not show the estimated time to discharge or recharge (as it used to).

Not sure if its kernel related or a incredible coincidence (since it was only the kernel that I have updated now).

By the way, it still feels really warm, almost hot.

[PresuntoRJ (fabio-tleitao)]

Oddly, I get to check the temperatures in Ubuntu-Classic (gnome) via the sensors-applet (some how they can still get temperatures from CPU and HDD) and it is quite hot CPU=63C and HDD=53C, not sure if I can read the GPU on this machine... but I think that is what is heating up the system... the fan is running, not at full throttle (?) and I can cope with the overheat for a while, with an external ventilation forcing air into the notebook ;) but I hope its not the new default state....

[PresuntoRJ (fabio-tleitao)]

Everything else seems pretty ok, so far... and no new freezes yet

[Bryce Harrington (bryce)]

Hmm, alright given the timing my suspicion would be a regression introduced by either the xserver or mesa updates that occurred shortly prior to this might have led to the freeze, but given that it seems to have happened only once, it'll be very hard to narrow in on an actual suspect.

Still, I've received a handful of other i965 gpu freeze regression reports similar to this one. Except for #760054, the error codes vary a bit so they're not perfect dupes. All hard to reproduce or one-time-only, so debugging this lockup could be quite hard.

I think what I'm going to do for now is dupe these bug reports together and forward a bug report upstream on the chance that they recognize the issue.