[i855] GPU hung (drm i915 intel) on Lucid Lynx

[gmud]

Binary package hint: xserver-xorg-video-intel

With the alpha 3 version of the Lucid Lynx live CD I get a GPU hung shortly after the splash screen of the the desktop (KDE) is shown.

the following messages are from dmesg:
[drm:i915_hangcheck_elapsed] * ERROR * Hangcheck timer elapsed... GPU hung
render error detected , EIR: 0x00000000
i915: waking up sleeping processes
reboot required

The loginmanager then tries to respawn the xserver several times until there is a non blinking cursor shown at the botton of the screen.

On some other boot tries with the live cd the xserver freezes completely and it is not possible to switch to another console.

Hardware info:
00:02.0 VGA compatible controller: Intel Corporation 82852/855GM Integrated Graphics Device (rev 02)
        Subsystem: ASUSTeK Computer Inc. Device 1712
        Flags: bus master, fast devsel, latency 0, IRQ 16
        Memory at f0000000 (32-bit, prefetchable) [size=128M]
        Memory at feb80000 (32-bit, non-prefetchable) [size=512K]
        I/O ports at dc00 [size=8]
        Capabilities: [d0] Power Management version 1
        Kernel driver in use: i915
        Kernel modules: i915

00:02.1 Display controller: Intel Corporation 82852/855GM Integrated Graphics Device (rev 02)
        Subsystem: ASUSTeK Computer Inc. Device 1712
        Flags: bus master, fast devsel, latency 0
        Memory at e8000000 (32-bit, prefetchable) [size=128M]
        Memory at fea80000 (32-bit, non-prefetchable) [size=512K]
        Capabilities: [d0] Power Management version 1

[Geir Ove Myhr (gomyhr)]

Thank you for reporting this bug. It's possible that this is the same issue as bug 511001, but we prefer to keep xorg bugs separate until proven otherwise. The Lucid kernel does not have the infrastructure in place to troubleshoot GPU hung bugs properly. The intel-drm-next kernel from [1] (see [2] for more information) has a patch that will dump the error state to /sys/kernel/debug/dri/0/i915_error_state. Could you copy this file and attach to this bug report. To get a consistent set of logs it would be nice if you could copy /var/log/Xorg.0.log and dmesg output at the same time and attach here.

There is some general information about this kind of bug at [3].

[1]: http://kernel.ubuntu.com/~kernel-ppa/mainline/drm-intel-next/
[2]: https://wiki.ubuntu.com/KernelTeam/MainlineBuilds
[3]: https://wiki.ubuntu.com/X/Troubleshooting/Freeze

[gmud (gmud)]

BTW: using "vesa" in xorg.conf and "nomodeset" at the kernel command line seems to be a workaround until the bug is fixed in the intel driver.

[Geir Ove Myhr (gomyhr)]

Originally reported by Gero Mudersbach at:
  https://bugs.launchpad.net/bugs/528467

[Problem]
Another GPU hang bug on 855GM with captured i915_error_state. I am not good enough at reading the output of intel_error_decode to tell if this a cache flushing problem like bug 26746.

[Original report]

Binary package hint: xserver-xorg-video-intel i915_

With the alpha 3 version of the Lucid Lynx live CD I get a GPU hung shortly after the splash screen of the the desktop (KDE) is shown.

the following messages are from dmesg:
[drm:i915_hangcheck_elapsed] * ERROR * Hangcheck timer elapsed... GPU hung
render error detected , EIR: 0x00000000
i915: waking up sleeping processes
reboot required

The loginmanager then tries to respawn the xserver several times until there is a non blinking cursor shown at the botton of the screen.

On some other boot tries with the live cd the xserver freezes completely and it is not possible to switch to another console.

Hardware info:
00:02.0 VGA compatible controller: Intel Corporation 82852/855GM Integrated Graphics Device (rev 02)
        Subsystem: ASUSTeK Computer Inc. Device 1712
        Flags: bus master, fast devsel, latency 0, IRQ 16
        Memory at f0000000 (32-bit, prefetchable) [size=128M]
        Memory at feb80000 (32-bit, non-prefetchable) [size=512K]
        I/O ports at dc00 [size=8]
        Capabilities: [d0] Power Management version 1
        Kernel driver in use: i915
        Kernel modules: i915

00:02.1 Display controller: Intel Corporation 82852/855GM Integrated Graphics Device (rev 02)
        Subsystem: ASUSTeK Computer Inc. Device 1712
        Flags: bus master, fast devsel, latency 0
        Memory at e8000000 (32-bit, prefetchable) [size=128M]
        Memory at fea80000 (32-bit, non-prefetchable) [size=512K]
        Capabilities: [d0] Power Management version 1

BTW: using "vesa" in xorg.conf and "nomodeset" at the kernel command line seems to be a workaround until the bug is fixed in the intel driver.

[Geir Ove Myhr (gomyhr)]

Created an attachment (id=33652)
i915_error_state

[Geir Ove Myhr (gomyhr)]

Created an attachment (id=33653)
dmesg output

[Geir Ove Myhr (gomyhr)]

Created an attachment (id=33654)
Xorg.0.log

[Geir Ove Myhr (gomyhr)]

Assigning to Chris Wilson, since he is into this kind of problems. Let me know if I shouldn't to this.

[Geir Ove Myhr (gomyhr)]

Gero, I have forwarded this bug to the intel developers upstream: https://bugs.freedesktop.org/show_bug.cgi?id=26808 . Please register at bugs.freesktop.org and add yourself to the CC field so that you may receive any questions from the intel developers.

[Geir Ove Myhr (gomyhr)]

One more thing: You write in bug 528491 that using the patches from https://bugs.freedesktop.org/show_bug.cgi?id=24789#c38 helps on Karmic. Have you tested this on Lucid? There is a patch at https://bugs.freedesktop.org/show_bug.cgi?id=26746#c10 which may be doing something similar. Could you try that?

You may add any information about your finding to the upstream bug report directly.

[gmud (gmud)]

Geir, the patch helped delaying the freeze and it was possible to switch to another console. In the long run it didn't prevent the freeze. I think it's still a flushing problem and the "big hammer" patch just makes it happen less often.

Until now it wasn't possible for me to test it with Lucid. I will test both and report back soon.

[Geir Ove Myhr (gomyhr)]

On Tue, Mar 2, 2010 at 10:46 AM, Gero Mudersbach wrote:
> Geir, the patch helped delaying the freeze and it was possible to switch
> to another console. In the long run it didn't prevent the freeze. I
> think it's still a flushing problem and the "big hammer" patch just
> makes it happen less often.

Yes, that is possible. Another possibility is that there is more than
one bug causing the GPU hangs and that the "big hammer" patch prevents
the flushing problems. If that is the case, troubleshooting with the
patch may help uncover the other problem.

> Until now it wasn't possible for me to test it with Lucid. I will test
> both and report back soon.

Thank you!

[Chris Wilson (ickle)]

It is not an obvious cache-flushing bug like the others, IPEHR matches the previous instruction. Nothing stands out as being the cause, if you can grab more i915_error_states that may help establish a pattern.

[Geir Ove Myhr (gomyhr)]

Gero, upstream has been waiting for more dumps for a while now in order to see if they can find a pattern. Can you provide that?

[Noobie1 (noobie1kenobie)]

I get these i915 GPU hung messages too with Lucid Beta 1 on a i845G card, but not sure if it's the same bug.
The last time was a complete freeze which necessitated a hard shut down.

I don't know that much but am willing to help test if someone lists what I need to do.

[Noobie1 (noobie1kenobie)]

Should have added that I can boot only with options "quiet" and "splash" off, and crash seems random.

Can we edit posts?

[Geir Ove Myhr (gomyhr)]

Noobie1, for the 845G GPU hangs look at bug 541492. There is a potential fix there.

No, as far as I know there is no way to edit comments.

[Chris Halse Rogers (raof)]

We're considering blacklisting KMS on this card to work around bug #511001, which I believe will probably work around this bug too. It would be useful if people who are encountering this problem could try booting with KMS disabled (by adding “i915.modeset=0” after “quiet splash” in the kernel boot line).

If there seems to be a consensus favoring blacklisting KMS, then we'll reassign this bug to the kernel and have andy blacklist for us.

If anyone finds that their i855 system works *worse* with KMS turned off for some reason, please shout; I wouldn't expect turning KMS to cause regressions so this would be a surprise, but with i8xx weirder things have been known to happen!

[Peter Husen (phusen)]

I for one find that my system works worse with KMS turned off. Whenever the system tries to start X, it comletely freezes with a black screen (so it was quite a hassle to install Lucid using the desktop iso). Furthermore, the vesa driver doesn't even work for me - it finds no usable modes by default and by some manual configuration (HorizSync and VertRefresh) I can get at most 1024x768 with 8 bit color or 640x480 with 16 bit color, apparently because it detects only 832 kb of memory, and the VideoRAM option seems to be ignored. When I set i915.modeset=1 (overriding the blacklisting) everything is fine with the intel driver - accelerated and everything. I haven't tested it for a long time yet, though. My system is a Dell Inspiron 510m.

[pertain (pertain-deactivatedaccount)]

I can reproduce this bug reliably with Ubuntu Lucid and a recent Debian Squeeze/SID on a Fujitsu-Siemens Lifebook E8010D with Intel 855GM Chipset. The problem occurs at random when browsing with firefox but X crashes for sure when I launch xine or mplayer (as soon as I start playing).

Here are two samples from the kernel ringbuffer:

[18907.328045] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[18907.328065] render error detected, EIR: 0x00000000
[18907.328100] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 2167633 at 2167616)

[14534.096050] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[14534.096068] render error detected, EIR: 0x00000000
[14534.096435] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 933567 at 933566)

Please let me know if you need further information!
Thank you.

[Łukasz Pińkowski (pinky-ok)]

I reproduced this bug on my toshiba portege m100 laptop on newest zenwalk 6.4 which is based on slackware so it's not only on ubuntu/debian systems

[Łukasz Pińkowski (pinky-ok)]

Try this :https://launchpad.net/~glasen/+archive/855gm-fix, it works for me

[Bob+freedesktop (bob+freedesktop)]

There is also some display corruption in the upper right, and in some instances the display is completely corrupted. Here is another i915_error_state from Ubuntu 10.04.

Time: 1274739074 s 997040 us
EIR: 0x00000000
  PGTBL_ER: 0x00000000
  INSTPM: 0x00000000
  IPEIR: 0x00000000
  IPEHR: 0xff17111b
  INSTDONE: 0xffe5fafd
  ACTHD: 0x00002e14
  INSTPS: 0x8011e032
  INSTDONE1: 0x000fffff

In one instance where X did not crash immediately, the Xorg.0.log says:

(WW) intel(0): i830_uxa_prepare_access: bo map failed

I'm downgrading to 9.10 now. This is unusable.

[gene (eugenios)]

Hi, ran into a similar problem: My desktop display froze and both keyboard and mouse did not work. I haven't checked if my webserver was running/available, since I prematurely powered off the pc.
My issue seems to be a little different from those reported though. The freeze has never occurred before since my upgrade to Lucid more than a week ago. The kernel
uname -a: Linux 2.6.32-22-generic #36-Ubuntu SMP Thu Jun 3 19:31:57 UTC 2010 x86_64 GNU/Linux
has been run for about a couple days before it happened.
I will continue using this very kernel (I do not want to try the ppa fix yet, since this intel machine used to be very effective before). After a next freeze I will try to grab some earlier/later versions of the kernel from ppa/kernel
Here are some specs:
sudo lshw:

 description: Mini Tower Computer
    product: Dell DM051
    vendor: Dell Inc.
    serial: B4KPL81
    width: 64 bits
    capabilities: smbios-2.3 dmi-2.3 vsyscall64 vsyscall32
..........................
description: VGA compatible controller
             product: 82945G/GZ Integrated Graphics Controller
             vendor: Intel Corporation
             physical id: 2
             bus info: pci@0000:00:02.0
             version: 02
             width: 32 bits
             clock: 33MHz
             capabilities: msi pm bus_master cap_list rom
             configuration: driver=i915 latency=0
             resources: irq:16 memory:eff00000-eff7ffff ioport:ecd8(size=8) memo
ry:d0000000-dfffffff(prefetchable) memory:efec0000-efefffff
And
lspci
00:00.0 Host bridge: Intel Corporation 82945G/GZ/P/PL Memory Controller Hub (rev 02)
00:01.0 PCI bridge: Intel Corporation 82945G/GZ/P/PL PCI Express Root Port (rev 02)
00:02.0 VGA compatible controller: Intel Corporation 82945G/GZ Integrated Graphics Controller (rev 02)
00:02.1 Display controller: Intel Corporation 82945G/GZ Integrated Graphics Controller (rev 02)
00:1b.0 Audio device: Intel Corporation N10/ICH 7 Family High Definition Audio Controller (rev 01)
00:1c.0 PCI bridge: Intel Corporation N10/ICH 7 Family PCI Express Port 1 (rev 01)
00:1d.0 USB Controller: Intel Corporation N10/ICH7 Family USB UHCI Controller #1 (rev 01)
00:1d.1 USB Controller: Intel Corporation N10/ICH 7 Family USB UHCI Controller #2 (rev 01)
00:1d.2 USB Controller: Intel Corporation N10/ICH 7 Family USB UHCI Controller #3 (rev 01)
00:1d.3 USB Controller: Intel Corporation N10/ICH 7 Family USB UHCI Controller #4 (rev 01)
00:1d.7 USB Controller: Intel Corporation N10/ICH 7 Family USB2 EHCI Controller (rev 01)
00:1e.0 PCI bridge: Intel Corporation 82801 PCI Bridge (rev e1)
00:1f.0 ISA bridge: Intel Corporation 82801GB/GR (ICH7 Family) LPC Interface Bridge (rev 01)
00:1f.1 IDE interface: Intel Corporation 82801G (ICH7 Family) IDE Controller (rev 01)
00:1f.2 IDE interface: Intel Corporation N10/ICH7 Family SATA IDE Controller (rev 01)
00:1f.3 SMBus: Intel Corporation N10/ICH 7 Family SMBus Controller (rev 01)
03:03.0 Modem: Intel Corporation FA82537EP 56K V.92 Data/Fax Modem PCI (rev 04)
03:08.0 Ethernet controller: Intel Corporation N10/ICH 7 Family LAN Controller (rev 01)

[gene (eugenios)]

There's one more detail that may be worth mentioning. My freeze happened right after the power-manager kicked in to out the monitor to sleep. So accordin to some comments from archlinux http://bugs.archlinux.org/task/17705 it is not a mere coincidence.

[gene (eugenios)]

I am running two instances of mplayer with -vo xvid (which was not possible before kms on i915).
VIDEO: [VP6F] 480x360 0bpp 25.000 fps 512.0 kbps (62.5 kbyte/s)

 The desktop-effects are on and work flawlessly with the video windows. The performance is awesome, like on an nVidea card! No freezes so far.

[gene (eugenios)]

I also tried
xset dpms force off
It still does not make the machine hang

[gene (eugenios)]

This get more interesting. As a matter of fact I experience an xserver crash. The kernel seems to be intact. After having run for 4 days
$uptime
 01:48:36 up 4 days, 13:21, 4 users, load average: 0.17, 0.46, 0.88
I get an X crash and a window in a minute saying that a low graphic session can only be run with different options none of which bring me back to drm/dri capable Xserver.
This probably indicates that I have to file a separate bug? However I get a similar entry in the kern.log

Jun 13 01:21:22 domus kernel: [392077.221266] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Jun 13 01:21:22 domus kernel: [392077.221276] render error detected, EIR: 0x00000000
Jun 13 01:21:22 domus kernel: [392077.221293] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 18764775 at 18764774)

Will get back tomorrow

[gene (eugenios)]

Along with the kern.log messages I see the following in my Xorg.[012].log:

Xorg.[02.]log

(==) intel(0): Silken mouse enabled
(II) intel(0): Initializing HW Cursor
(II) intel(0): No memory allocations

Xorg.1.log:

(EE) FBDEV(0): FBIOPUTCMAP: Invalid argument

Last message floods the log file I get
grep '(EE) FBDEV' /var/log/Xorg.1.log | wc -l
144356
!!!!!!!!

[gene (eugenios)]

I have probably to file a new bug report, however, after having done this many times - this is far from being an automated process and I do not want to fiddle with https://launchpad.net/~glasen/+archive/855gm-fix, so
I will be testing the latest kernel from kernel-ppa instead , i.e.,

uname -a
Linux domus 2.6.34-020634-generic #020634 SMP Mon May 17 19:27:49 UTC 2010 x86_64 GNU/Linux

[gene (eugenios)]

https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/593463

[lirel (m8r-lcodw5)]

marked #593643 as duplicate because of same behaviour.

[lirel (m8r-lcodw5)]

this still happens with i915.modeset=0 when launching the antscreensaver.

[lirel (m8r-lcodw5)]

doesn't work better nor worse with "i915.modeset=0".

[Christian (christian-bley2)]

when i begin to play "Alien Arena" this error happens:

my kernel.log sais:
"kernel: [115407.680041] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
kernel: [115407.680055] render error detected, EIR: 0x00000000
 kernel: [115407.680091] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 16246926 at 16246920)"

uname -a:
"Linux xxxxxxx@xxxx-xxxx 2.6.32-22-generic #36-Ubuntu SMP Thu Jun 3 19:31:57 UTC 2010 x86_64 GNU/Linux"

lsb_release -a:
"No LSB modules are available.
Distributor ID: Ubuntu
Description: Ubuntu 10.04 LTS
Release: 10.04
Codename: lucid"

[Geir Ove Myhr (gomyhr)]

Christian, we're all better off if you report your problem separately instead of adding it as a comment on another bug report. Comments on a bug saying essentially "I have this problem..." are at best ignored and at worst very confusing if we start mixing up the commenters problem with the problem of the original reporter (and the difference may be very subtle).

I don't mean to be harsh, but we only focus on the problem of the original reporter in each bug report and we ask that all comments also focus on that problem. For xorg-related bugs we much prefer duplicate bug reports over "me too" comments, since issues that look the same often turn out to be different underlying problems.

[Chris Wilson (ickle)]

Gah, mishandling of tiling. We told the GPU that we were rendering to a tiled buffer, but did not allocate sufficient space for one. Old bug.

[David Sauvage (pariakanet)]

I had the same bug using latest Ubuntu Lucid (i386).

I fix the problem after installing latest maverick kernel from ppa:kernel-ppa/ppa
(linux-image-generic-lts-backport-maverick 2.6.35-12-generic)

sudo add-apt-repository ppa:kernel-ppa/ppa
sudo aptitude update
sudo aptitude install linux-image-generic-lts-backport-maverick

[David Sauvage (pariakanet)]

I spoke too quickly, the problem still happen with kernel 2.6.35-12,

sorry for that

[Chris Halse Rogers (raof)]

Ok! Ladies and Gentlemen! Chris Wilson of upstream fame has done some work to re-integrate a legacy driver for the Intel cards that have been hard done by in the GEM transition.

The xserver-xorg-video-intel packages in https://edge.launchpad.net/~raof/+archive/aubergine have a GEM-less legacy driver re-integrated which is activated when KMS is disabled - which it is by default on your card for Lucid.

It would be useful if you could remove any work-arounds you've used to get your system more stable and test the drivers from this PPA. They should hopefully end up slightly more stable than the drivers in Ubuntu 9.04 (Jaunty), which were the last set of drivers to not use the GEM memory manager, and so significantly more stable than the drivers in Ubuntu 9.10 and Ubuntu 10.04 (Karmic and Maverick).

Could you please test the drivers from this PPA and report your experiences with them?

[tplus (tripleplussolutionz)]

@ Chris : Installed the package from ppa now. The display transition has become slow. Also the graphics response is slow. For instance played chess on chess.com (in the browser. would suggest to have a look at the graphics scheme there). Could not move the pieces smoothly by dragging them. No hung gpu till now in 20 minutes

sudo lshw:

picodtp
    description: Desktop Computer
    product: PROD00000000
    vendor: OEM00000
    width: 32 bits
    capabilities: smbios-2.2 dmi-2.2 smp-1.4 smp
    configuration: boot=normal chassis=desktop cpus=1
  *-core
       description: Motherboard
       product: 845GL-W83637F
       physical id: 0
     *-firmware
          description: BIOS
          vendor: Phoenix Technologies, LTD
          physical id: 0
          version: 6.00 PG (03/25/2004)
          size: 128KiB
          capacity: 448KiB
          capabilities: isa pci pnp apm upgrade shadowing cdboot bootselect socketedrom edd int13floppy360 int13floppy1200 int13floppy720 int13floppy2880 int5printscreen int9keyboard int14serial int17printer int10video acpi usb agp ls120boot zipboot
     *-cpu
          description: CPU
          product: Intel(R) Pentium(R) 4 CPU 2.40GHz
          vendor: Intel Corp.
          physical id: 4
          bus info: cpu@0
          version: 15.3.3
          serial: 0000-0F33-0000-0000-0000-0000
          slot: Socket 478
          size: 2400MHz
          capacity: 3066MHz
          width: 32 bits
          clock: 133MHz
          capabilities: boot fpu fpu_exception wp vme de pse tsc msr pae mce cx8 apic mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe constant_tsc up pebs bts pni dtes64 monitor ds_cpl cid
          configuration: id=0
        *-cache:0
             description: L1 cache
             physical id: 8
             slot: Internal Cache
             size: 32KiB
             capacity: 32KiB
             capabilities: synchronous internal write-back
        *-cache:1
             description: L2 cache
             physical id: 9
             slot: External Cache
             size: 1MiB
             capacity: 1MiB
             capabilities: synchronous external write-back
     *-memory
          description: System Memory
          physical id: 1b
          slot: System board or motherboard
          size: 1GiB
          capacity: 2GiB
        *-bank:0
             description: DIMM SDRAM Synchronous
             physical id: 0
             slot: A0
             size: 512MiB
             width: 64 bits
        *-bank:1
             description: DIMM SDRAM Synchronous
             physical id: 1
             slot: A1
             size: 512MiB
             width: 64 bits
     *-pci
          description: Host bridge
          product: 82845G/GL[Brookdale-G]/GE/PE DRAM Controller/Host-Hub Interface
          vendor: Intel Corporation
          physical id: 100
          bus info: pci@0000:00:00.0
          version: 03
          width: 32 bits
          clock: 33MHz
          configuration: driver=agpgart-intel
          resources: irq:0 memory:d0000000-d7ffffff(prefetchable)
        *-display UNCLAIMED
             description: VGA compatible controller
           ...

[Warren E. Downs (vwdowns)]

The goal of a re-integrated legacy driver for Intel cards is great. Unfortunately, so far it doesn't work with my card (lspci output):
00:02.0 VGA compatible controller: Intel Corporation Mobile 915GM/GMS/910GML Express Graphics Controller (rev 03)
00:02.1 Display controller: Intel Corporation Mobile 915GM/GMS/910GML Express Graphics Controller (rev 03)
With KMS remaining enabled (the default for my card and kernel 2.6.32-23-generic #37-Ubuntu SMP) the system appears to work but X crashes quite easily e.g. when un-minimizing the Google Chrome browser, or starting synaptic.

Disabling KMS seemed like the way to go. When I did that, it did revert to behavior I remembered from Kubuntu 9.04 or 9.10 (strange flickering upon startup of KDE). But the computer hard-locked X during the KDE startup, before the desktop appeared. It is not actually freezing the computer, as if I switch away from X before it locks, I can login at the console. (Which I did in order to recover).

So now, I am back to the original default Kubuntu 10.04 xorg, which works for me for days at a time, but then will hard-freeze.

Warren

[b52 (b52-entrap)]

My system crashes every time the monitor goes in standby. I am using Xubuntu and tried to deactivate any power saving options without success. This heavy bug makes me rebooting every day *g*

Kernel 2.6.32-24-generic #38-Ubuntu SMP Mon Jul 5 09:22:14 UTC 2010 i686 GNU/Linux
Distributor ID: Ubuntu
Description: Ubuntu 10.04.1 LTS
Release: 10.04
Codename: lucid

Is anyone working on this? Or is there a workaround available? Thanks

[gene (eugenios)]

Hi b52. I guess that using kernels with versions 34 and higher should fix this. You can get it from ppa kernel.ubuntu.com/~kernel-ppa/mainline/ Try grabbing 35*, it is stable now. This problem does not happen to me since I installed a 34 version:
uname -a
Linux 2.6.34-020634-generic #020634 SMP Mon May 17 19:27:49 UTC 2010 x86_64 GNU/Linux. My uptimes are long now, occasionally I have to reboot because of some power outages. Hope it helps
I did turn off screensaver/monitor sleep and use "xset dpms 300 300 300" instead in the beginning of every X session

--
Please do not send me Microsoft Office/Apple iWork documents. Send OpenDocument instead! http://fsf.org/campaigns/opendocument

[gene (eugenios)]

You all guys seem to be using 2.6.32* kernels. Can you try upgrading your kernel to, say, 2.6.34* It helped me, although, the issue sounds a little different (#593463)

[Stenten (stenten)]

>
> Is anyone working on this? Or is there a workaround available? Thanks
>

See Comment #36.

[Cliff Wells (cliff-develix)]

I installed the legacy driver mentioned in #36 last night and not only did it appear to solve my issue, I don't really notice any performance degradation (no games, medium-high compiz settings).

Thanks.

[Warren E. Downs (vwdowns)]

Regarding ppa kernel.ubuntu.com/~kernel-ppa/mainline/, I tried adding the source but it wasn't recognized. I noticed that there is no 'dists' directory under it like apt expects. Is there another source or do I have to install the kernel manually (which I can do but I hate the maintenance headache when I update packages later).

[Geir Ove Myhr (gomyhr)]

On Mon, Aug 9, 2010 at 1:59 AM, Warren E. Downs wrote:
> Regarding ppa kernel.ubuntu.com/~kernel-ppa/mainline/, I tried adding
> the source but it wasn't recognized.  I noticed that there is no 'dists'
> directory under it like apt expects.  Is there another source or do I
> have to install the kernel manually (which I can do but I hate the
> maintenance headache when I update packages later).

kernel.ubuntu.com/~kernel-ppa/ is not a normal PPA, so you cannot add
it to /etc/apt/sources or sources.d. You have to install them
manually, but since the kernels install in parallel with the other
kernels you have, there is not really any maintenance headache. See
https://wiki.ubuntu.com/Kernel/MainlineBuilds .

[Kamil Páral (kamil.paral)]

I have 82852/855GM and my system crashed everytime I tried to play video in Totem. It also didn't want to boot, but enabling i915.modeset helped with that.

After installing driver update from comment #36 and disabling modeset workaround the system won't boot again. When enabling the modeset workaround, I still can't play videos in Totem, but this time Totem crashes, not the whole system. That's an improvement. But screen redrawing seems slower overall.

[b52 (b52-entrap)]

Ok, this issue gets really really annoying..

As suggested in comment #36 I installed the drivers from https://edge.launchpad.net/~raof/+archive/aubergine. => xserver didn't start anymore. I tried to rollback and was not able to remove this ppa without destroying the while system. ppa-purge did not work correctly here!

After cleaning and reainstalling this packages manually I tried to install this ppa kernels.

2.6.32-23-generic didn't work, as I knew, the X-Server dies.
2.6.35-020635rc1 doesn't boot at all. I did not even find anything in the logs.
2.6.34-020634rc7 is booting successfully, but as in 2.6.32-23 after some while the drm dies and the whole system freezes!
2.6.34-020634 is booting successfully, but the same here, if drm dies after some time, the whole system freezes.

So I am back to 2.6.32-23-generic in which I am at least able to login remotely if the X dies *g*

Any more suggestions, solutions or workarounds here? Please? .... or do I have to switch back to karmic?

Thanks.

[Jonathan (jonathan00)]

Hi b52,

right now, a new kernel-packages 2.6.32-24.41 was installed on my system, but since i installed glasen's package from #20 some month ago, i had no problems anymore, mplayer, xine and vlc works fine, in kaffeine i have only sound, no video, but that doesn't matter...

[mattlach (mattlach)]

I also have this problem. It appears to randomly occur when I am away from the computer. It doesn't occur every time the display goes to sleep.

It seems to happen more frequently since the latest kernel update I just installed.

This is on my Dell Latitude E4310 using integrated graphics on a Core i5 m520.

Is upstream still looking for dumps? If so, how can I help?

[mattlach (mattlach)]

Oh, and here is my lspci|grep -i vga

VGA compatible controller: Intel Corporation Core Processor Integrated Graphics Controller (rev 02)

[mattlach (mattlach)]

Definitely much bigger of the problem since the last kernel update. It's happening all the time now.

I'm not quite sure how to display a log of recent package updates, or I would show it here.

I am running AMD64, btw.

[Chris Halse Rogers (raof)]

On Mon, 2010-08-23 at 04:27 +0000, mattlach wrote:
> I also have this problem. It appears to randomly occur when I am away
> from the computer. It doesn't occur every time the display goes to
> sleep.

You're definitely seeing a different bug, since…

>
> It seems to happen more frequently since the latest kernel update I just
> installed.
>
> This is on my Dell Latitude E4310 using integrated graphics on a Core i5
> m520.

…your Core i5 is more than 10 years newer than the i855 cards this bug
applies to.

Feel free to file a new bug, particularly if apport is offering to
submit a GPU hang for you.

[ilia (ilia)]

freedesktop bug 26808 is marked as resolved now. Is this going to be fixed in Lucid?

[juky (juraj-belina)]

Installed customized Ubuntu image from Stephen Glasenhardt (http://files.yoschi.cc/glasen/ubuntu-10.04.1-855gm-desktop-i386.iso) with KMS enabled and updated Intel drivers. However, whenever some video playback is started (In VLC, webpage,etc..), the system freezes, and only hard reset helps.

In syslog, I can find the following: [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung

Please let me know, if I can output some additional information.

[Duy Truong (jimreynold2nd)]

I'm confirming the problem still exists with Maverick x64 kernel 2.6.35-22 running KDE4. System is Thinkpad T400 using Intel 4500. Sometimes a hard reset is the only way, sometimes I was able to switch to a VT and get those message around (but not exactly) every second (based on the timestamps):

[4200.123456] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[4200.789012] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 264334 at 263373)

The number "263373" stays constant, while the number before it ("264334") increments by 2 everytime a message is printed.

Most of the time this happen when some desktop effect is used. I also noticed that most of the time the effect was "Present Windows", which is also buggy (i.e. window thumbnails disappear after around a second)

[Alvin (alvind)]

Ubuntu 10.04 with Intel 82845G/GL is unusable
Only X with rdesktop (from .Xsession) is started, but crashes several times a day.

[drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
render error detected, EIR: 0x00000000
[drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 10621 at 10620)
[drm:i915_gem_idle] *ERROR* hardware wedged

[tplus (tripleplussolutionz)]

i confirm the bug again. Its not resolved. I find a similarity in most
of the confimations - the bugs are all mostly on 845g/gl brookdale
architecture. my machine abruptly hangs every 2-3 hours.

On 10/21/10, Alvin <email address hidden> wrote:
> Ubuntu 10.04 with Intel 82845G/GL is unusable
> Only X with rdesktop (from .Xsession) is started, but crashes several times
> a day.
>
> [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
> render error detected, EIR: 0x00000000
> [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting
> 10621 at 10620)
> [drm:i915_gem_idle] *ERROR* hardware wedged
>
> --
> [i855] GPU hung (drm i915 intel) on Lucid Lynx
> https://bugs.launchpad.net/bugs/528467
> You received this bug notification because you are a direct subscriber
> of a duplicate bug (593643).
>
> Status in X.org xf86-video-intel: Fix Released
> Status in “xserver-xorg-video-intel” package in Ubuntu: Confirmed
>
> Bug description:
> Binary package hint: xserver-xorg-video-intel
>
> With the alpha 3 version of the Lucid Lynx live CD I get a GPU hung shortly
> after the splash screen of the the desktop (KDE) is shown.
>
> the following messages are from dmesg:
> [drm:i915_hangcheck_elapsed] * ERROR * Hangcheck timer elapsed... GPU hung
> render error detected , EIR: 0x00000000
> i915: waking up sleeping processes
> reboot required
>
> The loginmanager then tries to respawn the xserver several times until there
> is a non blinking cursor shown at the botton of the screen.
>
> On some other boot tries with the live cd the xserver freezes completely and
> it is not possible to switch to another console.
>
> Hardware info:
> 00:02.0 VGA compatible controller: Intel Corporation 82852/855GM Integrated
> Graphics Device (rev 02)
> Subsystem: ASUSTeK Computer Inc. Device 1712
>
> Flags: bus master, fast devsel, latency 0, IRQ 16
>
> Memory at f0000000 (32-bit, prefetchable) [size=128M]
>
> Memory at feb80000 (32-bit, non-prefetchable) [size=512K]
>
> I/O ports at dc00 [size=8]
>
> Capabilities: [d0] Power Management version 1
>
> Kernel driver in use: i915
>
> Kernel modules: i915
>
>
> 00:02.1 Display controller: Intel Corporation 82852/855GM Integrated
> Graphics Device (rev 02)
> Subsystem: ASUSTeK Computer Inc. Device 1712
>
> Flags: bus master, fast devsel, latency 0
>
> Memory at e8000000 (32-bit, prefetchable) [size=128M]
>
> Memory at fea80000 (32-bit, non-prefetchable) [size=512K]
>
> Capabilities: [d0] Power Management version 1
>
> To unsubscribe from this bug, go to:
> https://bugs.launchpad.net/xserver-xorg-video-intel/+bug/528467/+subscribe
>

[Luigi Capriotti (l.capriotti)]

Running on updated maverick, experiencing the same issue when using desktop effects in KDE (4.5):

lspci | grep -i vga:

00:02.0 VGA compatible controller: Intel Corporation 4 Series Chipset Integrated Graphics Controller (rev 03) (prog-if 00 [VGA controller])

uname -a
Linux test-desktop 2.6.35-22-generic #35-Ubuntu SMP Sat Oct 16 20:36:48 UTC 2010 i686 GNU/Linux

syslog entries:

syslog.1:Oct 20 11:48:03 test-desktop kernel: [10180.024091] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 1152263 at 1149353)
syslog.1:Oct 20 11:48:03 test-desktop kernel: [10180.388073] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung

[Alvin (alvind)]

Is this bug related to Desktop Environments, special effects, sleep or running certain applications? Just running plain X is enough to crash X after a while for me.

[Josh Knight (jgknight)]

I'm also getting this error. Running Ubuntu Maverick 10.10, 2.6.35-22 gnome 2.32.0 on a toshiba satellite a505-6005 with intel i3. I thought it was 3d screensavers causing the problem, powersave mode, something related. But I've switched all of these off and I still get the error. It usually happens sometime after I lock my screen to go someplace. It's also happened once while I was at the computer.

[emvy (mate-varga)]

Same here. i5-540, Kubuntu Maverick 10.10;
Linux XXX 2.6.35-22-generic #35-Ubuntu SMP Sat Oct 16 20:45:36 UTC 2010 x86_64 GNU/Linux

[Andrey Levkovich (levkovich-a)]

uname -a
Linux artem 2.6.32-24-generic #39-Ubuntu SMP Wed Jul 28 06:07:29 UTC 2010 i686 GNU/Linux

lsb_release -a
No LSB modules are available.
Distributor ID: Ubuntu
Description: Ubuntu 10.04.1 LTS
Release: 10.04
Codename: lucid

same error:
Nov 29 13:30:30 artem kernel: [16347.687400] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Nov 29 13:30:30 artem kernel: [16347.687408] render error detected, EIR: 0x00000000
Nov 29 13:30:30 artem kernel: [16347.687528] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 4558966 at 4558931)
Nov 29 13:30:31 artem kernel: [16348.207279] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Nov 29 13:30:31 artem kernel: [16348.207288] render error detected, EIR: 0x00000000
Nov 29 13:30:31 artem kernel: [16348.207324] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 4558967 at 4558931)

[Karoly Negyesi (karoly)]

I can reproduce this as reported http://code.google.com/p/chromium/issues/detail?id=67177 here. Just use Google Chrome and open bodybrowser.googlelabs.com

[Johan Vervloet (jvervloet)]

A workaround seems to be disabling 'kernel mode setting'. (https://wiki.ubuntu.com/X/KernelModeSetting)

Since I am a clueless user ;-), I am not sure how to do this. I tried these two things:

1. execute this command (as root):

echo options i915 modeset=0 > /etc/modprobe.d/i915-kms.conf

2. edit /etc/default/grub; change the line starting with GRUB_CMDLINE_LINUX_DEFAULT to

GRUB_CMDLINE_LINUX_DEFAULT="splash quiet nomodeset"

and then: update-grub

[Johan Vervloet (jvervloet)]

I forgot to mention that I didn't experience the problem any more after doing the things I described in #72.

[gene (eugenios)]

Hi all,
I have a question, did you guys test 2.6.36* kernels yet (2.6.37 has other issues)? I used to have "[drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung" in my logs too. However, the problem was not as severe as yours, gnome-session would crash, X would just drop dri/drm capabilities and vesa appears, bug #593463
I am not seeing this message in the logs anymore and experience no crashed, although I keep power-manager disabled and use dpms with "xset dpms 300 300 300"
uname -a:
Linux domus 2.6.36.1-mine #1 SMP Sat Dec 4 12:56:31 CST 2010 x86_64 GNU/Linux
(self-built kernel, you can just grab a similar one from here http://kernel.ubuntu.com/~kernel-ppa/mainline/ )

[gene (eugenios)]

Forgot to tell that on the recent kernels *3[67] I also see a short initial hang when the gdm window comes up. It takes 30 or so seconds before the pc becomes 100% usable again.

[Johan Vervloet (jvervloet)]

OK, foret what I told in #72 and #73. The problem persists.

[Chris Halse Rogers (raof)]

I'm marking this as fix released - upstream has identified that this was caused by an old libdrm bug.
You will still, of course, hit the GTT incoherency crashes (bug #541511), which are a separate bug.