8086:2a42 [GM45] GPU Hang when enabling KDE desktop effects

I have exactly the same GPU, and I don't see this hang. Could you please try disabling effect plugins to find the culprit?

Also, the /sys/kernel/debug/dri/0/i915_error_state file should contain useful data after the crash; could you please attach that file?

ulysses@locris:~$ cat /sys/kernel/debug/dri/0/i915_error_state
no error state collected

I disabled the desktop effects, now try to reproduce the error.

You'll have to collect i915_error_state after the GPU has locked up - you should be able to switch VT and grab it from the console. Alternatively, apport _should_ catch that and offer to report a bug for you (it'll appear to be from apport-gpu-error-intel).

I am running Ubuntu 10.04.1 LTS (netbook edition) on a Samsung N220 plus with Gnome 2.30.2, Xorg 1:7.5+5ubuntu1
I saw the same behavior (flickering display, xserver stopping, unable to restart X. reboot necessary) a couple of times in the last weeks, but could pin it down to any single action.
/sys/kernel/debug/dri/0/i915_error_state was not holding any information (rebooted in the mean time)

Cheers
Malte

(posting my first bug, don't be too harsh on me for missing or imprecise information =)

syslog excerpt:
kernel: [ 5594.224067] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
kernel: [ 5594.224089] render error detected, EIR: 0x00000000
kernel: [ 5594.224148] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 437467 at 437459)
NetworkManager: <info> (eth1): device state change: 8 -> 3 (reason 38)
NetworkManager: <info> (eth1): deactivating device (reason: 38).
NetworkManager: <info> (eth1): canceled DHCP transaction, dhcp client pid 3242
NetworkManager: <WARN> check_one_route(): (eth1) error -34 returned from rtnl_route_del(): Sucess#012
avahi-daemon[739]: Withdrawing address record for 192.168.1.103 on eth1.
avahi-daemon[739]: Leaving mDNS multicast group on interface eth1.IPv4 with address 192.168.1.103.
avahi-daemon[739]: Interface eth1.IPv4 no longer relevant for mDNS.
wpa_supplicant[858]: CTRL-EVENT-DISCONNECTED - Disconnect event - remove keys
kernel: [ 5598.971112] [drm] Big FIFO is enabled
acpid: client 951[0:0] has disconnected
acpid: client connected from 3398[0:0]
acpid: 1 client rule loaded
kernel: [ 5602.563540] [drm] Big FIFO is enabled
kernel: [ 5602.906838] [drm] Big FIFO is enabled
acpid: client 3398[0:0] has disconnected
acpid: client connected from 3415[0:0]
acpid: 1 client rule loaded
kernel: [ 5603.162410] [drm] Big FIFO is enabled
kernel: [ 5603.479334] [drm] Big FIFO is enabled
gdm-binary[750]: WARNING: GdmDisplay: display lasted 0,571364 seconds
acpid: client 3415[0:0] has disconnected
acpid: client connected from 3422[0:0]
acpid: 1 client rule loaded
kernel: [ 5603.740346] [drm] Big FIFO is enabled
kernel: [ 5604.059952] [drm] Big FIFO is enabled
gdm-binary[750]: WARNING: GdmDisplay: display lasted 0,578838 seconds
(repeated multiple times, only reboot helps)

Here is the debug output you requested.

Just upgraded from Kubuntu 10.04.1 to Kubuntu 10.10, and the bug is still present. The system freezes so hard that I can't grab any error log, I have to shutdown with the power button.

Confirmed on Kubuntu 10.10 w/ KDE 4.5.1, ThinkPad T400.

I also get this problem on Kubuntu 10.10 w/ KDE 4.5.1, Lenovo ThinkPad X301 2776-C1G 8GB

Mouse locks up on login. Caps lock light no longer works, although I believe the system is still alive at some level, as when I press the power button the hard drive light comes back on for a while, at least before I give up waiting and hold it down again for 8 seconds.

Workaround was to remove Desktop Effects by booting into single user mode, editing ~/.kde/share/config/kwinrc and commenting out the following lines:

[Plugins]
kwin4_effect_blurEnabled=true
kwin4_effect_coverswitchEnabled=true
kwin4_effect_desktopgridEnabled=true
kwin4_effect_dialogparentEnabled=true
kwin4_effect_explosionEnabled=true
kwin4_effect_fadeEnabled=true
kwin4_effect_loginEnabled=true
kwin4_effect_logoutEnabled=true
kwin4_effect_magiclampEnabled=true
kwin4_effect_maketransparentEnabled=true
kwin4_effect_minimizeanimationEnabled=true
kwin4_effect_presentwindowsEnabled=true
kwin4_effect_scaleinEnabled=true
kwin4_effect_shadowEnabled=true
kwin4_effect_slideEnabled=true
kwin4_effect_taskbarthumbnailEnabled=true
kwin4_effect_thumbnailasideEnabled=true
kwin4_effect_wobblywindowsEnabled=true

System works normally after that, so perhaps one of the above plugins are the cause.

I have the same problem here with :
00:02.1 Display controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 03)
        Subsystem: AOPEN Inc. Device 062d
        Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
        Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
        Latency: 0
        Region 0: Memory at fde00000 (64-bit, non-prefetchable) [size=1M]
        Capabilities: [d0] Power Management version 3
                Flags: PMEClk- DSI+ D1- D2- AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-)
                Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-
                Bridge: PM- B3+

There is no total lock but after a while the desktop responds less and less rapidly.

I have this problem with my Fujitsu Lifebook S760 running Kubuntu 10.10.

I tried Puke's fix of disabling kde desktop effects, and it reduced the frequency of the problem but did not fix it.

Same here, I confirm that the GPU hang can also happen with desktop effects disabled (although way less often).

I see this on Lenovo W500 as well when using the intel graphics chip (W500 has both integrated intel chipset and discrete ATI graphics).

So, a little summary of the bug:
 * It doesn't matter that you enable or disable the desktop effects
 * If you upgrade from Lucid, you get this bug
 * Newly installed system is affected also
 * It seems that the bug affects only integrated Intel VGA chip sets
There are two choices now:
 * Using an older release
 * Switching to an other distribution
What's the problem with these alternatives?
 * There are two supported release of Kubuntu now: 9.10 Karmic Koala and 10.04 Lucid Lynx. Both contains „old” KDE release. As an upstream KDE translator it is important me to have an up to date KDE (and I like fresh packages:)
 * I am a Kubuntu Member, I want to use my favorite distribution

If you need any help to fix the bug, just tell me.

I tried the live session from CD with the „nomodeset” option, the result was bad:
 * Choosing „Try Kubuntu” dropped me to the tty2 console, „sudo service kdm restart” dropped me to tty7, I see the cursor but nothing else, so I can't test this workaround
 * Choosing „Install Kubuntu” started the installer
Anyway, the screen resolution was wrong on a 15,6" screen due to the nomodeset option

An xserver-xorg-video-intel patch came through today:

"Disable Xv code on sandybridge generation GPU's only, support does not exist in this driver release and it attempts to use an old code path that causes a GPU hang every time it is used."

Fingers crossed.

Any possibility this is a duplicate of #654876? The recent patch was applied to fix that bug and the symptoms are similar. https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/644740

Sorry, the correct link for comment #17 is https://bugs.launchpad.net/ubuntu/maverick/+source/xserver-xorg-video-intel/+bug/654876

Hmmm... maybe the recent patch is unrelated. Sandy Bridge is not used in the W500.

Hi,

I reported bug 653093 which is a dublicate of this one, and for me this bug is fixed. Desktop effects are enabled by default and I havn't had the "hangcheck" eror any more.

So, whatever you did: Thanks :-)

Hi all,
I think this can help.
I have got same problem after working an hour with my extra monitor. Driver works well with my display laptop but not with two monitors.
My new laptop was bought 1 week ago: Dell vostro v130
Im using ubuntu maverick 10.10
lspic -vv:
00:02.0 VGA compatible controller: Intel Corporation Core Processor Integrated Graphics Controller (rev 18) (prog-if 00 [VGA controller])
        Subsystem: Dell Device 049c
        Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+
        Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
        Latency: 0
        Interrupt: pin A routed to IRQ 46
        Region 0: Memory at fe000000 (64-bit, non-prefetchable) [size=4M]
        Region 2: Memory at c0000000 (64-bit, prefetchable) [size=256M]
        Region 4: I/O ports at f080 [size=8]
        Expansion ROM at <unassigned> [disabled]
        Capabilities: <access denied>
        Kernel driver in use: i915
        Kernel modules: i915

Thx in advance for reading me.

Cheers

I have also seen this problem. I have ubuntu 10.10 (not kubuntu). I upgraded from Lucid. Ubuntu and gnome work fine, but I decided to give kde a try and I installed kubuntu-desktop via synaptic. I got the same problem. I have a lenovo w500 laptop and lspci gives

Display controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07)

Same problem. I get the hangcheck error when effects are enabled. Gnome still works fine and I use full effect there.

The same happens for me on ubuntu 10.10 with the desktop effects disabled.
lspci -vv
0:02.0 VGA compatible controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 0c) (prog-if 00 [VGA controller])
        Subsystem: Lenovo T61
        Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+
        Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
        Latency: 0
        Interrupt: pin A routed to IRQ 47
        Region 0: Memory at f8100000 (64-bit, non-prefetchable) [size=1M]
        Region 2: Memory at e0000000 (64-bit, prefetchable) [size=256M]
        Region 4: I/O ports at 1800 [size=8]
        Expansion ROM at <unassigned> [disabled]
        Capabilities: <access denied>
        Kernel driver in use: i915
        Kernel modules: i915

Kristóf Kiszel, 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? If so, could you please test for this with the latest development release of Ubuntu? ISO images are available from http://cdimage.ubuntu.com/daily-live/current/ .

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

apport-collect -p linux <replace-with-bug-number>

Also, could you please test the latest upstream kernel available (not the daily folder) following https://wiki.ubuntu.com/KernelMainlineBuilds ? It will allow additional upstream developers to examine the issue. Once you've tested the upstream kernel, please comment on which kernel version specifically you tested. If this bug is fixed in the mainline kernel, please add the following tags:
kernel-fixed-upstream
kernel-fixed-upstream-VERSION-NUMBER

where VERSION-NUMBER is the version number of the kernel you tested. For example:
kernel-fixed-upstream-v3.13-rc1

This can be done by clicking on the yellow circle with a black pencil icon next to the word Tags located at the bottom of the bug description. As well, please remove the tag:
needs-upstream-testing

If the mainline kernel does not fix this bug, please add the following tags:
kernel-bug-exists-upstream
kernel-bug-exists-upstream-VERSION-NUMBER

As well, please remove the tag:
needs-upstream-testing

Once testing of the upstream kernel is complete, please mark this bug's Status as Confirmed. Please let us know your results. Thank you for your understanding.

[Expired for linux (Ubuntu) because there has been no activity for 60 days.]