[855GM] GPU hang when an application tries to use XVideo

[LGB [Gábor Lénárt]]

Binary package hint: xorg

xorg crashes when an application tries to use xv. Then screen flashes (I guess gdm tries to start again and again) at the end only black screen I can got, and only hard reset can help.

ProblemType: Bug
DistroRelease: Ubuntu 10.04
Package: xserver-xorg 1:7.5+3ubuntu1
ProcVersionSignature: Ubuntu 2.6.32-19.28-generic 2.6.32.10+drm33.1
Uname: Linux 2.6.32-19-generic i686
Architecture: i386
Date: Sat Apr 3 23:25:47 2010
Lsusb:
 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
 Bus 001 Device 002: ID 0ace:1215 ZyDAS WLA-54L 802.11bg
 Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
MachineType: Hewlett-Packard compaq nx9020 (PG662ES#AKC)
PccardctlIdent:
 Socket 0:
   no product info available
PccardctlStatus:
 Socket 0:
   no card
ProcCmdLine: BOOT_IMAGE=/boot/vmlinuz-2.6.32-19-generic root=UUID=dcdc2473-3fd8-4e42-8132-6612dc8c9bf9 ro quiet splash
ProcEnviron:
 LANG=en_US.utf8
 SHELL=/bin/bash
SourcePackage: xorg
dmi.bios.date: 08/18/2004
dmi.bios.vendor: Hewlett-Packard
dmi.bios.version: F.10
dmi.board.name: 3084
dmi.board.vendor: Quanta
dmi.board.version: 41.09
dmi.chassis.type: 10
dmi.chassis.vendor: Quanta
dmi.chassis.version: N/A
dmi.modalias: dmi:bvnHewlett-Packard:bvrF.10:bd08/18/2004:svnHewlett-Packard:pncompaqnx9020(PG662ES#AKC):pvrRev1:rvnQuanta:rn3084:rvr41.09:cvnQuanta:ct10:cvrN/A:
dmi.product.name: compaq nx9020 (PG662ES#AKC)
dmi.product.version: Rev 1
dmi.sys.vendor: Hewlett-Packard
system:
 distro: Ubuntu
 codename: lucid
 architecture: i686
 kernel: 2.6.32-19-generic

[LGB [Gábor Lénárt] (lgb)]

I've attached the xorg.log to this comment.

Linux orion 2.6.32-19-generic #28-Ubuntu SMP Wed Mar 31 17:46:20 UTC 2010 i686 GNU/Linux

version of xserver-xorg is 1:7.5+3ubuntu1, 2:2.9.1-3ubuntu1 for xserver-xorg-video-intel

Video:

00:02.0 VGA compatible controller: Intel Corporation 82852/855GM Integrated Graphics Device (rev 02)
        Subsystem: Hewlett-Packard Company Device 3084
        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 10
        Region 0: Memory at e8000000 (32-bit, prefetchable) [size=128M]
        Region 1: Memory at e0000000 (32-bit, non-prefetchable) [size=512K]
        Region 2: I/O ports at 1800 [size=8]
        Capabilities: [d0] Power Management version 1
                Flags: PMEClk- DSI+ D1+ D2- AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-)
                Status: D0 PME-Enable- DSel=0 DScale=0 PME-
        Kernel driver in use: i915
        Kernel modules: i915

[LGB [Gábor Lénárt] (lgb)]

I've found this in the kernel log:

Apr 3 23:00:26 orion kernel: [ 575.444041] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Apr 3 23:00:26 orion kernel: [ 575.444056] render error detected, EIR: 0x00000000
Apr 3 23:00:26 orion kernel: [ 575.444070] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 27995
 at 27994)

[LGB [Gábor Lénárt] (lgb)]

Applications tries which caused crash: mplayer (with xv output vo) and miro (using xine backend to play video). I am attaching the output of the xvinfo command.

[LGB [Gábor Lénárt] (lgb)]

Btw I have had a crash during normal usage of the desktop (no xv apps - as far as I know), the xorg.log and kernel log indicated that it's the same problem (exactly the same messages). I am not sure however that it was not caused by some xv apps, but as far as I know it wasn't that at that time at least ...

[braseb (braseb37)]

I confirm the bug, when i launch a video.The bug isn't present before the beta 2 with the xserver-xorg-video-intel 2.9.1-3ubuntu1 if this can help you.

[Chris Halse Rogers (raof)]

The 855 chipset has been particularly troublesome after the recent -intel driver work. For the Lucid release we have turned off KMS support for this chipset, in the hopes of not triggering the serious stability issues you and others have found.

It would therefore be useful if you could test this with kernel-modesetting disabled. You can do this either by adding “i915.modeset=0” to the kernel boot options, on the same line after “quiet splash”, or by following https://wiki.ubuntu.com/X/KernelModeSetting

Alternatively, if you update to the most recent kernel (2.6.32-21.32) kms will be disabled by default.

If this is resolved by disabling kernel modesetting, then we can close this bug. Please comment whether or not disabling KMS makes this problem better, worse, or no different.

[LGB [Gábor Lénárt] (lgb)]

I don't have the notbook with me now, but a co-worker of mine has the same issue with identical hardware. It seems there is no problem with this kernel: http://kernel.ubuntu.com/~kernel-ppa/mainline/v2.6.34-rc3-lucid/ (just checked out on his notebook). It means that for example mplayer can play videos with XVideo output (mplayer -vo xv ...) without a crash. Also kernel package just has been upgraded to 2.6.32-21.32. With that kernel not even login screen got just black screen and no disk I/O after a while ... (even with boot options nosplash - btw even with noplash there is splash -, and quiet option is removed too). According to /var/log/kern.log the kernel log ends at the MTRR stuffs in the "middle of a line" and logs of the new boot (with that 2.6.34 kernel which works) starts with the date/time not at a new line ... xorg.log files in /var/log/ does not give many informations the only difference some kind of acpi socket opening problems.

[LGB [Gábor Lénárt] (lgb)]

In recovery mode system boots but if I try to launch X from root console, system freezes totally.

[braseb (braseb37)]

I test with the option i915.modeset=0, with the kernel before the kernel 2.6.32-21.32 on a live cd, and i have a black screen after the plymouth's animation.

[LGB [Gábor Lénárt] (lgb)]

So I have the same (but) with kernel 2.6.32-21.32 (with out without that i915.modeset=0) as braseb commented ...

[reidi (ian-reid)]

"It would therefore be useful if you could test this with kernel-modesetting disabled. You can do this either by adding “i915.modeset=0” to the kernel boot options, on the same line after “quiet splash”, or by following https://wiki.ubuntu.com/X/KernelModeSetting"

Lucid 2.6.32.21-generic #32-Ubuntu SMP Fri Apr 16 08:10:02 UTC
00:02.0 VGA compatible controller: Intel Corporation 82852/855GM Integrated Graphics Device (rev 02)

Adding "i915.modeset =0" to kernel boot options did NOT solve this for me.

Abandoning that option, I followed directions in KernelModeSetting:

Intel step 1 did NOT help: "echo options i915 modeset=0 > /etc/modprobe.d/i915-kms.conf"
Nor did step 2: "...need to run sudo update-initramfs -u"

The only thing that helped wasstep 3: add xorg.conf, specifying: Driver "vesa". (I did not try this alone, I just stepped through the list in the KernelModeSetting document, this being the "last resort" step...

[LGB [Gábor Lénárt] (lgb)]

It seems that the current (and previous) lucid kernel simply cannot work well enough to be usable on my notebook at least. Using http://kernel.ubuntu.com/~kernel-ppa/mainline/v2.6.34-rc3-lucid/ works, but that's not the default kernel for lucid, of course. 2.6.32.21-31 makes xorg crashes on XVideo but also at some other cases according to my experience, switching off KMS does not help. The newer kernel in lucid since then (2.6.32.21-32) is even more worse, since the xorg does not even start, and machine freezes completly. So for me at least this bug makes lucid unusable on my hardware which is a show-stopper. Currently I am using 2.6.34 kernel from that PPA to be able to use lucid.

[xanmoo (jeanpatrick-pommier)]

On my IBM thinkpad R50e 1834 running Lucid beta, I have a black screen after plymouth and before gdm. I have to start in failsafeX mode.

 lspci yields:

~$ lspci
00:00.0 Host bridge: Intel Corporation 82852/82855 GM/GME/PM/GMV Processor to I/O Controller (rev 02)
00:00.1 System peripheral: Intel Corporation 82852/82855 GM/GME/PM/GMV Processor to I/O Controller (rev 02)
00:00.3 System peripheral: Intel Corporation 82852/82855 GM/GME/PM/GMV Processor to I/O Controller (rev 02)
00:02.0 VGA compatible controller: Intel Corporation 82852/855GM Integrated Graphics Device (rev 02)
00:02.1 Display controller: Intel Corporation 82852/855GM Integrated Graphics Device (rev 02)
00:1d.0 USB Controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) USB UHCI Controller #1 (rev 01)
00:1d.1 USB Controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) USB UHCI Controller #2 (rev 01)
00:1d.2 USB Controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) USB UHCI Controller #3 (rev 01)
00:1d.7 USB Controller: Intel Corporation 82801DB/DBM (ICH4/ICH4-M) USB2 EHCI Controller (rev 01)
00:1e.0 PCI bridge: Intel Corporation 82801 Mobile PCI Bridge (rev 81)
00:1f.0 ISA bridge: Intel Corporation 82801DBM (ICH4-M) LPC Interface Bridge (rev 01)
00:1f.1 IDE interface: Intel Corporation 82801DBM (ICH4-M) IDE Controller (rev 01)
00:1f.3 SMBus: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) SMBus Controller (rev 01)
00:1f.5 Multimedia audio controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) AC'97 Audio Controller (rev 01)
00:1f.6 Modem: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) AC'97 Modem Controller (rev 01)
02:00.0 CardBus bridge: Texas Instruments PCI1510 PC card Cardbus Controller
02:02.0 Network controller: Intel Corporation PRO/Wireless 2200BG [Calexico2] Network Connection (rev 05)
02:08.0 Ethernet controller: Intel Corporation 82801DB PRO/100 VE (MOB) Ethernet Controller (rev 81)

I noticed since Karmic, a bug freezing the computer, when trying to run aport, related to intel driver ?

[robertpeter (rob-antenna)]

I Have the same problems with my Thinkpad R50 as Xanmoo.
even the live cd 10.4 RC does not startup and gives a black screen.

[dan robinson (dgrobinson)]

I don't know if this is the same problem. I was able to putty into the machine and captured the files /sys/kernel/debug/dri/0

I have had problems since upgrading to 9.10. My display would freeze. This may be an associated problem or not.

What files should I capture when I get this problem again?

[royden (ryts)]

For the record, Bug #558633 and it's duplicate Bug #568716 appear to be duplicates of this bug.

[Wijbrand Schaap (wijbrand-schaap)]

Have the same problem here, however in kernel version 2.6.32.20, everything works fine. It's only the .21 version that breaks evrything

[LGB [Gábor Lénárt] (lgb)]

For me: .20 mostly works, only crashes on XVideo usage, .21: black screen after boot, nothing works. 2.6.34 rc3 kernel (not official lucid kernel! from ppa): everything works, even XVideo.

[Chris Halse Rogers (raof)]

The underlying problem here is highly likely to be the same as bug #554835. There's a kernel fix for that in the works, but it's not yet pushed upstream, and then it would need to be backported to our kernel. We're hoping that we can do that for a post-release update.

[LGB [Gábor Lénárt] (lgb)]

Ahaa, I see, the only problem that .21 made things even more worse (no idea about .22 yet) so lucid will come out (today?) as not being able to show me even a login screen (so looks like no successfull boot, only black screen) on some (?) intel GPUs. Sure, I can work-around that with using that 2.6.34 rc3 kernel from PPA, true ... Thanks for the comment!

[AM (macchi)]

My test laptop a Sony Vaio VGN-B1XP with Intel 855 graphics crashes badly with the 10.04 Lucid release candidate kernel 2.6.32-21. The symptom is only black screen and no activity upon attempt to boot /start. I get no error logs.

Please assign a reasonably high priority for solving this regression on the LTS release. There are lots of systems with 855 graphics that may fail completely upon upgrade.

[royden (ryts)]

@Chris Halse Rogers: did you mean Bug #541511?

@ LGB etc: My system also will not work without KMS. In fact, the 2.6.34 rc3 kernel with modeset=0 will not complete boot either. One peculiarity on my system with kernel 2.6.32-21 is that setting modeset=1 via /etc/modprobe.d/i915-kms.conf does not work.

However, placing i915.modeset=1 on the boot line in grub does work. Go figure - as devs have commented this is a highly contextual bug.

@Chris/Bryce: want the above on the wiki?

[Chris Halse Rogers (raof)]

On Thu, 2010-04-29 at 15:25 +0000, royden wrote:
> @Chris Halse Rogers: did you mean Bug #541511?

Whoops. Yes, that's exactly what I meant.

>
> @ LGB etc: My system also will not work without KMS. In fact, the 2.6.34
> rc3 kernel with modeset=0 will not complete boot either. One peculiarity
> on my system with kernel 2.6.32-21 is that setting modeset=1 via
> /etc/modprobe.d/i915-kms.conf does not work.
>
> However, placing i915.modeset=1 on the boot line in grub does work. Go
> figure - as devs have commented this is a highly contextual bug.

I suspect that you'll find that setting the option
in /etc/modprobe.d/i915-kms.conf will work if you run “sudo
update-initramfs -u” after editing the file. I've added that to the
wiki. This is needed if Ubuntu is loading the i915 driver from the
initramfs, before trying to mount the root filesystem. It will do this,
for example, if you've set up an encrypted filesystem for which you need
to enter the passphrase before it can be mounted.

For the users who are concerned that this bug renders Ubuntu 10.04
unusable on a lot of systems: yes, so are we. However, there are
similar problems for other 855 users with kernel modesetting turned *on*
- there just doesn't seem to be a set of defaults we could use where all
855 cards would work.

[mungewell (simon-mungewell)]

I am unable to start either the Ubuntu or Xubuntu LiveCDs (Live mode or install mode) on my Thinkpad X40 with Intel 855GM. The whole machine locks up when starting X, last visible message is about starting cups server.

Following advice in this thread I added to the grub boot parameters:
i915.modeset=0 - gives no improvement, X still locks up machine
i915.modeset=1 - allows X to start and presents first installer window.

For Ubuntu CD this requires careful timing of the 'Esc' key to bring up the grub menu, which normally whizes past. Too late now to run through complete install, but hopefully this might help others who are pulling their hair out.

Cheers and Happy Release Day!
Mungewell.

[royden (ryts)]

On Fri, 2010-04-30 at 02:53 +0000, Chris Halse Rogers wrote:

> I suspect that you'll find that setting the option
> in /etc/modprobe.d/i915-kms.conf will work if you run “sudo
> update-initramfs -u” after editing the file. I've added that to the
> wiki. This is needed if Ubuntu is loading the i915 driver from the
> initramfs, before trying to mount the root filesystem. It will do this,
> for example, if you've set up an encrypted filesystem for which you need
> to enter the passphrase before it can be mounted.

Thanks, that worked. Never occurred to me as I don't have an encrypted
filesystem. Strangely, it knocks out plymouth (abs no big deal that)
whereas i915.modeset=1 on boot line does not.

[oscar (oacg7max)]

I have had the same problem. I am not sure if is related to this bug or is the same. However I am sure it has made lucid unstable on my computer. None of the solutions listed above has worked for me. Sometimes when I'm doing nothing particular in lucid, I get a black screen and the system completely freezes, or it enters an endless loop as if it is about to reboot or shutoff, but never does it. Here's what lspci says,

00:00.0 Host bridge: Intel Corporation 82845G/GL[Brookdale-G]/GE/PE DRAM Controller/Host-Hub Interface (rev 01)
00:02.0 VGA compatible controller: Intel Corporation 82845G/GL[Brookdale-G]/GE Chipset Integrated Graphics Device (rev 01)
00:1d.0 USB Controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) USB UHCI Controller #1 (rev 01)
00:1d.1 USB Controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) USB UHCI Controller #2 (rev 01)
00:1d.2 USB Controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) USB UHCI Controller #3 (rev 01)
00:1d.7 USB Controller: Intel Corporation 82801DB/DBM (ICH4/ICH4-M) USB2 EHCI Controller (rev 01)
00:1e.0 PCI bridge: Intel Corporation 82801 PCI Bridge (rev 81)
00:1f.0 ISA bridge: Intel Corporation 82801DB/DBL (ICH4/ICH4-L) LPC Interface Bridge (rev 01)
00:1f.1 IDE interface: Intel Corporation 82801DB (ICH4) IDE Controller (rev 01)
00:1f.5 Multimedia audio controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) AC'97 Audio Controller (rev 01)
05:08.0 Ethernet controller: Intel Corporation 82801DB PRO/100 VE (LOM) Ethernet Controller (rev 81)
05:09.0 Communication controller: Conexant Systems, Inc. HSF 56k HSFi Modem (rev 01)

Here's what kern.log says, and the same goes for the xorg log (this is the log of the last bug event),

May 9 16:54:34 oscar-desktop kernel: [ 418.992019] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
May 9 16:54:34 oscar-desktop kernel: [ 418.992038] render error detected, EIR: 0x00000000
May 9 16:54:34 oscar-desktop kernel: [ 418.992060] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 44302 at 44294)
May 9 16:54:37 oscar-desktop kernel: [ 421.888019] [drm:i915_gem_idle] *ERROR* hardware wedged
May 9 16:54:43 oscar-desktop kernel: [ 427.915019] [drm:i915_gem_entervt_ioctl] *ERROR* Reenabling wedged hardware, good luck
May 9 16:54:43 oscar-desktop kernel: [ 428.248012] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
May 9 16:54:43 oscar-desktop kernel: [ 428.248022] render error detected, EIR: 0x00000000
May 9 16:54:43 oscar-desktop kernel: [ 428.248921] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 44304 at 44294)
May 9 16:54:45 oscar-desktop kernel: [ 430.348017] [drm:i915_gem_idle] *ERROR* hardware wedged
May 9 16:54:46 oscar-desktop kernel: [ 431.682383] [drm:i915_gem_entervt_ioctl] *ERROR* Reenabling wedged hardware, good luck
May 9 16:54:47 oscar-desktop kernel: [ 432.016012] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
May 9 16:54:47 oscar-desktop kernel: [ 432.016023] render error detected, EIR: 0x00000000
May 9 16:54:47 oscar-desktop kernel: [ 432.016917] [drm:i915...

[sasa becircic (sasaenator)]

I have the same problem and i915.modeset=1 works but only if I don't try to play video.I have the latest 2.6.32-22 kernel.I can cause system to crash or freeze when I try to play video or when I run system testing utility,test bars and static!I haven't try mainline kernel yet!

[mungewell (simon-mungewell)]

> Following advice in this thread I added to the grub boot parameters:
> i915.modeset=0 - gives no improvement, X still locks up machine
> i915.modeset=1 - allows X to start and presents first installer window.

Just a follow up, this is still a problem. Using the above grub option I was able to get Xubuntu installed but the install still suffered the 'crash on X start' issue. Even when all updates were applied, and with both available kernels.

--
$ dpkg --list | grep linux-image
ii linux-image-2.6.32-21-generic 2.6.32-21.32 Linux kernel image for version 2.6.32 on x86/x86_64
ii linux-image-2.6.32-22-generic 2.6.32-22.33 Linux kernel image for version 2.6.32 on x86/x86_64
--
$ dpkg --list | grep xserver|grep intel
ii xserver-xorg-video-intel 2:2.9.1-3ubuntu5 X.Org X server -- Intel i8xx, i9xx display driver
--

Setting the 'i915.modeset=1' in /etc/defaults/grub and running 'update-grub', got the machine to a workable start, however any XV activity will hard crash machine, such as:
--
$ mplayer -vo xv dp-hop2yab-sample.avi
--

In order to prevent this I have generated a xorg.conf and adjusted the modules section:
--
Section "Module"
        Disable "XVideo"
        Disable "extmod"
        Load "dri"
        Load "dri2"
        Load "record"
        Load "dbe"
        #Load "extmod"
        Load "glx"
EndSection
--

which completely disables XV on the system. Now at least the above mplayer line will just fail to display anything and the machine continues to run OK.

Cheers,
Mungewell.

[axelsvag (darinsson)]

I got exactly the same problem. The computer is unsable with the kernel 2.6.32 the screen goes black in the middle of the boot process. WIth the old kernel 2.6.31 everything work as it should and if I use the 2.6.34 rc kernel the computer boots OK but after 10 minutes the screen goes black.
I have tried all your advices. If I use the i915=0 or=1 the boot works but the screen flicker constantly.

[ar (arjenmeijernl)]

See allso https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bug/575103.

[Alex K (alex-khazin)]

I see this problem on my ThinkPad R50e in both 2.6.32-21 and 2.6.32-22 upgrades . The problem isn't exist on 2.6.32-20 release . I'm not sure if I understood the resolution of this bug ???
i915=0 - should it work for any laptop ?
In addition I got the feeling that in -22 release everything will be fixed . Maybe I didn't understand the blog well enough

[braseb (braseb37)]

my ubuntu lucid is on a 2.6.34-lucid kernel of the ppa ubuntu kernel and vidéo is OK
when i test the maverick kernel (2.6.34-2) the bug is present. Alway with intel drivers of lucid.
The problem is in a change of the ubuntu version kernel.

[Vincent Caron (vincent-zerodeux)]

I have mostly the same problems than other people. I have an Inspiron 510M with an ICH4/855GM chipset (from 2005).

With Jaunty I used the "i195.modeset=0" to fix the X hard crash while closing lid (which happened whatever action was configured on LID closed, and could be repeated without any X or acpid running).

While upgrading to Karmic, the "i195.modeset=0" was kept in GRUB settings and X totally crashed the system at startup: no VT switching, no sysrq to the rescue. I changed the kernel setting to "i915.modeset=1" then X started properly and seems to run fine, except for immediate hardlock as soon as XV is used. This running Karmic's 2.6.32-22.33 (-generic on a Pentium M).

I have asked at least gstreamer not to use XV (eg: http://www.thinkwiki.org/wiki/Intel_Graphics_Media_Accelerator_950#Issues).

I can try other kernels or settings to help narrow down that XV crash if that helps.

[MarcH (marc-h38)]

I found the following wiki page useful: https://wiki.ubuntu.com/X/Bugs/Lucidi8xxFreezes

It seems to basically contains the information already posted in this bug, but presented (in a much clearer way).

I am surprised that this bug is "undecided". Maybe I do not understand the meaning of this field.

I am also surprised that any distribution keeps distributing such new, totally untested software from Intel. Every time I upgrade Ubuntu it is becoming more and more difficult to make my laptop usable.

[oscar (oacg7max)]

I've had many system crashes because of this bug. Here are some lines I took from the syslog and kernel log. I hope this helps.

syslog,

May 26 20:26:23 oscar-desktop anacron[1399]: Updated timestamp for job `cron.daily' to 2010-05-26
May 26 20:28:50 oscar-desktop kernel: [ 464.464015] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
May 26 20:28:50 oscar-desktop kernel: [ 464.464032] render error detected, EIR: 0x00000000
May 26 20:28:50 oscar-desktop kernel: [ 464.465142] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 11778 at 11775)
May 26 20:28:52 oscar-desktop kernel: [ 466.380017] [drm:i915_gem_idle] *ERROR* hardware wedged
May 26 20:28:52 oscar-desktop x-session-manager[1158]: WARNING: Detected that screensaver has left the bus
May 26 20:28:53 oscar-desktop acpid: client 770[0:0] has disconnected
May 26 20:28:53 oscar-desktop acpid: client connected from 1529[0:0]
May 26 20:28:53 oscar-desktop acpid: 1 client rule loaded
May 26 20:28:53 oscar-desktop pulseaudio[1526]: pid.c: Stale PID file, overwriting.
May 26 20:28:53 oscar-desktop pulseaudio[1526]: main.c: Unable to contact D-Bus: org.freedesktop.DBus.Error.NoServer: Failed to connect to socket /tmp/dbus-snjWMwup0Y: Conexión rechazada
May 26 20:28:54 oscar-desktop kernel: [ 468.349487] [drm:i915_gem_entervt_ioctl] *ERROR* Reenabling wedged hardware, good luck
May 26 20:28:54 oscar-desktop kernel: [ 468.680014] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
May 26 20:28:54 oscar-desktop kernel: [ 468.680026] render error detected, EIR: 0x00000000
May 26 20:28:54 oscar-desktop kernel: [ 468.680048] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 11782 at 11775)
May 26 20:28:57 oscar-desktop kernel: [ 470.740018] [drm:i915_gem_idle] *ERROR* hardware wedged
May 26 20:28:57 oscar-desktop acpid: client 1529[0:0] has disconnected
May 26 20:28:57 oscar-desktop acpid: client connected from 1541[0:0]
May 26 20:28:57 oscar-desktop acpid: 1 client rule loaded
May 26 20:28:57 oscar-desktop bonobo-activation-server (oscar-1546): could not associate with desktop session: Failed to connect to socket /tmp/dbus-snjWMwup0Y: Conexión rechazada
May 26 20:28:58 oscar-desktop kernel: [ 472.074908] [drm:i915_gem_entervt_ioctl] *ERROR* Reenabling wedged hardware, good luck
May 26 20:28:59 oscar-desktop kernel: [ 472.904014] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
May 26 20:28:59 oscar-desktop kernel: [ 472.904025] render error detected, EIR: 0x00000000
May 26 20:28:59 oscar-desktop kernel: [ 472.904048] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 11784 at 11775)
May 26 20:29:01 oscar-desktop kernel: [ 474.960019] [drm:i915_gem_idle] *ERROR* hardware wedged
May 26 20:29:01 oscar-desktop kernel: Kernel logging (proc) stopped.

May 26 21:21:49 oscar-desktop udev-configure-printer: parent devpath is /devices/pci0000:00/0000:00:1d.1/usb3/3-1
May 26 22:11:16 oscar-desktop AptDaemon: INFO: Initializing daemon
May 26 22:11:18 oscar-desktop kernel: [ 6114.736013] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GP...

[liorda (liorda)]

i can confirm this bug.

running on R50e 1834 with intel 855 chipset, even when disabling DRI, whenever i try to play a video using xv, the system hangs

[Jiri Trnka (jiri-trnka-79)]

I also confirm this bug with the following experiences:

[Laptop, Intel 855 GM]

When I upgraded from Hardy to Lucid, laptop did not boot until I re-enabled KMS (according to https://wiki.ubuntu.com/X/Bugs/Lucidi8xxFreezes ). Everything else then worked almost fine, except it few times unexpectedly crashed (screen flashed -> hard freeze). Video worked fine with some minor problems on external monitor.

I started to use a program with 3D graphics and computer was crashing frequently when using external monitor. Then I tried to switch video driver to the older one from PPA X-retro repos. It did not help much (only video with external monitor got better), so I switched back to official Ubuntu driver. After that xv always crashes.

Now I use retro - driver (see https://edge.launchpad.net/~ubuntu-x-swat/+archive/x-retro ) with quite success. Problems with external monitor still remain (flickering with login screen, occassional crashes).

This steps I had to perform to make my laptop work under Lucid:

1. re-enable KMS ( at least to be able to boot )
2. downgrade the video driver to avoid video crashes ( it works generally better even when original driver works also)
3. Enable FrameBuffer Device in xorg.conf to avoid crashes on VT switch ( it was the same with Hardy )
4. avoid using 3D acceleration on external monitor

[Colm (colmmchugh)]

A solution to this issue is posted on Ubuntu forums.

See post titled "Stuck with black screen on booting after 10.04 upgrade" See comments #7, 8,9 and 10 especially. http://ubuntuforums.org/showthread.php?t=1467235

The solution people found is to boot into the older kernel, and then install a kernel that's even newer than the one that ships with the Lucid 10.04 iso. Seems everything works then.

[Colm (colmmchugh)]

Another Launchpad member has pointed out that this solution suggests installing an mainline (upstream) kernel. It will boot, but is not a perfect solution because those kernels aren't modified for specific Ubuntu releases (see Limitations on https://wiki.ubuntu.com/KernelTeam/MainlineBuilds). Looks like we'll have to hope for the 2.6.34 kernels to be submitted to the repositories for a clean solution.

[Jiri Trnka (jiri-trnka-79)]

The problem seems to be resolved (for me) simply by downgrading the kernel, at least 2.6.31-22 (from karmic repos) works properly even with the external monitor. I still use the x-retro driver, maybe it would work with the official driver too.

[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

[peakit (guntasg)]

Just upgraded from 8.04.4 LTS to 10.04.1 LTS, and suddenly my Ubuntu experience has been runined because of this frequent freezes/crashes.

The kern.log says:
Aug 19 23:13:21 ubuntu kernel: [ 2440.380012] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Aug 19 23:13:21 ubuntu kernel: [ 2440.380027] render error detected, EIR: 0x00000000
Aug 19 23:13:21 ubuntu kernel: [ 2440.380044] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 214801 at 214797)
Aug 19 23:13:24 ubuntu kernel: [ 2442.824019] [drm:i915_gem_idle] *ERROR* hardware wedged
Aug 19 23:13:31 ubuntu kernel: [ 2450.568583] [drm:i915_gem_entervt_ioctl] *ERROR* Reenabling wedged hardware, good luck
Aug 19 23:13:32 ubuntu kernel: [ 2450.880012] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Aug 19 23:13:32 ubuntu kernel: [ 2450.880023] render error detected, EIR: 0x00000000
Aug 19 23:13:32 ubuntu kernel: [ 2450.880045] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 214803 at 214797)
Aug 19 23:13:34 ubuntu kernel: [ 2453.112033] [drm:i915_gem_idle] *ERROR* hardware wedged
Aug 19 23:13:35 ubuntu kernel: [ 2454.217610] [drm:i915_gem_entervt_ioctl] *ERROR* Reenabling wedged hardware, good luck
Aug 19 23:13:35 ubuntu kernel: [ 2454.532015] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Aug 19 23:13:35 ubuntu kernel: [ 2454.532025] render error detected, EIR: 0x00000000
Aug 19 23:13:35 ubuntu kernel: [ 2454.532052] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 214805 at 214797)
Aug 19 23:13:37 ubuntu kernel: [ 2456.588020] [drm:i915_gem_idle] *ERROR* hardware wedged
Aug 19 23:13:38 ubuntu kernel: [ 2457.764415] [drm:i915_gem_entervt_ioctl] *ERROR* Reenabling wedged hardware, good luck
Aug 19 23:13:39 ubuntu kernel: [ 2458.080013] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Aug 19 23:13:39 ubuntu kernel: [ 2458.080025] render error detected, EIR: 0x00000000
Aug 19 23:13:39 ubuntu kernel: [ 2458.080048] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 214807 at 214797)

And the machine details are:
00:00.0 Host bridge: Intel Corporation 82845G/GL[Brookdale-G]/GE/PE DRAM Controller/Host-Hub Interface (rev 01)
00:02.0 VGA compatible controller: Intel Corporation 82845G/GL[Brookdale-G]/GE Chipset Integrated Graphics Device (rev 01)
00:1d.0 USB Controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) USB UHCI Controller #1 (rev 01)
00:1d.1 USB Controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) USB UHCI Controller #2 (rev 01)
00:1d.2 USB Controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) USB UHCI Controller #3 (rev 01)
00:1d.7 USB Controller: Intel Corporation 82801DB/DBM (ICH4/ICH4-M) USB2 EHCI Controller (rev 01)
00:1e.0 PCI bridge: Intel Corporation 82801 PCI Bridge (rev 81)
00:1f.0 ISA bridge: Intel Corporation 82801DB/DBL (ICH4/ICH4-L) LPC Interface Bridge (rev 01)
00:1f.1 IDE interface: Intel Corporation 82801DB (ICH4) IDE Controller (rev 01)
00:1f.3 SMBus: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) SMBus Controller (rev 0...

[peakit (guntasg)]

BTW, does anyone know how to downgrade from 10.04.1 to 8.04.4 without losing any data.

Thanks.

[peakit (guntasg)]

Tried the mainline kernel 2.6.34-020634-generic #020634 on Lucid, and this time again it suffered the same fate

-----------
Kern.log:
-----------
Aug 22 19:37:19 ubuntu kernel: [ 3892.348015] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Aug 22 19:37:19 ubuntu kernel: [ 3892.348319] render error detected, EIR: 0x00000000
Aug 22 19:37:19 ubuntu kernel: [ 3892.348340] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 436087 at 436086)

Chris, any ETA on when can we expect this fixed?

[MarcH (marc-h38)]

Just tried kernel 2.6.32-24-generic and everything is still crashing the exact same. So I'll keep running some 2.6.34 PPA kernel as a workaround.

[Chris Halse Rogers (raof)]

I'm unassigning myself from this bug; I've got the needed feedback for Maverick, and we've gone with the safe option of fbdev.

I'll leave this bug open; there's still a reasonable chance we can get a proper fix, and apparently some i8xx documentation has just been released.