MASTER: [i845] GPU lockup

[Geir Ove Myhr]

Binary package hint: xserver-xorg-video-intel

This is a MASTER bug report, i.e. not a real bug report, but a tool to help manage other bug reports.

Most bug reports on i845 are probably due to the CPU/GPU incoherency problem that is now consolidated upstream at http://bugs.freedesktop.org/show_bug.cgi?id=26345 . For now, we mark all automatically reported GPU lockups as duplicates of this unless there is a reason not to.

To use the available fix, run the following commands:

apt-add-repository ppa:glasen/855gm-fix
apt-add-repository ppa:brian-rogers/graphics-fixes
apt-add-repository ppa:glasen/intel-driver
aptitude update
aptitude install linux 855gm-fix-dkms
aptitude dist-upgrade

There is a similar master bug report for i855 at bug 541511.

[Odin Hørthe Omdal (velmont)]

It happens all the time now. Sometimes when I click a new link, sometimes when I open a new program or whatever. The screen doesn't get signal from the GPU anymore and complains.

It's pretty reliable, after 1-5 hours I always get a crash. Hope it can be fixed soon.

[jerrylamos (jerrylamos)]

Haven't had the bug recently. Now on

Linux version 2.6.32-17-generic (buildd@vernadsky) (gcc version 4.4.3 (Ubuntu 4.4.3-3ubuntu3) ) #26-Ubuntu SMP Fri Mar 19 23:58:53 UTC 2010

Do note this Lucid Beta is running with KMS and default xorg. Finally.

Jerry

[MasterNetra (designerkline)]

Still recieve the crashes. Just updating.

[useResa (rdrijsen)]

Have posted my experiences with the trial version (2:2.10.903-4-g0c47195.1) in my original report, bug #545201
Unfortunately caused more issues than the current official version

[Odin Hørthe Omdal (velmont)]

So far, so good with the patch. :-)

[istoyanov (istoyanov)]

The integrated video on a Fujitsu-Siemens Scenic P300 desktop PC suffered from random lock-ups and problems detecting the correct refresh rate and resolution of the monitor attached to it. Had the problem with 9.04, 9.10 and a clean install of 10.04-beta-1, but after getting all the updates as of 31. March I the issue seems to be resolved and I can use the optimal resolution and refresh rate of the (now correctly detected) the monitor without a problem so far.

---
Hardware details:

00:00.0 Host bridge: Intel Corporation 82845G/GL[Brookdale-G]/GE/PE DRAM Controller/Host-Hub Interface (rev 01)
 Subsystem: Fujitsu Technology Solutions Device 1003
 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 e0000000 (32-bit, prefetchable) [size=256M]
 Capabilities: <access denied>
 Kernel driver in use: agpgart-intel
 Kernel modules: intel-agp

00:02.0 VGA compatible controller: Intel Corporation 82845G/GL[Brookdale-G]/GE Chipset Integrated Graphics Device (rev 01)
 Subsystem: Fujitsu Technology Solutions Device 1003
 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 16
 Region 0: Memory at d8000000 (32-bit, prefetchable) [size=128M]
 Region 1: Memory at d0000000 (32-bit, non-prefetchable) [size=512K]
 Capabilities: <access denied>
 Kernel modules: i915

[jerrylamos (jerrylamos)]

On i845 video graphics updated as of 31 March got three black screen crashes today. The first one was with default the other two with nomodeset. One of them seemed to point to gnome-panel which looked a bit different so I filed a bug on that. (I'm on another computer and don't have the bug # here)

Jerry

[Bryce Harrington (bryce)]

Btw I've just now uploaded a new intel-gpu-tools package. I see it
includes a few updates for improving debugging on i8xx.

[istoyanov (istoyanov)]

Unfortunately, after my previous optimistic report, I have experienced several random crashes (combined with wrong resolution signal to the monitor) similar to these reported by Jerry. Basically, I use only the "nomodeset" boot option (i.e. removed the default "splash" and "quiet" to follow the boot process) but sometimes the gdm screen received a non-standard resolution that was reported by the monitor as "user setting" (and not as "1024x768"). This has led me to experiment a little bit by adding "Modes 1024x768" and "Modeline FREQ1-FREQ2" to a newly-created xorg.conf. Thereafter, the switch from the framebuffer to the gdm login screen seems to happen with the specified resolution applied, but sometimes, when clicking on the logout button or scrollling over the volume control icon X crashes and the monitor starts to switch between a black-white striped upper 1/2 (or third) and a completely blank screen. In this situation I cannot switch to a virtual console and the only way out is the ON/OFF button on the PC (this seems to trigger a clean switch-off). I'll test the updates that Bryce mentioned and will report here any further problems. Or should I open a separate bug-report for this?

[jerrylamos (jerrylamos)]

Ivailo, I was getting the alternating upper screen black white striped and blank in two incidents yesterday as well. The screen is a 1280x1024. Only way out is power off. Could not ssh in from another pc on the same network.

How do I use intel-gpu-tools package after it is installed?

Thanks, Jerry

[Odin Hørthe Omdal (velmont)]

I have also had crashes after a full day of working with no problem. Next day, crash, crash.

The wall paper was "low resolution", the rest was OK. Every alternating line was wallpaper, and the other was old programs I've closed that "hung around, stuck" into the wallpaper!

I changed wallpaper and played with the settings -> BAM -> crash. No apport from this. Then after two restarts everything was fine, and I worked on the computer 4 hours yesterday with no problems... So it's still quite shaky.

[istoyanov (istoyanov)]

After the update today, the "nomodeset" boot parameter seems to be non-functional and at the stage where the gdm screen should appear, the monitor switches off by "entering in power save mode". Switching to a virtual console wakes it up again, but after changing the driver to "vesa" this option doesn't work anymore.

[Bryce Harrington (bryce)]

On Thu, Apr 01, 2010 at 07:42:02AM -0000, Ivailo Stoyanov wrote:
> Unfortunately, after my previous optimistic report, I have experienced
> several random crashes (combined with wrong resolution signal to the
> monitor) similar to these reported by Jerry.

Jerry has reported some crashes as well as freezes. This bug report is
about a freeze rather than a crash.

If what you are seeing is a crash there will be a backtrace in your
Xorg.0.log (usually, or check gdm logs). Also be aware that the
backtraces in your Xorg.0.log are fairly useless - instead you need to
collect a full backtrace using either apport or gdb. See
http://wiki.ubuntu.com/X/Backtracing for details.

On the other hand, if you're seeing a freeze (graphics are frozen, but
maybe the mouse works) you need to follow the directions on
http://wiki.ubuntu.com/X/Troubleshooting/Freeze to collect data needed
for these kinds of issues.

However, I think it's pretty well known at this point that upstream's
code is fairly buggered with i845, so if you have that video card, if
you have a freeze it's probably not worth putting in more bug reports
about it at this point. You can track the upstream bug report as it
progresses, and help them with testing things. Be aware you'll probably
be expected to patch and build new kernels in order to do this.

Beyond that, assuming a patch doesn't come to light, the one other knob
we have to fiddle with at the distro level is to blacklist KMS from this
hardware. See http://wiki.ubuntu.com/X/KernelModeSetting for more info
about this and how to configure it off (basically boot i915.modeset=0)
so it uses UMS instead.

What we need is a convincing consensus among 845 owners that UMS works
better than KMS for the hardware, and then we can request the kernel
team to blacklist it.

[Bryce Harrington (bryce)]

On Thu, Apr 01, 2010 at 11:39:03AM -0000, jerrylamos wrote:
> Ivailo, I was getting the alternating upper screen black white striped
> and blank in two incidents yesterday as well. The screen is a
> 1280x1024. Only way out is power off. Could not ssh in from another
> pc on the same network.
>
> How do I use intel-gpu-tools package after it is installed?

See https://wiki.ubuntu.com/X/Troubleshooting/Freeze which is where this
should be documented. The info is kind of scant though (please help us
flesh it out better - it's a wiki, you can edit.)

Basically, you run the tool `intel_gpu_dump` to capture dumps.

[Odin Hørthe Omdal (velmont)]

Sorry, I was seeing a freeze. At least I think. I don't remember correctly anymore, the screen went black anyway.

But KMS is the way forward right? So once the bugs has been ironed out that's what we want? I've lost access to the computer for some time now.

[jerrylamos (jerrylamos)]

Bryce, installed the ppa see this grub.cfg entry:

linux /boot/vmlinuz-2.6.33-997-generic root=UUID=fcf85719-2242-4b65-a162-a41060902f45 ro quiet splash nomodeset

and cat /proc/version:

Linux version 2.6.33-997-generic (root@zinc) (gcc version 4.2.3 (Ubuntu 4.2.3-2ubuntu7)) #201003091338 SMP Tue Mar 9 14:32:16 UTC 2010

Do note even though it says to boot "nomodeset" KMS is actually active. That just started happening today with the 1 April aptitude -q update safe-upgrade.

If the next crash is such that "ssh" will work I'll try the wiki directions.

Thanks, Jerry

[istoyanov (istoyanov)]

I have also installed the 2.6.33-997-generic kernel from the PPA and was able to get a batchbuffer dump after one of the incidents.

[Noel Arzola (noel.arzola)]

Still receiving. Went to cycle through screensavers to duplicate the issue and received the crash as soon as it went to display one. I've only being updating through the update manager haven't tried anything else yet. Disabling my screensaver has allowed me to keep my system up without crashing so often though.

[jerrylamos (jerrylamos)]

Bryce, on 3 of my 4 test pc's Lucid Beta with updates from end of March through 1 April were dead or not really usable with KMS problems until I saw in the forums to use "i915.modeset=0" for the i830 and i845, even though they are not i915. The i830 (the one I'm using here) wouldn't even boot in "recovery" mode only rescued by doing ssh from another pc.

The ati Mobility 7500 also failed within 2 minutes of logging in to AOL mail until I put in "radeon.modeset=0".

All three were running O.K. on Karmic, with KMS on on the i845 and ati but "nomodeset" on the i830.

Jerry

[Geir Ove Myhr (gomyhr)]

> saw in the forums to use "i915.modeset=0" for the i830 and i845, even
> though they are not i915.

Jerry, the i915 in the i915.modeset=0 kernel parameter is the name of
the kernel driver, not the chipset. I don't know the historical
reasons for it being named i915 (possibly it was only used for i915
and newer in the beginning), but at least nowadays it is used on all
chipsets, including i830, 845G, and 855GM. The DDX driver used to be
i810, even though it was used on i915 and i945 as well, but now it has
been renamed intel (this is the one packaged in
xserver-xorg-video-intel).

[Geir Ove Myhr (gomyhr)]

Ivailo, I attached your i915_error_state to the upstream bug report in case it is useful there. At least this does not seem to be the kind of freeze that would be fixed with xserver-xorg-video-intel from my PPA (I uploaded a new version based on the latest Lucid version today). For freeze bugs, dmesg output is more useful that Xorg.0.log (but both are fine). Especially with the kernel parameter drm.debug=0x02 does a lot of information about what is going on before the freeze end up in dmesg output.

[Bryce Harrington (bryce)]

We should consider if it makes sense to blacklist KMS for this chipset.

Jerry has indicated that by booting his system with i915.modeset=0 that it worked around the issue. I would like to see several more people confirming that it did similarly for them. Please test this.

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 i845 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!

[Bryce Harrington (bryce)]

One consideration to take into account, is if we move to UMS for i845, it means we will be entirely unable to get further upstream support on this chip, since any new patches they generate are going to be only for KMS cases. However 8xx has typically not received a high level of support priority upstream anyway, so i845 may be better off in UMS anyway. I have hopes we'll see things better squared away for i8xx KMS in Meerkat, but we'll see...

[Chris Halse Rogers (raof)]

It looks like upstream is not going to have a fix in time for Lucid. What we can do instead is disable KMS for this chip - but we only want to do that if disabling KMS actually makes the driver more stable!

If you are experiencing these GPU freezes, could you please try and see if adding “i915.modeset=0” to your boot options after “quiet splash”.

[jerrylamos (jerrylamos)]

Installed 20100407 which ran for 5 hours with KMS on i845 video graphics and was entering a comment here when screen flashed black then to command line like Ctrl-Alt-F1 a couple times then crashed hard. Could not even ssh in.

I don't think Ubuntu had time to record anything useful but I ran ubuntu-bug xorg see Bug#558613 after booting up.

I'd strongly recommend "i915.modeset=0" default for both i845 and i830. Most of the time over the last year while Ubuntu has been working on KMS it has not worked on these video graphics, or if it does briefly a "dread update" makes them unstable.

If someone wants to turn KMS on and try it, O.K., but in my experience not for an ordinary user as default.

Jerry

[pakraticus (pakraticus)]

Chris Rogers,
The following is my current kludge around. An xorg.conf only containing.

Section "Device"
  Identifier "my-self-configured-device"
  Driver "intel"
 Option "DRI" "Off"
EndSection

GL apps still work, but are painfully slow, but it beats not working at all. The above kludge was actually derived from the troubleshooting documentation.

At this point in time I am seeing the system crash/lock to the affected system is no longer on the network

00:02.0 VGA compatible controller: Intel Corporation 82845G/GL[Brookdale-G]/GE Chipset Integrated Graphics Device (rev 01)
 Subsystem: IBM Device 0267
 Flags: bus master, fast devsel, latency 0, IRQ 16
 Memory at 88000000 (32-bit, prefetchable) [size=128M]
 Memory at 80000000 (32-bit, non-prefetchable) [size=512K]
 Capabilities: <access denied>
 Kernel modules: i915

With i915.modeset=0 added to the kernel command line I keep seeing white vertical bars flash across the top half-third of the screen and iterations of
[ 167.472853] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 1770 at 1733)
[ 169.524015] [drm:i915_gem_idle] *ERROR* hardware wedged
[ 170.647918] [drm:i915_gem_entervt_ioctl] *ERROR* Reenabling wedged hardware, good luck
[ 171.008012] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[ 171.008023] render error detected, EIR: 0x00000000

in dmesg. And the machine stays on the network with the extra parameter.

The problem happens with both 2:2.9.1-3ubuntu2~gomyhr1~clipsolids and 2:2.9.1-3ubuntu1 on 2.6.32-19-generic.

If someone wouldn't mind reminding me how to setup a serial console for debugging this kind of annoyance I'll gladly collect as much debug information as upstream wants.

My two cents of things to do before Lucid ships are
1) Prominently include in the release notes that xserver-xorg-video-intel is broken.
2) Prominently include in the release notes the work around xorg.conf.
3) Repeat #1 and #2 in a recognizeably named file in /usr/share/doc/xserver-xorg-video-intel along with this bug tracking number and upstream's bug tracking number.

If someone is feeling very creative, try having the crash hooks install that stub xorg.conf and reboot the system.

[jerrylamos (jerrylamos)]

At Beta 2 level, with "i915.modeset=0", I'm getting the "hardware wedged" error so often I can't even do a launchpad "Add comment". This is i845 with Lucid 20100407 updated as of 8 April.

I can do ssh in and do "sudo service gdm stop" so Ubuntu isn't completely dead, it just can't run Gnome gdm any length of time at all from boot, and won't restart gdm.

My experience with failures pretty much in line with pakraticus comment above.

Let me try "vesa"

Jerry

[jerrylamos (jerrylamos)]

This Add a comment is being done with i845 with "i915.modeset=0" and "vesa".

Attached is a dump.tar with
i915_error_state
dmesg
Xorg.0.log

Doubt if it has any more info than is already posted.

Recent experience with i915.modeset=0 is flashing black screen and was able to ssh in. Without i915.modeset=0 it crashes hard.

Currently trying vesa...

Jerry

[Chris Halse Rogers (raof)]

Let me try and get this straight:
1) Booting with i915.modeset=0 *does* improve things - the system frequently hits a “hardware wedged” error, but is still responsive to ssh.
2) Without i915.modeset=0 the system will frequently freeze, and fail to respond to ssh.
3) Disabling DRI makes the system freeze less, but it will still freeze.

Have I misinterpreted any part of that?

[useResa (rdrijsen)]

Basically I can only confirm wat has been posted recently.
IMHO the "i915.modeset=0" option added gives me more headaches than WITHOUT.

Without the option I have currently 1 or 2 hard crashes a day, which is not pleasant but I can live with it since my desktop is not my "production" machine.
With the option the situation described above happens and at a much more frequent rate.

[Brian Rogers (brian-rogers)]

I don't expect disabling KMS to help overall. The nature of this bug is that it's timing-dependent, so just about any change can affect the stability for better or worse. On an affected system, I would often get a repeatable freeze on bootup after upgrading an unrelated package just because the new disk layout changed the timing of the boot process.

So you might get reports that disabling KMS helps, but I'd expect it to break as many systems as it helps.

[pakraticus (pakraticus)]

Chris Rogers,

#3) With DRI disabled, I have only experienced the wedge condition a couple times when I switch VTs while a GL application is running. Those have only occurred while unintentionally booted to a backlevel 2.6.32 kernel.

[Odin Hørthe Omdal (velmont)]

Oh, this is happening much more often to me now. :-/

I tried both with KMS on and off. With it off, I see the white zebra stripes as told by another.

How do I disable DRI?

[jerrylamos (jerrylamos)]

With default Lucid Beta, KMS on, this i845 intel video graphics gets black screen hard crash at random intervals. I usually have Firefox up of course.

With "i195.modeset=0", I get random X failures as described above. Lucid is still barely running but it cannot recover the desktop.

With "i915.modeset=0" as well as:

/etc/X11/xorg.conf

Section "Device"
 Identifier "Configured Video Device"
 Driver "vesa"
EndSection

I haven't seen any X failures (yet?).

Jerry.....

[Bryce Harrington (bryce)]

Hrm, bumping all the way down to -vesa seems rather extreme, but if that's what it takes to get it working... Can others confirm whether going with vesa seems like the best approach?

With vesa, that means no 3D acceleration, no kms, and probably no resolutions other than the plain VGA modes (1024x768, etc.)

[Chris Halse Rogers (raof)]

The attached patch disables DRI on i845 and i855 cards. This should make the systems usable, and will be better than falling all the way back to vesa.

[Launchpad Janitor (janitor)]

This bug was fixed in the package xserver-xorg-video-intel - 2:2.9.1-3ubuntu3

---------------
xserver-xorg-video-intel (2:2.9.1-3ubuntu3) lucid; urgency=low

  [Christopher James Halse Rogers]
  * debian/patches/107_disable_dri_on_845_855.patch:
    + Disable DRI on i845 and i855 chips. Works around the stability problems
      these chips have in Lucid (LP: #541492, LP: #541511).
 -- Bryce Harrington <email address hidden> Tue, 13 Apr 2010 15:25:23 -0700

[Geir Ove Myhr (gomyhr)]

Chris, with this way of disabling DRI, it would not be possible to re-enable DRI to test a new kernel without replacing the xserver-xorg-video-intel package. Would it be possible/easy to disable it by default, but let

Section "Device"
  Identifier "my-self-configured-device"
  Driver "intel"
 Option "DRI" "On"
EndSection

in xorg.conf re-enable it? Or is it easier to put up a PPA with an xserver-xorg-video-intel with DRI enabled?

[Geir Ove Myhr (gomyhr)]

Bug reporters, it would be nice to have some confirmation that the workaround works as intended. The expected behaviour is that with 2:2.9.1-3ubuntu3, the freezes will stop and the performance (at least 3D and video) will be horrible.

[Daniel Baumann (dnjl)]

On my system, a Dell X300 with 82852/855GM, this release produces many more lockups. Now it's really really unstable.
Last version (2:2.9.1-3ubuntu1) also lockups sometimes, but not so many times.

[Geir Ove Myhr (gomyhr)]

Daniel, do you have the possibility to ssh into the locked up system and grab the output of `dmesg`, the output of `sudo intel_gpu_dump`, /sys/kernel/debug/dri/0/i915_error_state, and /var/log/Xorg.0.log. This is so that we can check the nature of the new lockups.

Also, you should be commenting on bug 541511 instead of this one, since that is the same symptom as this one on 855GM hardware. It has already been verified that the underlying problems on these two chipsets are different.

[jerrylamos (jerrylamos)]

Geir, is there a bug # for i830? This is a Thinkpad R31 which runs Lucid Beta2 just fine with Driver "vesa". GtkPerf runs in 33 seconds O.K. for a 1 Ghz Celeron.

With DRI off GtkPerf is twice as slow at 65 seconds. Synaptic says xserver-xorg-video-intel 2:2.9.1-3ubuntu3 that's on the R31 as of this morning update. Noticeably sluggish response. The test "Circles" runs at 35 seconds vs. 7 seconds with "vesa".

All this is with i915.modeset=0 of course since it is black screen otherwise, either driver.

Jerry

[jerrylamos (jerrylamos)]

i845 with KMS on and xserver-xorg-video-intel 2:2.9.1-3ubuntu3 ran for a half hour before flashing black screen and hard crash. I was entering a bug comment on launchpad and also running youtube video at the time.

This is 1280x1024 screen.

 I was logged in ssh however that didn't respond since Beta2 was dead stopped.

After booting this time with i915.modeset=0 did get
i915_error_state
Xorg.0.log
dmesg
as attached. This was after booting up after the crash so I'm not sure what use it is.

Will try running with KMS off and xserver-xorg-video-intel 2:2.9.1-3ubuntu3 to see if that combination will crash. I'm logged in with ssh in case it works.

Jerry

[jerrylamos (jerrylamos)]

i845 with KMS off but with xserver-xorg-video-intel 2:2.9.1-3ubuntu3 ran for maybe 20 minutes then X crashed and flashing black screen.

Ubuntu was still up so ssh in while it was constantly flashing black screen fence stripes etc. and got dump of
i915_error_state
Xorg.0.log
dmesg
which are attached.

So with KMS on and intel driver on i845 X fails Beta 2 gets hard crash.

With KMS off and intel driver Beta 2 on i845 X fails but still able to ssh in.

I think I'll switch back to "vesa" since it is faster than "intel" and doesn't crash all day. Let me know if I can test anything else out, looks like 2:2.9.1-3ubuntu3 crashes frequently in case any other info would be useful.

Thanks, Jerry

[jerrylamos (jerrylamos)]

Got another Xorg failure, KMS off, intel driver on i845, and took another dump from ssh I presume it is like the previous comment, so I won't attach it unless someone is interested.

Jerry

[Bryce Harrington (bryce)]

Reopening since this is a master bug; while disabling DRI might have helped, it seems there's still problems.

[Geir Ove Myhr (gomyhr)]

On Wed, Apr 14, 2010 at 5:33 PM, jerrylamos <email address hidden> wrote:
> i845 with KMS off but with xserver-xorg-video-intel 2:2.9.1-3ubuntu3 ran
> for maybe 20 minutes then X crashed and flashing black screen.

Thank you for the report.

I'm wondering if the DRI disablement didn't work as intended. Did you
get the same problem when turning DRI off in xorg.conf as in comment
#27? If you haven't tried, you may use the the
xserver-xorg-video-intel in my standard PPA:
https://launchpad.net/~gomyhr/+archive/standard since that doesn't
have the DRI disablement patch. Could you check if you get the same
erratic behaviour with that?

Looking at the source code, it seems that the way that DRI is disabled
in other cases is by setting pI830->directRenderingType = DRI_DISABLED
in i830_driver.c. But I'm by no means an expert on this.

PS: Jerry, I don't know about a bug report for i830. I think adding
that to the mix here (where we're dealing with i845) will only make
everything more confusing.

[Bryce Harrington (bryce)]

I wish we had more testing evidence to base this decision on, but I've posted a kernel bug report requesting KMS disablement on three of the older 8xx cards: lp #563277

We've already sent up a fair plentitude of bug reports to upstream, so I'm hopeful that they'll come up with fixes to this and to KMS, so we can re-enable in meerkat, or maybe even in 10.04.1, but we'll have to see how things go.

[Rick Spencer (rick-rickspencer3)]

This is an unfortunate situation. There is a non-trivial number of users with 845 and 855 chips who are impacted by a regressions in stability in the current x stack when running 3d and KMS.

We have opted for a "stability first" approach for these users. We will disable 3D and KMS for these chips in Lucid final release. This will have the unfortunate effect of disabling compiz. This will introduce a functional regression. So we will be sacrificing functionality for these users in favor of stability. This is a painful choice to make, but we feel that stability must trump functionality when we are forced to make such choices.

We will be pursuing functional fixes. However, we will do this outside the main release, for example in a PPA. If we are able to provide a fix that delivers stability and functionality, we will consider this a potential SRU in 10.04.1.

[jerrylamos (jerrylamos)]

I'm an "ordinary user" running 1280x1024 on i845 primarily:
Full screen internet email, searches, news, politics health, forums like Ubuntu
Full screen Open Office spreadsheets, taxes etc.
Full screen Picasa digital photographs
Full screen Open Office word documents
Full screen (or nearly so) video's like You Tube, BBC, ABC Australia, Medline Health Day, ...
(I don't do FaceBook or Twitter)
For TV-viewer I run on an ati Express 200 which has had occasional trouble with Ubuntu X but nothing like as much as the intel's.
On my wife's PC we do Skype to Australia.

Now I don't know the insides of Ubuntu, seems to me all Compiz would do in these cases is insert code into the mailine with no help to my applications. Perhaps Compiz might be showy on a more powerful pc with screens bigger than I have, 1280x1024 and 1440x900 and of course 1024x768 on the two laptops.

With KMS on, when X fails so does Ubuntu and I'm at power off/on time. Unstable.

With KMS off, Ubuntu command line is still up for trying to ssh, resurrect the gdm, getting dumps, etc.

Anyway thanks for all the help people are trying to do for us users.

Geir, I'll give a try at the PPA for DRI on/off. Once booted up, is there any way to tell if DRI is on or off?

Jerry

p.s. Stability first. If there are fancy showy options, let them be options not defaults.

[Geir Ove Myhr (gomyhr)]

> Geir, I'll give a try at the PPA for DRI on/off.  Once booted up, is
> there any way to tell if DRI is on or off?

There will usually be a line or two in Xorg.0.log giving it away. In
one of the logs you posted the log stopped before it got that far. If
in doubt, include your Xorg.0.log with your result.

[jerrylamos (jerrylamos)]

Geir,
i845 with
i915.modeset=0
xorg.conf
Section "Device"
        Identifier "Configured Video Device"
# Driver "vesa"
        Driver "intel"
        Option "DRI" "Off"
EndSection
and kernel 2.6.32-21 with xserver-xorg-video-intel 2:2.9.1-3ubuntu3
stayed up all night. Previously without the Option it black screened in a half hour.
GtkPerf was 46.69 seconds much slower than "vesa" with 34.29 seconds.

What benefit do I get for my application centered operations is there with video intel which is by that measure 36% slower than vesa?

Thanks, Jerry

[Geir Ove Myhr (gomyhr)]

On Thu, Apr 15, 2010 at 1:26 PM, jerrylamos wrote:
> Geir,
> i845 with
> i915.modeset=0
> xorg.conf
> Section "Device"
>        Identifier      "Configured Video Device"
> #       Driver          "vesa"
>        Driver          "intel"
>        Option          "DRI" "Off"
> EndSection
> and kernel 2.6.32-21 with xserver-xorg-video-intel 2:2.9.1-3ubuntu3
> stayed up all night.  Previously without the Option it black screened in a half hour.
> GtkPerf was 46.69 seconds much slower than "vesa" with 34.29 seconds.

So in this case, turning DRI off in xorg.conf actually made a
difference, even though it is already disabled. This suggests that the
disabling is not done in the right place. Could you attach your
Xorg.0.log from this session here (which is now Xorg.0.log.old if you
restarted xorg once since you ran this)? I would like to see how the
DRI disabling messages interact.

> What benefit do I get for my application centered operations is there
> with video intel which is by that measure 36% slower than vesa?

For you there will apparently be no reason to run with intel, since
vesa is much better. But then, I suppose you help us with this in
order to help everyone else running ubuntu :-)

The main problem with the VESA driver is that it has a very limited
set of resolutions available. If the native resolution of your monitor
is one of those, then fine. The VESA resolutions are all in 4:3
format, so nobody with a 16:9 monitor will be able to use the native
resolution. I think also high-resolution monitors will have problems.
This is the main reason we don't want to use VESA by default.

[jerrylamos (jerrylamos)]

Geir, I'm retired from IBM where among other things I did system testing, so here to try out new Ubuntu releases I run 4 test pc's i830, i845, ati Mobility 7500, and ati Express 200. Within my linux capabilities and time I can run some tests where people are interested.

Here's the Xorg.0.log since I haven't re-booted yet. I'll do an aptitude update, aptitude safe-upgrade shortly.

Thanks for your explanations. Mostly from what I see Ubuntu changes "just appear" where someone in the know must understand why the changes are being done and what they are good for.

I do multi-boot so I can compare previous Ubuntu's usually when the latest update breaks things. My impression is Lucid Beta 2 is quite responsive to the keyboard with nice boot up and shutdown times and is my preferred version.

If what "vesa" lacks is 16:9 resolution since I'm not a coder it would seem to me to be easier to add that than whatever development effort the "intel" driver is taking. My narrow view I'm the only user so there are just two accounts, root and jerry, and I never switch accounts, so I'm not sure what KMS would do for me. With the amount of work and trouble KMS must have some value to someone.

Thanks, Jerry

[Geir Ove Myhr (gomyhr)]

Thank you for the Xorg.0.log file. It confirms that different things
happen when the xorg.conf option is used.

In the Xorg.0.log from comment #44 (DRI disabled in driver, not in
xorg.conf) there's first
(WW) intel(0): i845 or i855 chip detected. Disabling DRI to prevent
system instability.
then a bit later
(II) intel(0): direct rendering: Failed
while in Xorg.0.log from the previous comment (DRI disabled in driver
and in xorg.conf), the first line does not show up and then
(II) intel(0): direct rendering: Disabled

The difference between Failed and Disabled is that in the first case
pI830->directRenderingType is DRI_NONE and in the second case it is
DRI_DISABLED (messages are from in I830ScreenInit() in the driver).

This part of I830PreInit() is skipped when directRenderingType is
DRI_DISABLED, but as far as I understand the current patch only
disables DRI inside the dri2 module that is loaded here.

   /* Load the dri2 module if requested. */
   if (xf86ReturnOptValBool(pI830->Options, OPTION_DRI, FALSE) &&
       pI830->directRenderingType != DRI_DISABLED) {
       xf86LoadSubModule(pScrn, "dri2");
   }

I hope this digging can be of some help to the developers.

[jerrylamos (jerrylamos)]

Oops, after 14 hours being on, I was composing a Yahoo message and X failed.

This is a continuation of Comment #53. Lucid Beta 2 had stayed up about 16 hours.

Since KMS was not on, I was able to ssh in and get a dump with i915 error state, Xorg.0.log, dmessage. With KMS on, when X fails, nothing will run not even ssh.

Rebooting took a couple tries. Even though /etc/default/grub specifies i915.modeset=0 this is NOT on the recovery mode kernel line so it comes up with KMS on, and of course startx fails on this i845.

Jerry

[Geir Ove Myhr (gomyhr)]

So with that comment, I suppose we can no longer say that there is a difference between using DRI "off" in xorg.conf and not with 2:2.9.1-3ubuntu3(?)

[Bryce Harrington (bryce)]

On Thu, Apr 15, 2010 at 11:59:24AM -0000, Geir Ove Myhr wrote:
> The main problem with the VESA driver is that it has a very limited
> set of resolutions available. If the native resolution of your monitor
> is one of those, then fine. The VESA resolutions are all in 4:3
> format, so nobody with a 16:9 monitor will be able to use the native
> resolution. I think also high-resolution monitors will have problems.
> This is the main reason we don't want to use VESA by default.

I wonder if most of the i8xx based laptops are 4:3 resolutions? If so,
then the main place this would show up as an issue is with external
monitors.

[Bryce Harrington (bryce)]

On Thu, Apr 15, 2010 at 12:46:49PM -0000, jerrylamos wrote:
> If what "vesa" lacks is 16:9 resolution since I'm not a coder it would
> seem to me to be easier to add that than whatever development effort the
> "intel" driver is taking.

It's not so simple. All monitors are required to support certain
standard resolutions, so we can always just assume those resolutions
exist. These are the VESA resolutions (VESA is the name of the
standards body that specified them, and this was done well before
monitors started being HD, so they're all 4:3).

Outside this, monitor manufacturers are free to implement any other
additional resolutions, and the timings for those are stored in the
monitor itself in a blob of firmware data called EDID, which certain
video drivers can query to learn more about available resolutions.

Anyway -vesa does not have any of that probing stuff, it just hard codes
the VESA resolutions it knows will be there.

> With the amount of work and trouble KMS must have some value to someone.

It enables various functionalities to be done at a lower level closer to
the hardware. Some of the benefits of this are showing up now, such as
improving boot behavior. Other benefits that depend on KMS are in the
pipelines but won't show up for a while yet, like being able to run X as
a non-root user, or being able to have external monitors automatically
re-configure themselves when plugged in, and so on.

[jerrylamos (jerrylamos)]

With xorg.conf specifying intel driver and DRI Off, the following are excerpted from the Xorg.0.log after the X failure:

Xorg.0.log:
.......
(II) Loading extension XFree86-DRI
(II) LoadModule: "dri2"
(II) Loading /usr/lib/xorg/modules/extensions/libdri2.so
(II) Module dri2: vendor="X.Org Foundation"
 compiled for 1.7.6, module version = 1.1.0
 ABI class: X.Org Server Extension, version 2.0
.....
(II) Loading extension DRI2
......
(**) intel(0): Option "DRI" "Off"
......
(II) intel(0): 0x007df000-0x07ffefff: DRI memory manager (123008 kB)
(II) intel(0): 0x07fff000-0x07ffffff: overlay registers (4 kB, 0x000000002ac4b000 physical
)
(II) intel(0): 0x08000000: end of aperture
(II) intel(0): BO memory allocation layout:
(II) intel(0): 0x007df000: start of memory manager
(II) intel(0): 0x00800000-0x00ffffff: front buffer (8192 kB) X tiled
(II) intel(0): 0x01000000-0x01004fff: HW cursors (20 kB)
(II) intel(0): 0x07fff000: end of memory manager
(WW) intel(0): ESR is 0x00000001, instruction error
(WW) intel(0): Existing errors found in hardware state.

On the surface it says the DRI module is loaded, it sees the "Off" option, then it does have DRI memory manager allocated anyway. Is DRI running or not?

I assume (II) messages are information on startup.
The (WW) look like messages from the crash.

The i915 error state says

Time: 1271353623 s 200345 us
EIR: 0x00000000
  PGTBL_ER: 0x00000000
  INSTPM: 0x00000000
  IPEIR: 0x00000000
  IPEHR: 0x18000001
  INSTDONE: 0x01ffffc1
  ACTHD: 0x07a95000

I have no clue on that.

The dmesg says
[61077.272028] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[61077.272047] render error detected, EIR: 0x00000000
[61079.508044] [drm:i915_gem_idle] *ERROR* hardware wedged
[61082.601057] [drm:i915_gem_entervt_ioctl] *ERROR* Reenabling wedged hardware, good luck
[61082.940014] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[61082.940025] render error detected, EIR: 0x00000000
[61082.941401] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 599825 at 599822)
[61085.008028] [drm:i915_gem_idle] *ERROR* hardware wedged
[61086.245669] [drm:i915_gem_entervt_ioctl] *ERROR* Reenabling wedged hardware, good luck
.....

Any way to tell if "DRI" was responsible for the "GPU hung"?

Let me give a try with i915.modeset=0 and default video driver on this i845...

Thanks, Jerry

[jerrylamos (jerrylamos)]

Bryce, thanks much for the explanation. With more and more dependence on X then stability in the X code is even more important.

Just booted i915.modeset-0 and default video driver:
Xorg.0.log
....
(WW) intel(0): i845 or i855 chip detected. Disabling DRI to prevent system instability.
(**) intel(0): Framebuffer compression disabled
......
(II) AIGLX: Screen 0 is not DRI2 capable
(II) AIGLX: Screen 0 is not DRI capable
.....
(II) Loading extension XFree86-DRI
(II) LoadModule: "dri2"
(II) Loading /usr/lib/xorg/modules/extensions/libdri2.so
(II) Module dri2: vendor="X.Org Foundation"
 compiled for 1.7.6, module version = 1.1.0
 ABI class: X.Org Server Extension, version 2.0
(II) Loading extension DRI2
(==) Matched intel as autoconfigured driver 0
....
(II) LoadModule: "intel"
(II) Loading /usr/lib/xorg/modules/drivers/intel_drv.so
(II) Module intel: vendor="X.Org Foundation"
 compiled for 1.7.6, module version = 2.9.1
 Module class: X.Org Video Driver
 ABI class: X.Org Video Driver, version 6.0
.....

On the surface seems to be what is expected except does load DRI modules anyway for some reason?

Jerry

[jerrylamos (jerrylamos)]

Bryce, referring to comment #59, this is a 1280:1024 monitor ration 5:4 or 1.25 a bit different from 1024:768 which has 4:3 or 1.33. Vesa has no problem with the 5:4. I do have a 1440:900 or 1.6 in the other room if I really got energetic I could tug it in here.....

Running i915.modeset=0 and default intel driver, right now the 3ubuntu4 version, running O.K. Got bored so I tried Glasen's 2:2.11.0+git20100416~glasen~ppa1 intel video driver. That ran for a couple minutes then the gdm froze. Ctrl-Alt-F1 worked so I took a dump of i915 error, dmesg, Xorg.0.log in case anyone wants it.

Jerry

[jerrylamos (jerrylamos)]

i845 GPU hung with latest update after about 4 minutes run time from boot. Looks similar to GPU hang after an hour and 44 minutes:

xorg-server 2:1.7.6-2ubuntu5 (Bryce Harrington <email address hidden>)

KMS is not on, i915.modeset=0, there is no xorg.conf so it is default intel driver.

Lets see if I can post this before X crashes again.

Jerry

[Felix Lechner (felix.lechner)]

My 845G also still crashes with the white stripes after setting 'i915.modeset=0'. Please let me know if I can contribute to the testing of any workarounds.

Only have one computer though, so no SSH.

[jerrylamos (jerrylamos)]

Bryce, Comment # 60, I tried "vesa" on Compaq with 1440:900 ration 16:10 ran fine gdm, FireFox, YouTube flash video. TV-Viewer didn't run because it couldn't find /dev/video0.

Preferences Monitor said it was unknown but did get the resolution correct at 1440:900. Visual appearance just like using ati driver.

GtkPerf ran 12.34 seconds with "vesa" which is faster than ati video driver at 14.01 seconds. It's ati Radeon Xpress 200.

Only reason I bring this up is over the years with Ubuntu "X" has not been stable many many times, and apparently the direction of Ubuntu is to depend more and more on "X" with KMS and you'd know what else is coming.

Jerry

[Purgatori (purgatori-sakkara)]

After implementing various workarounds that failed to resolve the issue, I reverted back to version 2.8.0-27 of the xserver-xorg-video-intel package. Since doing so, I have not experienced another crash/freeze.

[Brian Rogers (brian-rogers)]

I have a kernel available in this PPA that might help: https://launchpad.net/~brian-rogers/+archive/experimental

[c_ellesley (archelegraph)]

intel 855GM 1024x768 lucid "lubuntu"
up to kernel 2.6.32.20 I was running with nosplash nomodeset using the fbdev driver (vesa doesn't work)
2.6.32.21 doesn't work anymore
2.6.34rc3-51.2 works great with the intel driver (but I need the fsam7400 module to switch my wireless on)
attached kern.log shows boot failures an the succesful init with the experimental kernel

[jerrylamos (jerrylamos)]

i845 video graphics Lucid Beta 2 trying
Brian Roger's experimental kernel:

Linux version 2.6.34rc3-51-generic (buildd@radon) (gcc version 4.4.3 (Ubuntu 4.4.3-4ubuntu5) ) #2-Ubuntu SMP Fri Apr 16 13:31:29 UTC 2010

No xorg.conf.
KMS is up, quiet splash.
ssh in from a Thinkpad T40 just in case. Previously if X fails, KMS fails, and Lucid is dead altogether and ssh can't work but we'll see.

Running two Firefox windows, one with YouTube video, other this "Add comment".
Also tried GtkPerf at the same time. Kinda sluggish as would be expected but still ran.

Downloading Picasa3 from Google. Went out of the room.

Oops, crashed all the way down to power off, the ssh got a broken pipe.

Tried booting a couple times with KMS and intel driver, got the complaint about low graphics mode and wouldn't run even though it had earlier.
Tried booting with i915.modeset=0 and intel driver, got the complaint about low graphics mode but got to a login and tried dumping some logs, attached.

Tried booting with kernel 2.6.32-21 and intel driver, crashed while trying to Add a comment on launchpad.

Now booted up with experimental kernel, i915.modeset=0, and "vesa" driver. I couldn't get it up without "vesa".

Jerry

[Bryce Harrington (bryce)]

I posted about the current status of the 8xx problems onto the i855
master bug, and it's equally relevant to i845.

It bears repeating here though: take care to know that this bug is very
conditions sensitive, so what might appear to "fix" it for you might not
fix it for another user with identical hardware to you, and might cause
massive breakage to someone else. There aren't going to be silver
bullets here, but there'll be plenty of lead.

I find Michael's observation that 2.8.0 resolved it quite interesting,
and would like to hear other 8xx owner's findings when reverting to
that. Might give another way to work around the problems, and/or a lead
on what caused them to begin with.

On Fri, Apr 16, 2010 at 10:49:01PM -0000, jerrylamos wrote:
> Only reason I bring this up is over the years with Ubuntu "X" has not
> been stable many many times, and apparently the direction of Ubuntu is
> to depend more and more on "X" with KMS and you'd know what else is
> coming.

[jerrylamos (jerrylamos)]

What steps are used to revert back to version 2.8.0-27 of the xserver-xorg-video-intel package?

I looked at https://wiki.ubuntu.com/ReinhardTartler/X/RevertingIntelDriverTo2.4
but couldn't figure out how to use the directions, example where did "0xce90d8983e731f79" come from?

Thanks, Jerry

[Michael Rickmann (mrickma)]

Yesterday I installed Ubuntu 10.04 on an elderly Dell desktop equipped with 82845G on board graphics and was hit by the stuck zebra crash reported here about every 5 min. The xserver-xorg-video-intel was disabling DRI when it recognized the chipset. The i915.modeset=0 workaround really worked for me - I sucessfully tested for more than one hour. Then I tried reverting the driver by installing http://ppa.launchpad.net/brian-rogers/graphics-testing/ubuntu/pool/main/x/xserver-xorg-video-intel/xserver-xorg-video-intel_2.8.0-27-g926c7e7+lucid_i386.deb . This driver works with KMS for more than 3 hours now and the OpenGL performance is more than twice of the DRI disabled one. So for me there are two workarounds possible.
Thanks
Michael

[Matt Toups (mtoups)]

Hi, I realize that this may or may not be useful information, but a while ago I was tired of the frequent crashes on my i845, and I noticed that they started happening after I rebooted for a kernel update. After that the crashes were very frequent.

So I tried going back to kernel 2.6.32-10-generic and I haven't had a crash -- current uptime is 28 days.

I don't know if this is a coincidence, but in case this is useful, here it is. Thanks for working on this tough problem!

[jerrylamos (jerrylamos)]

Installed latest Ubuntu build 20100419, has not been updated last 2 days.

i845 ran for 1.5 hours before GPU lockup with quiet, splash, default intel driver.

Only benefit was even with KMS the login screen stayed up so I could ssh in. I have the usual dump but it looks like there are lots already.

I presume i915.modeset=0 and "vesa" will work as usual but let me try 2.8.0-27 as in comment #73.

Jerry

[jerrylamos (jerrylamos)]

i845 Lucid install of today's Daily Build has been running
http://ppa.launchpad.net/brian-rogers/graphics-testing/ubuntu/pool/main/x/xserver-xorg-video-intel/xserver-xorg-video-intel_2.8.0-27-g926c7e7+lucid_i386.deb
for over three hours now....KMS, quiet, splash. sure did NOT do that with daily build as installed.

Wasn't intel driver 2.8.0 on Karmic last summer? Looks like Brian Rogers has it packaged for Lucid.

Thanks so far.....

Jerry

[Purgatori (purgatori-sakkara)]

Yep, the above ppa is what I used to to revert the driver back as well -- sorry for not replying sooner. For those wishing to test whether reverting will help them, I recommend that you also have a look here: http://www.ubuntugeek.com/intel-graphics-performance-guide-for-ubuntu-904-jaunty-users.html -- and read the instructions on how to configure xorg.conf to improve performance, as I did encounter quite a major drop in performance after reverting from the driver in Lucid (which worked great, apart from the crashing), and I figure that others might too.

On another note, it has been several days now, and still no crashes or hangs. Hopefully if more people test this potential fix and report on the results, it might throw some light on a more permanent fix, as Bryce said.

[Purgatori (purgatori-sakkara)]

Well, so much for that. Rebooted my system about 1/2hr ago, and as soon as I load up scrotwm, I can see screen corruption (horizontal bar filled with white noise), and then X locked up completely. Tried re-jigging my xorg.conf (had previously been mucking around with vesa and fbdev), but to no avail... every time I loaded X, the same thing occured. Eventually, after removing my xorg.conf altogether, I concluded that something was awry with the driver itself, and so I removed it, and installed the one from the Lucid repos... which means I'll be due for a crash anytime soon, but at least it works for more than a few minutes.

I'm kinda at a loss here. I don't understand how the karmic driver worked so well for several days, and then bam, it decides to go crazy.

[jerrylamos (jerrylamos)]

i845 ran for a total of 5 hours no problem with intel driver 2.8.0 see comment $76. I didn't try exercising KMS but it was up, with boot parameters "quiet, splash".

i830 running Lucid updated as of 20100420 with intel driver 2.8.0 for two hours O.K. It won't run hardly at all with the Lucid level intel driver. Vesa works fine. I don't think KMS is active with i830, will check to see.

Jerry

[jerrylamos (jerrylamos)]

KMS not active on i845 and i830 as far as I can tell (I know little about it) with driver 2.8.0.

Still running fine.

Jerry

[Bryce Harrington (bryce)]

On Thu, Apr 22, 2010 at 01:03:56PM -0000, Michael Audas wrote:
> I'm kinda at a loss here. I don't understand how the karmic driver
> worked so well for several days, and then bam, it decides to go crazy.

Well, the fact is that xserver-xorg-video-intel is not like an
all-in-one driver, it depends on lots of stuff from other packages like
mesa, libdrm, the kernel, etc. With KMS this is even moreso true.

This is why I found downgrading to 2.8.0 to be intriguing. But the
further testing suggests it was just tweaking the conditions. Like I
mentioned this bug is very sensitive to conditions, so it's not
unexpected that you could have a working system, change one thing,
change it back to how it was (as far as you know), and no longer have a
working system.

[Bryce Harrington (bryce)]

On Thu, Apr 22, 2010 at 06:47:17PM -0000, jerrylamos wrote:
> KMS not active on i845 and i830 as far as I can tell (I know little
> about it) with driver 2.8.0.
>
> Still running fine.

Dunno if 2.8.0 is going to be an adequate solution for most people, but
figured it's worth at least making it easy for people to test, so I've
dusted off an old 2.8.0 package of mine (still not sure where the
aforementioned -27 comes from) and pushed it to x-retro:

  https://edge.launchpad.net/~ubuntu-x-swat/+archive/x-retro

(I don't plan to maintain this since it's such old code, but if it's
completely broken do let me know so I can pull it.)

[Bryce Harrington (bryce)]

On Wed, Apr 21, 2010 at 05:49:38PM -0000, Matt Toups wrote:
> So I tried going back to kernel 2.6.32-10-generic and I haven't had a
> crash -- current uptime is 28 days.

Can you write up what you did (step-by-step directions would be cool) to
this wiki page?

  https://wiki.ubuntu.com/X/Bugs/Lucidi8xxFreezes

It could help others wishing to work around this until we find a proper
solution.

[Felix Lechner (felix.lechner)]

The x-retro package worked well for a day. Then there were a couple of crashes in rapid succession. One thing is definitely better: The X server restarts and offers to run in low res mode. So no more zebra stripes. I can also get to the gettys.

I can supply more information next time if someone points out the relevant items.

[jerrylamos (jerrylamos)]

i845 worked O.K. since putting retro on 4/22. Just now booted up was doing an aptitude update with Firefox up and black screened no zebra stripes. ssh worked so tar is attached.

Fatal server error:
Failed to submit batchbuffer: Input/output error

Does not look like the usual GPU lockup.

Jerry
?field.comment=i845 worked O.K. since putting retro on 4/22. Just now booted up was doing an aptitude update with Firefox up and black screened no zebra stripes. ssh worked so tar is attached.

Fatal server error:
Failed to submit batchbuffer: Input/output error

Does not look like the usual GPU lockup.

Jerry

[jerrylamos (jerrylamos)]

On previous comment, GPU is hung see excerpt from dmesg:

[ 98.664029] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[ 98.664048] render error detected, EIR: 0x00000000
[ 98.664092] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 3401 at 3400)
[ 100.344064] [drm:i915_gem_idle] *ERROR* hardware wedged
[ 102.518761] [drm:i915_gem_entervt_ioctl] *ERROR* Reenabling wedged hardware, good luck

Was not successful at sudo service gdm stop, sudo startx.

Jerry

[timosha (timosha)]

Here are my five cents:

On an IBM Thinkpad R51 - 2887 with:
VGA compatible controller: Intel Corporation 82852/855GM Integrated Graphics Device (rev 02)

and kernel 2.6.32-21 + xserver-xorg-video-intel 2:2.9.1-3Ubuntu1 everything works perfectly, including Compiz. No X crahes at all. So, I pinned xserver-xorg-video-intel 2:2.9.1-3Ubuntu1 to avoid that the "workaround" xserver-xorg-video-intel 2:2.9.1-3Ubuntu3 gets installed and disables DRI.

[Launchpad Janitor (janitor)]

This bug was fixed in the package xserver-xorg-video-intel - 2:2.9.1-3ubuntu5

---------------
xserver-xorg-video-intel (2:2.9.1-3ubuntu5) lucid; urgency=low

  * Drop debian/patches/107_disable_dri_on_845_855.patch:
    + This attempt to work around the crashes on i845 and i855 documented in
      LP: #541492 and LP: #541511 simply shuffled the brokeness around. It
      hasn't helped enough, and it unconditionally disables 3D which worked for
      some users before.
 -- Christopher James Halse Rogers <email address hidden> Mon, 26 Apr 2010 10:14:02 +1000

[alexg (asg1290)]

I just upgraded to the latest "fix" and I am still having this problem

dpkg -l|grep xorg-video-intel
ii xserver-xorg-video-intel 2:2.9.1-3ubuntu5

[ 2397.948014] [drm:i915_gem_idle] *ERROR* hardware wedged
[ 2399.064724] [drm:i915_gem_entervt_ioctl] *ERROR* Reenabling wedged hardware, good luck
[ 2399.392014] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[ 2399.392026] render error detected, EIR: 0x00000000
[ 2399.392887] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 39152 at 38087)
[ 2401.448014] [drm:i915_gem_idle] *ERROR* hardware wedged
[ 2402.563316] [drm:i915_gem_entervt_ioctl] *ERROR* Reenabling wedged hardware, good luck
[ 2402.892012] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[ 2402.892023] render error detected, EIR: 0x00000000
[ 2402.892872] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 39154 at 38087)

[Ralf Schulze (ralf-schulze)]

I can confirm that the bug is still present:

!> dpkg -l|grep xorg-video-intel
ii xserver-xorg-video-intel 2:2.9.1-3ubuntu5 X.Org X server -- Intel i8xx, i9xx display driver

~> dmesg |grep drm
[ 0.000000] Linux version 2.6.32-21-generic (buildd@yellow) (gcc version 4.4.3 (Ubuntu 4.4.3-4ubuntu5) ) #32-Ubuntu SMP Fri Apr 16 08:09:38 UTC 2010 (Ubuntu 2.6.32-21.32-generic 2.6.32.11+drm33.2)
[ 10.173005] [drm] Initialized drm 1.1.0 20060810
[ 10.388609] [drm] MTRR allocation failed. Graphics performance may suffer.
[ 10.388709] [drm] set up 31M of stolen space
[ 11.337849] fb0: inteldrmfb frame buffer device
[ 11.388092] [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0
[79060.540065] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[79060.540143] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 11055012 at 11055011)
[79061.322511] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[79061.322526] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 11055016 at 11055011)
[79062.112728] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[79062.112801] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 11055023 at 11055011)
[79062.902728] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[79062.902793] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 11055035 at 11055011)
[79063.682728] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[79063.682794] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 11055040 at 11055011)

[jerrylamos (jerrylamos)]

Strange. i845 video. I have 2 Lucid partitions both updated to 26 April for all packages including the xserver-xorg-video-intel 2:2.9.1-3ubuntu5 and both with just "quiet" in the boot line

KMS was on (?) in the one from RC. It was up 4 hours with intermittent use and went black screen as I was attempting to Add comment on this bug.

KMS was off on the one updated from Alpha. I haven't used it as much.

With 3ubuntu5 should KMS default on or off for an i845?

Jerry

[Abe Lau (abelau+launchpad)]

Here is the combinations I have tried (with options i915 modeset=1 & with options i915 modeset=0 both tested)

and all failed after a while (from 5 mins to 2 hours):

--With Lucid RC install
--With xserver-xorg-video-intel 2:2.9.1-3-ubuntu4
--With xserver-xorg-video-intel_2.8.0-27-g926c7e7+lucid_i386.deb as supplied in #76
--With xserver-xorg-video-intel 2:2.9.1-3-ubuntu5

[Christian (c-pradelli)]

I'm having this bug in all my old onboard intel 845 cpu. One workaround I found is to set acpi=off in boot kernel option. This completely fix the video problems.

[Geir Ove Myhr (gomyhr)]

> With 3ubuntu5 should KMS default on or off for an i845?

Jerry, the KMS defaults are set in the kernel, not the DDX driver.
With 2.6.32-20 KMS was on by default, and with 2.6.32-21 it was turned
off on i830, i845, and i855.

One complicating factor is that xserver-xorg-video-intel (the DDX
driver) at version 2.9.1-3ubuntu3 and earlier included a file that
would override the kernel default. Therefore, with 3ubuntu3 and
earlier, KMS would be on by default for any kernel, while 3ubuntu4 and
later will let the kernel decide.

[jerrylamos (jerrylamos)]

Lucid nstable on i845. I've tried everything in this launchpad bug except 2.6.32-10 and all attempts fail except "vesa".

Anyone know how to install the package for 2.6.32-10? I haven't found it on any of the ubuntu URL's that I've tried.

Thanks, Jerry

[Bryce Harrington (bryce)]

On Tue, Apr 27, 2010 at 08:56:43PM -0000, jerrylamos wrote:
> Lucid nstable on i845. I've tried everything in this launchpad bug
> except 2.6.32-10 and all attempts fail except "vesa".

Jerry, are you indicating that turning KMS off did not fix the problem
for you on your systems?

I had understood from your earlier reports that with KMS turned off your
systems would work... if this is not correct we need to re-think the KMS
blacklist.

[linuxcentre (linuxcentre)]

I get exactly the same as #90. Same version, same errors. Tried disabling KMS, acpi=off, disabling DRI,
Using a Dell 2400 desktop equipped with 82845G on board graphics.
Only vesa mode proves stable so far.

[kshitijmmmec (jalashu-tripathi)]

I am using lucid lynx from its release candidate version on my i845 architecture desktop. I am regularly facing crashes due to black screen on the the monitor. Also i could not enable my visual effects to either "Normal" or "Extra" as it starts searching for available drivers and reports "Desktop effects could not be enabled".

[hfzorman (hfzorman)]

I can confirm linuxcentre (his machine is the same as mine) and kshitijmmmec. The crashes keep going on and sometimes I get the very frequently others I can be using the computer for hours without problems. Vesa proves stable.

[Michael Rickmann (mrickma)]

A new combination for testing, perhaps it is a sign for an upcoming fix: I am writing this on a 82845G/GL (rev 01) under Lucid with the following software installed:
linux-image-2.6.34-997-generic_2.6.34-997.201004191525_i386.deb from http://kernel.ubuntu.com/~kernel-ppa/mainline/drm-intel-next/2010-04-19-lucid/
libdrm-intel1_2.4.20~glasen~ppa1_i386.deb and libdrm2_2.4.20~glasen~ppa1_i386.deb from http://ppa.launchpad.net/glasen/libdrm/ubuntu/pool/main/libd/libdrm/
xserver-xorg-video-intel_2.11.0+git20100427~glasen~ppa1_i386.deb from http://ppa.launchpad.net/glasen/intel-driver/ubuntu/pool/main/x/xserver-xorg-video-intel/
I will report tomorrow whether this combination can run a gl-screensaver overnight. So far none of the workarounds did pass that test on my machine.

[hfzorman (hfzorman)]

KMS has been blacklisted in today's kernel updates. Look:

"* SAUCE: i915 KMS -- blacklist i845g
    - LP: #541492, #563277"

I'm going to install and check what happens.

[hfzorman (hfzorman)]

Ok, after a few hours I let the computer alone and found it crashed, constantly trying to turn on and off X.org. So the KMS blacklist attempt on the new Linux Kernel didn't worked.

[jdorwart (jdorwart)]

No matter how far I back up to older kernels (within the Lucid family) I
still get crashes of X that require power off and power back on. I have
tried, after enabling the key combos, to cnt - alt - bcksp to restart X but
that does not seem to do anything either. There does not appear to be any
rhyme or reason. I always get a good start up. IT may be an hour or a
whole day. I may click the address bar in Chromium or I may try to save in
Gimp. Any action (always seem to be when I click the mouse for something) I
can get a crash. Never a guarantee either way.

I am working on a Dell SX260 mini desktop with I845.

jeff

On Thu, May 6, 2010 at 2:54 PM, hfzorman <email address hidden> wrote:

> Ok, after a few hours I let the computer alone and found it crashed,
> constantly trying to turn on and off X.org. So the KMS blacklist attempt
> on the new Linux Kernel didn't worked.
>
> --
> MASTER: [i845] GPU lockup (apport-crash) (Should KMS be blacklisted?)
> https://bugs.launchpad.net/bugs/541492
> You received this bug notification because you are a direct subscriber
> of a duplicate bug.
>

--
Jeffrey Dorwart
U.S. Coast Guard
Aviation Engineer
CG-711/DD Airstation Clearwater FL.
[phone] cell

ASW-20 N26307
A&P/IA

[hfzorman (hfzorman)]

@jdorwart: Your problem is the same as mine. I have tried Workaround G from https://wiki.ubuntu.com/X/Bugs/Lucidi8xxFreezes and it's working well. Shall you try?

[Brian Rogers (brian-rogers)]

The kernel patch from this bug might help:
https://bugs.freedesktop.org/show_bug.cgi?id=27187

I placed a kernel with the newest version of the patch in my experimental PPA:
https://launchpad.net/~brian-rogers/+archive/experimental

If it helps, you can provide feedback in the upstream bug.

[jdorwart (jdorwart)]

I will try tonite. Let you know if it works.&nbsp;

Thanks.
Jeff

-- Sent from my Palm Pre
On May 11, 2010 12:15 AM, hfzorman &lt;<email address hidden>&gt; wrote:

@jdorwart: Your problem is the same as mine. I have tried Workaround G

from https://wiki.ubuntu.com/X/Bugs/Lucidi8xxFreezes and it's working

well. Shall you try?

--

MASTER: [i845] GPU lockup (apport-crash) (Should KMS be blacklisted?)

https://bugs.launchpad.net/bugs/541492

You received this bug notification because you are a direct subscriber

of a duplicate bug.

[Sam Diener (sam-samdiener)]

Hey all,

I know I am late joining in this bug but I have been following closely.
Running Ubuntu 10.04.

Using workaround G from hforzman above. System runs a lot longer but now instead of the horizontal bars/zebra lockup... I get a black screen freeze.

Hardware:

0:02.0 VGA compatible controller: Intel Corporation 82845G/GL[Brookdale-G]/GE Chipset Integrated Graphics Device (rev 01)

Kern.log:

ay 19 07:39:28 sam-desktop kernel: [ 17.206726] vboxnetadp: no symbol version for RTR0Init
May 19 07:39:28 sam-desktop kernel: [ 17.206733] vboxnetadp: Unknown symbol RTR0Init
May 19 07:39:28 sam-desktop kernel: [ 17.207060] vboxnetadp: no symbol version for RTErrConvertFromErrno
May 19 07:39:28 sam-desktop kernel: [ 17.207064] vboxnetadp: Unknown symbol RTErrConvertFromErrno
May 19 07:39:28 sam-desktop kernel: [ 17.207200] vboxnetadp: no symbol version for RTErrConvertToErrno
May 19 07:39:28 sam-desktop kernel: [ 17.207203] vboxnetadp: Unknown symbol RTErrConvertToErrno
May 19 07:39:28 sam-desktop kernel: [ 17.207408] vboxnetadp: no symbol version for RTLogRelDefaultInstance
May 19 07:39:28 sam-desktop kernel: [ 17.207411] vboxnetadp: Unknown symbol RTLogRelDefaultInstance
May 19 07:39:28 sam-desktop kernel: [ 17.207931] vboxnetadp: no symbol version for RTR0Term
May 19 07:39:28 sam-desktop kernel: [ 17.207934] vboxnetadp: Unknown symbol RTR0Term
May 19 07:39:29 sam-desktop kernel: [ 18.603219] ADDRCONF(NETDEV_UP): wlan0: link is not ready
May 19 07:39:29 sam-desktop kernel: [ 18.634640] render error detected, EIR: 0x00000010
May 19 07:39:29 sam-desktop kernel: [ 18.634650] [drm:i915_handle_error] *ERROR* EIR stuck: 0x00000010, masking
May 19 07:39:29 sam-desktop kernel: [ 18.634664] render error detected, EIR: 0x00000010
May 19 07:39:36 sam-desktop kernel: [ 25.501753] ADDRCONF(NETDEV_UP): wlan0: link is not ready
May 19 07:39:36 sam-desktop kernel: [ 25.671231] ADDRCONF(NETDEV_UP): wlan0: link is not ready
May 19 07:39:39 sam-desktop kernel: [ 29.011757] wlan0: direct probe to AP 00:23:69:ac:bb:e6 (try 1)
May 19 07:39:39 sam-desktop kernel: [ 29.014596] wlan0: direct probe responded
May 19 07:39:39 sam-desktop kernel: [ 29.014601] wlan0: authenticate with AP 00:23:69:ac:bb:e6 (try 1)
May 19 07:39:39 sam-desktop kernel: [ 29.016341] wlan0: authenticated
May 19 07:39:39 sam-desktop kernel: [ 29.016371] wlan0: associate with AP 00:23:69:ac:bb:e6 (try 1)
May 19 07:39:39 sam-desktop kernel: [ 29.018709] wlan0: RX AssocResp from 00:23:69:ac:bb:e6 (capab=0x411 status=0 aid=3)
May 19 07:39:39 sam-desktop kernel: [ 29.018713] wlan0: associated
May 19 07:39:39 sam-desktop kernel: [ 29.019666] ADDRCONF(NETDEV_CHANGE): wlan0: link becomes ready
May 19 07:39:40 sam-desktop kernel: [ 29.044046] padlock: VIA PadLock not detected.
May 19 07:39:40 sam-desktop kernel: [ 29.824221] __ratelimit: 9 callbacks suppressed
May 19 07:39:40 sam-desktop kernel: [ 29.824226] type=1503 audit(1274269180.800:15): operation="open" pid=1258 parent=1222 profile="/sbin/dhclient3" requested_mask="r::" denied_mask="r::" fsuid=0 ouid=0 name="/var/lib/wicd/dhclient.conf"
May 19 07:39:50 sam-desktop kernel: [ 39.632010] wl...

[hfzorman (hfzorman)]

Thanks Sam Diener for letting me know what after all it freezes. It has happened also to me that X.org crashed but I wanted to make sure it wasn't me only so I could edit the wiki. The last crash I had was trying using a game in Lucid and I got this error:

starting postfix Mail Transport Agend postfix
[113.644035] [drmi915_hangcheck_elapsed]
*Error* Hangcheck timer elapsed... GPU hung
[113.644118] render error detected, EIR: 0x80's

[Christian (c-pradelli)]

I'm still having this problem in my 20 LTSP clients that have i845 onboard card, they randomly crash up after 2 or 3 hours working. Setting acpi=off in kernel clients solves the problem for me, has anybody tried this?

[nunogt (nunogt)]

Christian, I've tried acpi=off and it does not solve the problem. It will crash as randomly and frequently as before.

[Andy (andy-xillean)]

I am having this problem on 2 new installs today. One is a HP P4 and the other is a Emachines P4 celeron. Fresh installs of 10.04 and black screen flashes. No 3D and only vesa driver works on the HP.

[Darxus (darxus)]

What's the short term fix for people who just want their computer to stop crashing?

Anything less drastic than installing jaunty?

[Odin Hørthe Omdal (velmont)]

On Wed, May 26, 2010 at 6:53 PM, Darxus <email address hidden> wrote:
> What's the short term fix for people who just want their computer to
> stop crashing?
>
> Anything less drastic than installing jaunty?

Agreed. I've tried VESA, but it doesn't work. I can't get any picture.
Guess my screen isn't supported.

--
Beste helsing,
Odin Hørthe Omdal <email address hidden>
http://velmont.no

[Brian Rogers (brian-rogers)]

There is a kernel patch in this upstream bug: https://bugs.freedesktop.org/show_bug.cgi?id=27187

Since 2.6.34 has been released and this patch also seems pretty stable, I've uploaded the final 2.6.34 release with drm-intel-next and this patch to my graphics-fixes PPA. It's based on Maverick's current kernel and includes Ubuntu-specific patches, unlike a typical upstream kernel build. The 32-bit build is building right now and should be available within a few hours.

The PPA is here: https://launchpad.net/~brian-rogers/+archive/graphics-fixes

This PPA also contains a fix for a 3D-specific hang in mesa.

[Olaf Lessenich (olaf-lessenich)]

> The PPA is here: https://launchpad.net/~brian-rogers/+archive/graphics-fixes

I just tried with the kernel from your PPA.

With KMS enabled, the system freezes immediately and blanks the screen before even writing anything to a logfile.
If KMS is disabled via 'i915.modeset=0', the system boots but X refuses to load. Disabling DRI makes no difference.

[Jānis Kangarooo (kangarooo)]

I did whats written in description.
sudo apt-add-repository ppa:brian-rogers/graphics-fixes
sudo apt-get update
sudo apt-get dist-upgrade
sudo apt-get install linux-image-2.6.34-52-generic

and this kernel wasnt even starting computer.
i also tryd linux-image-2.6.34-020634rc7-generic_2.6.34-020634rc7_i386.deb
that started computer but crash was now different. if i had previusly crash immidiatly and with colors then with rc7 it was harder to get crash and also all screen was black.

[MikeG0 (michael-gorham)]

[Possible Workaround]
Dell Inspiron 1100 -intel 82845G/GL (on Lucid 10.04)
[Intel Corporation 82845G/GL[Brookdale-G]/GE Chipset Integrated Graphics Device (rev 03)]

Hacked at this for several days now. Been working mainly in gnome-terminal & firefox for more than 12 hours straight w/out a single crash. I think I have a workaround from a fresh apt-get dist-upgrade at 6AM this morning. These are the only changes I've made:

Reinstalled xulrunner (might not matter) ...
apt-get install --reinstall xulrunner-1.9.2
apt-get install --reinstall xulrunner-1.9.2-gnome-support
apt-get remove xulrunner-1.9.1

Added to xorg.conf under Monitor0 (might not matter) ...
HorizSync 31.5-48.5
VertRefresh 59-75

added to grub ...
video=LVDS-1:e video=VGA-1:d i915.modeset=0

Turned off Screen Saver and Power Management

And finally ... the big workaround ...

Removed the >>> Shockwave Flash <<< plugin from Firefox 3.6.3.

firefox-bin was flooding .session-errors ... since removing flash, very few firefox-bin errors. I'm still getting random .xsession-errors from polkit-gnome-authentication-agent, gnome-terminal, firefox-bin, rhythmbox ... but 12 hours straight uptime is pretty good compared to the immediate X startup crash.

Regardless of the .xsession-errors, there's no observable system degradation. I'll report back if I crash ... fingers crossed.

Still looking forward to a fix for all those .xsession-errors.

Hope this helps,

Mike

[MikeG0 (michael-gorham)]

Oh yeah, also forgot ... removed compiz

[Eduardo Ribeiro Rodrigues (edu-pobox)]

I think the key here is the 3D stuff (OpenGL). In my case, what I noticed as the culprit was the screensaver. Once I disabled it, even with power management enabled, putting the monitor to sleep and despite the fact I use a KVM that most of the time is not selected to the Ubuntu machine, I haven't get any X crashes.

[Eduardo Ribeiro Rodrigues (edu-pobox)]

Just applied the PPA patch and re-enabled my GL screensaver. Let's see...

[sam (samuel-j-1993)]

I re-enabled kms which seemed to be disabled which seemed to stop the crashing a bit
then i tried to install a game through wine which kept crashing the system halfway through
i think this is related...

Compiz doesnt work on my install with 845gm
is that part of this bug?

[sam (samuel-j-1993)]

Originally I was getting no compiz and every time i tried to run something graphics intensive the system would crash and black and white lines would flash
tried to apply patch but system would not boot

[Michael Rickmann (mrickma)]

The same here as in comment #115: kernel 2.6.34-52 (2.6.34-52.1) from Brian Rogers' ppa freezes the system.

Dell Optiplex GX60
00:02.0 VGA compatible controller: Intel Corporation 82845G/GL[Brookdale-G]/GE Chipset Integrated Graphics Device (rev 01)

libdrm 2.4.20~glasen~ppa4
xserver-xorg-video-intel 2:2.11.0+git20100529~glasen~ppa2
mesa 7.7.1-1ubuntu4~gpufix1

The freeze happens at the first modeswitch, i.e. so early during startup that I cannot dump an i915_error_state. My favourite kernel on this system is a drm-intel-next one from the Ubuntu Kernel Team 2.6.34-997.201004191525, i.e. without the V9 patch. The later drm-intel-next kernels from http://kernel.ubuntu.com/~kernel-ppa/mainline/drm-intel-next also freeze the system at the first mode switch. The 201004191525 kernel shows an i915 error at about the same time when its successors freeze but it continues:

dmesg:
......
[ 19.940083] intel8x0_measure_ac97_clock: measured 54713 usecs (2636 samples)
[ 19.940091] intel8x0: clocking to 48000
[ 21.583337] render error detected, EIR: 0x00000010
[ 21.583352] [drm:i915_handle_error] *ERROR* EIR stuck: 0x00000010, masking
[ 21.583385] render error detected, EIR: 0x00000010
[ 29.516031] eth0: no IPv6 routers present
[ 42.412078] end_request: I/O error, dev fd0, sector 0
[ 42.472101] end_request: I/O error, dev fd0, sector 0
end of dmesg

i915_error_state:
Time: 1275242616 s 587221 us
PCI ID: 0x2562
EIR: 0x00000010
  PGTBL_ER: 0x00000049
  INSTPM: 0x00000000
  IPEIR: 0x00000000
  IPEHR: 0x01000000
  INSTDONE: 0x00ffffc0
  ACTHD: 0x00000048
seqno: 0x00000002
--- ringbuffer = 0x00fdf000
00000000 : 02000000
00000004 : 00000000
00000008 : 18000001
0000000c : 01500001
00000010 : 01500024
00000014 : 00000000
00000018 : 02000004
0000001c : 00000000
00000020 : 10800001
00000024 : 00000080
00000028 : 00000001
0000002c : 01000000
00000030 : 02000000
00000034 : 00000000
00000038 : 10800001
0000003c : 00000080
00000040 : 00000002
00000044 : 01000000
00000048 : 00000000
...... all 00000000

[sam (samuel-j-1993)]

Tried the kernel mentioned in the last post (http://kernel.ubuntu.com/~kernel-ppa/mainline/drm-intel-next) latest build

Crashes decreased in frequency but still crashes after being used for a while

[Darxus (darxus)]

Povray's graphical output is causing crashes the same as web browsers. My
exact command was: "povray -iall.pov -otrans.png". Ran it with -D (no
graphical output) and it didn't crash.

[Darxus (darxus)]

Based on comment #117 I removed flash, and haven't had a crash / hang while running chrome since (when it had been a few times a day previously). Still not long enough to be certain.

Exact steps:
sudo aptitude remove flashplugin-installer flashplugin-nonfree
sudo mv /opt/google/chrome/libgcflashplayer.so /opt/google/chrome/libgcflashplayer.so.bak

I did this right after a reboot, and did not restart anything after.

[MikeG0 (michael-gorham)]

Just had a crash after hitting wunderground.com in chrome. However, looks like the flashplugin-installer package and libgcflashplayer.so were still available. Removed/renamed these and the page loads fine now. wunderground.com loads a flash ad in an iframe which probably tries installing flash in an untested manner.

Thanks for the additional tips. I'm psyched ... this dell is getting pretty stable.

[Darxus (darxus)]

Just crashed when I ctrl-shift-clicked on a link in chrome. Damn.

Verified the flash stuff was still removed.

Link was "General Motors EV1" on
http://en.wikipedia.org/wiki/Automobile_drag_coefficient

[jdorwart (jdorwart)]

Unfortunately I have neither of these packages installed and I still get the
crash. I do not get the Zebra stripe thing, just a black screen that the
only way I can seem to recover from is by pressing the power button.
 Fortunately the boot process is so much faster I have not thrown the whole
thing out the window but it is getting old.

I rarely use Firefox, but that does not seem to impact my crashing
positively or negatively. I get crashes at sporadic times. Sometimes it is
five minutes, sometimes it is two days.

jeff

On Wed, Jun 2, 2010 at 11:43 PM, MikeG0 <email address hidden> wrote:

> Just had a crash after hitting wunderground.com in chrome. However,
> looks like the flashplugin-installer package and libgcflashplayer.so
> were still available. Removed/renamed these and the page loads fine
> now. wunderground.com loads a flash ad in an iframe which probably
> tries installing flash in an untested manner.
>
> Thanks for the additional tips. I'm psyched ... this dell is getting
> pretty stable.
>
> --
> MASTER: [i845] GPU lockup (apport-crash) (Should KMS be blacklisted?)
> https://bugs.launchpad.net/bugs/541492
> You received this bug notification because you are a direct subscriber
> of a duplicate bug.
>

--
Jeffrey Dorwart
U.S. Coast Guard
Aviation Engineer
CG-711/DD Airstation Clearwater FL.
[phone] cell

ASW-20 N26307
A&P/IA

[Tim Hallett (tim-hallett)]

Let me see if I understand....
We have a "fixed" kernel for i845 that has never successfully booted a machine...
The Intel corporate team officially leading the fix have contributed no code since this bug was released in January...
All the "workarounds" proposed involve removing fairly major areas of functionality, and don't seem to work anyway....

I have an i845 machine and would be happy to perform tests, but is there anything right now worth testing?

[Aether (raymond-dick)]

I think I have the same problem - for me X crashes and even restarting kdm doesn't fix it. Running linux image 2.6.32-22 on a Dell Inspiron mini 10v with an i915. Interestingly enough an Arch linux user reports a fix by starting X later in the boot process http://bbs.archlinux.org/viewtopic.php?id=94206

It sounds a lot like my issue but not sure how to try similar in ubuntu. I see the same things the Arch linux user does - it drops back to text on vty2 rather than vty7. I've tried the patch and I've tried turning off KMS, none of which help the issue.

[useResa (rdrijsen)]

Interesting link and read as provided by Aether (comment #131).
Unfortunately Ubuntu does no longer use inittab and I know to little on how to control the startup order.

I know it is controlled by the S and the number following it in the /etc/rcX.d (where X represents the runlevel) that is preceeding the file names. However, it seems to me that the X environment is started from /etc/rcS.d directory and I can not really see when that kicks in. If this is at the end of the boot process my understanding is that X is already started as last.

Can someone please confirm and test this (or explain how I can test it) since until now none of the suggested workarounds have provided a stable environment for me.

[Michael Rickmann (mrickma)]

In Brian Rogers' ppa two knew kernels have been build, linux-image-2.6.34-drm-generic and linux-image-2.6.34-v9patch-generic. Unfortunately those too freeze my Dell Optiplex GX60 with 82845G/GL[Brookdale-G]/GE (rev 01) the same way as the 2.6.34-52 kernel did, i.e. at the first mode switch. As I lost my patience a bit I started bisecting drm-intel-next. The culprit for my i845 specific freezes is commit 7fff400 Tue, 20 Apr 2010 03:16:04 Merge branch 'drm-fbdev-cleanup' into drm-core-next. I will have a look at the commits in that branch.

[jerrylamos (jerrylamos)]

i845G here which only runs Lucid with no KMS and vesa. KMS and/or intel driver cause X crashes randomly.

i845G does work on Karmic with its intel driver and KMS.

Now, this is Meerkat Alpha 1 and I think KMS and intel driver are working. ? How do I tell for sure? (May be fleeting, the dread Alpha update can strike any time.)

Linux version 2.6.34-5-generic (buildd@palmer) (gcc version 4.4.4 (Ubuntu 4.4.4-4ubuntu1) ) #14-Ubuntu SMP Fri Jun 4 06:18:40 UTC 2010

dpkg -l | grep xserver-xorg-video-intel
ii xserver-xorg-video-intel 2:2.9.1-3ubuntu5 X.Org X server -- Intel i8xx, i9xx display d

Thanks, Jerry

[fedsed (fedor-s)]

I use the drivers https://launchpad.net/~glasen/+archive/intel-driver. Crashes have stopped.

[Nandhakumar (nandhakumar-n87)]

Hi to all,

2010/6/16 fedsed <email address hidden>

> I use the drivers https://launchpad.net/~glasen/+archive/intel-driver<https://launchpad.net/%7Eglasen/+archive/intel-driver>
> .
> Crashes have stopped.
>
> I am using ubuntu 10.04 . My system configuration is intel p4 processor,
845GV chipset motherboard with 512+128MB RAM.

I too had the wierd problem of vertical black and white bars appearing when
I was using ubuntu and firefox.

I used the above driver and installed the xserver-xorg-video-intel package.

I was using xfce environment when I was installing this package. Then I
restart the system.

I could see the difference from the ubuntu splash screen itself. It looked
very clear and nice.

Now, I am using gnome session and it works fine. I had the same problem when
I was using firefox. So, I am using firefox again to check whether the same
problem occurs again.

> --
> MASTER: [i845] GPU lockup (apport-crash) (Should KMS be blacklisted?)
> https://bugs.launchpad.net/bugs/541492
> You received this bug notification because you are a direct subscriber
> of a duplicate bug.
>
> Status in X.org xf86-video-intel: Confirmed
> Status in “xserver-xorg-video-intel” package in Ubuntu: Triaged
> Status in “xserver-xorg-video-intel” source package in Lucid: Triaged
>
> Bug description:
> Binary package hint: xserver-xorg-video-intel
>
> This is a MASTER bug report, i.e. not a real bug report, but a tool to help
> manage other bug reports.
>
> Most bug reports on i845 are probably due to the CPU/GPU incoherency
> problem that is now consolidated upstream at
> http://bugs.freedesktop.org/show_bug.cgi?id=26345 . For now, we mark all
> automatically reported GPU lockups as duplicates of this unless there is a
> reason not to.
>
> A fixed kernel is available at
> https://launchpad.net/~brian-rogers/+archive/graphics-fixes<https://launchpad.net/%7Ebrian-rogers/+archive/graphics-fixes>
>
> To use this fixed kernel, run the following commands:
>
> sudo apt-add-repository ppa:brian-rogers/graphics-fixes
> sudo apt-get update
> sudo apt-get dist-upgrade
> sudo apt-get install linux-image-2.6.34-52-generic
>
> There is a similar master bug report for i855 at bug 541511.
>
>
>
> To unsubscribe from this bug, go to:
> https://bugs.launchpad.net/xserver-xorg-video-intel/+bug/541492/+subscribe
>

I hope ubuntu works fine from now.

Thanks for the great help .

Ubuntu Rocks.

Regards

N.Nandhakumar

[Tim Hallett (tim-hallett)]

Could newbies get detailed instructions on implementing this new driver (including correct apt-get commands) to install the new Glasenhardt driver? Perhaps this possible fix and said instructions should be added to the guide page for this bug.
https://wiki.ubuntu.com/X/Bugs/Lucidi8xxFreezes

Regards
Tim Hallett
Linux Neophyte

[sam (samuel-j-1993)]

Downloaded the intel driver from ppa in comment 136 and it seems to make a difference
the crashes no longer happen and the framerate has gone up to 1099
the only thing is compiz is blacklisted for the card and there is no easy way to overwrite in lucid that i can find out

[Tim Hallett (tim-hallett)]

Installed the Glasenhart 855-drm patch (stable) and then the intel driver. Improved overall stability and performance, but Google Earth still brought system down, as did OpenGL screensavers with frequency. Then added Brian Roger's Mesa fix, after which Google Earth functioned perfectly, although there seem to still be intermittent freezes on some screensavers (of which I am only aware of due to being unable to wake up the screen after screen OFF (assuming system has frozen. REISUB is ineffective so it is a fairly deep fault rather than with X). Which log files should I access at the next freeze to make a more useful report? (specifically, which ones, and where are they located.
Would also installing the Rogers kernel build be advisable? O r should I leave things as is for now?

[dbclinton (dbclin)]

I installed the Glasenhart patch on my Edubuntu thin client server - two of the clients have i845 chips. There was no improvement in the client experience: it still makes it to the login screen but crashes as soon as the user completes the login process. I haven't had the guts to install Brian Roger's patched kernel yet (this is a production server).

[Tim Hallett (tim-hallett)]

@dbclinton
Did you install both the 855 patch and the intel driver package (ther'e on different repositories on Glasenhart's site)?

[dbclinton (dbclin)]

I first added ppa:glasen/intel-driver to my repositories and then installed libdrm (actually it was called libdrm2 - marked for upgrade - in my synaptic) and xserver-xorg-video-intel (also marked for upgrade). I assume that would have got the job done.
Is there something I missed?
Thanks

[Nigel Atkinson (nigel-atkinson)]

I've tried the driver from ppa:glasen/intel-driver on two identical machines with i845 (from Memory - not at them at present). One seems to now work fine and can even run Blender without crashing (which would instantly previously). It has the server kernel (server + gnome-desktop).

On the other, a standard 10.04 desktop install (and Parents machine to hard to test on) - I'm not sure if the driver made things more stable as I had to revert due to a problem with pictures, usually photos in Firefox. The pictures would initially show as a black rectangle, but usually show the picture once moused over. Some showed a previous picture slightly offset. It looks to me like the initial texture coping is not working - and having the mouse cursor over that part of the screen forces an update.

Here's hoping this info gives someone an 'A ha!' moment. :-)

[Tim Hallett (tim-hallett)]

@dbclinton
Yes, you also need the fixed kernel modules from ppa:glasen/855gm-fix

[Tim Hallett (tim-hallett)]

@Nigel
Yes I have the same behavior with flash-embedded photos in Firefox as well. IIRC it goes way if the 855 experimental modules are used. Also, the GL screensavers will still take down the system intermittently.

[dbclinton (dbclin)]

@Tim
Thanks for the correction. However, I see that 855gm-fix-exp-dkms comes with a bit of a threat of its own:

"Stability may suffer with this version. If you experience frequent crashes than please install the non-"exp"-version of this package."

And 855gm-fix-dkms isn't perfect either:

"Known side-effect is a slowdown of some Gtk+-functions."

I think I need more information to be comfortable committing to this fix. In particular, what does Gtk control?

Thanks

[Nandhakumar (nandhakumar-n87)]

Hi

2010/6/21 Nigel Atkinson <email address hidden>

> I've tried the driver from ppa:glasen/intel-driver on two identical
> machines with i845 (from Memory - not at them at present). One seems to
> now work fine and can even run Blender without crashing (which would
> instantly previously). It has the server kernel (server + gnome-
> desktop).
>
> On the other, a standard 10.04 desktop install (and Parents machine to
> hard to test on) - I'm not sure if the driver made things more stable as
> I had to revert due to a problem with pictures, usually photos in
> Firefox. The pictures would initially show as a black rectangle, but
> usually show the picture once moused over. Some showed a previous
> picture slightly offset. It looks to me like the initial texture coping
> is not working - and having the mouse cursor over that part of the
> screen forces an update.
>
> Here's hoping this info gives someone an 'A ha!' moment. :-)
>
> --
> MASTER: [i845] GPU lockup (apport-crash) (Should KMS be blacklisted?)
> https://bugs.launchpad.net/bugs/541492
> You received this bug notification because you are a direct subscriber
> of a duplicate bug.
>
> Status in X.org xf86-video-intel: Confirmed
> Status in “xserver-xorg-video-intel” package in Ubuntu: Triaged
> Status in “xserver-xorg-video-intel” source package in Lucid: Triaged
>
> Bug description:
> Binary package hint: xserver-xorg-video-intel
>
> This is a MASTER bug report, i.e. not a real bug report, but a tool to help
> manage other bug reports.
>
> Most bug reports on i845 are probably due to the CPU/GPU incoherency
> problem that is now consolidated upstream at
> http://bugs.freedesktop.org/show_bug.cgi?id=26345 . For now, we mark all
> automatically reported GPU lockups as duplicates of this unless there is a
> reason not to.
>
> A fixed kernel is available at
> https://launchpad.net/~brian-rogers/+archive/graphics-fixes<https://launchpad.net/%7Ebrian-rogers/+archive/graphics-fixes>
>
> To use this fixed kernel, run the following commands:
>
> sudo apt-add-repository ppa:brian-rogers/graphics-fixes
> sudo apt-get update
> sudo apt-get dist-upgrade
> sudo apt-get install linux-image-2.6.34-v9patch-generic
>
> There is a similar master bug report for i855 at bug 541511.
>
>
>
> To unsubscribe from this bug, go to:
> https://bugs.launchpad.net/xserver-xorg-video-intel/+bug/541492/+subscribe
>

+1

Regards

N.Nandhakumar

[Michael Rickmann (mrickma)]

On one of ours Dell Optiplex GX60 with
00:02.0 VGA compatible controller: Intel Corporation 82845G/GL[Brookdale-G]/GE Chipset Integrated Graphics Device (rev 01)
I just tested the recommended fix from https://launchpad.net/~brian-rogers/+archive/graphics-fixes the "2.6.34-v9patch-generic #2-Ubuntu SMP Wed Jun 9" kernel and the mesa fix. All the other libs I have reverted to the standard lucid ones making shure that KMS was used. It really feels solid, congratulations and thanks. I really tried hard to crash it. The only flaw I could detect was a crash when visiting http://www.ardmediathek.de/ with firefox as already reported upstream in http://bugs.freedesktop.org/show_bug.cgi?id=27187 #179 and following.
Admittedly the i845 is old and slow but I wish to use the machines as LTSP thin clients. Now I will dare to upgrade our server.

[Matte Silver (ruf10)]

The patched kernel fixes problem with i915 crashing but now processes can't readlink /proc/$$/(exe|cwd|root) (access denied)

00:02.0 VGA compatible controller: Intel Corporation Core Processor Integrated Graphics Controller (rev 12)
 Subsystem: ASUSTeK Computer Inc. Device 1252
 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 32
 Region 0: Memory at e3400000 (64-bit, non-prefetchable) [size=4M]
 Region 2: Memory at c0000000 (64-bit, prefetchable) [size=256M]
 Region 4: I/O ports at e080 [size=8]
 Capabilities: [90] Message Signalled Interrupts: Mask- 64bit- Queue=0/0 Enable+
  Address: fee0a00c Data: 41a1
 Capabilities: [d0] Power Management version 2
  Flags: PMEClk- DSI+ D1- D2- AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-)
  Status: D0 PME-Enable- DSel=0 DScale=0 PME-
 Capabilities: [a4] PCIe advanced features <?>
 Kernel driver in use: i915
 Kernel modules: i915

[Darxus (darxus)]

Can hardware acceleration please be blacklisted again so I can go back to going a whole day without my computer crashing without going back to Jaunty?

That's all it takes, right? Is that hard?

Two months after release. This is really messed up.

[Tim Hallett (tim-hallett)]

The Glasenhart patches and Rogers Mesa patch have resulted in good performance but with occasional freezes at start of some GL screensavers (which I cannot identify because they never get a chance to start. Which log files should I send to Stefan in order to help him rectify this, and what software should I use for backtracing?

@Darxus
Have you tried the Glasenhart 855-patch plus the libdrm and intel driver patches along with Brian Rodger's Mesa patch (installed in that order). Fixed everything almost on my i845 box.

[Darxus (darxus)]

This is ridiculous.

A version that does not hang or crash at all can easily be released by blacklisting the hardware acceleration, right? How?

Do that. A release that works.

Then play with the rest.

[Darxus (darxus)]

After following the instructions in the description to install linux-image-2.6.34-v9patch-generic, it still crashed.

[jerrylamos (jerrylamos)]

Maverick Alpha 2 on intel i845G
Linux version 2.6.35-6-generic (buildd@vernadsky) (gcc version 4.4.4 (Ubuntu 4.4.4-6ubuntu1) ) #9-Ubuntu SMP Wed Jun 30 23:09:46 UTC 2010
still fails. Maverick Linux 2.6.34-5 did run as installed with KMS and with intel driver all Linux's since fail.

Alpha 2 will run only if i915.modeset=0 and Driver "vesa" are used.

Do note, CD live ran with defaults, but not the install.

Jerry

[Tim Hallett (tim-hallett)]

Mint Isadora (Lucid) graphics performance has been greatly improved with the Glasenhart drivers and libdrm. However, certain GL-screensavers (for example, Ant Inspect, Ant Maze, Atlantis) instantly bring down the system. I would like to be able to do a backtrace on these crashes, but do not know how to configure strace and xtrace in the terminal. Any help would be appreciated.
A solution seems very close, we now have flash stability and GoogleEarth operates without problems on i845, there simply seem to be some remaining issues with open-GL (Admittedly this could be a problem with Mesa, rather then the i915 driver. I have no way of knowing.)

[Darxus (darxus)]

After several more crashes yesterday, I finally broke down and spent a few hours installing Jaunty. Turns out wpa_supplicant doesn't work to get me on the lan at work under Jaunty.

So my remaining options are:
1) Vesa driver, at 1024x768 instead of 1200x1920.
2) Buy new hardware.
3) Switch operating systems.
4) Continued crashes.

This sucks.

[Tim Hallett (tim-hallett)]

Darxus,

Have you tried the following, in order, exactly? Somehow, I doubt it because this actually works.
1. In Synaptic package manager, go to Linux, select the item that just says "Linux", this will download and install the latest stable Ubuntu kernel, which will install without reformatting your drive or killing your network connectivity (as is the case with the Rogers' patch, not being fully implemented for Ubuntu, no I do not want to re-compile the rt2870 drivers). This by itself stops 95% of the crashes. Just try this for a while, if you still have occasional crashes, do the following.
2. Install the 855-fix at https://launchpad.net/~glasen/+archive/855gm-fix, if you do not know how to install a new repository, well there are detailed directions at that site. Either one works, but I use the experimental.
3. Install the Mesa fix at https://launchpad.net/~brian-rogers/+archive/graphics-fixes, or if you can get 7.8.2 use that.
4. Go to https://launchpad.net/~glasen/+archive/intel-driver, and install the libraries, not the Intel driver. The current version 712 of the intel driver is IMHO, inferior to v.710. 710 had no crashes at all, 712 has very very intermittent ones, overall graphics performance is great with both.
5. DO NOT install the Compiz patch unless you know how to operate your computer in recovery mode.

Try this, it works. I can see all my screensavers, Google Earth works great, haven't tried any of the 3-d games, but I'm confident now. Flash and online apps are rock solid now.

[Darxus (darxus)]

Tim: I did not have the linux meta package installed, but I did have the linux-image-generic meta package through which it depends on the latest actual kernel package, so I've been running 2.6.32-23-generic #37-Ubuntu SMP.

I haven't done the rest, I'll do that now.

[Darxus (darxus)]

Intel driver v.710 is in xserver-xorg-video-intel version 2:2.9.1-3ubuntu5?

[Felix Lechner (felix.lechner)]

Darxus,

Please try this before you buy a new computer:

My graphics performance seems a bit sluggish at times (but I also have a
slow network). My system is stable with a 1680x1050 resolution on
82845G/GL. Crashes have pretty much stopped. Every once in a while my
system freezes when I try to log on, but that is workable for me. Google
Earth also works.

I use the the Brian Rogers kernel linux-image-2.6.34-v9patch and Stefan
Glasenhardt's driver xserver-xorg-video-intel
2:2.12.0+git20100712~glasen~ppa1 and also his libraries libdrm2 and
libdrm-intel 2.4.21~glasen~ppa5.

Both are available from their PPAs.
https://launchpad.net/~brian-rogers/+archive/graphics-fixes
https://launchpad.net/~glasen/+archive/intel-driver

I am also going to look into Tim's suggestions but I have been using this
combination for some time now.

Good luck,
Felix

On Wed, Jul 14, 2010 at 10:16 AM, Tim Hallett <email address hidden>wrote:

> Darxus,
>
> Have you tried the following, in order, exactly? Somehow, I doubt it
> because this actually works.
> 1. In Synaptic package manager, go to Linux, select the item that just says
> "Linux", this will download and install the latest stable Ubuntu kernel,
> which will install without reformatting your drive or killing your network
> connectivity (as is the case with the Rogers' patch, not being fully
> implemented for Ubuntu, no I do not want to re-compile the rt2870 drivers).
> This by itself stops 95% of the crashes. Just try this for a while, if you
> still have occasional crashes, do the following.
> 2. Install the 855-fix at https://launchpad.net/~glasen/+archive/855gm-fix,
> if you do not know how to install a new repository, well there are detailed
> directions at that site. Either one works, but I use the experimental.
> 3. Install the Mesa fix at
> https://launchpad.net/~brian-rogers/+archive/graphics-fixes, or if you can
> get 7.8.2 use that.
> 4. Go to https://launchpad.net/~glasen/+archive/intel-driver, and install
> the libraries, not the Intel driver. The current version 712 of the intel
> driver is IMHO, inferior to v.710. 710 had no crashes at all, 712 has very
> very intermittent ones, overall graphics performance is great with both.
> 5. DO NOT install the Compiz patch unless you know how to operate your
> computer in recovery mode.
>
> Try this, it works. I can see all my screensavers, Google Earth works
> great, haven't tried any of the 3-d games, but I'm confident now. Flash
> and online apps are rock solid now.
>
> --
> MASTER: [i845] GPU lockup (apport-crash) (Should KMS be blacklisted?)
> https://bugs.launchpad.net/bugs/541492
> You received this bug notification because you are a direct subscriber
> of the bug.
>
> Status in X.org xf86-video-intel: Confirmed
> Status in “xserver-xorg-video-intel” package in Ubuntu: Triaged
> Status in “xserver-xorg-video-intel” source package in Lucid: Triaged
>
> Bug description:
> Binary package hint: xserver-xorg-video-intel
>
> This is a MASTER bug report, i.e. not a real bug report, but a tool to help
> manage other bug reports.
>
> Most bug reports on i845 are probably due to the CPU/GPU incoherency
> problem that is now consolidated upstream at
>...

[Darxus (darxus)]

Crashed again. I have installed:

linux 2.6.32.23.24
855gm-fix-dkms 0.7.5~glasen~ppa2
libgl1-mesa-dri 7.7.1-1ubuntu4~gpufix2
libdrm-intel1 2.4.21~glasen~ppa5
xserver-xorg-video-intel 2:2.9.1-3ubuntu5

Tim: I believe this is everything you recommended?

[Tim Hallett (tim-hallett)]

No, when I referred to 710, I was referring to the git number on the Glasenhardt version, now unavailable, instead install
xserver-xorg-video-intel 2:2.12.0+git20100712~glasen~ppa1

[Tim Hallett (tim-hallett)]

Thanks Felix
IMHO, stay with the combination you have. Glasenhardt should have Git 714 built within a few hours (either today or tomorrow) but aside from that upgrade, you are the same as me. When I spoke of 710, I was referring to Git 710 in Glasenhardt, which just worked really well, no crashes, (random screensavers continued for days without the black screen of death) clean bootup, everything seemed just about perfect, though apparently it wasn't, because there remained work to be done. But hopefully git714 will restore the stability of git710, along with maybe some performance improvements.

[Darxus (darxus)]

Thanks. So this does what you suggest?

apt-add-repository ppa:glasen/855gm-fix
apt-add-repository ppa:brian-rogers/graphics-fixes
apt-add-repository ppa:glasen/intel-driver
aptitude update
aptitude install linux 855gm-fix-dkms
aptitude dist-upgrade

[Darxus (darxus)]

One full work day without a crash!

What needs to happen for this stuff to get into Maverick?

[Julien Olivier (julo)]

I confirm that, with the solution in comment #164, the crashes seem gone. And even suspend works flawlessly! For now, at least...

[Christian (c-pradelli)]

Darxus, did you test this in Lucid?

[Darxus (darxus)]

Christian: Yes.

[Darxus (darxus)]

xserver-xorg-video-intel:
  Installed: 2:2.12.0+git20100712~glasen~ppa1
  Candidate: 2:2.12.0+git20100714~glasen~ppa1

Running the upgrade now.

[Felix Lechner (felix.lechner)]

Tim,

Thank you for your advice. Are you also still running the old
2.6.34-v9patch kernel from Brian Rogers? Why don't I need the updated
modules from Glasenhardt's '855gm-fix' like everyone else?

Anyone following the new instructions in the sticky is taken to Brian's new
kernel 2.6.35rc4-131-ge467e10-v9patch-generic. The dkms failed for the old
kernel. As an experiment, I now installed the new kernel and
'855gm-fix-exp' (the one with intel_wait_for_canary_flocks() turned off).
 Why should one stick with the old kernel, if that is what you recommended?

My apologies if this is not the right forum. And thanks to everyone who is
working on this bug.

Felix

On Wed, Jul 14, 2010 at 7:55 PM, Tim Hallett <email address hidden>wrote:

> Thanks Felix
> IMHO, stay with the combination you have. Glasenhardt should have Git 714
> built within a few hours (either today or tomorrow) but aside from that
> upgrade, you are the same as me. When I spoke of 710, I was referring to Git
> 710 in Glasenhardt, which just worked really well, no crashes, (random
> screensavers continued for days without the black screen of death) clean
> bootup, everything seemed just about perfect, though apparently it wasn't,
> because there remained work to be done. But hopefully git714 will restore
> the stability of git710, along with maybe some performance improvements.
>
> --
> MASTER: [i845] GPU lockup (apport-crash) (Should KMS be blacklisted?)
> https://bugs.launchpad.net/bugs/541492
> You received this bug notification because you are a direct subscriber
> of the bug.
>
> Status in X.org xf86-video-intel: Confirmed
> Status in “xserver-xorg-video-intel” package in Ubuntu: Triaged
> Status in “xserver-xorg-video-intel” source package in Lucid: Triaged
>
> Bug description:
> Binary package hint: xserver-xorg-video-intel
>
> This is a MASTER bug report, i.e. not a real bug report, but a tool to help
> manage other bug reports.
>
> Most bug reports on i845 are probably due to the CPU/GPU incoherency
> problem that is now consolidated upstream at
> http://bugs.freedesktop.org/show_bug.cgi?id=26345 . For now, we mark all
> automatically reported GPU lockups as duplicates of this unless there is a
> reason not to.
>
> A fixed kernel is available at
> https://launchpad.net/~brian-rogers/+archive/graphics-fixes
>
> To use this fixed kernel, run the following commands:
>
> sudo apt-add-repository ppa:brian-rogers/graphics-fixes
> sudo apt-get update
> sudo apt-get dist-upgrade
> sudo apt-get install linux-image-2.6.34-v9patch-generic
>
> There is a similar master bug report for i855 at bug 541511.
>
>
>
> To unsubscribe from this bug, go to:
> https://bugs.launchpad.net/xserver-xorg-video-intel/+bug/541492/+subscribe
>

[Tim Hallett (tim-hallett)]

Felix,

I'm not using the Rogers kernel at all. Because it is not cleared for Ubuntu release it is still somewhat poorly integrated into Lucid (Prime example is the Ralink wireless drivers no longer working properly because integration of the rt2870 and rt2860 have been lost. Yes, I could recompile them, but I have read the directions, and frankly , its a risk-filled PiTA, instead I have installed the 2.6.32.23, which is the latest Ubuntu kernel available in the regular repositories. That, plus the 855 path, the Rogers Mesa, and the Glasenhardt driver git716 (newly released, and IMHO, working really well, 16 hours of random screensavers and no crashes). I simply have found this recipe to work, and be easy to implement for a dummy like me.

[Tim Hallett (tim-hallett)]

Passable stability with drover git716, random screensavers cycle for several hours, with eventual freezes on open GL (implying that perhaps upgrade to Mesa 7.8.2 might be in order). Upon freeze, X does not crash, and desktop can be recovered, however, attempting to open the screensaver control panel will crash the system. On other occasion, X will be brought down (green light on monitor, X cannot be recovered), or system will be brought down (monitor indicates no signal). In either event system is working well under applications such as google Earth. Lucid solution is close, hopefull this will be fully resolved in Maverick. (I can assure that i815 system works flawlessly with perfect stability on current git716 driver)

[Michael Rickmann (mrickma)]

Stimulated by the previous posts I tested one of our Dell Optiplex GX60 desktops with
00:02.0 VGA compatible controller: Intel Corporation 82845G/GL[Brookdale-G]/GE Chipset Integrated Graphics Device (rev 01) again. I found a combination which is almost perfect for this kind of machine without fancy hardware.
linux-image-2.6.35rc4-131+ge467e10+nopatch-generic (!! nopatch from Brian Rogers' repository)
xserver-xorg-video-intel 2:2.12.0+git20100716~glasen~ppa1
libdrm-intel1 2.4.21~glasen~ppa5
libdrm2 2.4.21~glasen~ppa5
plus the gpufix2 mesa packages from Brian Rogers,
otherwise up to date lucid.
So it is similar to what has been recommended here lately, except of the kernel.
What I did for testing, frequently switching VTs, logging in as a second user (slow but safe), atlantis xscreensaver, voluntarily reboot, normal openoffice work, watching 2h xvid video with totem player, visiting heavily flash overloaden web sites with firefox and chromium. The machine is up now for more than a day and the only flaw I could detect was in dmesg during system startup,
[ 34.782896] render error detected, EIR: 0x00000010
[ 34.782909] [drm:i915_report_and_clear_eir] *ERROR* EIR stuck: 0x00000010, masking
[ 34.782941] render error detected, EIR: 0x00000010
which apparently is harmless.
I am writing this from a second of these i845 computer on which I just have installed above fixes. It really feels more responsive than with the 2.6.34-v9patch kernel.
Would the "nopatch" in the name of the kernel which I use mean, that I do not need the V8/V9 patch on my i845 hardware any longer with recent xserver-xorg-video-intel and libdrm ?
Regards
Michael

[Tim Hallett (tim-hallett)]

Are steps being taken to have this kernel incorporated in the Lucid release? I'd use it not but my i845 uses ralink2870 wireless drivers (currently uncompiled in Rodger's kernel), and the compilation process is about 12-steps long (why ralink rt2870 is not in the repositories is a mystery).

[Darxus (darxus)]

Tim: As far as I can tell nothing is being done to get any of this into any Ubuntu release.

Confirm no crashes with xserver-xorg-video-intel 2:2.12.0+git20100714~glasen~ppa1.

[Darxus (darxus)]

Source packages I'm using to prevent crashes which are not in the official archives:

855gm-fix
 - https://launchpad.net/~glasen/+archive/855gm-fix

linux
mesa
 - https://launchpad.net/~brian-rogers/+archive/graphics-fixes

libdrm
xserver-xorg-video-intel
 - https://launchpad.net/~glasen/+archive/intel-driver

[Michael Rickmann (mrickma)]

Hi Tim,
kernel 2.6.35-rc5 is just a week old. Maverick will use 2.6.35 and after its release, I guess, we can expect support for lucid as well, see http://ubuntuforums.org/showthread.php?t=1498079 . For the meantime, the module you wish for your wlan is presumably rt2860sta.ko or rt2870sta.ko, and they are in the 2.6.35rc4-131+ge467e10+xxpatch-generic kernels from Brian Rogers. So either they do not work yet or other modules are loaded instead, see http://ubuntuforums.org/archive/index.php/t-1485772.html . Try to blacklist them if they show up in lsmod, see http://ubuntuforums.org/archive/index.php/t-1485772.html .
Michael

[Darxus (darxus)]

It's nice to see this issue is apparently in the Desktop team's plan for Maverick:
http://theravingrick.blogspot.com/2010/05/track-desktop-and-une-in-maverick.html

[Tim Hallett (tim-hallett)]

Thanks Mike,
First thing I did was try the blacklisting route, in all possible permutations, but the best I could do was to get the rt2800usb to come online, at 1kb/s, yuk! No I know that the rt2870.ko module works, but there was no operational sign of it anywhere on the system with Rodgers' kernel. Anyhow, we can manage as is for now until 2.6.35 comes officially online. I'll try Brian's kernel again if I could get some more specific assistance in getting rt2870 online. I've read the directions dor direct compilation, and it is just too much of a pain.

[Michael Rickmann (mrickma)]

Folow up to comment #173
Yesterday evening I had a GPU hang after 24h of squeezing the i845. First symptom was that totem player was shutting down without comment after it had opened an video which I had already watched before. Then I logged out and instead of showing the login screen the computer became unresponsive. I still could shutdown via ACPI. Unfortunately had no ssh installed so I could not draw an i915_error_state. I try to attach the critical part of syslog.
I have one freze at a total uptime of 69h on two i845 based computers now with the 2.6.35rc4-131+ge467e10+nopatch-generic or 2.6.35rc5-nopatch1-generic kernels and updated drivers. This is certainly much better than what I got with the 2.6.34-v9patch-generic kernel and standard lucid drivers
Michael

[Michael Rickmann (mrickma)]

Hi Tim,
unfortunately I do not have your hardware, so I have to imagine what I would do. First I would blacklist all (!!!) drivers found in the /lib/modules/xxx/kernel/drivers/net/wireless/rt2x00 directories of Brian Rogers kernels, where xxx is e.g. 2.6.35rc5-nopatch1-generic for one of his latest build. I.e. you make a file /etc/modprobe.d/rt2870.conf containing
blacklist rt2x00lib
blacklist rt2x00pci
...
and so on, including the rt2800usb and reboot.
If one of these steps works you are done and do not have to continue.
If your wlan stick is not recognized, unplug it and plug it in again.
If that has not worked load the module by hand, i.e. sudo modprobe rt2870sta (the module's name is rt2870sta also usb_core claims to have loaded the rt2870 driver)
If that has worked you may add rt2870sta to /etc/modules
if not I don't know.
Michael

[Tim Hallett (tim-hallett)]

Thanks Mike, I'll give the modprobe command a try. If the module is really loaded in 2.6.35 then it should come online.
Hopefully the wireless driver integration will be up to the usual standards by time of release.

[Johannes Hessellund (osos)]

Shouldn't this bug be targeted at the upcoming 10.04.1 ?

It has been postponed for 12. august according to http://www.phoronix.com/scan.php?page=news_item&px=ODQzMA

My old IBM Netvista crashes even on the live-cd, rendering it unable to install and fix via graphical interfaces.

[Tim Hallett (tim-hallett)]

Xserver-Xorg-Driver-Intel Gitxx722 exhibited some stability regression especially when compared to Git710. I have also noticed that kernel upgrades tend to obliterate the add-on packages for the fix, and that these need to be re-installed to be effective. I currently have the fixes applied to new kernel 2.6.32.24. Performance is very good in terms of speed and rendering quality, and I have high hopes that it will pass tonight's screensaver challenge. It was however, as noted above, necessary to re-apply all the fixes as 2.6.32.24 by itself initially crashed almost immediately upon GL screensaver start up.

[Tim Hallett (tim-hallett)]

Survived overnight test. Had one Open-GL crash, but stability is clearly much improved under kernel 2.6.35.24 (after patches are re-installed. stability was poor prior to re-installation but I cannot pinpoint which patch was lost in the kernel upgrade (driver? 855-patch?). Does anyone know any method whereby I can monitor the screensaver program in order to generate error data that may help in developing an even more stable fix?

[Natarajan (natarajansr-mdu)]

Today I have installed the patch . After that I am able to use the system normally.

I have collected the troubleshooting information and submitted through bug tracking system for your reference.

The new Bug #610294 contains the data collected from my system.

Thanks

R.Natarajan

[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?

[Dan Munckton (munckfish)]

Hi Chris. Thanks. Unfortunately I can confirm that these new drivers worsen the situation on my config (see bug #578167) and gave me consistent blank screen on 2 subsequent reboots. With current official Lucid drivers I can get to the desktop on boot but get blanks whenever gdm is restarted for any reason. Thanks for your help.

[jerrylamos (jerrylamos)]

i845G here on IBM ThinkCentre A30 Celeron 2.0 gHz.
VGA compatible controller: Intel Corporation 82845G/GL[Brookdale-G]/GE Chipset Integrated Graphics Device (rev 01)

Running Maverick Alpha 2:
Linux version 2.6.35-13-generic (buildd@palmer) (gcc version 4.4.5 20100723 (prerelease) (Ubuntu/Linaro 4.4.4-7ubuntu3) ) #18-Ubuntu SMP Sat Jul 31 01:45:46 UTC 2010

From Ubuntu Maverick forum thread Call for Testing: i8xx legacy driver:

Install the xserver-xorg-video-intel and libdrm packages:

      sudo add-apt-repository ppa:raof/aubergine
      sudo apt-get update && sudo apt-get upgrade

Boots and runs with default Grub line, just "quiet", with KMS on.
I can switch to command line with Ctrl-Alt-F1, verify the tiny print KMS, log on, exit, then do Ctrl-Alt-F7 to switch back to gdm O.K.
No xorg.conf so it is running with default intel driver I assume the new one? I'm not good at reading Xorg.0.log so it is attached.

It boots to blank screen with nomodeset
It boots to blank screen with i915.modeset=0
It boots just fine without specifying mode and without xorg.conf.

Only way I've been able to run previously is with i915.modeset=0 and xorg.conf specifying driver vesa, which works just fine.

Thanks, it's running for the moment, but there's always the next dread Alpha update...

Jerry

[Joachim Schwender (jschwender)]

Thanks for the effort, here is my feed back:
Intel Corporation 82845G/GL[Brookdale-G]/GE Chipset Integrated Graphics Device (rev 01)
Linux pharge 2.6.32-24-generic #38-Ubuntu SMP Mon Jul 5 09:22:14 UTC 2010 i686 GNU/Linux
New Xorg drivers installed like described above.
Boots fine, X starts with short with a shortly visible vertical fuzzy stripe pattern. I can also switch to tty1 ... tty6 back and forth without problems. I found this in dmesg:
[ 14.147835] [drm] initialized overlay support
...
[ 14.877139] [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0
...
[ 15.412141] intel8x0_measure_ac97_clock: measured 55935 usecs (2695 samples)
[ 15.412148] intel8x0: clocking to 48000
[ 16.708918] render error detected, EIR: 0x00000010
[ 16.708927] [drm:i915_handle_error] *ERROR* EIR stuck: 0x00000010, masking
[ 16.708948] render error detected, EIR: 0x00000010

For now it seems to work, but i will watch it, as i had sporadic complete freezes with the workarounds before....

[Joachim Schwender (jschwender)]

I got a complete freeze with black screen again. Another problem is that vlc shows only black screen when playing a video file. No matter what output method i choose, it does not show video any more. This is a show stopper for me.

[jerrylamos (jerrylamos)]

Maverick Alpha 3 here,
Linux version 2.6.35-14-generic (buildd@rothera) (gcc version 4.4.5 20100728 (prerelease) (Ubuntu/Linaro 4.4.4-8ubuntu1) ) #19-Ubuntu SMP Mon Aug 2 01:43:35 UTC 2010

i845G video graphics
Intel Corporation 82845G/GL[Brookdale-G]/GE Chipset Integrated Graphics Device (rev 01)

Running default "quiet" with no xorg.conf
Driver Version: 2:2.12.0-1ubuntu2

KMS is active, able to Ctrl-Alt-F1 to tiny print, login, exit, switch back to gdm Ctrl-Alt-F7

YouTube flash video running O.K. (HD a bit slower, this is a 2.0 gHz Celeron)

I did get a crash when booting up. Apport wasn't able to process it:

Problem in apt-xapian-index
The problem cannot be reported
This is not a genuine Ubuntu package
                                               Close

ls /var/crash
_usr_sbin_update-apt-xapian-index.0.crash
xserver-xorg-video-intel.2010-08-05_17:43:30.137447.crash

Wierd. I just installed daily build 20100803 which must contain things which are not a "genuine Ubuntu package"?

Anyway, attached is crash report. Anyone know how I can process it?

Thanks, Jerry

[jerrylamos (jerrylamos)]

Here's the xapian crash report, whatever xapian is?

Jerry

[Jean-Marie PIVETEAU (jean-marie-piveteau)]

Hello,

Yesterday I installed the "raof/aubergine PPA" package per https://wiki.ubuntu.com/X/Bugs/Lucidi8xxFreezes: I have initial symptom back again: (X hangs with white vertical stripes blinking on half upper screen). Crash occurred only about an hour after I applied the PPA.

Before this, having applied workarounds A and E, my system was much more stable, having the system working correctly for several days (more than one week last time).

I can now see those lines repeating in the yesterday kern.log, which match the crash:
Aug 15 22:23:37 jmp-desktop kernel: [ 3837.708030] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Aug 15 22:23:37 jmp-desktop kernel: [ 3837.708049] render error detected, EIR: 0x00000000
Aug 15 22:23:37 jmp-desktop kernel: [ 3837.708094] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 107902 at 107901)
Aug 15 22:23:40 jmp-desktop kernel: [ 3840.220048] [drm:i915_gem_idle] *ERROR* hardware wedged
Aug 15 22:23:45 jmp-desktop kernel: [ 3845.668207] [drm:i915_gem_entervt_ioctl] *ERROR* Reenabling wedged hardware, good luck
Aug 15 22:23:46 jmp-desktop kernel: [ 3846.284013] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Aug 15 22:23:46 jmp-desktop kernel: [ 3846.284026] render error detected, EIR: 0x00000000
Aug 15 22:23:46 jmp-desktop kernel: [ 3846.284059] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 107908 at 107901)
Aug 15 22:23:48 jmp-desktop kernel: [ 3848.432027] [drm:i915_gem_idle] *ERROR* hardware wedged
Aug 15 22:23:49 jmp-desktop kernel: [ 3849.731140] [drm:i915_gem_entervt_ioctl] *ERROR* Reenabling wedged hardware, good luck
Aug 15 22:23:49 jmp-desktop kernel: [ 3850.048026] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Aug 15 22:23:49 jmp-desktop kernel: [ 3850.048038] render error detected, EIR: 0x00000000
Aug 15 22:23:49 jmp-desktop kernel: [ 3850.048078] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 107910 at 107901)
Aug 15 22:23:52 jmp-desktop kernel: [ 3852.112020] [drm:i915_gem_idle] *ERROR* hardware wedged
Aug 15 22:23:53 jmp-desktop kernel: [ 3853.392056] [drm:i915_gem_entervt_ioctl] *ERROR* Reenabling wedged hardware, good luck
Aug 15 22:23:53 jmp-desktop kernel: [ 3853.708029] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Aug 15 22:23:53 jmp-desktop kernel: [ 3853.708042] render error detected, EIR: 0x00000000

does this help ?

"sudo apt-get update && sudo apt-get upgrade": I did not really understand when they said "DO NOT force that upgrade; it will break things.". what did they mean ? I had only the question "do you want to continue", so I answered "Yes" because I did not feel I was about to force anything.

Then only the package libldap has been updated.
The following packages have been kept: "libdrm-dev libdrm-intel1 libdrm-nouveau1 libdrm-radeon1 libdrm2 xserver-xorg-video-intel"

My computer is an IBM Netvista S42, type (or model) is 8319-LGV.

$ lspci | grep VGA:
00:02.0 VGA compatible controller: Intel Corporation 82845G/GL[Brookdale-G]/GE Chipset Integrated Graphics Device (rev 01...

[John Anthony Kazos Jr. (jakj)]

Why can't you just put the Intrepid driver back into a PPA somewhere? I was using that driver for the entire duration of Karmic and it worked fine, but now obviously it isn't available. I know free is free, but this is simply unusuable, and thank god it's my off-box I just use for email and not my main box.

The initial drivers earlier in the cycle would turn my screen on and off repeatedly, and if I hadn't been sitting right there when it happened, you would have blown out my screen and I would have had to buy a new one. The most recent I have been putting up with total X lockups requiring a hard powerdown (why do they not put reset buttons on machines nowadays?), and for the past week I have not one single time gotten up in the morning without having to hard-boot the thing.

Now I just tried the raof/aubergine stuff and all it managed to do was send my X/gdm into some sort of infinite crashing loop (which is strange, because before it went into that loop, it correctly displayed my current desktop image, but before it could draw the toolbars, kaboom). So now I have to go in and reset the blasted thing.

Just put the Intrepid driver back up. Problem solved. So what if this is "old" hardware? It's still making the entire operating system completely unusuable on it, and for how long this has gone on, I doubt I'm the only one thinking other distributions might be a good idea.

[Felix Lechner (felix.lechner)]

A few months ago, I was so frustrated that I played around with other
distributions. Suse had similar X problems. Fedora only booted in
text mode. After many hours of trying, I now have a Ubuntu combo w/o
lockups.

It would be nice to know if there are alternatives. Perhaps you could
let us know which distributions work for you. I might also switch, or
maybe we can adapt their solution to Ubuntu.

On Thu, Aug 19, 2010 at 3:37 AM, John Anthony Kazos Jr.
<email address hidden> wrote:
> Why can't you just put the Intrepid driver back into a PPA somewhere? I
> was using that driver for the entire duration of Karmic and it worked
> fine, but now obviously it isn't available. I know free is free, but
> this is simply unusuable, and thank god it's my off-box I just use for
> email and not my main box.
>
> The initial drivers earlier in the cycle would turn my screen on and off
> repeatedly, and if I hadn't been sitting right there when it happened,
> you would have blown out my screen and I would have had to buy a new
> one. The most recent I have been putting up with total X lockups
> requiring a hard powerdown (why do they not put reset buttons on
> machines nowadays?), and for the past week I have not one single time
> gotten up in the morning without having to hard-boot the thing.
>
> Now I just tried the raof/aubergine stuff and all it managed to do was
> send my X/gdm into some sort of infinite crashing loop (which is
> strange, because before it went into that loop, it correctly displayed
> my current desktop image, but before it could draw the toolbars,
> kaboom). So now I have to go in and reset the blasted thing.
>
> Just put the Intrepid driver back up. Problem solved. So what if this is
> "old" hardware? It's still making the entire operating system completely
> unusuable on it, and for how long this has gone on, I doubt I'm the only
> one thinking other distributions might be a good idea.
>
> --
> MASTER: [i845] GPU lockup (apport-crash) (Should KMS be blacklisted?)
> https://bugs.launchpad.net/bugs/541492
> You received this bug notification because you are a direct subscriber
> of the bug.
>
> Status in X.org xf86-video-intel: Confirmed
> Status in “xserver-xorg-video-intel” package in Ubuntu: Triaged
> Status in “xserver-xorg-video-intel” source package in Lucid: Triaged
>
> Bug description:
> Binary package hint: xserver-xorg-video-intel
>
> This is a MASTER bug report, i.e. not a real bug report, but a tool to help manage other bug reports.
>
> Most bug reports on i845 are probably due to the CPU/GPU incoherency problem that is now consolidated upstream at http://bugs.freedesktop.org/show_bug.cgi?id=26345 . For now, we mark all automatically reported GPU lockups as duplicates of this unless there is a reason not to.
>
>
> To use the available fix, run the following commands:
>
> apt-add-repository ppa:glasen/855gm-fix
> apt-add-repository ppa:brian-rogers/graphics-fixes
> apt-add-repository ppa:glasen/intel-driver
> aptitude update
> aptitude install linux 855gm-fix-dkms
> aptitude dist-upgrade
>
>
> There is a similar master bug report for i855 at bug 541511.
>
> To unsubscribe from this bug, go to:
> https://bugs.launc...

[jerrylamos (jerrylamos)]

Maverick A3 i845G been running for the last couple weeks O.K.
All updates applied daily.
default Intel Driver 2:2.11.0-1ubuntu2
Note boots O.K. with KMS with just "quiet". i915.modeset anything gets black screen or crippled video mode degraded or whatever.

Got 2 screen "freeze" today within an hour, with Firefox 3.6.8

Ctrl-Alt-F1 worked so logged in and got a tar dump of
i915_error_state
Xorg.0.log
dmesg
.xsession-errors

Jerry

[peakit (guntasg)]

The same crash happens even with using the mainline kernel 2.6.34-020634-generic #020634 on Lucid:

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)

This is causing so much of inconvenience.
8.04 was great ! I am repenting on my decision to upgrade to 10.04 LTS

Do we have any "workable" workarounds? Upgrading to the mainline kernel did n't work.

[peakit (guntasg)]

Chris Halse Roger, can you consider assigning this to someone more competent?

6 months on and you are still not able to fix this neither provided any workaround for this.

Thanks.

[Stenten (stenten)]

peakit,

Please consider reading the Ubuntu Code of Conduct [1] before making any further comments.

Also please see the Lucid i8xx Freezes wiki page [2] for a list of patches and workarounds to try.

[1]: http://www.ubuntu.com/community/conduct
[2]: https://wiki.ubuntu.com/X/Bugs/Lucidi8xxFreezes

[peakit (guntasg)]

Chris Halse Roger, I am sorry for making a comment (#199) which was not in the spirit of Ubuntu. Thanks Stenten for correcting me.

Regarding the Lucid i8xx Freezes wiki page, I did try using a different kernel than 2.6.32 but, it did n't work as mentioned in #198.

Though now I have downgraded my System to Ubuntu 8.04 LTS. But could you please update the wiki and remove the Workaround D#Upgrading as this does not seem to work (#192, #198)

Thanks.

[Jean-Marie PIVETEAU (jean-marie-piveteau)]

Hello,

Eight days ago, I applied "Workaround C" (upgading to the 2.11.0 xserver-xorg-video-intel driver, from https://wiki.ubuntu.com/X/Bugs/Lucidi8xxFreezes).
Note that upon reboot I could not get the normal Gnome logon screen but instead had the following errors:
"
! Ubuntu fonctionne en mode graphique dégradé.
L'erreur suivante est survenue. Vous aurez peut-être besoin de mettre à jour votre configuration pour résoudre cela.
(EE) Intel (0): No kernel modesetting driver selected.
(EE) Screen(s) found, but none have a usable configuration.
"
the "No kernel modesetting driver selected" message made me think to the "Workaround A", which I applied immediately. Then after rebooting the logon screen returned to normal state.

Since then I had about 4 hangs, and I am using my computer only about 1h30 a day... The 3 or 4 last hangs were different: the GUI was completely frozen: it did not answer the mouse or keyboard clicks (the mouse pointer was movable though). However I could CTRL-ALT-F1 and get a text logon and then could issue a "sudo reboot": may be there is a smarter thing to do than reboot ?

before rebooting I did a "dmesg": the relevant errors are the same than in my previous post (#194):
"
[17866.396057] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[17866.396072] render error detected, EIR: 0x00000000
[17866.396121] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 439601 at 439600)
[17937.484046] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[17937.484174] render error detected, EIR: 0x00000000
"

My kernel is up to date (ubuntu update line - 2.6.32-24.41).

Best regards,
Jean-Marie

[Jean-Marie PIVETEAU (jean-marie-piveteau)]

Hello,

I want to had a detail: between hangs, I always use the "hibernate" function to get my computer down (not shutdown - unless I just did an update which requires it, e.g. kernel update).

Best regards,
Jean-Marie

[Brian Rogers (brian-rogers)]

Developer Chris Wilson has published a branch of xserver-xorg-video-intel which accesses the graphics card differently and avoids the kinds of operations that cause problems with the old chipsets.

I created a PPA for it: https://launchpad.net/~brian-rogers/+archive/intel-shadow

This branch is based off a new enough version of the driver that it requires Xorg 1.8, while Lucid has version 1.7.6. Therefore this PPA depends on the xorg-edgers PPA. To use it, you need to add both:

sudo add-apt-repository ppa:xorg-edgers
sudo add-apt-repository ppa:brian-rogers/intel-shadow

You will also need to edit your xorg.conf file to enable shadow buffer mode, since it is not enabled by default:

sudo gedit /etc/X11/xorg.conf

If you do not already have a xorg.conf file, you will be editing a blank document and you can put the following in it:

Section "Device"
    Identifier "GPU"
    Option "Shadow" "True"
EndSection

If you already have xorg.conf file with contents, find the "Device" section and insert the 'Option "Shadow" "True"' line like above. Or if there is no existing "Device" section, copy the one above to the end of the file.

I welcome reports of how this works and how it compares to the other solutions.

[Skip Guenter (skip)]

Brian, I booted up the Inspiron 1100 and got the usual v10.04 black screen.

I then did Ctrl-Alt-F1 and:

sudo add-apt-repository ppa:xorg-edgers
sudo add-apt-repository ppa:brian-rogers/intel-shadow

I then created the /etc/X11/xorg.conf via nano with only:
Section "Device"
    Identifier "GPU"
    Option "Shadow" "True"
EndSection

I then did sudo apt-get update
and sudo apt-get upgrade

Installed so much I had to go upgrade the HDD (no just kidding)

next: sudo reboot

at Grub I select the top entry (current stock v10.04) and once I heard the noise that tells me I'm at the logon screen I had a vertical blue (pretty bright blue) bar and a about a 1 inch tall horizontal white bar. Ctrl-Alt-F1 and sudo service gdm restart only faded the white bar and moved it to the top of the screen.

I'm gonna reinstall v9.10 which for reasons unknown didn't give me problems on this machine as I need to get it off to my granddaughter soon.

Thanx, Skip

[Brian Rogers (brian-rogers)]

Skip, if you haven't wiped your 10.04 system yet, what happens if you remove the 'Option "Shadow" "True"' part? I'm curious if the shadow buffer mode is introducing a new problem or just not fixing an existing problem. If normal 10.04 fails to even show the login screen, then how did you get that version on there? Did you start with 9.10 and upgrade to 10.04?

A suggestion: If you set the system up to dual boot between 9.10 and 10.04, you can have a partition for testing while you keep another partition ready for use.

[John Anthony Kazos Jr. (jakj)]

Making me install the xorg-edgers repository would be useless. I have already tried that, and only ppa-purge saved me from having to wipe the entire freaking system to reinstall. I never saw any Unix system produce a more violent crash/lockup/retry cycle than that repository did to my machine.

Using a stock Dell Dimension 2400, by the way.
00:02.0 VGA compatible controller [0300]: Intel Corporation 82845G/GL[Brookdale-G]/GE Chipset Integrated Graphics Device [8086:2562] (rev 01)

[Brian Rogers (brian-rogers)]

John, comments like yours aren't useful. If you don't want to test something that might solve your problem, you're completely free to keep quiet.

[John Anthony Kazos Jr. (jakj)]

I'm simply saying that it is entirely untestable on my machine, because just running the packages in that repository on that hardware made me think my monitor was about to blow its tube.

isn't it possible to get the old Intrepid driver working with Lucid's X version, or has it changed too much? That driver seemed to work in Karmic. I believe I was using xserver-xorg-video-intel-2.4 from https://launchpad.net/~siretart/+archive/ppa?field.series_filter=jaunty back when I had Jaunty. If I'm not mixing the two up in my mind. It's been a while.

(I would gladly just use vesa on this machine and be done with it, but the driver doesn't support above 640x480 at 16-bit color. Which I might actually be willing to try to use if the Ubuntu interface weren't designed with the idea of "nobody would ever run anything at 640x480 ever again" in mind. If anyone could get me working at at least 800x600, that would at least get me past the point of trying to decide between crash-modes that try to fry my screen or crash-modes that force me to hard-reboot the machine.)

[jerrylamos (jerrylamos)]

Brian,

I'd like to test it. I've got i845 and i830 which have had a LOT of trouble with Maverick, Lucid, ....

At the moment both of them are running Maverick Alpha 3 updated to 31 August with default:

2.6.32-24-generic #42-Ubuntu SMP Fri Aug 20 14:24:04 UTC 2010 i686 GNU/Linux

dpkg -s xserver-xorg-video-intel | grep Version: Version: 2:2.9.1-3ubuntu5

As of a week or two ago they would hang randomly. Not lately. Yay!

I do ordinary user internet email, web browse, network printing, Open Office, test Ubuntu Alpha etc., .... I don't do games or eye candy. Of course the next update might tank them again, for the nth time, starting with Intrepid. Then it's back to i915.modeset=0 and xorg vesa.

I have had poor results on trying a fix to back fixes off to what it was before.
Also I'm not sure the fixes will be compatible with the usual torrent of daily updates?
I intend to do a second Maverick install on the i845G which has enough disk space and then try the fix. That way I don't disturb the current working version and have it to fall back on.

Thanks for working on this long persisting problem area.

Jerry

[Joachim Schwender (jschwender)]

I have made a strange observation, it may be helpful to others:
Since i double boot, i have no more hang. Double boot means: I interrupt the booting by hitting ctrl+alt+del during the kernel initialization phase, which causes the machine to reboot, and the second time i let it start normally. Others may try if that helps on their machine too or not.
Intel Corporation 82845G/GL[Brookdale-G]/GE Chipset Integrated Graphics Device (rev 01)

[Jean-Marie PIVETEAU (jean-marie-piveteau)]

Brian,

I did the:
sudo add-apt-repository ppa:xorg-edgers
sudo add-apt-repository ppa:brian-rogers/intel-shadow
and then the apt-get update & apt-get upgrade commands.

because (I guess) KMS was still enabled I always had a black screen after bootup. The sequence was this:
   IBM BIOS screen
   Grub loading stage1.5.
   Starting up ...(bold)
   Starting up ...(thin)
   Ubuntu (white) + the five red dots
   black screen + fixed cursor at upper left corner (monitor LED is fixed green)
   the little drum sound (which says the "GUI" is ready)
but I could almost always do CTRL-ALT-F1 and (blindly) to a text login and then reboot.

then I disabled KMS and I can now have the GNOME logon screen but however is sticks at 640x480, i.e. using gnome-display-properties (because you cannot have access to the "System" menu at 640x480) you cannot change anything; the detect monitor" button does nothing.

I am right now adding this comment using this VGA resolution...

Here attached more details on my current configuration.

Tell me if you need more information.

Best regards,
Jean-Marie

[jerrylamos (jerrylamos)]

Just installed 20100831 on this i845G
VGA compatible controller: Intel Corporation 82845G/GL[Brookdale-G]/GE Chipset Integrated Graphics Device (rev 01)
which has had black screen, hangs, ... on Maverick (and Lucid, Karmic, ...) frequently, depending on what updates have done to it lately.

With:
Linux jerry-desktop 2.6.35-19-generic #28-Ubuntu SMP Sun Aug 29 06:36:51 UTC 2010 i686 GNU/Linux
and intel version: Version: 2:2.12.0-1ubuntu3

it boots normally using "quiet" and no xorg.conf to gdm with KMS active. I can Ctrl-Alt-Del-F1, log in, do some stuff, and switch back to gdm O.K.

Now I don't know if it will hang after a while. A couple of weeks ago it would, but then updates seem to fix that. By the way, my i830 all up to date on Maverick hasn't hung or anything lately either. Early on in Maverick it needed "vesa" and i915.modeset=0. Not today.

Let me try the Brian Rogers fix. I added his ppa and the xorg.conf. Now to do update. I assume Maverick has the latest Xorg (how do I tell) so I didn't do that.

Jerry

[Jean-Marie PIVETEAU (jean-marie-piveteau)]

Brian,

I did an important omission: the Xorg.0.log file. Here it is.

It shows that X cannot work with my intel driver because it is not compiled for the X server version 1.8. (as you mentioned in your post #204 offering your ppa).
Attempting to solve my i845 X problems (subject of this master bug report) I did install (see my post #202) the 2.11 intel driver (which is compiled for the version 1.7.6 of X) and I am still using it... and I don't know how to "detach" from it: I did a "sudo ppa-purge ppa:ubuntu-x-swat/x-updates" but when asking me if I accept the proposed solution, I did not know what to do (the "score" was -8386: what does this mean ?)

anyway, the vesa driver is then loaded automatically BUT it configures itself at the VGA resolution and cannot do better.

what do I need to do to either:
- install the intel driver from your ppa "intel-shadow"?
- or (in the meantime), make the vesa able to display upper resolutions e.g. 1024x760 ?

Best Regards,
Jean-Marie

[jerrylamos (jerrylamos)]

The Brian Rogers fix was put on Maverick installed from CD Live 20080831.
Version: 2:2.12.0+shadow+77.gb933ead-maverick
There is an xorg.conf to turn shadow on.
This is on i845G.

Booted up twice, it runs, but gets a crash report:
xserver-xorg-video-intel.2010-09-01_18:39:09.163479.crash

I have no idea how to process that report? I tried selecting the applet on the top panel however it said it couldn't process....I tried browsing the crash report and didn't notice anything.

I don't notice anything in Xorg.0.log either.

xsession-errors has this info at the end:
.....
(polkit-gnome-authentication-agent-1:1243): GLib-GObject-WARNING **: cannot register existing type `_PolkitError'

(polkit-gnome-authentication-agent-1:1243): GLib-CRITICAL **: g_once_init_leave: assertion `initialization_value != 0' failed

** (bluetooth-applet:1242): WARNING **: Could not open RFKILL control device, please verify your installation
Present: 0, Powered: 0, Default: 0

(bluetooth-applet:1242): GLib-GObject-WARNING **: invalid cast from `GtkToggleAction' to `GtkWidget'

(bluetooth-applet:1242): Gtk-CRITICAL **: IA__gtk_widget_set_sensitive: assertion `GTK_IS_WIDGET (widget)' failed
Window manager warning: Failed to read saved session file /home/jerry/.config/metacity/sessions/1098a9e1245b304f8b128338072912787100000010480026.ms: Failed to open file '/home/jerry/.config/metacity/sessions/1098a9e1245b304f8b128338072912787100000010480026.ms': No such file or directory
Unable to find a synaptics device.
Initializing nautilus-gdu extension
** Message: applet now removed from the notification area

(nautilus:1240): GConf-CRITICAL **: gconf_value_free: assertion `value != NULL' failed
** Message: applet now embedded in the notification area

(nautilus:1240): GConf-CRITICAL **: gconf_value_free: assertion `value != NULL' failed

(nautilus:1240): GConf-CRITICAL **: gconf_value_free: assertion `value != NULL' failed
** (update-notifier:1421): DEBUG: Skipping reboot required
/var/crash/xserver-xorg-video-intel
** (update-notifier:1421): DEBUG: fire up the crashreport tool

[jerrylamos (jerrylamos)]

Got a GPU hang (I think) twice with the "shadow-intel" fix on Maverick and on re-booting ran ubuntu-bug xorg which generated bug #628556.

Version: 2:2.12.0+shadow+77.gb933ead-maverick

Without the fix's xorg.conf it boots without a crash...let me see if it hangs.

Note, prior to the fix, Maverick 20100831 wasn't having any trouble with this i845G, limited running ...

Jerry

[Jean-Marie PIVETEAU (jean-marie-piveteau)]

Brian,

Follow-up of my #214: I tried many times figuring out how to make vesa driver make use of the 1024x768, to no avail. Each time modifying the xorg.conf file and adjusting the screen and monitor sections with tricks I found on the "change resolution/refresh rate in Xorg" HOWTO (http://ubuntuforums.org/showthread.php?t=83973). After all may be vesa cannot use higher resolution than 640x480, but most probably I haven't understood some basics about how all these things works :-)

Anyway, I just realized about an hour ago that I could upgrade from the 2.11 driver to the version 2.12 provided by your package.
And by issuing the command "sudo apt-get install xserver-xorg-video-intel=2:2.12.0+shadow+77.gb933ead-lucid2" I could upgrade to the 2.12 (compiled for the 1.8.2 X server!)

then I had again to re-enable KMS using workaround A from https://wiki.ubuntu.com/X/Bugs/Lucidi8xxFreezes#Workaround%20A:%20Re-enable%20KMS, otherwise I got the "intel(0): No kernel modesetting driver detected" error from the X server and a black screen (CTRL-ALT-F1 usable though).

I am up again with my preferred resolution 1024x768 !!

My X server is now therefore 1.8.2 and intel driver version is 2.12.0 (from your package).

Let's hope it will be stable :-)

My configuration as of now:
- uname -a:
Linux jmp-desktop 2.6.32-24-generic #42-Ubuntu SMP Fri Aug 20 14:24:04 UTC 2010 i686 GNU/Linux
- cat /etc/lsb-release:
DISTRIB_ID=Ubuntu
DISTRIB_RELEASE=10.04
DISTRIB_CODENAME=lucid
DISTRIB_DESCRIPTION="Ubuntu 10.04.1 LTS"
- head -2 /var/log/Xorg.0.log | tail -1:
X.Org X Server 1.8.2
- Here attached my last X server boot log.

Best Regards,
Jean-Marie

[jerrylamos (jerrylamos)]

Maverick Beta was running yesterday with i845G, KMS on.
Today's update included kernel 2.6.35-20 plus usual pile of other stuff.
Now gdm won't come up with either that kernel or yesterday's, i915.modeset=0 no help.
I'm up with driver "vesa".
Command line was still running so I did ubuntu-bug which is attached.

Jerry

[jerrylamos (jerrylamos)]

Ref comment #218
Not sure what's in the apport report so here's a tar dump of:
i915_error_state
Xorg.0.log
dmesg
xsession-errors
uname-a
Intel Version

Jerry

[Chris Halse Rogers (raof)]

X is running properly with vesa, yes? If you have no xorg.conf, you get
the vesa driver and X works properly?

If you set i915.modeset=1 and have no xorg.conf, you should get the
intel driver.

Are both of these statements true?

[jerrylamos (jerrylamos)]

I get "vesa" by specifying it in xorg.conf.
With no xorg.conf ran fine yesterday, KMS and all. Did updates today and X fails. Only get a command line mode and startx won't. Lots of complaints in .xsession-errors and Xorg.0.log see attachments to comment #219.

I tried i915.modeset=0 with no xorg.conf and X failed.

It runs now with "vesa" and just specifying "quiet" in the grub boot line. I don't have any use for "splash".

I'm not a developer so I don't know which of the many updates caused the trouble. I tried going back to kernel 2.6.35-19 and it doesn't work either although it did yesterday. I looked at the intel driver version and it was the same today as yesterday.

I did and do have system, preferences, appearance, visual effects, none. I haven't used Compiz since it broke intel video graphics in Intrepid.

?

Jerry

[useResa (rdrijsen)]

As jerrylamos I am also running Maverick. I have been running it since Alpha 1 on my desktop (Dell Dimension 2400) with an i845G graphics card and was pleased that the intel graphics card issues seemed to have been resolved.

However ... experienced exactly the same as jerrylamos that the upgrade to the 2.6.35-20 kernel resulted in a black screen. But I have been able to use the 2.6.35-19 kernel as before the upgrade so no issues there (at least not until about 2 hours ago).

Is there any additional information I can provide or tests I can perform?

[Brian Rogers (brian-rogers)]

2.6.35-20 blacklists KMS on the problematic cards. X is then supposed to fall back to vesa. If it doesn't, a bug should be filed for that. If falling back to vesa has enough problems of its own, that decision might be reversed.

i915.modeset=1 will override the blacklist, so you can always turn KMS back on with the new kernel. Also, my own graphics-fixes kernel undoes the blacklist.

[Brian Rogers (brian-rogers)]

Also, to catch up with an older comment...

John, Intrepid's driver can't be built for a newer system. I'm also not sure it would solve the problem if someone forward-ported it. As far as I can tell, the driver has always been broken for these chipsets, but was being used in just the right circumstances on older systems that it in practice didn't crash that often. That's why I had a hard time bisecting the issue. Everybody got different results for which versions were unstable, so I had to give up on that approach.

[useResa (rdrijsen)]

Until now Maverick worked fine on my desktop without having to alter the boot options.
From comment #223 I understand that something has been altered with the coming of the 2.6.35-20 kernel now resulting in a black screen at startup.
Using the i915.modeset=1 indeed solved the issue for me now and gave me a problemless start up. So will add this to the default boot options until .... ??

Do I understand correctly that a separate bug is to be filed for the fact that with the 2.6.35-20 kernel the system does not start up properly?

[Brian Rogers (brian-rogers)]

Yeah, a separate bug should be filed because it means either the system failed to fallback to vesa when KMS wasn't available or the vesa driver didn't work right.

[jerrylamos (jerrylamos)]

Brian, bug # is 633593 which I think is what you were asking for. This is i845G.

i915.modeset=1, boots fine, KMS on. Runs O.K.

xorg.conf with "vesa" boots fine. Runs O.K.

Default gets "no screen found". Command line works so Xorg.0.log is available.
Relevant section (I think) of Xorg.0.log is in bug #633593:
[ 17.647] (==) Matched intel as autoconfigured driver 0
[ 17.647] (==) Matched vesa as autoconfigured driver 1
[ 17.647] (==) Matched fbdev as autoconfigured driver 2
[ 17.647] (==) Assigned the driver to the xf86ConfigLayout
[ 17.647] (II) LoadModule: "intel"
[ 17.648] (II) Loading /usr/lib/xorg/modules/drivers/intel_drv.so
[ 17.672] (II) Module intel: vendor="X.Org Foundation"
[ 17.672] compiled for 1.8.99.905, module version = 2.12.0
[ 17.672] Module class: X.Org Video Driver
[ 17.672] ABI class: X.Org Video Driver, version 8.0
[ 17.672] (II) LoadModule: "vesa"
[ 17.673] (II) Loading /usr/lib/xorg/modules/drivers/vesa_drv.so
[ 17.673] (II) Module vesa: vendor="X.Org Foundation"
[ 17.673] compiled for 1.8.99.905, module version = 2.3.0
[ 17.673] Module class: X.Org Video Driver
[ 17.673] ABI class: X.Org Video Driver, version 8.0
[ 17.673] (II) LoadModule: "fbdev"
[ 17.674] (II) Loading /usr/lib/xorg/modules/drivers/fbdev_drv.so
[ 17.677] (II) Module fbdev: vendor="X.Org Foundation"
[ 17.677] compiled for 1.8.99.905, module version = 0.4.2
[ 17.677] ABI class: X.Org Video Driver, version 8.0
[ 17.677] (II) intel: Driver for Intel Integrated Graphics Chipsets: i810,
 i810-dc100, i810e, i815, i830M, 845G, 852GM/855GM, 865G, 915G,
 E7221 (i915), 915GM, 945G, 945GM, 945GME, Pineview GM, Pineview G,
 965G, G35, 965Q, 946GZ, 965GM, 965GME/GLE, G33, Q35, Q33, GM45,
 4 Series, G45/G43, Q45/Q43, G41, B43, Clarkdale, Arrandale
[ 17.679] (II) VESA: driver for VESA chipsets: vesa
[ 17.679] (II) FBDEV: driver for framebuffer: fbdev
[ 17.679] (++) using VT number 7

[ 17.683] (WW) Falling back to old probe method for vesa
[ 17.683] (WW) Falling back to old probe method for fbdev
[ 17.684] (II) Loading sub module "fbdevhw"
[ 17.684] (II) LoadModule: "fbdevhw"
[ 17.684] (II) Loading /usr/lib/xorg/modules/libfbdevhw.so
[ 17.693] (II) Module fbdevhw: vendor="X.Org Foundation"
[ 17.693] compiled for 1.9.0, module version = 0.0.2
[ 17.693] ABI class: X.Org Video Driver, version 8.0
[ 17.694] (EE) open /dev/fb0: No such file or directory
[ 17.707] (EE) intel(0): No kernel modesetting driver detected.
[ 17.707] (II) UnloadModule: "intel"
[ 17.708] (EE) Screen(s) found, but none have a usable configuration.
[ 17.708]
Fatal server error:
[ 17.708] no screens found

[useResa (rdrijsen)]

jerrylamos thanks for filing the other bug.
I have indicated that I am also affected with the same bug and have added my Xorg.0.log file for reference.

[jerrylamos (jerrylamos)]

i830 gets same problem with' Beta 2.6.35-20. That level of Beta is supposed to default to "vesa" according to Brian Rogers but does not, skips the intel driver on purpose, skips vesa by mistake, then fbdevhw gets "Screen(s) found, but none have a usable configuration". I'll update bug #633593 as well.

So I specified i915.modeset=1 which bypasses the "KMS blacklist", gets the default intel driver, KMS is up and running, and for ordinary use even internet videos running O.K.

Frankly, Maverick Beta you tube videos running better than any previous Ubuntu on this 1 gHz Celeron. HD rather jerky of course.

For my i830 and my i845G running the default beta intel driver Version: 2:2.12.0-1ubuntu3 running fine for ordinary use. "vesa" also runs O.K. on this i830.

I'll be testing Beta updates to see if the bug of not defaulting to "vesa" gets fixed.

Jerry

[Jean-Marie PIVETEAU (jean-marie-piveteau)]

Hello,

Following my comment #217:
I had no problem since this comment until tonight.
I was looking a video from http://www.youtube.com/NASATelevision and found the video very jerky, however it was already jerky before my upgrade to X 1.8.2 and the 2.12 intel video driver from Brian package. I think this jerky effect became very pronounced since I upgraded from 8.04 to
10.04.
Hence, I wanted to know if hardware acceleration was working or not. Looking at ubuntu forum I found I could use the command: glxinfo | grep "direct rendering" to have the answer.
So I typed the glxinfo command above and the result was (all three in a row):
1 - the GUI disappeared, replaced by many error text lines
2 - a black screen (cursor at upper left corner)
3 - the GUI reappeared but offering me to logon ...

By looking at /var/log/Xorg.0.log.old, I can see a segmentation fault occured (see attached file).
Can't I use the "glxinfo" command ?
What can I do to:
- not have this error
- have a video of better quality

Best Regards,
Jean-Marie

[Brian Rogers (brian-rogers)]

My shadow branch PPA is now obsolete because the code was committed upstream. So now you only need xorg-edgers and the xorg.conf changes to enable a shadow buffer.

Jean-Marie, do you observe the glxinfo-triggered segfault if you remove my shadow branch PPA and switch the xserver-xorg-video-intel package back to the version in xorg-edgers? You can select the package in synaptic and choose 'Package -> Force version' to change back.

If you still experience a crash from glxinfo after making that change and restarting, then go ahead and file a new bug for it and post the link here, and I'll take a look at it.

[E. Rinehart (aer0usa-gmail)]

Hi,
Would more log dumps and system info help, or is this problem well-defined enough that it wouldn't help? I can provide info if it would help.

Also, since this trouble seems to be linked to Intel display drivers, has anyone "solved" this problem by adding a cheap video card to their system?

Thanks,
Eric

[dbclinton (dbclin)]

On Mon, 2010-09-13 at 16:08 +0000, E. Rinehart wrote:
> Also, since this trouble seems to be linked to Intel display drivers,
> has anyone "solved" this problem by adding a cheap video card to their
> system?
>
I've tried adding different video cards but I believe the problem is
with the on-board controller rather than the card itself.

[Anthony Protopapas (aprotopapas)]

I have revision 1 of the 845 chip.

I tested the Shadow option in xorg.conf using the latest updates from the xorg-edgers PPA but found no improvement. This same hardware has been running Ubuntu with no problems for 4 years until I upgraded from 8.04 to 10.04 a couple of days ago.

With kernel 2.6.35-20 the GPU hangs within a few minutes.
With kernel 2.6.32-25 the GPU hangs after anywhere from half an hour to a few hours with the following error:

[ 1568.344011] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[ 1568.344024] render error detected, EIR: 0x00000000
[ 1568.344044] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 9298 at 9297)

Before this test, I was running kernel 2.6.32-25 with KMS enabled and DRI disabled with latest updates from swat-X PPA with GPU hanging every few hours.

As a workaround for us 845 folks, is it feasible to roll back the graphics architecture that relies on mesa, dri, etc and use the old i810 driver?

@Jean-Marie: I have also tried but failed to downgrade to the vesa driver. It seems there is a problem somewhere between vesa and the 845 video BIOS so that X retrieves a video buffer size of only 832Kb even if you increase that value to 8Mb from the system BIOS.

[peakit (guntasg)]

> As a workaround for us 845 folks, is it feasible to roll back the graphics architecture that relies on mesa, dri, etc and use the old i810 driver?

I liked the proposal made by Anthony. Seems like all we are trying to do in this request is somehow "suppress" the calls/features which are resulting in this "GPU hung" situation. But what we are missing here is doing the suppression is impacting the user experience (as mentioned in Comment #230 - Jean is finding the youtube videos "jerky" after installing the patch)

Shall we start thinking in rolling back to the i810 driver? Can someone explain me why we are not doing that?

Thanks!

[Jean-Marie PIVETEAU (jean-marie-piveteau)]

Brian,

Per your comment #231, I removed the shadow branch PPA: ok
I installed the version "2:2.12.0+git20100913.08c2caca-0ubuntu0sarvatt~lucid": I guess this is the version from xorg-edgers.
Then I shutdown my computer and powered on again.
I tested glxinfo | grep "direct rendering": answer is "direct rendering: Yes".
I can also run my Celestia program (which uses OpenGL I think), so this seems stable.

So have not the segmentation fault anymore. Thank you for the help.

@Anthony: thank you for your info. Maybe you are right. I am not expert enough with this.

As I told before the youtube video seems significantly more jerky than before, but I can't tell you for sure if this is since upgrading to 10.04, or upgrading my X server to 1.8.2, sorry for this. Maybe there exist some performance measurement tool that would reflect more objectively this video "fluidity downgrade" ...

Best Regards,
Jean-Marie

[Brian Rogers (brian-rogers)]

Yeah, that's the xorg-edgers version. Do you still have Option "Shadow" "True" in the device section of your xorg.conf? Because with shadow buffer mode enabled, OpenGL isn't supposed to work (but it shouldn't crash, it should just revert to software rendering).

If glxinfo says "direct rendering: Yes", then it sounds like shadow buffer mode isn't enabled. If you still have it enabled in xorg.conf, then can you attach Xorg.0.log? If it wasn't enabled, then can you check if there's still a crashing bug when running glxinfo or other OpenGL programs with it enabled, and if so, file a bug?

[Jean-Marie PIVETEAU (jean-marie-piveteau)]

Brian,

I do not have the Option "Shadow" "True" in my xorg.conf.
If I insert it, restart the GUI and run xglinfo, actually yes I do have a crash !
The GUI disapears and the screen seems in a logged off position but is black with the mouse pointer in the middle of the screen.

When I launch the ubuntu-bug to create the bug, first ubuntu-bug asked me to create a crash file, so I started apport, then re-ran ubuntu-bug, but ubuntu-bug then seems to trigger the X crash when collecting the files ...

I have right now the crash file in /var/crash/_usr_bin_Xorg.0.crash available.

As ubuntu-bug cannot complete, I tried to file the bug with "sudo apport-cli -c /var/crash/_usr_bin_Xorg.0.crash" but this command also crashes the X server ...

I opened the bug directly via Launchpad: number is #640058
I cannot upload the crash file, even gzipped (about 2,5MB) ... too big ?

Best Regards,
Jean-Marie

[sam (samuel-j-1993)]

Is there any chance that maverick will be any more stable than lucid in regards to this

from the comments i would say it would be?

[sam (samuel-j-1993)]

Is there any chance that maverick will be any more stable than lucid in regards to this bug?

typo in comment 239 correction

[Brian Rogers (brian-rogers)]

Maverick now defaults to using the fbdev xorg driver by default instead of intel for i8xx chips, and this should work around the freezing bug.

[useResa (rdrijsen)]

@sam,
I have been running Maverick since Alpha 1. Initially with the bleeding edge X but since Alpha 3 with the "default" X environment. I have replaced Lucid with Maverick due to this bug and I have to admit that (besides a few "bumps") Maverick has been much more stable than Lucid ever was on my desktop.

So from my perspective it seems that they have it under control in Maverick.

[peakit (guntasg)]

Good to hear that at least Maverick does not suffer from this problem (though I have n't tested that myself and can't confirm).

Given that Lucid (10.04) was an LTS release (3 years of support!), it is very very important to get this fixed. Many i845 folks are awaiting for this bug to get fixed, before jumping onto Lucid. And rightly so this bug has been marked as 'Critical'.

[sam (samuel-j-1993)]

Good to hear that maverick will have some more stability than lucid
i spend a fair bit of time trying fixes to get my GX260 stable enough on lucid

Any Change Compiz may work again in the future? Though i must say i havent actually tried compiz on an 845gm so i dont know how it would work

[jerrylamos (jerrylamos)]

fbdev by itself didn't work on my i845G on the update that included 2.6.35-21. "No screens found."

KMS+fbdev did work on my i845G on yesterday's update that included 2.6.350-22.

Haven't done much testing. The couple internet videos I tried did work.

Jerry

[jerrylamos (jerrylamos)]

KMS+fbdev still working on i845G updated 20 September kernel 2.6.35-22. Some info from Xorg.0.log. Amazing to me how Ubuntu notes some problems coming up and then recovers to run O.K.:

Linux jerry-desktop 2.6.35-22-generic #33-Ubuntu SMP Sun Sep 19 20:34:50 UTC 2010 i686 GNU/Linux

Intel Corporation 82845G/GL[Brookdale-G]/GE Chipset Integrated Graphics Device (rev 01)

[ 15.949] (EE) VESA: Kernel modesetting driver in use, refusing to load
[ 15.949] (WW) Falling back to old probe method for vesa
[ 15.949] (II) Loading sub module "fbdevhw"
[ 15.949] (II) LoadModule: "fbdevhw"
[ 15.950] (II) Loading /usr/lib/xorg/modules/libfbdevhw.so
[ 15.950] (II) Module fbdevhw: vendor="X.Org Foundation"
[ 15.950] compiled for 1.9.0, module version = 0.0.2
[ 15.951] ABI class: X.Org Video Driver, version 8.0
....
        "Default Screen Section" for depth/fbbpp 24/32
[ 15.951] (==) FBDEV(0): Depth 24, (==) framebuffer bpp 32
[ 15.949] (EE) VESA: Kernel modesetting driver in use, refusing to load
[ 15.949] (WW) Falling back to old probe method for vesa
[ 15.949] (II) Loading sub module "fbdevhw"
[ 15.949] (II) LoadModule: "fbdevhw"
[ 15.950] (II) Loading /usr/lib/xorg/modules/libfbdevhw.so
[ 15.950] (II) Module fbdevhw: vendor="X.Org Foundation"
[ 15.950] compiled for 1.9.0, module version = 0.0.2
[ 15.951] ABI class: X.Org Video Driver, version 8.0
[ 15.951] (**) FBDEV(0): claimed PCI slot 0@0:2:0
[ 15.951] (II) FBDEV(0): using default device
[ 15.951] (II) FBDEV(0): Creating default Display subsection in Screen section
        "Default Screen Section" for depth/fbbpp 24/32
[ 15.951] (==) FBDEV(0): Depth 24, (==) framebuffer bpp 32
[ 15.949] (EE) VESA: Kernel modesetting driver in use, refusing to load
[ 15.949] (WW) Falling back to old probe method for vesa
[ 15.949] (II) Loading sub module "fbdevhw"
[ 15.949] (II) LoadModule: "fbdevhw"
[ 15.950] (II) Loading /usr/lib/xorg/modules/libfbdevhw.so
[ 15.950] (II) Module fbdevhw: vendor="X.Org Foundation"
[ 15.950] compiled for 1.9.0, module version = 0.0.2
[ 15.951] ABI class: X.Org Video Driver, version 8.0
[ 15.951] (**) FBDEV(0): claimed PCI slot 0@0:2:0
[ 15.951] (II) FBDEV(0): using default device
[ 15.951] (II) FBDEV(0): Creating default Display subsection in Screen section
        "Default Screen Section" for depth/fbbpp 24/32
[ 15.951] (==) FBDEV(0): Depth 24, (==) framebuffer bpp 32
[ 16.008] (==) FBDEV(0): Backing store disabled
[ 16.010] (EE) FBDEV(0): FBIOPUTCMAP: Invalid argument
[ 16.010] (EE) FBDEV(0): FBIOPUTCMAP: Invalid argument
....
many many screensful of the FBIOPUTCMAP Invalid line
....
[ 16.081] (EE) FBDEV(0): FBIOPUTCMAP: Invalid argument
[ 16.082] (==) FBDEV(0): DPMS enabled
[ 16.082] (==) RandR enabled
[ 16.082] (II) Initializing built-in extension Generic Event Extension
....
[ 16.083] (II) Initializing built-in extension GESTURE
[ 16.150] (II) AIGLX: Screen 0 is not DRI2 capable
[ 16.150] (II) AIGLX: Screen 0 is not DRI capable
[ 16.167] (II) AIGLX: Loaded and initialized /usr/lib/dri/swrast_dri.so
[ 16.171] (II) GLX: Initialized ...

[peakit (guntasg)]

Very excited to read that the patch is working fine!

> Good YouTube video except HD jerky as expected on this pc.

Jerry,

Can you please let us know that whether you were experiencing this "jerkiness" before Lucid? Or is this loss in user experience due to this patch?

Thanks.

[sam (samuel-j-1993)]

Maverick 2.6.35.22-generic

Working quite well. some youtube videos a little slow and jittery \
framerate seems a bit low and cant manage to play games.

no crashes yet though.

so far so good!

[jerrylamos (jerrylamos)]

i830 running KMS+fbdevhw & shadow O.K. I think, reading from Xorg.0.log at
2.6.35-22-generic #32-Ubuntu SMP Wed Sep 15 23:39:25 UTC 2010 i686 GNU/Linux

Had a couple kernel crashes bugs #543717 and #643759 but X running O.K. including YouTube videos.

Thanks much, all

Jerry

[jerrylamos (jerrylamos)]

peakit,

This is a 1 gHz Celeron with integrated i830 video Thinkpad which does a usable job at BBC video.

My 2 ghz Celeron i845G video tower is better. HD video is jerky with that hardware as expected.

I also test with a 3.3 gHz Celeron ati radeon Xpress 200 tower which goes as fast as my internet service will go.

They're all on Maverick Beta multibooted with Lucid, Lubuntu, ... Lubuntu Maverick Beta with LXDE & Chromium appears to be faster than Gnome Firefox haven't done any tests. I prefer running Lubuntu on my two slower machines while Gnome Firefox is fine on my 1.5 gHz P4 ati Thinkpad T40 and the 3.33 gHz Celeron.

General impression is KMS+fbdevhw running O.K. for the two days it has been available on 2.6.35-22. I haven't done any Gtkperf video test comparisons yet vs. vesa and i915 drivers.

Jerry

[jerrylamos (jerrylamos)]

peakit,

KMS+fbdevhw "shadow" i845G performance with Gtkperf about 34 seconds, about same as "vesa" i925.modeset=0. This hardware not very fast but perfectly usable. My 3.3 gHz Celeron is 3 times faster, between 9 and 12 seconds depending on the release of Ubuntu.

That presumes I'm reading Xorg.0.log on which drivers are actually in use.

intel driver i915.modeset=1 definitely slower taking 50 seconds to run the same Gtkperf suite as KMS+fbdevhw ran in 34 seconds.

Jerry

[peakit (guntasg)]

Sam, thanks for testing the patch.

> some youtube videos a little slow and jittery framerate seems a bit low and cant manage to play games
This seems to be scaring me. So far 2 people have applied the patch and, both of them are reporting the jerkiness/slowness in viewing videos. Seems its quite observable. Slowness of few milliseconds is enough to spoil the user experience. I will let Ubuntu developers to take a note of this.

[Chris Halse Rogers (raof)]

Video playback will suffer - there is no hardware acceleration in fbdev.
On the other hand, actually using the hardware video scaling in Intel
has been one excellently reproducible way to hang the GPU on some
people's chips.

Video scaling is the main case where hardware acceleration is faster
than doing it in software.

[peakit (guntasg)]

Thanks Chris Halse Rogers!

So, now (after the patch) the video scaling is happening in software rather than hardware. And this is causing the jerkiness. Hmm.. got it.

Chris, before Lucid the same Intel chips used to work quite well even with HD youtube videos. Could you please explain us what so fundamental changed in this release which is resulting into this behavior? And also why there is no way to rollback to the previous drivers on Lucid?

Would greatly appreciate this,

Thanks.

[jerrylamos (jerrylamos)]

peakit, as a user/tester "what so fundamental changed:"

Back on Intrepid Compiz decided to use some video code paths no one else does so wham black screen. Easy fix, don't run Compiz eye candy. Doesn't have anything to do with applications as they run.

Next the dread "KMS" showed up where the kernel as it was booting up turned on video graphics mode. Oops, X windows uses video graphics mode too. Seems to me lots of problems co-ordinating the kernel use of video graphics vs. X windows use. Some of the developers claim KMS is a big advantage switching users since no graphics mode reset is needed. I never ever switch users. Ever. Also switching to command line with Ctrl-Alt-F1 and back to X with Ctrl-Alt-F7 works better too. Only time I ever do that is for chasing bugs where Ubuntu is broken big time. The developers also allude to some future plans which I haven't seen elaborated.

At various update levels the "KMS" code has killed my ati video graphics to say nothing of the many trials and tribulations on my two intel pc's.

That's my viewpoint just looking at Ubuntu as a "black box". I haven't the faintest idea why developers want to affect video graphics all the time as it runs just so switching users will go more smoothly. Must be some other reason....

Jerry

[sam (samuel-j-1993)]

Running Maverick With Latest Updates and using whatever graphics driver is default
did switch to intel driver which worked for a while though games were still slow but ended up crashing so i switched back to default

[Douglas Radcliffe (doughoist)]

I am just asking. Why did Ubuntu 9.04 work so well and there was no issue
with my hard ware, but ever since, beginning with 9.10, this problem has
persisted. What Changed and why not go back to that version of X?

Doug

On Fri, Oct 1, 2010 at 12:14 AM, sam <email address hidden> wrote:

> Running Maverick With Latest Updates and using whatever graphics driver is
> default
> did switch to intel driver which worked for a while though games were still
> slow but ended up crashing so i switched back to default
>
> --
> MASTER: [i845] GPU lockup (apport-crash) (Should KMS be blacklisted?)
> https://bugs.launchpad.net/bugs/541492
> You received this bug notification because you are a direct subscriber
> of a duplicate bug (456902).
>
> Status in X.org xf86-video-intel: Confirmed
> Status in “xserver-xorg-video-intel” package in Ubuntu: Triaged
> Status in “xserver-xorg-video-intel” source package in Lucid: Triaged
>
> Bug description:
> Binary package hint: xserver-xorg-video-intel
>
> This is a MASTER bug report, i.e. not a real bug report, but a tool to help
> manage other bug reports.
>
> Most bug reports on i845 are probably due to the CPU/GPU incoherency
> problem that is now consolidated upstream at
> http://bugs.freedesktop.org/show_bug.cgi?id=26345 . For now, we mark all
> automatically reported GPU lockups as duplicates of this unless there is a
> reason not to.
>
>
> To use the available fix, run the following commands:
>
> apt-add-repository ppa:glasen/855gm-fix
> apt-add-repository ppa:brian-rogers/graphics-fixes
> apt-add-repository ppa:glasen/intel-driver
> aptitude update
> aptitude install linux 855gm-fix-dkms
> aptitude dist-upgrade
>
>
> There is a similar master bug report for i855 at bug 541511.
>
> To unsubscribe from this bug, go to:
> https://bugs.launchpad.net/xserver-xorg-video-intel/+bug/541492/+subscribe
>

[jerrylamos (jerrylamos)]

Douglas, from a user standpoint what changed was adding the KMS graphics feature to the kernel so the command line kernel operates in graphics mode as well as the regular gdm graphics display mode. Presumably this allows faster switching between users and I don't know what else. I never switch between users so I don't get any benefit. Also, switching to the kernel by doing Ctrl-Alt-F1 is faster because there isn't any graphics mode resetting. I only ever do that when I'm attempting to deal with a bug or doing bug reporting. Time saved by not resetting the video graphics mode is miniscule to me.

From the user's view, the kernel graphics mode and the gdm don't seem to be well coordinated at all. The kernel fires up graphics mode, you can tell with the tiny print, then X windows starts later setting the graphics mode again and the two do not seem to be working together so crash!

Developers have also alluded to some future KMS things they want to do that I'm not aware of.

At the moment both my i830 integrated video graphics and my i845G integrated video graphics run with the latest Release Code level of ubuntu, no changes required. Note I only use "quiet" and not "quiet splash" since splash is always the same and does not add any new info, and has a history of causing video graphics problems. So booting up I just look at a black screen with a blinking underline in the top left corner for a few seconds until gdm starts.

First real trouble I had with video graphics was Compiz on Intrepid so it is blacklisted. Compiz is "gdm eye candy" and has nothing to do with my running applications which I do full screen anyway.

Hang in there. Jerry

[oblong (bob-oblong)]

System/hardware:
10.04 (lucid) 2.6.32-25-generic

1024x768

$ lspci|grep "VGA"
00:02.0 VGA compatible controller: Intel Corporation 82845G/GL[Brookdale-G]/GE Chipset Integrated Graphics Device (rev 03)

This reproduces the problem consistently (gnome-screensaver is installed):

System|Preferences|Screensaver.

By default, AntSpotlight is selected. After a couple of seconds the screen goes blank, sometimes with vertical zebra stripes.

syslog contains:

Oct 2 12:26:41 ubuntu-desktop kernel: [ 5531.425838] [drm:i915_gem_entervt_ioctl] *ERROR* Reenabling wedged hardware, good luck
Oct 2 12:26:41 ubuntu-desktop kernel: [ 5531.740027] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Oct 2 12:26:41 ubuntu-desktop kernel: [ 5531.740038] render error detected, EIR: 0x00000000
Oct 2 12:26:41 ubuntu-desktop kernel: [ 5531.740880] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -5 (awaiting 4120 at 1135)
Oct 2 12:26:43 ubuntu-desktop kernel: [ 5533.792017] [drm:i915_gem_idle] *ERROR* hardware wedged

So after a reboot (pressing the power button seems to more-or-less cleanly shutdown), at GRUB, I've pressed 'e' to edit and after "quiet splash" I've added "i915.modetest=1" (no quotes).

After that, I can go to the screen saver preferences and there is no black screen/"crash". Screensaver comes in after the specified time and works, probably slowly, though.

However, I then realised that "modetest" should have been "modeset". So, not surprisingly,
syslog reports:

kernel: [ 18.768163] i915: Unknown parameter `modetest'

If I use modeset=0, or modeset=1, then the blanking/"crash" still happens when I go to the screensaver preferences.

So, modetest=0 or modetest=1 (either one avoids the blanking) is doing *something*. But what, and why?

Anyway, I'm happy to live with this until the official release of Maverick, which hopefully will solve things.

[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.

[Chris Halse Rogers (raof)]

@oblong: What adding the “modetest” option to i915 is doing is causing an error which prevents it from loading - the module doesn't recognise that option, so fails to load.

This ends up in the X server selecting a different driver, probably vesa, which won't have the same problems the intel driver has.

[peakit (guntasg)]

Has the fix been released to 10.04? If not, could you please let me know the tentative release date?

Eager to jump to 10.04 LTS..

Thanks!

[Warwick Bruce Chapman (warwickchapman)]

@Chris, I have a Lenovo X301 with Mobile Intel GM45 Express Chipset and Intel Graphics Media Accelerator 4500MHD. In 10.10, I get load rocking up to 6/8 at times, laggy, stuttering and even crashing video playback, flash performance is very poor (including no fullscreen). Is this the bug causing that issue?

[Darxus (darxus)]

Upstream (xf86-video-intel) bug was just closed fixed:

--- Comment #120 from Chris Wilson <email address hidden> 2010-12-30 08:11:43 PST ---
After applying

commit 15056d2c06862627ead868e035fcacc59dce1b1a
Author: Chris Wilson <email address hidden>
Date: Tue Dec 21 17:04:23 2010 +0000

    drm/i915: Flush pending writes on i830/i845 after updating GTT

    There is an erratum on these two chipsets that causes the wrong PTE
    entries to be invalidate after updating the GTT and when used from the
    BLT engine. The workaround is to flush any pending writes before those
    PTEs are used by the BLT.

    Signed-off-by: Chris Wilson <email address hidden>

this reduces to the general i8xx incoherency, bug 27187. For which I have a
patch which appears to work on my i845; passing both the wtf test and Daniel's
cache-coherency checker!

[Geir Ove Myhr (gomyhr)]

Note that the new patch upstream fixes the 845G-specific part of this bug. The 845G also suffers from the same i8xx incoherency that the 855GM suffers from (LP bug 541511). There is a patch for that as well at https://bugs.freedesktop.org/show_bug.cgi?id=27187#c281, but this has not been commited yet and probably needs more testing.

[Geir Ove Myhr (gomyhr)]

For completeness, I should add that the upstream fix introduces another bug which may cause some corruption. This one is fixed in this commit:

commit dc3bfebcf77d943b7e8495d30d0ee3d01b3042a5
Author: Chris Wilson <email address hidden>
Date: Thu Dec 30 18:02:21 2010 +0000

   drm/i915: Don't skip ring flushes if only invalidating

   Commit 15056d2 tried to optimize away a flush if there were no
   outstanding writes on a ring (in order to prevent a too-early-flush
   during ring init). However, this has the unfortunate side-effect of
   eliminating the texture cache invalidation, and so causing rendering
   artefacts.

   Reported-by: Alexey Fisher <email address hidden>
   Signed-off-by: Chris Wilson <email address hidden>

[Brian Rogers (brian-rogers)]

I've made a kernel with the latest fix available in this PPA:
https://launchpad.net/~brian-rogers/+archive/graphics-fixes-testing

If it gets good feedback, I'll copy it to my regular graphics-fixes PPA.

This kernel is based on Natty's 2.6.37 kernel, has the changes in drm-intel-next applied, and the patch at https://bugs.freedesktop.org/show_bug.cgi?id=27187#c291 is added. I've produced builds for Lucid, Maverick, and Natty. It has a cache coherency checker, so you'll see periodic messages in dmesg reporting on the number of flushes and whether any problems are detected.

From Daniel Vetter's patch description:
> Poke HIC bit + wbinv + cache coherency checker
>
> Chris Wilson's latest patch with my cache coherency checker added. Spills the
> number of chipset flushes regurlarly into the dmesg and bails loudly if one
> fails.
>
> Tested-by lines (like for the previous patch attempts by me) highly welcome.

Feedback about the patch can be sent directly to the upstream bug report at
https://bugs.freedesktop.org/show_bug.cgi?id=27187

If you have issues relating to installing/booting this kernel, report them here.

[Geir Ove Myhr (gomyhr)]

Note that the upstream 845G bug report has been reopened. It turned out that the fixes (which are contained in Brian's PPA kernel) are insufficient to fix the 845G-specific problem completely. The kernel will probably improve stability on 845G, but I think adding results from 845G to the upstream bug https://bugs.freedesktop.org/show_bug.cgi?id=27187 will only be considered as noise. The upstream bug for the 845G-specific issue is http://bugs.freedesktop.org/show_bug.cgi?id=26345.

[FactTech (launchpad-facttechnologies)]

It's not clear to me what the status of this bug is right now, but for anyone experiencing the issue and/or still working on it, I thought I would add some details that might be useful.

I'm running Ubuntu Lucid Lynx on an eMachines T2824.

kernel = 2.6.32-31-generic
graphics = Intel Corporation 82845G/GL[Brookdale-G]/GE Chipset Integrated Graphics Device (rev 01)

I have the lucid-proposed repository enabled with all updates as of today applied. No special PPAs or other software sources are enabled.

Machine will reliably lock up with zebra-stripes/flashing/non-recoverable symptoms as described by other users but can ssh in and usually will be given the option to revert to a console on the affected machine.

I have found that if I disable the screensaver, screen effects, and power management, the machine will run for a very long time without problems (days to weeks), but I do not use this machine for anything graphic-intensive.

Since it seems like the precise cause of this issue is not well understood, the one piece of information I wanted to point out for developers is that I found if I set the Appearance Preferences to use Clearlooks, this same machine will reliably lock up fairly quickly (usually within the first 60 minutes after login). On reverting to Ambiance, this behavior stopped. That doesn't make a lot of sense to me, but it seems reproducible and might be helpful in identifying the exact trigger. Trying to use various screensavers (especially OpenGL-based ones) will also reliably cause a crash.

If there's anything I can do to help resolve this issue, I am happy to do so. This note is mainly for other people who are experiencing the issue and want a workaround that doesn't involve installing non-standard packages.

Also: Anyone who has their screensaver stuck on a "bad" one and can't change it back, see the instructions at http://ubuntuforums.org/showthread.php?t=651868 to use gconf-editor to change it to blank only. From there, you can open the screensaver settings and disable it.

[Don Daniels (dondaniels)]

Had three GPU errors on first boot up this morning. After installing this morning's update I rebooted and did not have any problems. Guessing something patched since yesterday fixed the problem.

[Rüdiger Kupper (ruediger.kupper)]

This bug affects almost 100 Thin clients equipped with intel cards at my school.
It seems that the intel driver is completely disabled for LTSP-Clients to avaid this bug. However, the clients then fall back to the really (really!) slow VESA driver and software rendering. This makes a really poor user experience on natty, with many of the software distributed with edubuntu near to unusable (e.g., celestia, stellarium).
This bug has a huge impact on user experience at my school.

[Abdul nazar P (nazarkanayi)]

I resolved this issue completely by installing kernel 3.0.0.12 from kernel ppa and enabling KMS in Lucid.I enabled glasen ppa and update all packages except intel driver from this ppa. I installed only intel driver from x update ppa. thus i completely solved this.
I can do every thing in Lucid including playing 3D games.

[Rüdiger Kupper (ruediger.kupper)]

Hello Alkis,
in what sense does this bug no longer affect ltsp? I still have massive GPU lockups with the intel driver on our ltsp clients. This is a huge problem for us.
Lockups do happen in almost every session, markedly when scrolling windows. The session freezes for some seconds, then windows contents appear scrambled with old content not erased and new content printed over it.

This problem does not seem to occur on a standalone system, but it does in ltsp sessions.

So please: this DOES affect ltsp.

[Rüdiger Kupper (ruediger.kupper)]

Dear all,
what is the recommended way to install the available fixes at his time? The bug description advises to activate those three ppas:

apt-add-repository ppa:glasen/855gm-fix
apt-add-repository ppa:brian-rogers/graphics-fixes
apt-add-repository ppa:glasen/intel-driver

However, ppa:glasen/855gm-fix cannot be reached, and ppa:brian-rogers/graphics-fixes only has packages for natty, not oneiric.
I tried to activate ppa:glasen/intel-driver alone, but it does not fix the problem (GPU hangs appear as before).

Thanks for helping.

[FactTech (launchpad-facttechnologies)]

I had this issue temporarily worked around by installing an ATI video card that used a different driver. However, about a week ago, I took the video card out to see if the lockup issue had been resolved.

So far, things have been running very smoothly on the same hardware and configuration that was having problems before (the eMachines T8284 with device "VGA compatible controller: Intel Corporation 82845G/GL[Brookdale-G]/GE Chipset Integrated Graphics Device (rev 01)"). Still running lucid, current kernel version is 2.6.32-41, and the i915 module is loaded. None of the optional PPAs listed in the bug description are enabled.

There doesn't seem to be much activity related to this bug, so maybe a fix for another bug took care of it? Other users subscribed to this bug may wish to retest and report results.

[Biswa (b.biswajit)]

Agreed, random lockup seems to have stopped on this hardware.
I'm using the 3.0 kernels on lucid now btw.

________________________________
 From: FactTech <email address hidden>
To: <email address hidden>
Sent: Friday, 13 July 2012 2:11 PM
Subject: [Bug 541492] Re: MASTER: [i845] GPU lockup

I had this issue temporarily worked around by installing an ATI video
card that used a different driver. However, about a week ago, I took the
video card out to see if the lockup issue had been resolved.

So far, things have been running very smoothly on the same hardware and
configuration that was having problems before (the eMachines T8284 with
device "VGA compatible controller: Intel Corporation
82845G/GL[Brookdale-G]/GE Chipset Integrated Graphics Device (rev 01)").
Still running lucid, current kernel version is 2.6.32-41, and the i915
module is loaded. None of the optional PPAs listed in the bug
description are enabled.

There doesn't seem to be much activity related to this bug, so maybe a
fix for another bug took care of it? Other users subscribed to this bug
may wish to retest and report results.

--
You received this bug notification because you are subscribed to the bug
report.
https://bugs.launchpad.net/bugs/541492

Title:
  MASTER: [i845] GPU lockup

To manage notifications about this bug go to:
https://bugs.launchpad.net/xserver-xorg-video-intel/+bug/541492/+subscriptions

[Rüdiger Kupper (ruediger.kupper)]

I confirm that graphics run a lot more stable now, however, the issue is not gone:

I had been working around the lockups by setting
  Option "DRI" "false"
  Option "Shadow" "true"
in my xorg.conf

Recently, I wondered if the problem still exists, and I removed these lines. The machines ran quite stable in day-to-day life, but I could still reproducibly crash the GPU running the demo of "celestia (Gnome)" (part of the edubuntu package).

The weird thing is that after approx. one week, crashes started to increase a lot, and I had to reenable the DRI/Shadow workaround. Note that these machines are LTSP clients, all graphics goes across the network.

I have the impression that GPU lockups occur when network traffic is high, i.e., X needs to wait for graphics data more often.

All in all, things have improved a lot, but the problem has not "magically disappeared". I don't believe it's fixed.

[madbiologist (me-again)]

To Rudiger and anyone else still having trouble - I think you will find this recent article very interesting and promising - "Intel Finally Delivers Stable i830GM/i845G Driver" - http://www.phoronix.com/scan.php?page=news_item&px=MTI1MzI

The bottom line is you will need xf86-video-intel (xserver-xorg-video-intel) 2.20.16. According to https://launchpad.net/ubuntu/+source/xserver-xorg-video-intel this is already available in the development version of the upcoming Ubuntu 13.04.

[Rüdiger Kupper (ruediger.kupper)]

Madbiologist, that sounds almost too good to be true. This problem has caused me so many sleepless nights over the last years...
Do you know if it is possible to install "xserver-xorg-video-intel-2.20.16" from the Raring Ringtail into Precise? Or would the package need backporting to work in precise?

[madbiologist (me-again)]

I don't have much experience in that area so I can't be sure, but I think it would need to be backported. It will probably show up in the xorg-edgers PPA soon. See https://launchpad.net/~xorg-edgers/+archive/ppa?field.series_filter=precise for detail on how to install and uninstall. Note in particular the warning to install the whole PPA and not just individual packages from it. The other option is to compile it from source on your machine - see https://help.ubuntu.com/community/CompilingSoftware

[madbiologist (me-again)]

The slightly newer xserver-xorg-video-intel 2:2.20.16+git20121221.3793ccf7-0ubuntu0sarvatt~precise is now available in the xorg-edgers PPA.

[Rüdiger Kupper (ruediger.kupper)]

Great! I'll give it a try, thanks!

[Jānis Kangarooo (kangarooo)]

Could this be put or backported to 10.04? Couse so old comp can have
only with gnome. And since 10.04 is LTS then it still has support for
few months.

[Rüdiger Kupper (ruediger.kupper)]

Nope, it's not working for me (precise). I activated the xorg-edgers PPA both on my server and clients, and I can't even start an OpenGL application. I get this error:

root@LTClient128-01:/localhome/linadmin# glxinfo
name of display: :7
X Error of failed request: GLXBadContext
  Major opcode of failed request: 153 (GLX)
  Minor opcode of failed request: 6 (X_GLXIsDirect)
  Serial number of failed request: 23
  Current serial number in output stream: 22

That's even worse than before: OpenGL applications were instable before, now they won't even run.

[Rüdiger Kupper (ruediger.kupper)]

This may be another, hopefully unrelated issue (regression in recent Xserver), as reported here:
https://bugs.freedesktop.org/show_bug.cgi?id=56835

[madbiologist (me-again)]

It likes they are actually still putting the finishing touches on i830GM/i845G stability - http://lists.freedesktop.org/archives/xorg/2012-December/055213.html announces the xf86-video-intel (xserver-xorg-video-intel) 2.20.17 driver and describes a change. It sounds as if this change is designed to work with an upcoming kernel, perhaps 3.8.

However the error you are now getting sounds different and more serious. A GTK patch which is claimed to fix this has been sent to the GNOME mailing list - see https://mail.gnome.org/archives/commits-list/2012-December/msg03994.html - I will also bring that up on the freedesktop bug.

[Stan Young (stan738)]

Today was a good day ! if you have an old Intel 82845G chipset try this ...
NO INTEL GPU HUNG in dmesg.
I started using the new Lubuntu iso from here ...
http://cdimage.ubuntu.com/lubuntu/daily-live/current/
I used my computer most of the day today with no GPU crashes.
Video very fast all day :)

[Rüdiger Kupper (ruediger.kupper)]

That's good news. I still am out of luck installing the Xorg-edgers version in precise with our Intel 82845G machines. The GLXBadContext error is still present. (comment #284, #285)

[Chris Wilson (ickle)]

It is fair to say that raring should have stable 845g support. And I think even the GLXBadContext have been fixed (xorg bug, not driver).

[user (meta1729-deactivatedaccount)]

Yes, in Raring Ringtail, graphics works like a charm! No GLXBadContext error at all. All I had to do was have the following as xorg.conf

Section "Device"
   Identifier "Intel Graphics"
   Driver "intel"
   Option "AccelMethod" "uxa"
EndSection

]$ lspci | grep VGA
00:02.0 VGA compatible controller: Intel Corporation 82845G/GL[Brookdale-G]/GE Chipset Integrated Graphics Device (rev 01)

[Rolf Leggewie (r0lf)]

lucid has seen the end of its life and is no longer receiving any updates. Marking the lucid task for this ticket as "Won't Fix".