[i845] karmic fails to initialise display on boot. 2.6.31-3 + drm + Intel 845G

[Twig]

A Karmic test box boots fine with 2.6.31-2 generic kernel. However, when booting with 2.6.31-3 generic kernel, there is no display. However, everything else appears to function correctly.

relevant dmesg output:

[ 1.507428] Linux agpgart interface v0.103
[ 1.512917] agpgart-intel 0000:00:00.0: Intel 830M Chipset
[ 1.513083] agpgart-intel 0000:00:00.0: detected 892K stolen memory
[ 1.515801] agpgart-intel 0000:00:00.0: AGP aperture is 128M @ 0x88000000
[ 1.556813] [drm] Initialized drm 1.1.0 20060810
[ 1.578196] i915 0000:00:02.0: PCI INT A -> Link[LNKA] -> GSI 11 (level, low) -> IRQ 11
[ 1.578209] i915 0000:00:02.0: setting latency timer to 64
[ 1.724218] [drm] fb0: inteldrmfb frame buffer device
[ 1.724239] [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0
[ 1.737531] Console: switching to colour frame buffer device 128x48
[ 1.808850] render error detected, EIR: 0x00000010
[ 1.808854] page table error
[ 1.808856] PGTBL_ER: 0x00000011
[ 1.808861] [drm:i915_driver_irq_handler] *ERROR* EIR stuck: 0x00000010, masking
[ 1.808866] render error detected, EIR: 0x00000010
[ 1.808869] page table error
[ 1.808870] PGTBL_ER: 0x00000011
[ 1.821865] [drm] DAC-5: set mode 1024x768 c

uname -a output:
Linux ub-desktop 2.6.31-3-generic #19-Ubuntu SMP Tue Jul 14 16:04:41 UTC 2009 i686 GNU/Linux

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

Please let me know what further information I can provide.

[Philip Muškovac (yofel)]

Thank you for taking the time to report this bug and helping to make Ubuntu better. This bug did not have a package associated with it, which is important for ensuring that it gets looked at by the proper developers. You can learn more about finding the right package at https://wiki.ubuntu.com/Bugs/FindRightPackage. I have classified this bug as a bug in xserver-xorg-video-intel.

When reporting bugs in the future please use apport, either via the appropriate application's "Help -> Report a Problem" menu or using 'ubuntu-bug' and the name of the package affected. You can learn more about this functionality at https://wiki.ubuntu.com/ReportingBugs.

[Bryce Harrington (bryce)]

Hi pfirth,

Please attach the output of `lspci -vvnn`, and attach your /var/log/Xorg.0.log (and maybe Xorg.0.log.old) file from after reproducing this issue. If you've made any customizations to your /etc/X11/xorg.conf please attach that as well.

[This is an automated message. Apologies if it has reached you inappropriately; please just reply to this message indicating so.]

[Twig (pfirth)]

Attachments as requested. I've made no changes to my xorg.conf. Let me know if you require anything else.

[Rebug (j-de-cuba)]

Same problem since 07-20-2009
Same Intel chip i845G
After login I get the Ubuntu background screen with moving cursor no panels and I think the keyboard is dead.
Can update/upgrade using safe-mode.
Will try to post files syslog and xorglog.
Thanks

[Rebug (j-de-cuba)]

I have searched for file "xorg.conf", not found. No such file in /etc/X11 folder.
Will upload results code:lspci -vvnn
Thanks

[Bryce Harrington (bryce)]

Twig, I've forwarded your issue upstream to https://bugs.freedesktop.org/show_bug.cgi?id=22980 - please subscribe to this bug in case upstream needs further information or wishes you to test something

[Rebug (j-de-cuba)]

07-29-2009
Have today installed karmic daily build alt-cd.
Installer only installs base system and errors at app install.
I could manage to install grub2 and reboot, login to shell prompt and uses apt-get to install gdm and ubuntu-desktop then reboot. This karmic koala daily build works fine. I will keep this version for testing if you need information pls let me know, thanks.

[Bryce Harrington (bryce)]

I'm betting that since it is booting fine on 2.6.31-2 but not -3, that it is a kernel regression in the changes between -2 and -3. Probably should have been filed against the kernel to begin with. Reassigning to the kernel for further analysis.

[Leann Ogasawara (leannogasawara)]

Hi Twig,

As I posted in the upstream bug, lets try to narrow down even more where this regression was introduced. First, can you try the 2.6.31-rc3 vanilla mainline kernel build (I've pasted the link below). The Ubuntu 2.6.31-3 kernel was rebased on this upstream kernel:

http://kernel.ubuntu.com/~kernel-ppa/mainline/v2.6.31-rc3/

Depending if the issue remains or not with the mainline 2.6.31-rc3 kernel, this will help determine if it's an Ubuntu specific change or not which introduced this regression.

If the issue does remain with the mainline 2.6.31-rc3 kernel, lets have you reconfirm the issue is resolved with the mainline 2.6.31-rc2 kernel. The Ubuntu 2.6.31-2 kernel was based on the 2.6.31-rc2 mainline kernel:

http://kernel.ubuntu.com/~kernel-ppa/mainline/v2.6.31-rc2/

Please let us know how all your testing goes and what the results are. We can then work together to try to narrow down the specific patch which introduced the regression and get this resolved. Thanks.

[Rebug (j-de-cuba)]

Linux-image-2.6.31-020631rc3-generic
This image runs black-screen right after boot start.
Recovery mode: all text boot to console and if I enter "exit" this Linux-image boots normal to login and I can use karmic koala.

Linux-image-2.6.31-020631rc2-generic
The image is installed and I can find it in synaptic however not on list grub.cfg or the grub-boot-menu so can not test.

My window version xp is on karmic koala list grub.cfg but not on grub-boot-menu so to start windows I have to edit grub-boot-menu at start.

Up to now all karmic koala Linux-images takes about 4 minutes to load hardware drivers, long boot time.

Karmic koala "gedit" crashes all the time so I use Ubuntu version 9.04 for viewing karmic koala file-system.

I am running Linux-image-2.6.31-020631rc3-generic with this post.
Thanks

[Twig (pfirth)]

Hi,

Just a quick note to say I don't have access to my test box for the next day or two but as soon as I do, I'll post my results.

Twig.

[Twig (pfirth)]

I had more time than I orginally thought today to do some further testing (and lots of googling) and have made some progress.

The mainline 2.6.31 rc3 kernel worked correctly. I can boot the problematic rc3 kernel by passing the nomodeset parameter.

If I pass the VGA = ask parameter to the rc2 kernel I get a list of VGA modes. If I do the same thing with the rc3 kernel I get the same list of VGA modes plus a second set of VESA modes (none of which appear to work).

That led me to this bug:

https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-vesa/+bug/401214

and I've tried turning debugging on as suggested with:

echo 1 | sudo tee /sys/module/drm/parameters/debug

Although didn't get any extra output.

A little more searching led to a very similar looking bug report:

http://bugs.freedesktop.org/show_bug.cgi?id=20115

with a patch attached at comment #50. I'm now at a point where I probably need to learn how to patch a kernel which I won't get chance to do today. However, at least it looks like this has been narrowed down to a problem with KMS. Alternatively, I could be headed in completely the wrong direction as this stuff is a bit over my head. Let me know if there's anything else I can try. I'll post any further results.

Twig.

[Twig (pfirth)]

Patch from previous comment applied but the problem still persists.

Twig.

[Leann Ogasawara (leannogasawara)]

Hi Twig,

Just a few questions. . . . When you applied and tested the patch you referenced, did you apply and test the patch on a mainline kernel or an Ubuntu kernel? The reason I ask is you mentioned in comment #16 that "The mainline 2.6.31 rc3 kernel worked correctly" which would lead me to believe this is not a bug in the mainline kernel. Just to be sure could you test the latest 2.6.31-rc5 mainline kernel?

http://kernel.ubuntu.com/~kernel-ppa/mainline/v2.6.31-rc5/

[Twig (pfirth)]

Hi,

Just to clarify: The mainline kernel works fine. I can test the latest mainline but it won't be for a day or so.

None of the Ubuntu kernels work since rc2. All patches which I've applied have been to the Ubuntu kernel source (by downloading the linux-source package). I hope this is the right way of doing this.

I've tried the patch I mentioned and also those suggested on upstream bug 22980 at comment #7 (although I'm not sure how to apply the third patch as it doesn't seem to be a patch to the kernel source code).

All kernels boot correctly if I pass the nomodeset parameter.

Thanks,
Twig.

[Leann Ogasawara (leannogasawara)]

Thanks Twig. Based on your information, if the mainline kernels are working fine it would seem this is an Ubuntu specific regression that was introduced. Thus, applying any patches to the Ubuntu kernel won't likely help. Let's try and narrow down the Ubuntu specific patch that introduced this issue. First, if you can boot back into the 2.6.31-2 Ubuntu kernel which works fine, I need you to look up and tell me the specific Ubuntu kernel version being used. You can find this by doing:

cat /proc/version_signature

Likewise if you can boot into the 2.6.31-3 Ubuntu kernel which fails and again look up and tell me the specific Ubuntu kernel version being used that would be great. That will at least provide the specific window of patches which introduced this issue. We can then work on narrowing down the specific patch.

Below is just a list of all Ubuntu specific patches between 2.6.31-2.15 and 2.6.31-3.19:

ogasawara@emiko:~/ubuntu-karmic$ git log --pretty=oneline Ubuntu-2.6.31-2.15..Ubuntu-2.6.31-3.19
5cbce77153bdad1d6ad8a6def60f995a6d5ce2f3 UBUNTU: Ubuntu-2.6.31-3.19
a9d1cc1cd06a456e8d41f33ce24d2d318d24846c UBUNTU: [Config] Enable AUFS
5b3fd84287d4708cb728e2da054703d72f3fec0b UBUNTU: AUFS -- track changes in v2.6.31
c5e07a9283af083bd9b2d0c51135c3332b6c5a85 UBUNTU: AUFS -- core filesystem
dc892f663366294b96996c521ddbde44d9a493f7 UBUNTU: AUFS -- export various core functions -- fixes
d36e7e7f27f406f548afffe92043c843b3b8f1fb UBUNTU: AUFS -- export various core functions
df755792654e07acc80a618e54d8a5b9d5d1f599 UBUNTU: Start new release
41897760d840d8f87bdccfda827ca9db36c3ea82 UBUNTU: SAUCE: fsnotify: fix inotify tail drop check with path entries
5caefb5f01820fc49df7cfcad16830ee6160de48 UBUNTU: SAUCE: inotify: check filename before dropping repeat events
4f0861b1631a636ef697c901e7bd7150d1b9072e UBUNTU: SAUCE: fsnotify: use def_bool in kconfig instead of letting the user choose
87d7c3a4bd40770c6d6cc764d17fb64d260fd14a Revert "UBUNTU: [Config] Disabled NDISWRAPPER"
b5f4102c0d38804426fe00307d9736b161d1dd07 UBUNTU: ndiswrapper -- fix i386 compilation failures on cmpxchg8b
2dcb28212c0d3e711d76b341c19ccf8e57296257 UBUNTU: Ubuntu-2.6.31-3.18
029ffb80bd230b86b068fc4a7f9694e81bb87579 UBUNTU: [Config] Disabled NDISWRAPPER
05fc2e175dd986b3309cfbdc87874b4d2604dd2b UBUNTU: [Config] Update configs after rebase to 2.6.31-rc3
1cc8754849910e475ff484f84d42e23fcbefe5e9 UBUNTU: [Config] Dropped ubuntu/misc/wireless/acx
2bf63a37aae6a811aa09e4825b18a47911791992 UBUNTU: [Config] Enable CONFIG_BLK_DEV_IDECD on powerpc
506e2287a3b4db6aecb18cbe037dbf671152d173 UBUNTU: SAUCE: Explicitly include header files to allow apparmor to build on powerpc
9f73338649d5dd089f69eea0400ac9e2da527e43 UBUNTU: [Config] Enable CONFIG_USB_ISP116X_HCD on sparc
0bd10cff57dd0f7486cf70048b4a7dc4d0fb33e1 UBUNTU: Bump ABI
e5d7d3423a11bd20f29edaa5cfbf65dddcec42da Revert "UBUNTU: Add sec_perm-2.6.24.patch from AUFS - export security_inode_permission"
43bb18b794cceee803940b52306eab7d0307f801 Revert "UBUNTU: Add put_filp.patch from AUFS to export a symbol needed by AUFS"
ab871980c9d7695f2b835cfe8479866cf6b2a78f Revert "UBUNTU: Add splice-2.6.23.patch from AUFS to export a symbol needed by AUFS"
ed...

[Twig (pfirth)]

Hi,

Sorry for the delay. rc2 version:

Ubuntu 2.6.31-2.17-generic

and rc3 version:

Ubuntu 2.6.31-3.19-generic

Thanks,
Twig.

[Leann Ogasawara (leannogasawara)]

Thanks Twig,

Can you try installing the 2.6.31-3.18 kernel and let me know if this bug happens with it or not?

https://edge.launchpad.net/ubuntu/+source/linux/2.6.31-3.18/+build/1117529

ie. https://edge.launchpad.net/ubuntu/+source/linux/2.6.31-3.18/+build/1117529/+files/linux-image-2.6.31-3-generic_2.6.31-3.18_i386.deb

[Rebug (j-de-cuba)]

Dell2400 Intel82845G/GL
Ubuntu KK-9.10/alpha-4/alt-cd
Installer could not configure network
This build: black screen after login, moving cursor and dead keyboard
Recovery-mode same issues
Ubuntu KK 9.10 alt-cd daily build of July 29 2009 (daily updated) was doing well up to and including today.
Will try an other UKK daily build sometime next week
Ubuntu 9.04 runs fine

[Twig (pfirth)]

The bug does appear with 2.6.31-3.18.

Twig.

[Twig (pfirth)]

Curiously enough, I've just tried the new rc6 based kernel:

Linux ub-desktop 2.6.31-6-generic #25-Ubuntu

and this works fine. Problem solved?

Twig.

[Leann Ogasawara (leannogasawara)]

Thanks for the update Twig. It seems 2.6.31-6.25 rebased with upstream 2.6.31-rc6, brought in quite a few config changes, and also dropped some patches. See https://edge.launchpad.net/ubuntu/+source/linux/ . Since you mention the issue now seems resolved with this newer 2.6.31-6.25 kernel, lets go ahead and mark this Fix Released for now. However, if this issue reappears with any future update, please reopen this bug by setting the status back to "New". Thanks!

[Rebug (j-de-cuba)]

For your information.
Dell2400D Intel-845G
Have installed Xubuntu kk-9.10 live-cd build 20090815
Smooth install, auto-login works, hardware driver load is slow therefore boot-time exceeds 4 minutes. This build comes with kernel 2.6.31-5 installed and runs fine except for xorg which draws high cpu power while running applications. The OS was updated with kernel 2.6.31-6. When running this kernel xorg draws 42% cpu power with no application running "clean desktop except for terminal running top". Cpu use hits 100% when one program is started and it eventually becomes unresponsive. I am using Xubuntu with this post, thanks.

[Leann Ogasawara (leannogasawara)]

@Rebug,

Please file a new bug report against xorg. The issue you described is different than what was originally reported here.

[Rebug (j-de-cuba)]

RE: Ubuntu Karmnic Koala alpha-4 and previous posts
My apology for the inconvenience.
I wonder why?!>>kernel 2.6.31-5 in Ubuntu shows a black screen after login while same identical kernel works fine under Xubuntu.
Thanks

[Rebug (j-de-cuba)]

Dell2400D/i845G/Celeron-2.6 GHz
Xubuntu KK 9.10 A4
Kernel 2.6.31-6
Boot time after last distribution update now 40 sec. Shutdown time 12 sec.
Progressing
Thanks

[th (lp-der-hammer)]

Same bug as thread starter but in newest ubuntu kernel:

Sep 3 04:40:26 localhost kernel: [ 2.043646] [drm] TV-14: set mode NTSC 480i 0
Sep 3 04:40:26 localhost kernel: [ 2.185182] [drm] fb0: inteldrmfb frame buffer device
Sep 3 04:40:26 localhost kernel: [ 2.185258] [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0
Sep 3 04:40:26 localhost kernel: [ 2.257825] page table error
Sep 3 04:40:26 localhost kernel: [ 2.257826] PGTBL_ER: 0x00000100
Sep 3 04:40:26 localhost kernel: [ 2.257830] [drm:i915_handle_error] *ERROR* EIR stuck: 0x00000010, masking
Sep 3 04:40:26 localhost kernel: [ 2.257836] render error detected, EIR: 0x00000010
Sep 3 04:40:26 localhost kernel: [ 2.257838] page table error
Sep 3 04:40:26 localhost kernel: [ 2.257840] PGTBL_ER: 0x00000100
Sep 3 04:40:26 localhost kernel: [ 2.265385] [drm] LVDS-8: set mode 1400x1050 18
Sep 3 04:40:26 localhost kernel: [ 2.305635] Console: switching to colour frame buffer device 175x65

uname -a:
Linux th 2.6.31-9-generic #29-Ubuntu SMP Sun Aug 30 17:39:23 UTC 2009 i686 GNU/Linux

[Leann Ogasawara (leannogasawara)]

@th did you see this with the 2.6.31-8 kernel? It's always the most helpful to note between which two kernel versions you saw a regression appear. It really helps us to isolate a set of patches to examine. You can 'cat /proc/version_signature' to know the specific kernel version being ran. Can you also test with the latest mainline kernel build - http://kernel.ubuntu.com/~kernel-ppa/mainline/v2.6.31-rc8/ .

@Twig, since you are the original bug reporter, can you comment if this has reappeared with the 2.6.31-9 kernel? Thanks.

[Twig (pfirth)]

This bug hasn't resurfaced on my system and I have the same kernel version: 2.6.31-9-generic #29-Ubuntu.

[th (lp-der-hammer)]

I get exactly the same error in the following versions:
 Ubuntu 2.6.31-6.26-generic
 Ubuntu 2.6.31-8.28-generic
 Ubuntu 2.6.31-9.29-generic
 and in mainline-rc8

[Leann Ogasawara (leannogasawara)]

@th, care to open a new bug since it seems Twig is not experiencing this any longer. It's best if we have a report specifically tracking this bug for you against your specific hardware. I'll continue to follow up with you at that new report. For now I'm marking this back to Fix Released based on Twig's feedback.

[Fabus (fabian-gebert-hh)]

Trying Karmic Koala Alpha 5 I get the same error: After the boot screen, nothing shows up and my display goes into standby. With Alpha 4 I at least got a screen (with the wrong resolution however).

[Id2ndR (id2ndr)]

Fabus, this bug have been marked as fixed release. If you still encountered a trouble with karmic RC, you should open an other bug because it may be an other issue. I've got a computer with 845G and it works well on karmic RC.

[Kai F. Lahmann (kfl)]

I'm sure, this bug is _not_ fixed, esp. as I get exactly the problem described by Rebug on 2009-07-23 with an intel 855 on the 9.10 final. It doesn't crash at a predictable time but random somewhere when the X-Server is fresh loaded. Changing to vesa is the only workarround.

[Twig (pfirth)]

This is now re-occurring with Lucid since very early in the development cycle and has been persistent over the last few kernel updates. I will try and isolate the kernel version in which this first occurred (although I suspect it was when the first 2.6.32 landed)

[Twig (pfirth)]

This occurred somewhere between kernel versions:

Ubuntu 2.6.31-14.48-generic

and:

Ubuntu 2.6.32-2.2-generic

[Andy Whitcroft (apw)]

@Twig -- so this does not occur with 2.6.31-14.28, and does with 2.6.32-2.2. It would be helpful to try and work out whether it occurs on the latest Karmic kernel 2.6.31-17.54 in karmic-proposed. If so it would be handy to see which release inbetween that the issue occured. If no then it would be somewhere between 2.6.31 and 2.6.32, so it would be helpful to see which of the 2.6.32-rcN kernels work and which do not. Those can be found in the mainline kernel archive:

    https://wiki.ubuntu.com/KernelTeam/MainlineBuilds

Please report any testing results here. Thanks!

[Twig (pfirth)]

Andy:

When this issue surfaced in the Karmic development cycle, the mainline kernel builds always worked. However, I'll try the mainline builds and the karmic-proposed and see if I can narrow down the issue further when I get in front of my test box - likely early next week.

[Twig (pfirth)]

Mainline kernels 2.6.32.2 and 2.6.33rc1 DO have the same problem.

If I pass the nomodeset parameter, the machine boots fine.

Where would I find the latest karmic-proposed kernel to test?

[Twig (pfirth)]

Quick note to say I haven't had much chance to look at this recently but can confirm that 2.6.33rc2 and 2.6.33rc3 also have the same issue.

[Leann Ogasawara (leannogasawara)]

Hi Twig,

For instructions on how to enable karmic-proposed, please refer to the following link. Thanks.

https://wiki.ubuntu.com/Testing/EnableProposed

[Twig (pfirth)]

Have enabled karmic-proposed and installed. Thanks.

Kernel 2.6.31-18-generic #55-Ubuntu works fine so this has been introduced with the 2.6.32 kernels.

[Twig (pfirth)]

I've tried all the latest mainline kernels up to and including 2.6.33rc6 and all have the same issue. I'm a bit stuck now and would be grateful if anyone could point me at what to try next.

[Adam (adam-leckron-deactivatedaccount-deactivatedaccount)]

Not completely sure I have the same problem, but I had issues with a Dell
Dimension 2400 that would consistently lock up on boot. Everything worked
fine in Jaunty. I even reloaded Karmic from scratch and still have
problems. I tried a lot of troubleshooting and the only thing that seems to
help for me is to remove the quiet and splash kernel options. Once I do
that, it boots up fine & seems to work okay.

[Jeremy Foshee (jeremyfoshee)]

Adam,
     You will need to open a new bug for your particular issue.

Twig,
    Thank you for the update. I'll work with the Kernel team to investigate this in Lucid.

Thanks,

-JFo

[Leann Ogasawara (leannogasawara)]

Hi Twig,

Can you do one more round of testing. Can you try the latest Lucid release available but update to the newest 2.6.32-16 kernel? The reason I ask is that we recently backported the entire 2.6.33 DRM stack back to the Lucid kernel as of 2.6.32-16. We'd appreciate any feedback you could give if this helps resolve the issue you reported here. Please let us know your results. Thanks.

[Twig (pfirth)]

Hi,

Updated to 2.6.32-16-generic #25-Ubuntu and still have the same problem. Interestingly, I'm not getting any drm errors in my dmesg output:

[ 9.934691] agpgart-intel 0000:00:00.0: Intel 830M Chipset
[ 9.934924] agpgart-intel 0000:00:00.0: detected 892K stolen memory
[ 9.947250] agpgart-intel 0000:00:00.0: AGP aperture is 128M @ 0x88000000
[ 10.146521] [drm] Initialized drm 1.1.0 20060810
[ 10.455902] i915 0000:00:02.0: PCI INT A -> Link[LNKA] -> GSI 11 (level, low) -> IRQ 11
[ 10.455914] i915 0000:00:02.0: setting latency timer to 64
[ 10.472169] [drm] set up 7M of stolen space
[ 10.514667] [drm] initialized overlay support
[ 10.700508] fb0: inteldrmfb frame buffer device
[ 10.700514] registered panic notifier
[ 10.700532] [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0
[ 11.163428] vga16fb: initializing
[ 11.163440] vga16fb: mapped to 0xc00a0000
[ 11.163943] fb1: VGA16 VGA frame buffer device
[ 11.551806] Console: switching to colour frame buffer device 160x64

Will keep digging. Let me know what other information you need.

[Twig (pfirth)]

Hi,

Just an update to say this is still a problem on my test box. Should this at least be targetted to try and resolve before Lucid is released? I imagine this will affect other users too.

Twig.

[Twig (pfirth)]

On my last boot received the following dmesg output:

[ 19.884199] [drm] i915 disabling kernel modesetting for known bad device.

and the system boots successfully! Have rebooted a few times to check. This can be marked as fixed for me now.

[Twig (pfirth)]

Hi,

It appears that this is a perennial problem with new kernels as have just upgraded to maverick with kernel version:

2.6.34-3-generic #10-Ubuntu

and the problem resurfaced. However, I don't receive the message about disabling drm in my dmesg as in comment #53 above and I imagine this is the problem.

Thanks,
Twig.

[Twig (pfirth)]

Further problems with the new 2.6.35 kernel - 2.6.35-1-generic #1-Ubuntu:

If I boot with no parameters passed, I am unable even to ssh to the box (as I have been able to with earlier kernels)

If I boot with nomodeset paramater, I get ubuntu running in low-graphics mode with the following errors:

(EE) intel(0): No kernel modesetting driver detected.
(EE) Screen(s) found, but none have a usable configuration.

followed by no display but I can ssh to the machine.

It seems that whatever fixed the problem in #53 hasn't been applied to newer kernels. If someone can show me where to start investigating this, I'd be happy to do so.

Thanks,
Twig

[Twig (pfirth)]

New kernel based on 2.6.35rc3 fixes this again (hopefully for good). I'll keep monitoring.

Thanks,
Twig.