[i945gme] Ubuntu falls back on vesa driver (no screens)

Thank you for reporting this bug in the development version of Ubuntu. Could you test some newer kernels for me so that we can decide if there is a fix upstream that we can pull, or it has to be fixed there as well? Information about testing mainline kernels can be found at [1]. The two kernels that I would like you to test are v2.6.33-rc4 and the newest drm-intel-next. If the problem is still present with those kernels we will collect some more information and submit a bug report to the intel developers. If it is fixed with one of the kernels, we will try and identify the fix and see if it can be pulled into Lucid.

[1]: https://wiki.ubuntu.com/KernelTeam/MainlineBuilds

@Geir

Sorry for my late response. Where do I find these two kernels?

There is a link at the page I referred to: http://kernel.ubuntu.com/~kernel-ppa/mainline/?C=M;O=D. There have been some new kernels since I wrote it. v2.6.33 just came out and you should test that instead of v2.6.33-rc4. Also, make sure your Lucid is up-to-date.

I will do that asap.

On Feb 25, 2010 1:40 PM, "Geir Ove Myhr" <email address hidden> wrote:

There is a link at the page I referred to: http://kernel.ubuntu.com
/~kernel-ppa/mainline/?C=M;O=D. There have been some new kernels since I
wrote it. v2.6.33 just came out and you should test that instead of
v2.6.33-rc4. Also, make sure your Lucid is up-to-date.

--
[i945gme] Ubuntu falls back on vesa driver (no screens)
https://bugs.launchpad.net/bugs/507196
...

@Geir

Unfortunately, neither kernel offered any improvement. I tested the latest mainline .33 kernel and the feb. 24. build of drm-intel-next.

Okay, we'll collect some information and send this on the the upstream developers. Could you add the kernel option drm.debug=0x04 [*] and then attach here
- the output of `dmesg > dmesg.txt`
- /var/log/Xorg.0.log

[*] If you can get to the Grub boot menu by press Esc at boot, you can edit the command options for the selected kernel by pressing 'e'. Add "drm.debug=0x04" after "splash quiet". It can be difficult sometimes to hit Esc at the right time, so alternatively you can edit the option that says "splash quiet" in /etc/default/grub and add drm.debug=0x04 there.

Here they are. With the latest Ubuntu kernel, and the latest Ubuntu Intel drivers.

Nicholas, the logs you provided are without drm.debug=0x04. From dmesg.txt:
[ 0.000000] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-2.6.33-020633-generic root=UUID=33f45800-d1dc-45aa-9098-839ef5477cac ro quiet splash nomodeset i8042.noloop=1
and Xorg.0.log:
Kernel command line: BOOT_IMAGE=/boot/vmlinuz-2.6.33-020633-generic root=UUID=33f45800-d1dc-45aa-9098-839ef5477cac ro quiet splash nomodeset i8042.noloop=1

You should also not use nomodeset, since the intel driver will not work with that (you may need nomodeset to make the vesa driver work, though). Did you add nomodeset instead of drm.debug=0x04 by mistake?

Well, without nomodeset, I won't be able to use my laptop at all. I guess
I'll run the command, and then get the files by chrooting in from a livecd.

On Mar 1, 2010 12:25 PM, "Geir Ove Myhr" <email address hidden> wrote:

Nicholas, the logs you provided are without drm.debug=0x04. From dmesg.txt:
[ 0.000000] Kernel command line:
BOOT_IMAGE=/boot/vmlinuz-2.6.33-020633-generic
root=UUID=33f45800-d1dc-45aa-9098-839ef5477cac ro quiet splash nomodeset
i8042.noloop=1
and Xorg.0.log:
Kernel command line: BOOT_IMAGE=/boot/vmlinuz-2.6.33-020633-generic
root=UUID=33f45800-d1dc-45aa-9098-839ef5477cac ro quiet splash nomodeset
i8042.noloop=1

You should also not use nomodeset, since the intel driver will not work
with that (you may need nomodeset to make the vesa driver work, though).
Did you add nomodeset instead of drm.debug=0x04 by mistake?

--
[i945gme] Ubuntu falls back on vesa driver (no screens)
https://bugs.launchpad.net/bugs/507196
...

On Mon, Mar 1, 2010 at 5:27 PM, Nicholas Christian Langkjær Ipsen wrote:
> Well, without nomodeset, I won't be able to use my laptop at all. I guess
> I'll run the command, and then  get the files by chrooting in from a livecd.

The easiest way to get the logs from a system that has an xorg which
is not working properly is to install the package openssh-server,
connect it to the network with ethernet, and ssh in from another
computer.

Of course, you can use nomodeset with the vesa driver in order to work
around the problems for daily use, but we need standard conditions for
the troubleshooting.

Here we go again. They should contain what you want this time.

Ignore the above Xorg.0.log, it's the old one. Here's the new one.

[This is an automatic notification.]

Thanks for reporting this bug to help making the Intel graphics driver
better. We hear from upstream that a number of bugs (possibly including
this one) have been fixed in the newer DRM code from the 2.6.33 kernel.
I don't know if your bug is one of the ones fixed in this release,
though, but we've prepared a PPA with this DRM update. Would you mind
installing this, rebooting, and testing if the original issue can be
reproduced with it or not?

The DRM PPA is here:

    https://edge.launchpad.net/~apw/+archive/red

Note there could be new bugs... please file these as new reports using
the command 'ubuntu-bug linux' (for kernel or DRM or KMS bugs) or
'ubuntu-bug xorg' if you suspect them to be X.org issues.

Nope, nothing changed using the kernel in the PPA with the latest Lucid Intel drivers.

This is pretty clearly a kernel bug. xserver-xorg-video-intel 2.9.1-1ubuntu3 through 1ubuntu10 and newer are compiled without UMS support so it's expected to fallback to vesa if you boot with nomodeset and is not a bug. The LVDS not getting detected properly under KMS is the main issue here as shown in the dmesg in comment #13.

Looks pretty clear that this machine needs to be added to the bad lid state quirk list . Can you compile a kernel by any chance? I put together a patch adding your machine to the blacklist which I will send upstream if it ends up working for you, but I am unable to compile a kernel for you at the moment.

patch against 2.6.32-15.22

I will compile a new kernel with your patch later today, and post the
results here.

Nicholas Ipsen

On 5 March 2010 08:41, Robert Hooker <email address hidden> wrote:

> patch against 2.6.32-15.22
>
> ** Patch added: "patch against 2.6.32-15.22"
>
> http://launchpadlibrarian.net/40229686/0001-drm-i915-Add-Gigabyte-M1028-bad-lid-status-quirk.patch
>
> --
> [i945gme] Ubuntu falls back on vesa driver (no screens)
> https://bugs.launchpad.net/bugs/507196
> You received this bug notification because you are a direct subscriber
> of the bug.
>
> Status in “linux” package in Ubuntu: Confirmed
>
> Bug description:
> Binary package hint: xserver-xorg-video-intel
>
> As the summary says, Ubuntu falls back to using VESA at boot. However, this
> only happens with drivers newer than xserver-xorg-video-intel
> 2:2.9.1-1ubuntu3, and even in those if I don't use nomodeset. With drivers
> later than the one mentioned above, if I do not use nomodeset, I simply get
> a totally dead, black screen.
>
> ProblemType: Bug
> Architecture: i386
> Date: Wed Jan 13 21:56:16 2010
> DistroRelease: Ubuntu 10.04
> EcryptfsInUse: Yes
> InstallationMedia: Ubuntu-Netbook-Remix 9.10 "Karmic Koala" - Release i386
> (20091028.4)
> MachineType: GIGABYTE M1028
> Package: xserver-xorg-video-intel 2:2.9.1-1ubuntu1
> ProcCmdLine: BOOT_IMAGE=/boot/vmlinuz-2.6.32-10-generic
> root=UUID=33f45800-d1dc-45aa-9098-839ef5477cac ro quiet splash
> i8042.noloop=1
> ProcEnviron:
> LANG=en_DK.UTF-8
> SHELL=/bin/bash
> ProcVersionSignature: Ubuntu 2.6.32-10.14-generic
> RelatedPackageVersions:
> xserver-xorg 1:7.5+1ubuntu1
> libgl1-mesa-glx 7.7-0ubuntu4
> libdrm2 2.4.17-0ubuntu1
> xserver-xorg-video-intel 2:2.9.1-1ubuntu1
> xserver-xorg-video-ati 1:6.12.99+git20091125.0061c4db-0ubuntu2
> SourcePackage: xserver-xorg-video-intel
> Tags: ubuntu-unr lucid
> Uname: Linux 2.6.32-10-generic i686
> XorgConf: Error: [Errno 2] No such file or directory: '/etc/X11/xorg.conf'
> dmi.bios.date: 07/17/2009
> dmi.bios.vendor: INSYDE
> dmi.bios.version: T1028 FB
> dmi.board.asset.tag: Base Board Asset Tag
> dmi.board.name: M1028E
> dmi.board.vendor: Gigabyte Technology Co.,Ltd.
> dmi.board.version: 2.0
> dmi.chassis.asset.tag: Chassis Asset Tag
> dmi.chassis.type: 1
> dmi.chassis.vendor: Insyde
> dmi.chassis.version: Chassis Version
> dmi.modalias:
> dmi:bvnINSYDE:bvrT1028FB:bd07/17/2009:svnGIGABYTE:pnM1028:pvr1:rvnGigabyteTechnologyCo.,Ltd.:rnM1028E:rvr2.0:cvnInsyde:ct1:cvrChassisVersion:
> dmi.product.name: M1028
> dmi.product.version: 1
> dmi.sys.vendor: GIGABYTE
> fglrx: Not loaded
> system:
> distro: Ubuntu
> architecture: i686kernel: 2.6.32-10-generic
>
> [lspci]
> 00:02.0 VGA compatible controller [0300]: Intel Corporation Mobile 945GME
> Express Integrated Graphics Controller [8086:27ae] (rev 03)
> Subsystem: Giga-byte Technology Device [1458:1999]
>
>
>
> To unsubscribe from this bug, go to:
> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/507196/+subscribe
>

How do I apply your patch to the kernel sources?

I seem to have the same problem, also a netbook, i945GME, no screens found. I can only boot with nomodeset, Lucid Lynx up-to-date as of today, kernel is 2.6.32-16-generic from ppa.launchpad.net/apw/red/

How do I get the DMI values?

The netbook is called "Datacask Jupiter 1014a", it's an OEM machine that has been sold under different names (e.g. "Point of View Mobii").

hwinfo --bios gives me:

  BIOS Info: #0
    Vendor: "INSYDE"
    Version: "D0.05T39"
    Date: "08/19/2008"
[...]
  System Info: #1
    Product: "PC-81005"
    Version: "1"
    Serial: "123456789"
    UUID: undefined, but settable
    Wake-up: 0x06 (Power Switch)
  Board Info: #2
    Product: "PC-81005"
    Version: "Base Board Version"
    Serial: "Base Board Serial Number"
    Asset Tag: "Base Board Asset Tag"

I have installed the kernel sources from ppa and I see my netbook is already mentioned, although with a different vendor name (starting at line 632):

        {
                .ident = "PC-81005",
                .matches = {
                        DMI_MATCH(DMI_SYS_VENDOR, "MALATA"),
                        DMI_MATCH(DMI_PRODUCT_NAME, "PC-81005"),
                },
        },

I have now copied the text block 's/MALATA/INSYDE/' and am compiling a custom kernel ATM.

Damn, doesn't work :-(

I also changed the ident string of course. Built a new kernel, initrd, booted all that stuff, no change. The netbook's screen turns black and that's it. I connected an external TFT via VGA, this works just fine. And boots blazingly fast!

Info from 'dmidecode -d /dev/mem' is identical to hwinfo stuff above btw.

What am I missing?

linux-image-2.6.32-16-generic from ppa.launchpad.net/apw/red is version 2.6.32-16.23~drm33v5 btw, said to contain the latest intel drm.

Malata (http://malata.com/en/products.aspx) is the original manufacturer btw, I found their product name to be PC-81005N about one year ago. The product page has vanished in between. Just in case anybody wants to know ;-)

Update: someone else (Zhenyu Wang) had the same idea: http://patchwork.kernel.org/patch/83555/

This conflicts (same ident!) with the patch from above that Greg Kroah-Hartman applied on 19 Feb 2010: http://lkml.org/lkml/2010/2/19/249

How about leaving out the vendor match? When I google for "PC-81005", I only seem to find references to the malata netbook and its offshoots.

Hello,

following Robert's advice (thanks alot!), I removed the VENDOR line because that string is empty in my DMI info.
So what remains is:

        {
                .ident = "PC-81005",
                .matches = {
                        DMI_MATCH(DMI_PRODUCT_NAME, "PC-81005"),
                },
        },

And now kernel modesetting works perfectly for me, both with internal LVDs an with a VGA screen attached!

Me very happy :-)

Could this please be patched into the kernel ASAP so that I can confirm whether or not it work for me?

Anything happening?

In the 2.6.32-19 kernel a patch went it to disable lid switch usage for all gpus of both i915 generation 2 and 3 [1]. Please test this kernel out to see if it fixes your issue.

Thanks

[1] http://kernel.ubuntu.com/git?p=ubuntu/ubuntu-lucid.git;a=commitdiff;h=02cf17d29ec2b659b7e058318721a1ad980d9b99;hp=2410233263bd7ccd4bd13b15e98f298684171ba1

@Nicholas Christian Langkjær Ipsen:

Please don't assign anyone else to bugs. Assignments mean that a specific person is actively working on a fix, and it's not fair to them to assign a bug without their knowledge. The best approach is to ensure that the person is subscribed to a bug. They will be able to follow the bug and assign it to themselves if they decide to work on it.

Thanks

My apologies. I thought he'd simply forgotten about this. I'll refrain from
doing this in the future.

2.6.32-19 works fine with my netbook using KMS, thanks alot!

Works fine on mine as well, I think we can close this bug.