[g45] X doesn't start with intel driver only vesa - "(EE) intel(0): No valid modes."

[w074n]

Binary package hint: xserver-xorg-video-intel

I've got a Intel G45 chipset with a X4500HD. It was installed with the vesa driver as already the desktop install cd failed to use the intel driver.
So I tried to use the intel driver manually, which always resulted in the X server not starting up giving this error message: "(EE) intel(0): No valid modes.". Because of that i thought it might be related with bug #194760. Therefore I added the information about the modes to xorg.conf and still it gave the same error.
I'll append the different xorg configurations and the log files respectively.

Kind regards

$lspci -nn | grep VGA
00:02.0 VGA compatible controller [0300]: Intel Corporation 4 Series Chipset Integrated Graphics Controller [8086:2e22] (rev 03)

ProblemType: Bug
Architecture: i386
DistroRelease: Ubuntu 9.04
Package: xserver-xorg-video-intel 2:2.6.3-0ubuntu9.3
ProcEnviron:
 LANG=de_DE.UTF-8
 SHELL=/bin/bash
ProcVersion: Linux version 2.6.28-13-generic (buildd@vernadsky) (gcc version 4.3.3 (Ubuntu 4.3.3-5ubuntu4) ) #45-Ubuntu SMP Tue Jun 30 19:49:51 UTC 2009
SourcePackage: xserver-xorg-video-intel
Uname: Linux 2.6.28-13-generic i686

[Bryce Harrington (bryce)]

Well, next I'd like to forward this issue upstream, but first it must be retested on latest development version of Karmic to make sure the issue still happens there with the latest code. ISO images are available at http://cdimages.ubuntu.com/releases/karmic/. If you can reproduce it in the LiveCD environment, you shouldn't need to modify your installed system. Also please attach a fresh Xorg.0.log and dmesg from this testing.

[w074n (w074n)]

As alpha 3 is released I tried it, unfortunately still the same.

[w074n (w074n)]

Then I connected the display via VGA, not DVI and everything is working just fine. Is there a way to force the intel driver to use DVI-port through xorg.conf, as a workaround? Btw. the display seems to be properly detected now. That's make me think this bug is related to bug #194760. Should it be marked as duplicate? I append one set of logs for the vga deployment from my jaunty system, but it worked fine karmic alpha 3, too.

[Bryce Harrington (bryce)]

Forwarding this bug from Ubuntu reporter w074n:
http://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/401131

[Problem]
Driver detects all outputs as disabled when VGA port being used. Works okay with vesa.

[Original Description]
I've got a Intel G45 chipset with a X4500HD. It was installed with the vesa driver as already the desktop install cd failed to use the intel driver.

So I tried to use the intel driver manually, which always resulted in the X server not starting up giving this error message: "(EE) intel(0): No valid modes.". Because of that i thought it might be related with bug #194760. Therefore I added the information about the modes to xorg.conf and still it gave the same error.

Then I connected the display via VGA, not DVI and everything is working just fine.

$lspci -nn | grep VGA
00:02.0 VGA compatible controller [0300]: Intel Corporation 4 Series Chipset Integrated Graphics Controller [8086:2e22] (rev 03)

Architecture: i386
DistroRelease: Ubuntu 9.04
Package: xserver-xorg-video-intel 2:2.6.3-0ubuntu9.3
ProcEnviron:
 LANG=de_DE.UTF-8ProcVersion: Linux version 2.6.28-13-generic (buildd@vernadsky) (gcc version 4.3.3 (Ubuntu 4.3.3-5ubuntu4) ) #45-Ubuntu SMP Tue Jun 30 19:49:51 UTC 2009
SourcePackage: xserver-xorg-video-intel
Uname: Linux 2.6.28-13-generic i686

[Bryce Harrington (bryce)]

Created an attachment (id=28085)
karmic (alpha 3)_Xorg.0.log

[Bryce Harrington (bryce)]

Created an attachment (id=28086)
karmic (alpha 3)_dmesg.log

[Bryce Harrington (bryce)]

Created an attachment (id=28087)
edid.dat

[Gordon Jin (gordon-jin)]

I wish this is some bug already fixed by Ling

[Michael Fu (michael-fu-intel)]

Gordon, it's DVI , not VGA, not detected.. should be a dup of bug# 21322

[Ling-ma (ling-ma)]

commit 2ded9e2747d0a390d281bb5b16ff7f640ec85f78
Author: <email address hidden> <email address hidden>
Date: Thu Jul 16 17:23:09 2009 +0800

fixed, please update your kernel version from drm-intel-next tree.
close issue please open it if you need.

Thanks
Ma Ling

[Bryce Harrington (bryce)]

According to the upstream bug report, this is actually a kernel bug, fixed with commit 2ded9e2747d0a390d281bb5b16ff7f640ec85f78 in the drm-next tree. Since it's a kernel bug, I'm reassigning to the kernel.

[w074n (w074n)]

Everything works perfect with the latest karmic kernel from git (2a429b24d060a34c1861ed8de1b5d46848a65959) patched with the fix from drm-intel-next tree (2ded9e2747d0a390d281bb5b16ff7f640ec85f78). Thanks so far! Will the jaunty kernel be patched?

[Leann Ogasawara (leannogasawara)]

Just adding additional info for the kernel team to reference.

ogasawara@emiko:~/drm-intel$ git log 2ded9e2747d0a390d281bb5b16ff7f640ec85f78
commit 2ded9e2747d0a390d281bb5b16ff7f640ec85f78
Author: <email address hidden> <email address hidden>
Date: Thu Jul 16 17:23:09 2009 +0800

    drm/i915: hdmi detection according by reading edid

    According to investigations from windows team ,hw team,
    and our test results on all 4x platofrms available
    (gm45, g45b, q45, g45a, g45c, g41a, and g41), we find
    currently Hot plug live status and Hot plug interrupt
    detection are not reliable, sometime the results from
    the two approaches are contradicts. So we chose edid
    detection for hdmi output.

    Signed-off-by: Ma Ling <email address hidden>
    Reviewed-by: Ian Romanick <email address hidden>
    Signed-off-by: Eric Anholt <email address hidden>

[Andy Whitcroft (apw)]

The commit referenced here is already merged into linus' mainline tree and will hit our tree on our next rebase which should be in the next few days. Will let you know when its released.

[w074n (w074n)]

With the latest kernel in karmic dvi port works. Are there plans to fix this issue in jaunty? For the moment I don't like to use karmic.

[Jeremy Foshee (jeremyfoshee)]

w074n,
      I can say for certain that this will not make it to Jaunty. I am closing this bug as fix released per your comment.

-JFo