[g45] No X display on laptop screen with Intel 4500MHD GMA graphics on Sony Vaio

[daniel]

Hello, I have a Sony laptop model VGN-SR140D. It came with the intel 4500MHD GMA. I downloaded the live CD Ubuntu 8.10 alpha 5 as well as today's daily build. Both of them get past the fist loading screen then after the loading bar and ubuntu logo disappear the screen just stays black. When I press alt+ctrl+f1 the console comes up and all of the command line tools and system resources seem to work fine. Then when I went back to ctrl+alt+f7 the screen was still blank. I then tried pressing ctrl+alt+backspace which then made the GDM sound play, but still showed just a blank screen.

[Justyn Butler (justyn)]

I'm seeing this issue on a Sony Vaio laptop with the same graphics chipset, Intel 4500MHD.
The model number of the laptop (in the UK at least) is VGN-SR19XN.

As Daniel describes the splash screen displays fine, then when X starts you see nothing but a blank screen.

X has started though and you can hear the Ubuntu desktop startup sound.

You can see that apart from this little (!) issue the driver is working fine by plugging in an external monitor (in Ubuntu 8.10 Beta). Even the desktop effects work beautifully. But the laptop screen is completely black.

I have tried with Ubuntu 8.04.1 and 8.10 Beta live CDs. 8.10 Beta uses xserver-xorg-video-intel version 2:2.4.1-1ubuntu6.

I've included my Xorg.0.log from booting 8.10 Beta. I can't see anything glaringly obvious in it.

[Justyn Butler (justyn)]

Both Daniel's problem and mine are a manifestation of the xorg bug here:
https://bugs.freedesktop.org/show_bug.cgi?id=17292

It appears to be a quirk affecting a particular model of Sony Vaio laptop that has the Intel Cantiga integrated graphics.

From this bug report and the upstream one, it seems that at least the following models of Sony laptop are affected
* VGN-SR140D
* VGN-SR19XN
* VGN-SR11M

As far as I can see these are simply international variants of exactly the same model of laptop.

[Justyn Butler (justyn)]

I should also point out that the title of this bug is incorrect.

We now know that X is starting fine, it is simply the LVDS display that is not receiving the video signal.

[daniel (daniel-somers)]

Yeah it looks like your right about X starting therefore the title of this bug is incorrect.

Also I've discovered the vesa driver will output to the laptops LCD and just an external monitor.

I think you're also right about international variants of exactly the same model of laptop, for instance I'm in Canada.

If I can provide any other information that might help just say so.

[ricardo_bcl (ricardo-bcl)]

Anybody knows if this bug will be fixed in the final release 8.10?

And is this problem exclusive of ubuntu, or it is for all other linux distributions?

Thanks.

[Justyn Butler (justyn)]

Hi,

This is an upstream problem so it affects all Linux distros:

See the upstream bug report here:
https://bugs.freedesktop.org/show_bug.cgi?id=17292

Unfortunately work is still ongoing so it will not be fixed for Ubuntu 8.10. Perhaps if it can be fixed soon it might be included in an update.

[ponar (joshworden)]

I have the exact same problem on my Vaio VGN-FW140E. I've installed 8.10 RC and compiled the 2.5.0 intel drivers, and the LVDS still doesn't work.

Any idea what's going on? Let me know if I can help.

[Mihai (mihaid)]

Hei guys!
Good news, someone found a fix. It has just been discovered so it will take some time until we get un upstream fix. Until then, if you manage to edit some file and compile/install from source you're done.

https://bugs.freedesktop.org/show_bug.cgi?id=17292

Enjoy,
Mihai

[daniel (daniel-somers)]

That's awesome, does anyone know if I can expect an update though the package manager that will fix it. Or should I just compile the driver from source myself?

[daniel (daniel-somers)]

I just applied the fix that was on the freedesktop.org bug tracker for that bug which was in xf86-video-intel-2.5.0/src/i830_lvds.c and re built the packages. It fixed the problem for me, just remember to put your xorg.conf back to the intel driver from vesa. I don't even need the xorg.conf now that everything is auto detected.

I attached a .tar.gz with the built deb packages in it and the source packages. So if anyone wants to use this fix and not have to build the patch and rebuild the source packages just uncompressed this attachment and install the deb packages in it. They are 32 bit packages though if anyone need 64 ask and I'll get them built as soon as I can.

[Bryce Harrington (bryce)]

Hi daniel,

Thanks for your work!

Would you mind attaching the patch you used, to this bug report? Then we can look at pulling it into ubuntu. Looks like the upstream bug is still in progress and hasn't gotten committed upstream yet.

[daniel (daniel-somers)]

I attached the patched file, i830_lvds.c, which is in xserver-xorg-video-intel-2.4.1/src/ of the source package. The only thing changed in the file is on line 390 before "if (on) {" I've added "if (!on) return;". Which fixes it this bug, but I have no idea if it will cause a regression on systems that don't have this bug.

Anyways I'm glade to help, so if their is anything else I can do let me know. This is the first package for ubuntu I've built though so I'm still getting myself familiar with how packaging and submitting bug fixes and reports are done. If I'm doing something wrong tell me. Oh and if I don't respond right away it's just because I have a number of exams I have to write between now and the ninth. I hope this helps.

[daniel (daniel-somers)]

I attached the patched file, i830_lvds.c, which is in xserver-xorg-video-intel-2.4.1/src/ of the source package. The only thing changed in the file is on line 390 before "if (on) {" I've added "if (!on) return;". Which fixes it this bug, but I have no idea if it will cause a regression on systems that don't have this bug.

Anyways I'm glade to help, so if their is anything else I can do let me know. This is the first package for ubuntu I've built though so I'm still getting myself familiar with how packaging and submitting bug fixes and reports are done. If I'm doing something wrong tell me. Oh and if I don't respond right away it's just because I have a number of exams I have to write between now and the ninth. I hope this helps.

[martink (martinkaethler)]

Hello Daniel,

  Could you compile a 64bit version for me... i seem to be having difficulty getting it compiled. I will work away at it... but if you have a moment.... thank you for all your work and for all the others who have worked on this!

[JunCTionS (junctions)]

I would like the 64bit version ass well.
since when I try to compile it I, when I run "./configure" I get:
No package 'xorg-server' found
No package 'fontsproto' found
it seems I'd have to install Xorg 7.
But I'd rather just install the driver so to keep my system with the stable kubuntu repositories.

[daniel (daniel-somers)]

Sorry for the late reply, I just finished my exams yesterday. Right now I'm setting up a 64 bit build environment, so I should have this package built tonight as long as I don't run into any hiccups.

[daniel (daniel-somers)]

Here is a tar with just the patched binary packages for 64 bit 8.10

[daniel (daniel-somers)]

Here is the source packages with the same binary ones.

[JunCTionS (junctions)]

Thank you so very much!.
I guess I'm not such an expert as to build these by myself.
They worked like a charm on a Sony VAIO VGN-FW270.
Thanks a lot.

[Nik A. Melchior (melchior+launchpad)]

Daniel,

Thanks for the packages! The 64-bit intel driver that you built fixes the video problems with the gm45 graphics on my Sony Vaio VGN-SR290. It even survived a hibernate/resume cycle, so I'm hopeful that this will continue providing XVideo and DRI 3D on my new laptop.

Many thanks!

[Bryce Harrington (bryce)]

Reviewing the patched file, it appears that this logic disables the else clause in i830SetLVDSPanelPower(), which disables the logic for handling when the display is powered off.

I believe this change would have a high likelihood of causing regressions for some users. At the least it would require extensive testing before it could be deployed.

[Bryce Harrington (bryce)]

[This is an automatic notice.]

We'd like to forward your bug upstream, however upstream requires
that you first test it against their newer driver code.

To save you the effort of building the driver from source, we've built
packages for the driver and its new dependencies.

So you have a couple options:

 1. Download and test .debs for intrepid, from:
     https://edge.launchpad.net/~intel-gfx-testing/+archive

 -or-

 2. Download and test the Jaunty alpha-2 (or newer) Live CD,
     (which includes a beta of the new xserver 1.6 as well).
     See http://cdimage.ubuntu.com/releases/9.04/ for ISOs

Thanks ahead of time! You can simply reply to this email to report your
findings.

P.S., if you wish to forward your bug upstream yourself, please follow
these directions to do so:
  http://intellinuxgraphics.org/how_to_report_bug.html

[Bryce Harrington (bryce)]

A patch for this is now available for testing on the upstream bug.

[Mihai (mihaid)]

I think it is more of a hack than a patch. The hack only works well on version 2.4.1 of the driver and, as i understand, prevents output from turning on/off endlessly. But if i am wrong please post a link to the new patch.

[RichardGT (gatechmathgodpi)]

Hi,

I'm running a Sony Vaio VGN-FW250J with the Intel 4500MHD chipset

Daniel's 64 bit package fixed all of my issues last month (thanks Daniel et al.!) but...

The latest update of the intel driver: "Version 2:2.4.1-1ubuntu10.1" breaks my system again, and I have to go uninstall the updated package and then reinstall Daniel's old "hacked" version to regain functionality. I guess I'll just ignore the updates until further notice?

also, I wouldn't exactly describe the failure as "turning off the display" It is more a trippy screen-saverish effect where the screen is black at first, but light creeps in from the bottom and diffuses upward to the rest of the screen until the whole screen is eventually bright white. The standard vesa drivers work fine, but only at 800x600.

[daniel (daniel-somers)]

I'll try and get packages built tonight (Canada eastern time). I was just looking at the upstream bug tracker again and it looks like they have a real fix working for the 2.6 intel driver so hopefully that's in the next ubuntu release. I'll try alpha 3 and see if it works or at vary lest if I can get the new patch to work.

[daniel (daniel-somers)]

Here's an archive with the 64 bit hacked binary debs and source.

[daniel (daniel-somers)]

Here's an archive with the 32 bit hacked binary debs and source.

[Daniel FitzGibbons (igknighted)]

Good news everyone! A (real) fix has been found upstream that should free us from the if(!on) hack. It should be merged to the official driver soon and hopefully be in the 2.6.1 release. Also, let us all hope that this makes it into Jaunty... it's not too late to get the 2.6 driver in is it?

[VOS (daniel-vosler)]

xf86-video-intel 2.6.1 Display Driver released on January 21, 2009.

http://cgit.freedesktop.org/xorg/driver/xf86-video-intel/

anyone have any luck? I get stuck at Line 12475: XINERAMA
any debs?

[Daniel FitzGibbons (igknighted)]

Latest update through synaptic fixed the issue for me... good work everyone!

[Paulo (pkhouri)]

I didn't get with a simple update, I have a Vaio FW170, using Jaunty.
And don't know how to use "auto-xorg-git -g -H hooks intel", some help?
 Thanx.

[Daniel FitzGibbons (igknighted)]

Jaunty is using a different series of drivers, I don't know if the fix has made it into jaunty yet. The very latest update for intrepid does have it.

[Paulo (pkhouri)]

I got update. it was made later, but I still cannot use it, my system does not recognize the intel driver, I read about a bug in kernel 2.6.28, was fixed in 2.6.29, related to it, and I don't know what should happen, but I think I'm gonna try Intrepid to get this working. Thanx!

[VOS (daniel-vosler)]

Jaunty didn't work for me so I went back to Intrepid. fixed worked for me (VAIO FW235J).

Now I need to get the audio working properly. anyone know whereto look.
card - Intel [HDA Intel]
device - ALC262 Analog [ALC262 Analog]

[Jaap Haitsma (jaap)]

I've just downloaded the latest daily of jaunty and I can't get it installed. It freezes somewhere in the beginning of the second loading bar ( the 1st is when the loading bar goes from left to right and back again a few times)

I've got at HP Elitebook 6930p with a 4500MHD. Intrepid is showing the same problem

[Bryce Harrington (bryce)]

Fix for upstream bug 17292 is committed to Jaunty (and Intrepid), which should solve this. Feel free to reopen if you continue to have problems.

https://bugs.freedesktop.org/show_bug.cgi?id=17292

[Jaap Haitsma (jaap)]

Is this fix in xserver-xorg-video-intel 2:2.6.1-1ubuntu1? That version was released on 23 january. I downloaded the live daily on the 26th. So I assume that that version is in there, but it does not work.

If I switch to the terminal window with Alt+F1 I see a couple of times
BUG: soft lockup - CPU#0 stuck for 61s [modprobe 2981]
on the screen

Is it maybe another problem that ubuntu has with the HP Elitebook 6930p

[Paulo (pkhouri)]

For my Vaio FW170J and Jaunty still doesn't work.

[VOS (daniel-vosler)]

Paulo,

"For my Vaio FW170J and Jaunty still doesn't work."

revert to intrepid. everything works fine.

[Paulo (pkhouri)]

I installed the intrepid, added the repositories and get it upgrade, but still doesn't work, I ran dpkg-reconfigure and nothing changed too. Some help?