OpenBSD Journal

Donation Request: Older Thinkpad X60 or X60s (X200, X200s, T400, T400s, T500)

Contributed by jcr on from the I-thinkpad-therefore-I-am dept.

Editor's Note: Thanks to everyone in the OpenBSD community who donated laptops to help the developers find and fix the issues with Intel graphics chipsets. Stefan Sperling (stsp@) just informed us that they have received enough donated laptops for the developers who needed access to the hardware.

OpenBSD developer, Stefan Sperling (stsp@), wrote in to tell us about the need of the project to receive some older but specific laptops.

I've just added an entry on want.html for Thinkpad X60 or X60s laptops:

A Thinkpad X60 or X60s laptop is needed ASAP in Berlin, Germany, or in Melbourne, Australia, for inteldrm(4) development. These models exhibit graphics corruption which we would like to fix in time for the 5.5 release. Contact stsp<at>openbsd.org or jsg<at>openbsd.org

The X60 line is somewhat outdated by now, so I hope that someone out there has a spare one they could donate. More of these laptops are needed by developers to tackle the issues, and fixing these bugs without access to hardware is very hard.

These machines contain the 945GM chipset which shows severe graphics corruption problems with inteldrm(4). I've seen some corruption happen on other intel graphics hardware, but none as severe as on these Thinkpad models.

Keep in mind that newer chipset generations can also benefit from any stability improvements for older chipsets. Some sales or marketing material may confusingly refer to the 945GM chipset by the name "Intel Graphics Media Accelerator 950" or "Intel GMA 950" as can be seen in the Intel documentation:

The Mobile Intel 945GM/945GME/GMS/GU/GSE, 943/940GML and Intel 945GT Express Chipsets come with the Generation 3.5 Intel® Integrated Graphics Engine, and the Intel® Graphics Media Accelerator 950, providing enhanced graphics support over the previous generation Graphics (GMCH) and Memory Controller Hubs (MCH).

In dmesg(8) output, the graphics chipset show up as:

  vga1 at pci0 dev 2 function 0 "Intel 82945GM Video" rev 0x03

And in /var/log/Xorg.0.log as:

  [    19.045] (--) intel(0): Integrated Graphics Chipset: Intel(R) 945GM

These machines also show the "GPU hung" problems people have been complaining about with various intel graphics chipsets. The screen ends up completely garbled and a reboot is required to unwedge it. They also show random rectangles of white and black lines across images in web browsers, most notably when browsing maps.

For example, x11/gnome looks like this:

Please contact Stefan Sperling (stsp@) in Europe or Jonathan Gray (jsg@) in Australia if you can help out.

Thanks!

Editor's Notes:

Nearly all Thinkpad X60 and X60s models were 32-bit systems, but there is also a at least one 64-bit version. For example, one of the 2.0GHz CPU's used in Thinkpad X60 and X60s models was 64-bit:
Intel T2500 2 GHz / 2 MB Cache / 32-bit
Intel T7200 2 GHz / 4 MB Cache / 64-bit

According to stsp@, jsg@, and Mark Kettenis (kettenis@), the display corruption on the 945GM chipset is known to exist and be mostly repeatable on the 32-bit X60 and X60s systems. It is presently unknown if the same bugs exist on the 64-bit versions of these systems, since they've never been tested. The 32-bit versus 64-bit question may not make any difference, but until it's tested, we won't know.

Unfortunately, the Intel graphics support issues are not limited to just the 945 series of chips. The various Intel 965 series of chips (965G, 965Q, 946GZ, 965GM) also have some bugs that need to be squashed. In addition to the X60 and X60s models, it also would be great to donate a Thinkpad X200, X200s, T400, T400s, T500, or (*gasp*) something more recent. All of the Thinkpad models listed in this article are older systems and are no longer even sold by Lenovo. Many of them are old enough to still be branded "IBM" even though they were actually made by Lenovo.

Without access to the needed hardware, adding support and debugging problems is nearly impossible. Even on the rare occasions when it is possible to debug problems without access to hardware, the result is a lot of developer time and effort gets unnecessarily drained away when it could have been spent doing something else. In a recent thread on the tech@ mailing list, Theo de Raadt (deraadt@) was once again kind enough to politely point out the obvious problem:

> It's not really surprising that your GM45-based system has a lot of
> these issues.  Neither jsg@ nor I have access to this hardware.  If
> somebody has a laptop with this chipset that they're willing to donate
> and ship to Australia or the Netherlands, please contact us.

  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

If you're able to help make our favorite operating system even better, but don't have one of the listed systems collecting dust in your attic, closet, basement, or garage, then the best answer is to search ebay and contact the developers before you purchase an inexpensive used system for them.

  • Stefan Sperling in Berlin, Germany: (stsp<at>openbsd<dot>org)
  • Jonathan Gray in Melbourne, Australia: (jsg<at>openbsd<dot>org)
  • Mark Kettenis in the Netherlands: (kettenis<at>openbsd<dot>org)

The magic ebay search phrases you'll want to use are as follows:

Thanks for your help!

(Comments are closed)


Comments
  1. By Blake (78.192.104.249) on

    I've got an MSI Wind PC (Atom) with what is branded as "GMA950 graphics" I could donate if the developers think it'll help...

    Comments
    1. By jcr (jcr) on http://designtools.org

      > I've got an MSI Wind PC (Atom) with what is branded as "GMA950 graphics" I could donate if the developers think it'll help...

      Thanks Blake. The best answer is to always contact the mentioned developers directly. They know more about the project needs than
      anyone.

Latest Articles

Credits

Copyright © - Daniel Hartmeier. All rights reserved. Articles and comments are copyright their respective authors, submission implies license to publish on this web site. Contents of the archive prior to as well as images and HTML templates were copied from the fabulous original deadly.org with Jose's and Jim's kind permission. This journal runs as CGI with httpd(8) on OpenBSD, the source code is BSD licensed. undeadly \Un*dead"ly\, a. Not subject to death; immortal. [Obs.]