OpenBSD Journal

Call for i386-laptops

Contributed by sean on from the time to dust off those old portables dept.

Alexander von Gernler writes: As pointed out in my mail to misc@, we are in search of currently available notebooks running with recent versions of OpenBSD.

So, if any of you want to support the project, why not submit your
info to

(Comments are closed)

  1. By phessler ( on

    what about non-i386 laptops?

    1. By Alexander von Gernler ( on

      As for non-i386 laptops, there is not supposed to be the big variety
      of machines that would justify own web pages for them.

      Mac iBooks and PowerBooks for example fit easily on macppc.html

      Sparc notebooks are not owned by many, and few people are interested
      in buying them, I presume. Anyway they can be handled by their respective
      platform homepages.

      i386 Notebooks are quite common however, and virtually every manufacturer
      company thinks it can swamp the market with more or less fine hardware.
      Therefore, the page is intendet to offer help to OpenBSD "customers".

      If you have reasons why this should be handled in an other
      way, please tell me.


      1. By Michael van der Westhuizen ( michael - at - on

        Hi Alexander,

        There is a bit of a call for more info on non-i386 hardware, or at least some more details on that hardware.

        I repartitioned my 15" 1.5GHz PowerBook over the weekend so that I could finally get around to installing OpenBSD on it - everything worked great, except for the fact that there is no way I could get the accelerated X driver working (and the console frame buffer is soooooooo slow). Given an xorg.conf on the macppc page, I could have just downloaded that and I'd be fine, however, I suspect that there's actually a problem running accelerated X on this model, but that would have been nice to know before wasting a day trying to fry my LCD. Apple hardware isn't exactly a fast moving target, so keeping this kind of info up to date should not be a problem.

        That aside, of course it worked a charm!

        -- Michael

        1. By Brad ( brad at comstyle dot com on

          The only thing really necessary on macppc systems is the modelines for the various sizes of LCDs. The rest of the information is very straightforward.

        2. By David ( on

          For the 15'' Aluminium PowerBook, you need this patch:

          And this is the config:


          1. By ajacoutot ( on

            I've been using this patch for a while now.
            While accelerated X works fine, when you get back to console mode, the display is unusable (you can still use the command line to but you won't see what you type). Another solution is to use current (from the CVS tree, not the OpenBSD one)... or wait until a new version gets imported into OpenBSD.

            1. By David ( on

              Is restore to console working with from cvs?
              Have you tried with the "iBookHacks" option enabled?

              1. By ajacoutot ( on

                The iBookHacks option is needed of course. And yes, from current CVS tree works fine when comming back to console mode.

          2. By Michael van der Westhuizen ( michael - at - on

            Hi David,


            I found the page on misleading, as it talks about XFree86 4.3 and -current as of an unknown time. It also talks about patches being accepted (or not, as the case may be), but doesn't say where they were or were not accepted to (*sigh*).

            There is no mention of this information still being current for OpenBSD 3.7, which made me mistrust this patch.

            I think this makes a case for Alexander to also accept entries for non-x86 laptop hints - having the correct information in the right places would cut down on a lot of frustration (if it's on the OpenBSD site, I trust it... if it's not...).

            Brad: thanks - On browsing the Xorg log I've also noticed that the ATI and Radeon drivers do pick up the modeline correctly with no coercion, so this may no longer be necessary with 3.7.

            I'll take a good dig around CVS and see if that patch works - it's a public holiday here on Thursday, so I'll have the time to play.

            -- Michael

            1. By David ( on

              --- radeon_driver.c 2005-06-14 21:35:43.000000000 +0200
              +++ radeon_driver.c~ 2005-06-14 21:39:43.000000000 +0200
              @@ -5086,9 +5086,9 @@

              - RADEON_CRTC_VSYNC_DIS |
              + ~(RADEON_CRTC_VSYNC_DIS |
              RADEON_CRTC_HSYNC_DIS |
              - RADEON_CRTC_DISPLAY_DIS);
              + RADEON_CRTC_DISPLAY_DIS));

              @@ -5163,6 +5163,7 @@
              RADEONInfoPtr info = RADEONPTR(pScrn);
              unsigned char *RADEONMMIO = info->MMIO;
              unsigned long tmp;
              + int i;

              OUTREG(RADEON_FP_CRTC_H_TOTAL_DISP, restore->fp_crtc_h_total_disp);
              OUTREG(RADEON_FP_CRTC_V_TOTAL_DISP, restore->fp_crtc_v_total_disp);
              @@ -5174,6 +5175,12 @@
              OUTREG(RADEON_FP_VERT_STRETCH, restore->fp_vert_stretch);
              OUTREG(RADEON_FP_GEN_CNTL, restore->fp_gen_cntl);

              + for (i = 0; i < 8; i++) {
              + OUTREG(RADEON_SURFACE0_LOWER_BOUND + 0x10*i, 0);
              + OUTREG(RADEON_SURFACE0_UPPER_BOUND + 0x10*i, 0x1f);
              + OUTREG(RADEON_SURFACE0_INFO + 0x10*i, 0);
              + }
              /* old AIW Radeon has some BIOS initialization problem
              * with display buffer underflow, only occurs to DFP
              @@ -6191,8 +6198,6 @@
              static Bool RADEONInitCrtcRegisters(ScrnInfoPtr pScrn, RADEONSavePtr save,
              DisplayModePtr mode, RADEONInfoPtr info)
              - unsigned char *RADEONMMIO = info->MMIO;
              int format;
              int hsync_start;
              int hsync_wid;
              @@ -6297,7 +6302,7 @@
              : 0));

              save->crtc_offset = 0;
              - save->crtc_offset_cntl = INREG(RADEON_CRTC_OFFSET_CNTL);
              + save->crtc_offset_cntl = 0;

              save->crtc_pitch = (((pScrn->displayWidth * pScrn->bitsPerPixel) +
              ((pScrn->bitsPerPixel * 8) -1)) /

              1. By Michael van der Westhuizen ( michael - at - on

                wow! thanks - that's great.

              2. By Anonymous Coward ( on

                Speaking of radeon... Does anyone know why suspend doesn't work with X running? Or if this will ever be fixed?

              3. By Michael van der Westhuizen ( michael - at - on

                Thanks for all your help David.

                X is now up and running, and looks good.

                I'm also experiencing screen corruption as described by ajacoutot, but that's not a serious concern for me - I'll just switch to -current with the next Xorg import.

                -- Michael

    2. By Anonymous Coward ( on

      what about non-laptop i386s???

      1. By bert ( on

        But what about non-lap83 6tops?

        1. By Anonymous Coward ( on

          Oh man. What if mine is a 7-top?

  2. By Anonymous Coward ( on

    yah, right. Once again? how many times you asked for the same thing? I submitted my confs at least 4 times, the i386 laptop page never included my confs neither confs of other guys that I know that send their confs too.

    1. By Alexander von Gernler ( on

      It is sad to hear that your input was not recognized earlier.
      There is possibly no excuse I can give you, since I am not that long
      in the project.

      However, if you take the time to look at the commit log of the page,
      you can see that I am really onto it. Perhaps this makes you change
      your mind and submit your notebook info after all.


      1. By Wodisch ( on

        How about amd64 laptops?
        Actually I bought one to be able to use the amd64 version of OpenBSD (and other OSs, of course), but the one I've got (HP nx9105) seems to have ACPI problems, I cannot even boot any *BSD CD (Free/ Net/ Open/ PC/ DragonFly). Most Linux-CDs do work, though, even the strange WinXP64.
        As soon as I got more time to build specialized kernels to investigate I'll tell you what is/was broken - until then I cannot even show a "dmesg" output...

        If anybody already knows about this model, I'll be happy to learn from them.

      2. By Anonymous Coward ( on

        I submitted a dmesg of my laptop long ago also. Back then it wasn't put online, probably because it's none of the major brands.

        Resubmitting today seems a bit stupid, as this model hasn't been produced anymore since about 2-3 years... (yeah, I need a new laptop, I know :p)

        It works perfectly though, and due to my taste for light applications it's not even that slow.

        1. By Alexander von Gernler ( on

          Just submit it to www@ anyway. We will see what we can do with it.
          Remember, people also buy notebooks on Ebay. What matters is that
          you have a current OpenBSD version running on it.

          As to amd64 and other notebooks mentioned further above --
          Yes. We will think about creating extra pages for them.
          This has to be discussed, though.

          In the meantime, I just work off the batch of notebook submissions
          I already got.


          1. By tony ( on

            I sent my info for a Dell Inspiron 700m into www@, and didnt get a response and i didnt see anything posted on the website.... not sure if you're swamped, but i had to fight to get openbsd on that laptop working as good as it is, and i know if someone else had posted in what i posted i wouldve saved a lot of time.

  3. By evild ( on

    one other thing that can be done to encurage submissions is to also have a list or forum for submitting requests for configurations. someone might be more inclined to submit their configuration if they saw a request for their particular laptop, especially if its an obscure model. while were on the subject, anyone get firewire working on a sony srx 87?

    1. By phessler ( on

      no firewire in openbsd. sorry dude.

  4. By Stop Hurting America ( on

    I have an old 386 AST laptop (no cd, nic, pcmcia). We'll see how it goes with a serial install. ;-) I'll post my info this weekend.

    1. By Anonymous Coward ( on

      Yeah, ignored again...


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 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.]