OpenBSD Journal

Newsforge article on AMD64, including OpenBSD

Contributed by grey on from the reviews gone right dept.

Noryungi writes:

Here is a nice article about the different free UN*Xes available on AMD64 machines.

They have a number of nice quotes on OpenBSD/AMD64, including this: 'nearly all of the programs in the Ports tree compile for 64-bit without any problems [...] The reason for the breakage being so minimal is not an accident -- the OpenBSD Ports contributors strive for quality as a primary goal.' (the last part being from Peter Valchev). Overall, a nice review.

(Comments are closed)

Comments

  1. By Chas (147.154.235.52) on

    While the theoretical speed advantage and expanded resources of 64-bit computing...

    Programs are not faster because they are 64-bit. AMD64 is generally faster because of the availability of extra registers, but it has been shown time and again (especially on sparc) that 32-bit architectures move less data and thus run faster.

    Comments

    1. By JM (142.179.203.230) on

      Horsecookies.

      A longer word length doesn't necessarily mean that it takes longer to move data. It depends on how that data is moved around, how wide is your data path, what's the bandwidth of the channel...

      Even then sometimes it makes sense from the point of actual instruction execution time to pass even larger chunks of data around. Enter SIMD, MIMD, vector processing...

      Comments

      1. By tedu (64.173.147.27) on

        what he means is that every pointer takes twice as much memory. if you have lots of pointers, it doesn't matter how big the region they point to is, you can still only fit in half as many. for amd64, 64 bits also makes a big difference as the page tables are bigger and have another layer of indirection.

    2. By Anonymous Coward (12.33.122.68) on

      if you are concerned so much about cpu power waste then run
      i386 port on it and be happy...

      but then it's not exactly lots of troubles that this kind
      of pollution gives to people

  2. By Noryungi (82.123.246.182) on

    Err... Sorry, the title should read 'Newsforge article on AMD64...'. My fault: I probably did not re-read this submission as I should have. All my apologies.

    And yes, it's lame correcting your own submission like this! ;-)

    Comments

    1. By dan (134.174.91.241) on

      I think its great that you are correcting your own mistakes. More people should do that :-) Proofreading late is better than never and, besides, if you let someone else do it, you may end up with something like, 'Come on you fscking moron! Can't you even proofread?'

    2. By grey (207.215.223.2) on

      Fixed the subject. We do proofread before posting, but even we overlook things sometimes. ;)

  3. By Anonymous Coward (67.64.89.177) on

    booh! openbsd is named under the same header as gentoo; if that isn't an insult i don't know what is. It is generally positive for openbsd but i fear the dude does not really know what he is talking about.

  4. By Anonymous Coward (68.202.41.228) on

    Getting into more specifics here, does anyone have any recommendations for well-supported mainboards for use with OpenBSD/amd64? Which is to say, all the onboard devices are supported? Many of the ones I've looked at do not fit that definition, so I'm curious what people are using.

    Comments

    1. By Roo (195.137.43.11) darkboong@hotmail.com on

      Well... I can't get my Asus A8N-SLI Deluxe and Athlon64 3000+ running. It locks up hard after printing out the BIOS rom messages. The hardware seems to be OK, after a bit of BIOS tweaking I got Fedora Core 3 to run well on it, and Windows XP seemed happy for the two brief hours it had before I wiped it. I've tried fiddling with the BIOS, disabling all the onboard devices and booting with a *known to work* ATA card etc... I've tried the PnP BIOS option too, and also upgraded the BIOS to ver 1006... I've tried memtest86 (ran 8 hours with not a single peep of an error), underclocking, OpenBSD 3.5, 3.6 and two 3.7 snapshots, no dice at all. It is rather frustrating because I love using OpenBSD, but ultimately I got what I deserved for buying hardware from anti-social hardware vendors like Nvidia (it's a NForce4 board). The Asus board and the Linux/BitKeeper debacle are a timely reminder that closed source inevitably bites you on the ass... Meanwhile I'll keep dreaming of genuinely competitive Open 3D hardware while I look at the multi GFLOP TI DSPs. :) Cheers, Roo

      Comments

      1. By roo (84.9.60.116) darkboong@hotmail.com on

        I was trying the i386 snapshots (I wanted to upgrade an existing 3.6 install that ran on a K7)... They didn't work, however I have managed to PXE boot the 27/APR/2005 OpenBSD amd64 snapshot on that A8N-SLI board ! The disk controllers aren't recognised, so I'm stuck with a couple of PATA sockets in PIO mode, but at least it boots and it recognises the Marvel Ethernet controller too !

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.]