Slashdot Mirror

← Back to Stories (view on slashdot.org)

After 22 Years, 386BSD Gets An Update (386bsd.org)

Posted by EditorDavid on from the jumpstarting-Jolix dept.

386BSD was last released back in 1994 with a series of articles in Dr. Dobb's Journal -- but then developers for this BSD-based operating system started migrating to both FreeBSD and NetBSD. An anonymous Slashdot reader writes: The last known public release was version 0.1. Until Wednesday, when Lynne Jolitz, one of the co-authors of 386BSD, released the source code to version 1.0 as well as 2.0 on Github.

386BSD takes us back to the days when you could count every file in your Unix distribution and more importantly, read and understand all of your OS source code. 386BSD is also the missing link between BSD and Linux. One can find fragments of Linus Torvalds's math emulation code in the source code of 386BSD. To quote Linus: "If 386BSD had been available when I started on Linux, Linux would probably never had happened."
Though it was designed for Intel 80386 microprocessors, there's already instructions for launching it on the hosted hardware virtualization service Qemu.

38 of 83 comments (clear)

  1. No Trump? Climate change? Why the techie stuff? by Anonymous Coward · · Score: 1, Funny

    Where's the news that matters?

  2. Bizarre and nonsensical summary as usual. by Anonymous Coward · · Score: 4, Insightful

    ... What? Somebody makes 386BSD and releases version 0.1... and then works not only on a full, shiny 1.0, but also on a whole new generation (2.0)... and don't release the two latter ones? Just develop them and sit on them? What?! That's the real news. Not even an attempt to explain it, of course, so one is left with a hundred questions instead of learning anything.

    1. Re:Bizarre and nonsensical summary as usual. by freeze128 · · Score: 2

      Maybe now it runs on a 386SX....

    2. Re:Bizarre and nonsensical summary as usual. by bkmoore · · Score: 2

      Maybe now it runs on a 386SX....

      Wasn't it possible with some 386 SX chips to drill a tiny hole in the CPU, and make it into a DX? Or was that an urban myth, like drilling a headphone jack into your new iPhone?

    3. Re:Bizarre and nonsensical summary as usual. by hairyfeet · · Score: 2

      Considering the SX had a 16 bit bus compared to the DX's 32 bit I really do not see how drilling a hole would magically change that.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    4. Re:Bizarre and nonsensical summary as usual. by Anonymous Coward · · Score: 2, Informative

      You'd be surprised at how much code is out there that haven't been released just because there is a minor bug to be fixed or some strange problem to be solved.
      A lot of the people I know have hundreds of projects that have been on hold or abandoned for things more interesting.

      Also, the difference between 0.1 and 1.0 might not be as big as the numbers hint at, they are just arbitrary version numbers after all.
      As we know from software like Windows, Foxit Reader and PSP it even happens that the best version isn't the one with the highest version number.

    5. Re:Bizarre and nonsensical summary as usual. by TheRaven64 · · Score: 4, Informative

      He's probably thinking of the 486DX. The original 486 had an FPU, but the yields were low so Intel split the line into the 486SX (no FPU) and the 486DX (with FPU). Some motherboards let you plug in a 487 as an external FPU, but this was often really just a 486DX that took over completely. The 486SX was identical to the DX, but had the FPU disabled. It was possible to reenable it, and it would typically work most of the time. For gaming, this was fine (the occasional floating point error probably didn't make a difference) and was a cheap way of getting much more performance.

      --
      I am TheRaven on Soylent News
    6. Re:Bizarre and nonsensical summary as usual. by Anonymous Coward · · Score: 1

      Kiddo, I am really serious concerned of your reading ability. SERIOUSLY. It clearly reads "The last known public release was version 0.1" It should be so easy to understand that 1.0 and 2.0 are versions that were not released to public until today ?? Finish your school before comment.

    7. Re:Bizarre and nonsensical summary as usual. by Jesus_666 · · Score: 1

      386BSD got a 1.0 release in 1994. It's possible that that release didn't contain the full source code or that TFS is wrong.

      --
      USE HOT GRITS WITH STATUE OF NATALIE PORTMAN (NAKED AND PETRIFIED)
    8. Re:Bizarre and nonsensical summary as usual. by Rockoon · · Score: 1

      I guess they did make a 1.0 release with Dr. Dobb's Journal - and it has been lost in history. But I am glad it's back now.

      Now to bring back Dr Dobb's....

      --
      "His name was James Damore."
    9. Re:Bizarre and nonsensical summary as usual. by 0dugo0 · · Score: 1

      I have the 1.0 ISO, it comes with enough source code to rebuild the installed system.

    10. Re:Bizarre and nonsensical summary as usual. by 0dugo0 · · Score: 2

      1.0 was released as CD-ROM, but by that time everyone was running NetBSD, FreeBSD or Linux on their x86 boxen. In the circle of maybe 10 people that actually care about historical stuff like this it was known there was a 2.0, but unknown where and if it was ever released by the Jolitzes.

    11. Re:Bizarre and nonsensical summary as usual. by 0dugo0 · · Score: 1

      1.0 isn't lost. Just that the number of people actually caring and running it (in emulation) is about 10. The source and binaries in the original distribution should be freely distributable, the full ISO probably isn't. Hence the reluctance to host it publicly. If you know the right people you can get it through a coughing man in a raincoat together with the wiped first distribution of NetBSD,

    12. Re:Bizarre and nonsensical summary as usual. by Bing+Tsher+E · · Score: 2

      I think there's an InfoMagic CD-ROM set with this 386BSD version and also a pre 1.0 NetBSD on it. It's over there somewhere in that pile against the wall here.

      The Jolitzes also wrote a book on the codebase that appears to still be available.

    13. Re:Bizarre and nonsensical summary as usual. by citizenr · · Score: 1

      >It was possible to reenable it

      no, it was not. 486SX were made using different mask set and had no FPU inside. Im guessing you are to young to ever own real hardware, and read about reenabling fpu on some random website :(

      --
      Who logs in to gdm? Not I, said the duck.
    14. Re:Bizarre and nonsensical summary as usual. by fisted · · Score: 1

      That was an April fool's joke in a German computer magazine, all including a drilling mask.

      --
      CLI paste? paste.pr0.tips!
    15. Re:Bizarre and nonsensical summary as usual. by unixisc · · Score: 1

      So how different/better/worse is this than the spinoff distros - FreeBSD, NetBSD and everything since?

    16. Re:Bizarre and nonsensical summary as usual. by hairyfeet · · Score: 1

      Were there any Intel processors you could re-enable features on? The only chips I can recall that you could do that were the old AMD Barton core Durons were you could roll the dice and re-enable the cache with a pencil trace and of course the socket AM2/AM2+ Athlon and Phenom X2/X3s that you could re-enable the disabled core and see if they were any good or not.

      I'm not saying that there wasn't ever any Intel chips that were moddable but I honestly can't remember any.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    17. Re:Bizarre and nonsensical summary as usual. by raremediumwelldone · · Score: 1

      All I can think of is this http://www.notebookreview.com/...

    18. Re:Bizarre and nonsensical summary as usual. by hairyfeet · · Score: 1

      Well that was one I never heard of and I thought I'd heard of most, hmm learn something new every day. the only thing I could think of WRT Intel isn't really a "chip mod" so much as a socket swap.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    19. Re:Bizarre and nonsensical summary as usual. by raremediumwelldone · · Score: 1

      I don't imagine it was terribly popular (the Pentium-M pin mod/overclock). I only knew about it because a co-worker (successfully) tried it on an old notebook his dad had given him. He couldn't get anyone to take it on craigslist and basically figured "If it dies oh well".

  3. Re: Finally by Anonymous Coward · · Score: 1

    At least it doesn't run sysvinit either.

  4. Hope for Hurd yet? by Mostly+a+lurker · · Score: 3, Funny

    Hurd comes from the same era. I think some people are still tinkering with it.

    1. Re:Hope for Hurd yet? by tengwar · · Score: 1

      Still going, and you can get a Debian distro. They've got rid of that 2 Gb limit on disk (originally it was linked to the addressable RAM), but it's still 32 bit and things like USB don't seem to be there yet. Most people seem to be running it in emulation rather than on the metal..

  5. Internet's early days by Neo-Rio-101 · · Score: 3, Interesting

    Interesting. An ISP I worked at used BSDi in it's formative years.... the proprietary version of BSD from which 386BSD originated.
    The ISP runs FreeBSD now, of course.

    Speaking of which.... FreeBSD 11 is due for release any day now....

    --
    READY.
    PRINT ""+-0
    1. Re:Internet's early days by augustw · · Score: 2

      386BSD and BSDi both originated, independently, from the NET & NET/2 releases of BSD from the CSRG.
      Neither was a derivative of the other.
      NetBSD and FreeBSD are descendents of 386BSD.

  6. Re: Why bother? by augustw · · Score: 1

    Not forks of System V Unix, but of the various NET BSD releases Berkley made, at the end of the CSRG.

  7. It's just a tribute to a 90's star! by aglider · · Score: 1

    You all insensitive GUI clods!

    --
    Sent as ripples into the electromagnetic field. No single photon has been harmed in the process.
    1. Re:It's just a tribute to a 90's star! by 0dugo0 · · Score: 1

      Hey, 1.0 and 2.0 come with X386, you insensitive CLI clod!
      https://github.com/386bsd/386b...

    2. Re:It's just a tribute to a 90's star! by aglider · · Score: 1

      I have got into graphics only after 2001, you pixelated insensitive clod!

      --
      Sent as ripples into the electromagnetic field. No single photon has been harmed in the process.
  8. congratulations by jmccue · · Score: 1

    I think this is pretty cool, will keep an eye on it.

    I poked around /usr/include, maybe time to change _TIME_T_ to at least "unsigned long" due to 2038 ?

    1. Re:congratulations by Gornkleschnitzer · · Score: 1

      Parent is a virgin!? How did that happen?

  9. Re:Finally by 0dugo0 · · Score: 1

    ..and none of that "shared libraries" crap.

  10. Re: Why bother? by 0dugo0 · · Score: 1

    For various legal reasons FreeBSD pretended it was based on 4.4BSD-Lite for a long time. The NetBSD guys went as far as settling out of court and deleting 386BSD stuff from their cvs history. See eg. http://cvsweb.netbsd.org/bsdwe...

  11. Re:Why bother? by 0dugo0 · · Score: 1

    From what I remember when I got 1.0 to boot in emulation they went their own way. There were a few very essential bits I had to pull from the 0.1 patch kits to get the 1.0 kernel running somewhat stable.

  12. 386BSD 0.1 2016 remix by RLiegh · · Score: 1

    So you know those albums that were recorded in the 60's and 70's that the bands or record companies suddenly re-release with shitty mixes or that now include "lost tapes" of the engineer belching or other detritus? You know ..the stuff that's released for no other reason than to extend the copyright.

    Something about this release brings that phenomenon to mind for some reason...

  13. Re:Finally by unixisc · · Score: 1

    Uh, what's wrong w/ any of the other BSD distros - FreeBSD, NetBSD, et al?

  14. Re:Quote Bug by erik.opnemer · · Score: 1

    I'm just not sharing it with you. Dumbass.

    BSD in a nutshell.