tedm at toybox.placo.com
Mon Jun 12 06:35:56 UTC 2006
>From: owner-freebsd-questions at freebsd.org
>[mailto:owner-freebsd-questions at freebsd.org]On Behalf Of Nikolas Britton
>Sent: Sunday, June 11, 2006 10:46 PM
>To: Ted Unangst
>Cc: Hámorszky Balázs; misc at openbsd.org; freebsd-questions at freebsd.org;
>netbsd-users at netbsd.org
>Subject: Re: wikipedia article
>On 6/12/06, Ted Unangst <ted.unangst at gmail.com> wrote:
>> On 6/11/06, Nikolas Britton <nikolas.britton at gmail.com> wrote:
>> > * IIRC NetBSD was a fork of FreeBSD
>> that's an interesting theory when you consider that the first netbsd
>> release came out 8 months before the first freebsd release.
>Yes as many others have noted, I cleary did not have my thinking cap
>on. Let me correct myself:
>NetBSD and FreeBSD both have deep roots in 4.3BSD NET/2, 386BSD, and
>4.4BSD Lite. NetBSD is not a fork of FreeBSD but OpenBSD is a fork of
>NetBSD. DragonFly BSD is a fork of FreeBSD 4.x, etc. etc.
>With all the inbreeding it's hard to remember who's your daddy. :-)
386BSD 0.1 was what started it all off. That was the "Jolitz"
port featured in Dr. Dobbs which was basically a Net/2 port. That
split into 2 forks, 1 was "386BSD 0.1+unofficial patchkit" the
other was "386BSD 0.1 + The Jolitz's personal set of patches
which they claimed were better but wouldn't let anyone see"
The second fork turned into something like 386BSD 0.2 which the
Jolitzes released years later and nobody paid any attention to.
The first fork became NetBSD and FreeBSD, the FreeBSD 0.X
and NetBSD 0.X code were virtually identical. (I think by the 1.0
versions of both those OS's they had started diverging enough to be
considered separate forks)
BSDI from what I gather was partly a fork from 386BSD, partly
a parallel port from Net/2, it is difficult even today to know
the truth since the source has always been "paid" source, and
is now owned by Wind River and is sitting in their vault somewhere.
Prior to the release of the 80386
the Intel processors didn't have memory protection which was a
of any processor running the BSD kernel.
4.4BSD Lite came much later, after the USL lawsuit. When it came
out all Net/2 descendents (ie: 386BSD/Jolitzes version, BSDI,
FreeBSD, NetBSD, plus whatever anyone was still doing with Net/2
on the VAXes) were told that if they didn't switch to 4.4BSD Lite
that they could be sued by USL directly, and could not use the
excuse that they were just using the University's copyright and
to go sue the University of California, Berkeley.
The commercial licensees of Net/2 and BSD 4.3 and such were basically
left to hang out and dry, since when 4.4BSD Lite was released the
university closed the CSRG. I think that Sun and HP both had already
signed source licenses with USL (since they were also selling the
AT&T UNIX source) and didn't give a rat's ass what CSRG did with
4.4BSD Lite. I think that BSDI rewrote the kernel files that were
at the heart of the lawsuit themselves, perhaps with some inspiration
from FreeBSD 2.0, perhaps not.
More information about the freebsd-questions