Subject: Re: NeXT?
To: None <firstname.lastname@example.org>
From: der Mouse <mouse@Collatz.McRCIM.McGill.EDU>
Date: 11/03/1994 13:32:29
>>> Is there a NetBSD implementation for the NeXT machines?
>> I've got a Cube ([...]) and am looking at getting NetBSD running on
> Hey, I never expected anyone to be interested in this, as NeXTStep
> already is a BSD flavour
Well, sort of. It's BSD to someone who does no systems programming,
perhaps. But try to compile anything that uses sbrk(), or try to add
stuff to libc (to name just two that occur to me offhand), and you will
rapidly discover it bears only a superficial resemblance to anything
that came out of Berkeley. (Except perhaps LSD. :-)
> and _in my view_ the look and feel of NeXTStep is much better than
> X11 (and if you still need X11 you can get it anyway).
I loathe NeXTStep; avoiding it is why I did the first X port to the
NeXT in the first place, way back in X11R4 days. And recently, while
investigating in aid of getting NetBSD to run, I learned enough about
the keyboard that I now have all knowledge necessary to make the X
server sidestep the NeXTStep window server entirely (except for sound
output, which doesn't bother me too much because I leave it turned off
anyway). [The X server uses the NeXT window server to get input
events, because there wasn't enough documentation to get it to speak
directly to the kernel drivers. More details available on request.]
But the NeXTStep GUI isn't why I'm doing NetBSD - as I say, I now know
enough to sidestep it entirely and run X "native". I'm doing it
because the NeXT OS is causing me no end of headaches. I've got
release 2.1, and no prospect of upgrading because it costs. And no gcc
newer than 2.3.3 runs on 2.1, because of assembler differences. And it
has a kernel bug that causes swapfile space to leak; until I got a new
disk, I had to reboot about weekly because of this. Binary-only
_anything_ is the pits, and I want to get rid of it.
> Did you already try and ask NeXT if they supply any information about
> their hardware?
Yes, though I didn't press the issue when I was told "no".
> Once you know enough about the hardware it shouldn't be too hard to
> get something running, as there are already Ports for the m68k
Yes; I'm hoping and expecting to leverage heavily off the other
68k-based ports. Once I have it booting and have a couple of drivers
written, I expect the rest to fall into place pretty fast.
>> Unfortunately, it's slow going, because I don't have a whole lot of
>> spare time, and this is strictly a spare-time project.
> I'd really like to help, if you're seriously interested in getting it
> to work.
I'm not sure how. This would probably be better talked about without
current-users in the cc: header; I'll send you a note that way.