I've assembled some notes from old manuals and other sources
on the formats used for on-disk file systems through the
Additional notes, comments on style, and whatnot are welcome.
(It may be sensible to send anything in the last two categories
directly to me, rather than to the whole list.)
Mine Brooder in Unix Dast ist!
I have noticed something about your v7 creation. When I try to use the C
compiler to compile fp support or any system structures(not structs but
components) I get an error /lib/c0 so there's something wrong with the c0
pass in libc. It was probably that way when the tapes were recovered. I have
managed to compile and assemble all the c source in the /usr/src/cmd/c
directory into object files so the assembler works. What should I do
manually with all these .o files? I need a working compiler.
I have I V7 system Warren that runs on PDP-11 that was created from some
of Keith Bostics's fileblock fragments. I can get this system up and running
but the C compiler seems to be broke. I get ***error 8 which I don't know
what that means but it's probably a pdp11 error code. I'm still trying to
learn about the pdps but do you know how I might regenerate this C compiler
from v7 that will fix c0? When I try to add floating point number emulation
to the C compiler and regen things I always get an error at c0. How could I
regenerate the c0 pass file? That seems to be the only thing that's stopping
me from going further. I don't know if the compiler can be rebuilt from
scratch if something like lib/c0 is broken.
OK, so I'm wwwwwaaaaaaayyyyyyy behind on reading TUHS.
I just wanted to say that if you can find a copy of the third edition
of "Unix In A Nutshell" (NOT the current fourth edition) you'll find
a chapter on the MM macros. It should be enough to make use of them,
as I did buy one of the SysIII licenses and I have a copy of this paper
that I referred to when writing that chapter.
And groff did do a good enough job formatting it that I was able
to print it out and it looked reasonable if not perfect. (Of course,
that was circa 1999...)
If I ever Get A Round Tuit I want to take the troff material from that
edition and do it as an ebook for O'Reilly. But I don't know when or
even if that'll ever happen.
> Date: Sat, 11 Mar 2006 15:25:49 +0100
> From: Gunnar Ritter <gunnarr(a)acm.org>
> Subject: Re: [TUHS] papers on the -mm macros?
> To: tuhs(a)minnie.tuhs.org, "A. Wik" <aw(a)aw.gs>
> "A. Wik" <aw(a)aw.gs> wrote:
> > I've found the documentation for most of the major
> > troff preprocessors and macros packages, but I can't
> > seem to find anything but occasional references to a
> > paper on the "Programmer's Memorandum Macros" (troff -mm)
> > by Smith and Mashey.
> The source code for this paper had been available as part
> of the System III distribution under the old (unfree) SCO
> In case you had applied for that license, and you still
> have an old PUPS archive CD at hand, you can find it in
> You will not be able to recover the original layout since
> PostScript font metrics are quite different from CAT ones,
> but Heirloom troff produces readable output at least.
> >But the point being made was that I've been around the block, I've worked
> >on and/or looked hard at many different Unix variants and I'm not at all sad
> >to see them go.
> Why are you here then?
Good question. I like it here, I like old Unix. I have little fondness
for all the commercial unices, see http://www.bitmover.com/lm/papers/srcos.html
for my reasons.
I think you may be confusing my dislike for commercial unix with a dislike
for unix. If so, that's mistake because I love Unix. I've dedicated a
huge portion of my life to helping unix as best I can.
Larry McVoy lm at bitmover.comhttp://www.bitkeeper.com
On Mon, 22 May 2006, 19:38:21 -0700, Larry McVoy wrote:
>But the point being made was that I've been around the block, I've worked
on and/or looked hard at many different Unix variants and I'm not at all sad
to see them go.
Why are you here then?
It's a fact that many of the big-gun Unix vendors have moved on but Unix
development continues to persist, so don't put it down yet. Unix is still
very much alive and kicking. Unix has been around forever and the Unicies
that remain still offer enough diversions to mix up the market and make
things interesting for us all. If Linux was the only Unix like system out
there then what would happen if Linux went belly-up. It could easily
happen if the big Linux vendors Redhat, Suse etc went to the dogs. Having
other Unix systems out their competing with each other as well as Linux is
Hello everyone, I'm just trying to get into PDP-11 Unix.
I have a couple older machines lying around not doing much (a SPARCstation 4
and an Ultra 1), and I've been fiddling around with the simh pdp11 and
2.11BSD on the Ultra 1. I can get the system to boot using the
211bsd.simhconfig file from the tarball here:
I can boot into what appears to be a workable system, but I'd like to have
networking and a larger hard drive. Can somebody help me out with getting
this set up? If anybody else out there has done the same thing, I'd like to
hear exactly what you did.
Thank you very much
Does anyone know how to compile gcc-3.4.6 for the pdp11? I use
the --target=pdp11 switch and the compiler runs for awhile then breaks. The
output says it's bulding for a pdp11-unknown something so there maybe
something I'm not using.
> Michael Sokolov, it was, that writted:
> You silly, twisted boy, you.
Indeed. Michael does not seem to have been taking his meds. Nice guy but
a bit out there.
> A good example would probably be SunOS 4 - we already know that Sun are
> quite interested in open sourcing stuff given OpenSolaris, but SunOS 4
> hasn't been, presumably because it is full of stuff-they-don't-own and has
> no commercial value at all.
I'm the guy who took SunOS 4.1.3 and removed all the non-free stuff from it
(which was 90% STREAMS) and demo-ed it to McNealy in effort to set it free.
A lot went into this: http://www.bitmover.com/lm/papers/srcos.html
There isn't much chance they'll release it and at this point it is so far
behind I'm not sure I see the point. Even though that is the one kernel
that I really loved.
> From: Peter Jeremy <peterjeremy(a)optushome.com.au>
> SMP support started earlier than 4.1.4. The sun4m machines (SS470,
> SS670) were the first SMP machines and ISTR they were supported in
Um, search google groups for lm@slovax - that was a 470. It was most
definitely not an SMP box though it was my favorite Sun machine. Great
machine, my home machine is still named slovax in honor of that box (which
was named slovax in honor of a Wisconsin 11/750 that held the 4.x BSD source
which taught me more than anything else).
And for those who care, slovax/470 now belongs to Theo Deraadt, I'm
ashamed to say that I sold it to him so I could buy some parts for my
VW van at the time. At the time I didn't have any money, if I could do
it over again I would have given it to him.
The 670 was an SMP, that's Chuck Narad's box. Pretty nice except that
bcopy performance was really bad.
But the bigger point I wanted to make was to react to all the stuff about
OSF/1 or Ultrix or Tru64 or AIX or whatever. Most of you probably have
no idea who I am or what we do. I run a company that makes a software
product which runs on all those old Unix platforms. We have all the
boxen with all the various Unix versions.
Other than SunOS 4.x, if they all fell off the face of the earth tomorrow
I couldn't be happier. They suck. And even SunOS sucks in some ways, it's
way behind Linux. I'm a file system guy, I'm the last guy who did anything
significant to UFS (ask Kirk), and I have to admit that the Linux guys are
in some ways running circles around the old school Unix guys. The one
exception (that I know of) is ZFS. That's pretty cool, the Linux guys
are unlikely to do anything that good, it's too complex.
But my point is that the love for the old unix versions is mostly
misplaced. V7, you bet. That teaches you "small" (as does Comer's
Xinu work). But all of the vendor Unices, even my beloved SunOS, pale in
comparison to Linux. Sad but true, I've spent a lot of time in the code.
And in some ways it isn't sad at all, it's cool. Linux is free.
The only sad part that I still see is maybe personal. I loved SunOS
because working in it, as a young kid, I didn't know shit. But there
I was, hacking away. When I started, wandering through the code made
me feel like I was in a fog, I couldn't see the next step. But as time
went on the fog cleared and I saw this very clear and clean architecture.
It became something that you could really see and see why it was that
way and see how you could extend it and see how you shouldn't extend it.
The generic kernel source (take away drivers and file system
implementations, but keep the VFS layer) is very small. I've lived for
many years in SunOS, I've lived in IRIX, I've lived in SCO (which is
more true to V7 than anything else), I've lived in Linux, I've read the
HP-UX code, I haven't read Ultrix, OSF/1 or AIX, but the ones I know,
they are all pretty simple. The only one that ever cleared the fog for
me was SunOS, all the other ones looked like a mess which is why I don't
share the sentiment that we should be crying over the loss of all the
I don't want to go back. Linux is pretty nice. Maybe they'll fuck it
up, that seems to be a Unix OS tradition, but so far so good.
Larry McVoy lm at bitmover.comhttp://www.bitkeeper.com