The Unix Heritage Society mailing list
 help / color / mirror / Atom feed
From: Clem Cole <clemc@ccc.com>
To: Will Senn <will.senn@gmail.com>
Cc: TUHS main list <tuhs@minnie.tuhs.org>
Subject: Re: [TUHS] Looking back to 1981 - what pascal was popular on what unix?
Date: Sat, 29 Jan 2022 14:59:02 -0500	[thread overview]
Message-ID: <CAC20D2Nh0t9dxVbRvXFxXkCfEaWkmp9MP1Xf+=YC3rfXf3SK3g@mail.gmail.com> (raw)
In-Reply-To: <0f83f174-eeca-30fb-7b98-77fb0da80f2e@gmail.com>

[-- Attachment #1: Type: text/plain, Size: 3966 bytes --]

On Fri, Jan 28, 2022 at 6:08 PM Will Senn <will.senn@gmail.com> wrote:

> 1. What edition of UNIX were they likely to be using?
>
Already answered ...

> 2. What versions of "Standard Pascal" were in vogue on UNIX at the time
> (1981)?
>
Remember in 1981, the 'Pascal Standard' was still in flight. ISO 7185 was
1983 and Pascal really does not standardize enough that real
portability was possible until the 1990 version and people actually started
implementing compilers that obeyed it.

>
> 3. What combinations of UNIX/Pascal were popular?
>
Depends the implementation...  there were a number of them... there were
the ones I had at that time

   - UCB Pascal -- pi being the most popular for the PDP-11, and later pc
   for the Vax
   - VU Pascal for the 11
   - The Zurich P4 compiler was around and of course eventually begat UCSD,
   but I don't remember that it was directly made to run on Unix[I did not
   have one], although it may/must have been.
   - The Similer2 compiler for 68000 [was one of Wirth's -- I think was the
   basis for the compiler that they used for Lilith - or maybe it was a fork
   that produced the Lilith and the Similer2 compiler -- I don't remember].
   That showed up on a couple of the 68K workstations in 1981.
   - Andy's Amsterdam Toolkit
   - There was a compiler that the RTS folks had at MIT, but I don't know
   its origin or how complete it was.
   - Purdue had something that had started on the CDCs -- that went to
   Tektronix and was the basis for some of the stuff the Logical
   Analizer folks were using.  It was cross compiler that ran on Unix and
   generated 8-bit microprocessor code for embedded systems.
   - Per Brinch Hansen had a Pascal that we was pushing, but I don't think
   he even moved it to Unix
   - Plus, the "Tunis" folks in Toronto had a Concurrent-Pascal and a
   UNIX-like system that ran on PDP-11s.

And the problem was 'standard' - lots of people messed with it and every
Pascal was a little different.  In 1981 at one of the HP/Tektronix
[Hatfield/McCoy parties - Stienhart probably remembers], Mike Zuhl, TW Cook
and I counted 14 different "Tek Pascal's" and over 25 "HP Basic's" -- each
was a little different.

UCSD Pascal was its own system, not Unix [sort of like Smalltalk -- a
walled garden within itself].  They started with the P4 compiler but they
made it work on 8-bit systems, so while it was around, it was not a UNIX
Pascal >>and<< it the Standard was still in flight.  The OMSI Pascal
compiler was popular for the PDP-11's, but it was RT-11 and RSX [maybe
RSTS, but I'm not sure].  The Zurich PDP-10 Compiler was very popular in
PDP-10 land.  DEC released the VAX/pascal for VMS, which they eventually
brought to Ultrix and Tru64.   I believe that that was a scratch rewrite
and did not use P4 or anything from the PDP-10s or CDC systems before it.

One of the big issues with P4 based stuff is that the original P2 porting
kit from Wirth assumed a 40 bit integer of the CDC system.   Making it work
with a 36 bit world was a tad easier than the 32 or 16 bit world, much less
the 8-bit micros.   Folks like DEC, UCB and Amsterdam which started over
with a new front end, tended to have an easier go of it and often had
better compiler tools [which were mostly C based BTW for UNIX and a
combination of Pascal and BLISS at DEC].

Sun later brought the UCB PI and PC to the SunOS, but pls Rob G/Larry
correct me here - I think they later did their own compiler when they did
their new C and Fortran.  Masscomp started with the Similer2 compiler, but
eventually did a new front end that matched their C and Fortran like DEC
had.


As for SWTinP running with FreePascal.  I looked into it and decided it was
doable, but never went very fair with my investigation since the need for
it, was less.  Back in the days of VMS, Pr1meOS and the like, SWT was a
Godsend.  But Windows and like have real Unix subsystems, so I never really
was motivated other than pure curiosity.

>

[-- Attachment #2: Type: text/html, Size: 7164 bytes --]

  parent reply	other threads:[~2022-01-29 19:59 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-01-28 23:07 Will Senn
2022-01-28 23:18 ` Dan Cross
2022-01-28 23:31   ` Will Senn
2022-01-29  0:03     ` Rob Pike
2022-01-29  0:40     ` Will Senn
2022-01-29 19:05       ` John Cowan
2022-01-29 19:36         ` arnold
2022-01-29 19:59 ` Clem Cole [this message]
2022-01-29 20:02   ` Jon Steinhart
2022-01-29 20:13   ` Bakul Shah
2022-01-29 20:30     ` Clem Cole
2022-01-29 20:34     ` Larry McVoy
2022-01-29 21:03       ` Al Kossow
2022-01-29 21:38         ` Larry McVoy
2022-01-29 22:06       ` Bakul Shah
2022-01-29 22:48         ` GREEN
2022-01-30  3:27           ` Larry McVoy
2022-01-30 16:57   ` David Barto
2022-01-30 18:07     ` [TUHS] Compilation "vs" byte-code interpretation, was " Dan Stromberg
2022-01-30 20:09       ` David Barto
2022-01-31  7:59         ` WEB
2022-01-30 22:51       ` Dan Cross
2022-01-30 23:57         ` Dan Stromberg
2022-01-31  0:23         ` Nemo Nusquam
2022-01-31  0:45           ` Steve Nickolas
2022-01-31 17:16             ` Paul Winalski
2022-01-31 20:00               ` Erik E. Fair
2022-01-31 22:45               ` Steve Nickolas
2022-02-02  4:53               ` Adam Thornton
2022-01-31  1:41       ` Phil Budne
2022-02-07  3:04   ` [TUHS] " Rob Gingell

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CAC20D2Nh0t9dxVbRvXFxXkCfEaWkmp9MP1Xf+=YC3rfXf3SK3g@mail.gmail.com' \
    --to=clemc@ccc.com \
    --cc=tuhs@minnie.tuhs.org \
    --cc=will.senn@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).