28 Jul
2014
28 Jul
'14
1:27 p.m.
From: Dave Horsfall <dave(a)horsfall.org>
I recall that there were other differences as well,
but only minor. In
my paper in AUUGN titled "Unix on the LSI-11/23" it will reveal all
about porting V6 to the thing.
I did a google for that, but couldn't find it. Is it available anywhere
online? (I'd love to read it.) I seem to recall vaguely that AUUGN stuff were
online, but if so, I'm not sure why the search didn't turn it up.
I vaguely remember that the LTC had to be disabled
during the boot
process, for example, with an external switch.
I think you might be right, which means the simulated 11/23 I tested on
wasn't quite right - but keep reading!
I remember being worried about this when I started doing the V6 11/23 version
a couple of months back, because I remembered the 11/03's didn't have a
programmable clock, just a switch. So I was reading through the 11/23
documentation (I had used 11/23s, but on this point my memory had faded),
trying to see if they too did not have a programmable clock.
As best I can currently make out, the answer is 'yes/no, depending on the
exact model'! E.g. the 11/23-PLUS _does_ seem to have a programmable clock
(see pg. 610 of the 1982 edition of "microcomputers and memories"), but the
base 11/23 _apparently_ does not.
Anyway, the simulated 11/23 (on Ersatz11) does have the LTC (I just checked,
and 'lks' contains '0177546', so it thinks it has one :-).
But this will be easy to code around; if no link clock is found (in main.c),
I'd probably set 'lks' to point somewhere harmless (054, say - I'm
using
050/052 to hold the pointer to the CSW, and the software CSW if there isn't a
hardware one). That way I can limit the changes to be in main.c, I won't have
to futz with clock.c too.
Noel
PS: On at least the 11/40 (and maybe the /45 too), the line clock was an
option! It was a single-height card, IIRC.
28 Jul
28 Jul
10:23 p.m.
New subject: First Unix that could run on a PDP-11 with QBUS
On Mon, 28 Jul 2014, Noel Chiappa wrote:
There was a project a few years ago to scan all issues of AUUGN
(Australian Unix Users Group Newsletter); the last I heard was that all
issues had been obtained, and handed over to some Google mob for
archiving. Apparently the scanning process is destructive but makes for
an ideal copy for as many as you like. The originals, being up to 40
years or so old, would have been in bad shape anyway.
A search for "auugn" reveals a few pointers, but AUUG itself dissolved a
few years ago because we had achieved our purpose i.e. bring Unix to the
mass market in Australia (its competition at the time was RSTS, RSX, and
PICK of all things). Guess which one survived? Concurrent CP/M never
really had a hold, MS-DOS thankfully died (I was still using CP/M at the
time; heck, I even had UUCP on it, which was pretty impressive considering
that the Microbee didn't have a serial port), and I predict that Windoze
will go the way of the Irish potato crop and for the same reason.
Warren may know more about the archived issues.
It was hilarious, in a morbid sort of way. I cottoned on when the
bootstrap process crapped itself for no apparent reason (it got
interrupted when no ISR was in place), and we'd occasionally forget to
enable it...
I recall that
there were other differences as well, but only minor. In
my paper in AUUGN titled "Unix on the LSI-11/23" it will reveal all
about porting V6 to the thing.
I did a google for that, but couldn't find it. Is it available anywhere
online? (I'd love to read it.) I seem to recall vaguely that AUUGN stuff
were online, but if so, I'm not sure why the search didn't turn it up. I vaguely
remember that the LTC had to be disabled during the boot
process, for example, with an external switch.
I think you might be right, which means the simulated 11/23 I tested on
wasn't quite right - but keep reading! I remember being worried about this when I started
doing the V6 11/23
version a couple of months back, because I remembered the 11/03's didn't
have a programmable clock, just a switch. So I was reading through the
11/23 documentation (I had used 11/23s, but on this point my memory had
faded), trying to see if they too did not have a programmable clock.
As best I can currently make out, the answer is 'yes/no, depending on
the exact model'! E.g. the 11/23-PLUS _does_ seem to have a programmable
clock (see pg. 610 of the 1982 edition of "microcomputers and
memories"), but the base 11/23 _apparently_ does not.
I never saw the -PLUS, so I can't help you there, and my shelf of DEC and
Unix etc manuals disappeared during several moves.
Anyway, the simulated 11/23 (on Ersatz11) does have
the LTC (I just
checked, and 'lks' contains '0177546', so it thinks it has one :-).
Quite likely. I came up with a battery of tests at boot time, in order to
determine just what sort of a model it was e.g. did it have the SLR and so
on. Same thing for illegal instructions, such as floating point. We had
/40s all over the place (some dedicated ones had no MMU, and ran a custom
program to talk 200-UT to a remote Cyber), two or three /70s (I had no
responsibility for those, but we shared code a lot), a /60 (interesting
box), and a sprinkling of /23s.
But this will be easy to code around; if no link clock
is found (in
main.c), I'd probably set 'lks' to point somewhere harmless (054, say -
I'm using 050/052 to hold the pointer to the CSW, and the software CSW
if there isn't a hardware one). That way I can limit the changes to be
in main.c, I won't have to futz with clock.c too.
Speaking of the CSW, we came up with some amusing idle patterns. The
boxes with the octal display displayed rotating 1s (I had to determine
whether it had an octal display or a real one somehow; I've long since
forgotten).
PS: On at least the 11/40 (and maybe the /45 too), the
line clock was an
option! It was a single-height card, IIRC.
Yeah; the aforementioned low-end /40s had quite an impressive program that
scheduled by the use of co-routines (no LTC either). It emulated the CDC
Remote Batch Station (we briefly had one of those too; it was S L O W).
Fun days!
-- Dave
29 Jul
29 Jul
9:46 a.m.
New subject: First Unix that could run on a PDP-11 with QBUS
The old AUUG newsletters are all at http://minnie.tuhs.org/Archive/Documentation/AUUGN/
Cheers, Warren
On 29 July 2014 08:23:12 AEST, Dave Horsfall <dave(a)horsfall.org> wrote:
On Mon, 28 Jul 2014, Noel Chiappa wrote:
anywhere
on
--
Sent from my Android phone with K-9 Mail. Please excuse my brevity.
> I recall that there were other differences
as well, but only minor.
In
> my paper in AUUGN titled "Unix on the
LSI-11/23" it will reveal all
about
porting V6 to the thing.
I did a google for that, but couldn't find it. Is it available online? (I'd love to read it.) I seem to
recall vaguely that AUUGN
stuff
were online, but if so, I'm not sure why the
search didn't turn it
up.
There was a project a few years ago to scan all issues of AUUGN
(Australian Unix Users Group Newsletter); the last I heard was that all
issues had been obtained, and handed over to some Google mob for
archiving. Apparently the scanning process is destructive but makes
for
an ideal copy for as many as you like. The originals, being up to 40
years or so old, would have been in bad shape anyway.
A search for "auugn" reveals a few pointers, but AUUG itself dissolved
a
few years ago because we had achieved our purpose i.e. bring Unix to
the
mass market in Australia (its competition at the time was RSTS, RSX,
and
PICK of all things). Guess which one survived? Concurrent CP/M never
really had a hold, MS-DOS thankfully died (I was still using CP/M at
the
time; heck, I even had UUCP on it, which was pretty impressive
considering
that the Microbee didn't have a serial port), and I predict that
Windoze
will go the way of the Irish potato crop and for the same reason.
Warren may know more about the archived issues.
I vaguely
remember that the LTC had to be disabled during the boot
process, for example, with an external switch.
I think you might be right, which means the simulated 11/23 I tested wasn't quite right - but keep reading!
It was hilarious, in a morbid sort of way. I cottoned on when the
bootstrap process crapped itself for no apparent reason (it got
interrupted when no ISR was in place), and we'd occasionally forget to
enable it...
I remember being worried about this when I
started doing the V6 11/23
version a couple of months back, because I
remembered the 11/03's
didn't
have a programmable clock, just a switch. So I
was reading through
the
11/23 documentation (I had used 11/23s, but on
this point my memory
had
faded), trying to see if they too did not have a
programmable clock.
As best I can currently make out, the answer is 'yes/no, depending on
the exact model'! E.g. the 11/23-PLUS _does_
seem to have a
programmable
clock (see pg. 610 of the 1982 edition of
"microcomputers and
memories"), but the base 11/23 _apparently_ does not.
I never saw the -PLUS, so I can't help you there, and my shelf of DEC
and
Unix etc manuals disappeared during several moves.
Anyway, the simulated 11/23 (on Ersatz11) does
have the LTC (I just
checked, and 'lks' contains '0177546', so it thinks it has one :-).
Quite likely. I came up with a battery of tests at boot time, in order
to
determine just what sort of a model it was e.g. did it have the SLR and
so
on. Same thing for illegal instructions, such as floating point. We
had
/40s all over the place (some dedicated ones had no MMU, and ran a
custom
program to talk 200-UT to a remote Cyber), two or three /70s (I had no
responsibility for those, but we shared code a lot), a /60 (interesting
box), and a sprinkling of /23s.
But this will be easy to code around; if no link
clock is found (in
main.c), I'd probably set 'lks' to point somewhere harmless (054, say
-
I'm using 050/052 to hold the pointer to the
CSW, and the software
CSW
if there isn't a hardware one). That way I
can limit the changes to
be
in main.c, I won't have to futz with clock.c
too.
Speaking of the CSW, we came up with some amusing idle patterns. The
boxes with the octal display displayed rotating 1s (I had to determine
whether it had an octal display or a real one somehow; I've long since
forgotten).
PS: On at least the 11/40 (and maybe the /45
too), the line clock was
an
option! It was a single-height card, IIRC.
Yeah; the aforementioned low-end /40s had quite an impressive program
that
scheduled by the use of co-routines (no LTC either). It emulated the
CDC
Remote Batch Station (we briefly had one of those too; it was S L O W).
Fun days!
-- Dave
_______________________________________________
TUHS mailing list
TUHS(a)minnie.tuhs.org
https://minnie.tuhs.org/mailman/listinfo/tuhs
9:56 a.m.
New subject: First Unix that could run on a PDP-11 with QBUS
On Tue, 29 Jul 2014, I wrote:
We had /40s all over the place (some dedicated ones
had no MMU, and ran
a custom program to talk 200-UT to a remote Cyber), two or three /70s (I
had no responsibility for those, but we shared code a lot), a /60
(interesting box), and a sprinkling of /23s.
Oops; upon re-reading my article, we had a sprinkling of /34s, with just
the one /23. I think.
I'd like to believe that I was the first in Australia to port V6 to the
/34, the /23, and the /60 (I did a paper on that as well), but if others
in the rest of the world beat me to it then I never heard about it.
-- Dave
2:33 p.m.
New subject: First Unix that could run on a PDP-11 with QBUS
I can not speak for Australia. I lived the Fifth, Sixth and Seventh
Edition changes in the mid-late 1970s.
At CMU, in the USA we had an early /34 running Fifth Edition in the EE
Department summer of 75/76 ish - right after the /34 came out. There were
lots of 40's around campus and at the time were popular because CMU hacked
them (i.e. the E or Extended version of the 40), but I remember Gordon Bell
got the EE Dept a "deal" on the new single board 11 - i.e. the /34 & it was
significantly cheaper.
We got V6 shortly there after when a couple folks made a road trip to
NJ to see Ken and the EE system did not run V5 very long [frankly, I do not
remember much about it]. I do remember that we had boot strapped /34 from
some 11/40E's in CS - I remember the lack of switch register issue, but I
also seem to remember it was easy fix in m40.s. I've forgotten all the
differences, but I recall that they were small. I do have memories of
going back and forth between the CS and EE bldgs a couple of times until we
got it right.
The bigger issue that bit us was we had to careful about the fact that
the CS folks had implemented CSAVE/CRET in the WCS of the 11/40Es and
hacked on the C compiler to generate that - so binaries from CS could not
move to EE unless that hack was turned off [BTW: That job in EE was my
first experience with C].
The late Ted Kowalski showed up at CMU a few weeks later for his OYOC
year and updated the EE machine to be V6++ system [and also he had copies
of the proofs from Dennis's upcoming book on C ]. Being CMU, we had a lot
of BLISS hackers at the time - so Ted singing C praises was quite a stir --
I was not yet really indoctrinated with the merits of C and remember
arguing with him since at the time the C compilers were not nearly as
polished as the CMU BLISS compilers - although it ran natively which
BLISS/11 could not do].
At the time, a big thing Ted did was introduce us to stdio. Until
then most of the our C code was pretty hackneyed - I seem to remember
something called the portable C library for V6 (I do not remember it as
part of V5 but have been). However, Ted's V6+ compilers used stdio and it
quickly because the EE standard. They must have migrated to CS, but I've
forgotten because by them most of my hacking was on the EE system.
Ted was working on this really cool program to fix the file system
[fsck] - although a number of us worked with him and it was working with
that code - that I started to see what a great language it was and ended up
writing way more C than anything else.
I would clone the EE /34 system for a new job at CMU's Mellon
Institute but by then DEC had created the /34A [I've forgotten the
differences]. Mellon was the first time I ran into a UNIX vs. XX war. One
of the grad students wanted to run RSX-11 - since DEC "supported it." As
a lowly undergrad I was really pleased I won when the EE prof
behind Mellon Institute agreed we "seemed to have something" working
really well in the Dept [plus it must have helped that the Biomed team,
also decided to clone the EE system for their research and not use RSX].
As for the 11/60, upon graduation, I spent my first weekend in Oregon
in late 1979, helping Steve Glaser update the Teklabs /60 to be more CMU
like - but that was likely based on V6. I do not remember when we cut V7
in at Tek, it had been released and I don't remember the details for CMU -
we had to be running V7 by the time I left. At Tek, by the end of 1979, we
had done such a good job with the /60 that we managed to get an 11/70
(without any budget for it which was quite a trick). So, I spent my first
Oregon Christmas bringing V7 up on that system - with the infinite storage
capacity of 3 RP06s [oh boy]. My faint memory is that Tek had a V7
license, but Steve had not yet managed to get V7 running on the /60, only
V6.
Also, early in Winter/Spring of '79 before I had left Dan Klein and I
had gone on strike to force CMU to buy a commercial license for Mellon
institute -- which was not using UNIX for teaching for paid research. CMU
would be the first University to get a commercial license, I believe
Case-Western followed suite the next fall when Fred Park returned from
Tek-Labs after heard me talk about what CMU did that summer.
Clem
On Tue, Jul 29, 2014 at 5:56 AM, Dave Horsfall <dave(a)horsfall.org> wrote:
On Tue, 29 Jul 2014, I wrote:
We had /40s all over the place (some dedicated
ones had no MMU, and ran
a custom program to talk 200-UT to a remote Cyber), two or three /70s (I
had no responsibility for those, but we shared code a lot), a /60
(interesting box), and a sprinkling of /23s.
Oops; upon re-reading my article, we had a sprinkling of /34s, with just
the one /23. I think.
I'd like to believe that I was the first in Australia to port V6 to the
/34, the /23, and the /60 (I did a paper on that as well), but if others
in the rest of the world beat me to it then I never heard about it.
-- Dave
_______________________________________________
TUHS mailing list
TUHS(a)minnie.tuhs.org
https://minnie.tuhs.org/mailman/listinfo/tuhs
1:10 p.m.
New subject: First Unix that could run on a PDP-11 with QBUS
On Jul 28, 2014, at 5:23 PM, Dave Horsfall <dave(a)horsfall.org> wrote:
[snip]
There was a project a few years ago to scan all issues
of AUUGN
(Australian Unix Users Group Newsletter); the last I heard was that all
issues had been obtained, and handed over to some Google mob for
archiving. Apparently the scanning process is destructive but makes for
an ideal copy for as many as you like. The originals, being up to 40
years or so old, would have been in bad shape anyway.
Sounds like what Vernor Vinge was describing in 'Rainbows End' -- a cautionary
tale for librarians about destructive digitization/digitisation, among other themes.
- Milo
3553
days inactive
3554
days old
5 comments
5 participants
participants (5)
-
Clem Cole -
Dave Horsfall -
jnc＠mercury.lcs.mit.edu -
Milo Velimirović -
Warren Toomey