3 May
2016
3 May
'16
11:52 a.m.
Dear Experts,
during discussing the Rolms I came accross the following question:
What was the first (Minicomputer) architecture which offered
memory- and IO protection? I'd define the minimum requirements as:
- Existence of a superuser mode (Rolm calls this Executive mode)
- Existence of a user mode (With at least two users, Rolm offers 4)
- In superuser mode, IO and memory protection for each user can be
set up individually.
- Any access violation is trapped and handeled by superuser code.
- Of course commands for mode switching and setting up the
memory and IO ranges must exist.
I have got a real machine (Rolm 1602) having this implemented
and dating from 1975. A document on this "Access Protection Module" as
Rolm calls it also is dated 1975. It consists of a microcode module
which realizes an extension of the 16 bit Nova instruction set and an
additinoal CPU module, taking care of the new modes and supervising
the IO- and memory accesses.
My question is not regarding virtual memory memory, but regarding
protection (IO and memory) to ensure capsulation of indivitual
processes - not necessarily for multi user environments but e.g.
for safety critical applications...
Probably OS/2 in 1987 was one of the first home computer OSes to
support memory protection (how about IO protection?), BSD on some
Digital PDP-* was earlier (1977?) but still after the 1602.
Any hints out there on other "Mini" architectures of that era
having someting similar?
Erik.
3 May
3 May
12:17 p.m.
On May 3, 2016, at 11:52 AM, Erik Baigar <erik at
baigar.de> wrote:
Dear Experts,
during discussing the Rolms I came accross the following question:
What was the first (Minicomputer) architecture which offered
memory- and IO protection? I'd define the minimum requirements as:
- Existence of a superuser mode (Rolm calls this Executive mode)
- Existence of a user mode (With at least two users, Rolm offers 4)
- In superuser mode, IO and memory protection for each user can be
set up individually.
- Any access violation is trapped and handeled by superuser code.
- Of course commands for mode switching and setting up the
memory and IO ranges must exist.
...
Probably OS/2 in 1987 was one of the first home computer OSes to
support memory protection (how about IO protection?), BSD on some
Digital PDP-* was earlier (1977?) but still after the 1602.
No, the PDP-11 offered this starting with the 11/45, in 1971.
In larger computers the feature is much older. Consider the CDC 6600 (1964). While not
all the properties you mentioned apply because I/O is in separate peripheral processors,
the notion of a privileged mode and address mapping is there. And even that isn't the
oldest example, I think.
paul
1:34 p.m.
On 2016-05-03 12:17 PM, Paul Koning wrote:
(1964). While
not all the properties you mentioned apply because I/O is
in separate peripheral processors, the notion of a privileged mode and
address mapping is there. And even that isn't the oldest example, I think.
Professor Per Brinch Hansen's 2001 book, "Classic Operating Systems":
[the Atlas computer at Manchester University in the early 1960s] was
the first system to exploit _supervisor calls_ known as "extracodes":
Extracode routines form simple extensions of the basic order code,
and also provide specific entry to supervisor routines.
... The Atlas supervisor has been called "the first recognisable
modern operating system" (Lavington 1980).
The book reprints the 1961 paper, "The Atlas Supervisor," Tom Kilburn,
R. Bruce Payne and David J. Howarth:
The fixed store contains about 250 subroutines which can be called in
On May 3, 2016, at 11:52 AM, Erik Baigar <erik
at baigar.de> wrote:
Dear Experts,
during discussing the Rolms I came accross the following question:
What was the first (Minicomputer) architecture which offered
memory- and IO protection? I'd define the minimum requirements as:
- Existence of a superuser mode (Rolm calls this Executive mode)
- Existence of a user mode (With at least two users, Rolm offers 4)
- In superuser mode, IO and memory protection for each user can be
set up individually.
- Any access violation is trapped and handeled by superuser code.
- Of course commands for mode switching and setting up the
memory and IO ranges must exist.
...
Probably OS/2 in 1987 was one of the first home computer OSes to
support memory protection (how about IO protection?), BSD on some
Digital PDP-* was earlier (1977?) but still after the 1602.
No, the PDP-11 offered this starting with the 11/45, in 1971.
In larger computers the feature is much older. Consider the CDC 6600 from an object program by single instructions called
extracodes. When
these routines are being obeyed, extracode control is used:
extracode
control is also used by the supervisor, which requires access to the
"private" stores. ...
The supervisor program controls all those functions of the system
that are not obtained merely by allowing the central computer to proceed
with obeying an object program. ...
Supervisor extracode routines (S.E.R.'s) form the principal
"branches" of the supervisor program. ... They are protected from
interference by object programs by using subsidiary store as working
space, together with areas of core and drum store which are locked out
in the usual way whilst an object program is being executed ... The
S.E.R.'s thus apply mutual protection between themselves and an object
program.
Great book, that repeatedly shows how many ideas we think of as "modern"
are actually quite old. :)
--Toby
paul
2:17 p.m.
At 01:34 PM 5/3/2016, Toby Thain wrote:
Great book, that repeatedly shows how many ideas we
think of as "modern" are actually quite old. :)
An example - the first dynamically refreshed capacitive memory system of which I am aware
dated from 1939. It was not RAM, as it was a rotating drum.
Dale H. Cook, Roanoke/Lynchburg, VA
Osborne 1 / Kaypro 4-84 / Kaypro 1 / Amstrad PPC-640
http://plymouthcolony.net/starcity/radios/index.html
1:55 p.m.
On Tue, 3 May 2016, Paul Koning wrote:
No, the PDP-11 offered this starting with the 11/45, in 1971.
OK, that is a hint - the 11/45 also had a MMU and obviosly must
have been a great machine for multi-user stuff.
In larger computers the feature is much older.
Consider the CDC 6600
(1964). While not all the properties you mentioned apply because I/O is
in separate peripheral processors, the notion of a privileged mode and
address mapping is there. And even that isn't the oldest example, I
I have been aware, that bigger machines offered this already
and therefore I explicitly asked for "Minis"... But yes, the CDCs
are very impressive machines!!!
Thanks for your reply,
Erik.
2:15 p.m.
On May 3, 2016, at 1:55 PM, Erik Baigar <erik at
baigar.de> wrote:
On Tue, 3 May 2016, Paul Koning wrote:
Keep in mind that mapping and protection are not required to build a multi-user system.
DEC did timesharing on the PDP-11/20 (RSTS-11) and on the PDP-8 without either. And
multiprogramming goes back much further, at least as far as the THE operating system in
1964, on the EL-X8. That OS is particularly interesting because it has virtual memory and
demand paging without any hardware help, without address mapping or protection.
In all these cases, you need help from the compilers to keep things safe. Burroughs
mainframes do the same sort of thing, though that's not so obvious: the OS security is
dependent on the fact that ordinary users can't write and execute ESPOL programs.
paul
No, the PDP-11 offered this starting with the 11/45, in 1971.
OK, that is a hint - the 11/45 also had a MMU and obviosly must
have been a great machine for multi-user stuff.
2:18 p.m.
On Tue, 3 May 2016, Paul Koning wrote:
further, at least as far as the THE operating system
in 1964, on the
EL-X8. That OS is particularly interesting because it has virtual
memory and demand paging without any hardware help, without address
mapping or protection.
Wow - and this machine still used a drum for secondary
storage; quite outstanding, you are right!
Erik.
3:10 p.m.
On May 3, 2016, at 2:18 PM, Erik Baigar <erik at
baigar.de> wrote:
On Tue, 3 May 2016, Paul Koning wrote:
It's quite a nice system. The internals are fairly extensively documented in
Dijkstra's early "EWD" documents (at the U Texas Austin archive), though a
fair fraction are in Dutch. Among other interesting aspects is spooling to virtual memory
for both input (paper tape) and output (printer, paper tape punch, plotter). And of
course the impressive design discipline documented in his famous paper "The
structure of the THE operating system".
For that matter, the machine is interesting. Not only is this the place where semaphores
were invented and first applied, but they aren't just a software concurrency control
mechanism. The I/O system uses semaphores, too -- one that counts pending I/O requests,
and another that counts completions and ties to the interrupt request.
paul
further, at least as far as the THE operating
system in 1964, on the EL-X8. That OS is particularly interesting because it has virtual
memory and demand paging without any hardware help, without address mapping or protection.
Wow - and this machine still used a drum for secondary
storage; quite outstanding, you are right! 
2:22 p.m.
On 5/3/16 11:15 AM, Paul Koning wrote:
Keep in mind that mapping and protection are not
required to build a multi-user system. DEC did timesharing on the PDP-11/20 (RSTS-11) and
on the PDP-8 without either.
The SDS-940 is an early small-ish timsharing system that had base-bounds memory
protection, circa mid-60s
It was the platform that a lot of early timesharing services used.
TSS/8 used 4K segmentation with protection (memory expansion/timeshare control option)
The early BASIC minicomputer timesharing systems (line 2000TSB) had virtual BASIC machines
for each user that swapped.
2:23 p.m.
On Tue, 3 May 2016, Al Kossow wrote:
The early BASIC minicomputer timesharing systems (line
2000TSB) had
virtual BASIC machines for each user that swapped.
Well, but is this not an software abstraction for a non existing
encapsulation feature of the underlying hardware? DGs Extended BASIC
also implemented time sharing without the hardware offering this
feature and probably many other examples exist (Even if you beat
me: Early Windows?)...
If the language does not offer any critical commands, this works
as well as users can not do nasty things.
2:46 p.m.
On 5/3/16 11:23 AM, Erik Baigar wrote:
Well, but is this not an software abstraction for a
non existing
encapsulation feature of the underlying hardware?
Yes, or you can design a language with array bounds checking and no
arbitrary pointer dereferencing. Burroughs Algol, HP SPL, Xerox Mesa,
or their modern reinventions, built on segmented stack architectures.
On the microcomputer side, companies built MMUs for microcomputers that
had bank switching and protection. Digital Research MPM, for example.
3:05 p.m.
Again, not minicomputer, but the Burroughs B5000 did have "invalid
address" detection, which would case the machine to switch from "normal"
to "control" state with an interrupt. The problem with comparing the
B5000 architecture to anything else is that it was quite liberally like
nothing else.
Some may argue that the IBM S/360 had I/O protection, but those of us
who played games with CCWs in user mode might take exception to that
statement.:) One of the favorite CCW-writing exploits was to ring the
1052 console bell then execute a TIC (transfer in channel) back to the
bell-ringing CCW. On one occasion that I'm aware of, this so panicked
the operator that he pulled "emergency stop"...
Ah, the good old days...
--Chuck
1:14 p.m.
New subject: Cleaning rubber goo
What's the best commonly available solvent for cleaning the rubber goo that used to be
pressure rollers, belts, feet etc.?
m
1:34 p.m.
New subject: Cleaning rubber goo
On 05/03/2016 10:14 AM, Al Kossow wrote:
something naptha/citrus based, like goo gone?
On 5/3/16 10:14 AM, Mike Stein wrote:
> What's the best commonly available solvent for cleaning the rubber
> goo that used to be pressure rollers, belts, feet etc.?
I start with detergent and water, which allows me to get much of the goo
off without smearing it all over the place. Then paint thinner
liberally mixed with cursing.
--Chuck
1:39 p.m.
New subject: Cleaning rubber goo
On Tue, 3 May 2016, Mike Stein wrote:
What's the best commonly available solvent for
cleaning the rubber goo
that used to be pressure rollers, belts, feet etc.?
FWIW I've used IPA with reasonable results; as recently as last weekend
last time. It doesn't seem to affect intact rubber, at least not readily,
and it's quite gentle to various plastics -- it does remove the sticky
remains though. I take it as an advantage actually.
Maciej
2:11 p.m.
New subject: Cleaning rubber goo
I've tried Goo Gone and IPA (among others) but neither seemed very effective; maybe
I'm just expecting too much and elbow grease is the answer.
m
----- Original Message -----
From: "Maciej W. Rozycki" <macro at linux-mips.org>
To: "General Discussion: On-Topic and Off-Topic Posts" <cctalk at
classiccmp.org>
Sent: Tuesday, May 03, 2016 1:39 PM
Subject: Re: Cleaning rubber goo
On Tue, 3 May 2016, Mike Stein wrote:
What's the best commonly available solvent
for cleaning the rubber goo
that used to be pressure rollers, belts, feet etc.?
FWIW I've used IPA with reasonable results; as recently as last weekend
last time. It doesn't seem to affect intact rubber, at least not readily,
and it's quite gentle to various plastics -- it does remove the sticky
remains though. I take it as an advantage actually.
Maciej
1:45 p.m.
New subject: Cleaning rubber goo
On May 3, 2016, at 1:14 PM, Mike Stein <mhs.stein
at gmail.com> wrote:
What's the best commonly available solvent for cleaning the rubber goo that used to
be pressure rollers, belts, feet etc.?
That depends on what the substrate is. If possible, I use lacquer thinner, which is a
very powerful solvent. For example, it removes label adhesive or rubber cement faster
than anything else I've tried. But if the substrate is some kind of plastic, it
probably objects to this, so something less potent (and less effective) is needed. A
label chemist told me that label adhesive residue can be removed, slowly but safely, with
WD-40. I haven't tried that on former rubber, but it might serve for that as well.
The key point is that most plastics don't mind WD-40.
If it matters a lot, test the proposed solvent first on an inconspicuous/noncritical part
of the substrate. Sometimes you get surprised. For example, ethanol is safe for nearly
all plastics, but it badly messes up clear acrylic ("plexiglas").
paul
2:14 p.m.
New subject: Cleaning rubber goo
I've been wondering if methyl salicylate (oil of wintergreen) would do
a decent job. Usually, it's used to restore plasticity to dried-out
rubber (one common treatment for hard typewriter platens is "Rubber
Renu", which is MS in a solution of xylol).
Nice fresh minty smell. Good for your arthritis. I haven't tried it
yet as a goo solvent.
--Chuck
4 May
4 May
12:18 p.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
Erik,
I'm not sure whether it qualifies for your full list, but the HP2100A (that came out
in 1971) had a "Memory Protect" hardware that
"Gives the security necessary to protect a defined area of memory from alteration by
a user program.
Priority: Second highest priority interrupt (shared with memory parity).
Operation: Initiated under program control; protects any amount of memory.
Fence Register: Set under program control; memory below fence is protected.
Interrupt: To trap cell for system routine when user program:
a) Attempts to alter a protected location
b) Attempts to jump into the protected area
c) Attempts to execute an 1/0 instruction
Violation Register: Contains memory address of violating instruction."
(from the 1972 HP2100A Processor Description)
The predecessor HP2116B (of 1968 vintage!) also had a "Memory Protect" board as
an option, I cannot confirm it had the same functionality as the HP2100A above.
IIRC, this was the main mechanism to protect the OS (RTE, DOS) from user code.
Gottfried
_____
Gottfried Specht | Gottfried at specht-online.com | +49 211 151695?+49 151 2911 2915
-----Urspr?ngliche Nachricht-----
Von: cctalk [mailto:cctalk-bounces at classiccmp.org] Im Auftrag von Erik Baigar
Gesendet: Dienstag, 3. Mai 2016 17:53
An: cctalk at classiccmp.org
Betreff: When did Memory- and IO Protection Emerge (Esp. in Minis)?
Dear Experts,
during discussing the Rolms I came accross the following question:
What was the first (Minicomputer) architecture which offered
memory- and IO protection? I'd define the minimum requirements as:
- Existence of a superuser mode (Rolm calls this Executive mode)
- Existence of a user mode (With at least two users, Rolm offers 4)
- In superuser mode, IO and memory protection for each user can be
set up individually.
- Any access violation is trapped and handeled by superuser code.
- Of course commands for mode switching and setting up the
memory and IO ranges must exist.
I have got a real machine (Rolm 1602) having this implemented and dating from 1975. A
document on this "Access Protection Module" as Rolm calls it also is dated 1975.
It consists of a microcode module which realizes an extension of the 16 bit Nova
instruction set and an additinoal CPU module, taking care of the new modes and supervising
the IO- and memory accesses.
My question is not regarding virtual memory memory, but regarding protection (IO and
memory) to ensure capsulation of indivitual processes - not necessarily for multi user
environments but e.g.
for safety critical applications...
Probably OS/2 in 1987 was one of the first home computer OSes to support memory protection
(how about IO protection?), BSD on some Digital PDP-* was earlier (1977?) but still after
the 1602.
Any hints out there on other "Mini" architectures of that era having someting
similar?
Erik.
4:43 p.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
My 1974 HP 21MX, descendant of the HP 2100A, sure inherited this Memory Protect card. One
register that you load, prevents access to any memory below the address of the register.
Marc
From: cctalk <cctalk-bounces at classiccmp.org> on behalf of Gottfried Specht
<Gottfried.Specht at t-online.de>
Reply-To: "cctalk at classiccmp.org" <cctalk at classiccmp.org>
Date: Wednesday, May 4, 2016 at 9:18 AM
To: "cctalk at classiccmp.org" <cctalk at classiccmp.org>
Subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
Erik,
I'm not sure whether it qualifies for your full list, but the HP2100A (that came out
in 1971) had a "Memory Protect" hardware that
"Gives the security necessary to protect a defined area of memory from alteration by
a user program.
Priority: Second highest priority interrupt (shared with memory parity).
Operation: Initiated under program control; protects any amount of memory.
Fence Register: Set under program control; memory below fence is protected.
Interrupt: To trap cell for system routine when user program:
a) Attempts to alter a protected location
b) Attempts to jump into the protected area
c) Attempts to execute an 1/0 instruction
Violation Register: Contains memory address of violating instruction."
(from the 1972 HP2100A Processor Description)
The predecessor HP2116B (of 1968 vintage!) also had a "Memory Protect" board as
an option, I cannot confirm it had the same functionality as the HP2100A above.
IIRC, this was the main mechanism to protect the OS (RTE, DOS) from user code.
Gottfried
5:02 p.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
Marc wrote...
-----
My 1974 HP 21MX, descendant of the HP 2100A, sure inherited this Memory Protect card. One
register that you load, prevents access to any memory below the address of the register.
----
That's just the fence register (memory access interception) capability. There's
also the I/O & HLT instruction interception, interrupt interception, etc. mentioned
earlier.
M.E.M. was standard option (ie. Included) with most E/F machines.
I don't know if the 2114/5/6 had the same capabilities in the fence area....
J
7:46 p.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
Marc wrote...
-----
My 1974 HP 21MX, descendant of the HP 2100A, sure inherited this Memory Protect card. One
register that you load, prevents access to any memory below the address of the register.
----
That's just the fence register (memory access interception) capability. There's
also the I/O & HLT instruction interception, interrupt interception, etc. mentioned
earlier. [...]
J
Yes, I didn't know it was more comprehensive. I just learned it from this
thread! Is this a good message board or what...
Marc
6:02 p.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
Is the CDC 1700 considered to be in the family of "minicomputers"?
(i.e. was the word invented before then?).
If so, the 1700 had a rather elaborate system of memory and peripheral
protection. Circa 1965 (at least that's the date on my manual).
--Chuck
8:06 p.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
In the not mini but very maxi category, I just learned that IBM implemented memory
protection as an RPQ (customer feature) at the request of the MIT folks that built the
first IBM time sharing system (CTSS, the predecessor of Multics), on their IBM 7094.
Around 1963, unless it was already implemented in the IBM 7090 which would have been 1961.
At least that's my cursory understanding of it from www.multicians.org/thvv/7094.html
. I was very surprised it was that early!
Marc
Sent from my iPad
On May 4, 2016, at 3:02 PM, Chuck Guzis <cclist at
sydex.com> wrote:
Is the CDC 1700 considered to be in the family of "minicomputers"?
(i.e. was the word invented before then?).
If so, the 1700 had a rather elaborate system of memory and peripheral
protection. Circa 1965 (at least that's the date on my manual).
--Chuck
5 May
5 May
4:15 a.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
I thought we were talking Mini Computers. The Ferranti/Manchester Atlas had
virtual memory of a sort which provided protection, and indeed IBM bought
the Virtual Memory patents from Manchester University. I gather the PDP/11
received Memory Mapping boards early in its life, didn't these offer some
protection?
Dave
-----Original Message-----
From: cctalk [mailto:cctalk-bounces at classiccmp.org] On Behalf Of Curious
Marc
Sent: 05 May 2016 01:07
To: General Discussion: On-Topic and Off-Topic Posts
<cctalk at
classiccmp.org>
Subject: Re: AW: When did Memory- and IO Protection
Emerge (Esp. in
Minis)?
In the not mini but very maxi category, I just learned that IBM
implemented
memory protection as an RPQ (customer feature) at the
request of the MIT
folks that built the first IBM time sharing system (CTSS, the predecessor
of
Multics), on their IBM 7094. Around 1963, unless it
was already
implemented
in the IBM 7090 which would have been 1961. At least
that's my cursory
understanding of it from www.multicians.org/thvv/7094.html . I was very
surprised it was that early!
Marc
Sent from my iPad
> On May 4, 2016, at 3:02 PM, Chuck Guzis <cclist at sydex.com> wrote:
>
> Is the CDC 1700 considered to be in the family of "minicomputers"?
> (i.e. was the word invented before then?).
>
> If so, the 1700 had a rather elaborate system of memory and peripheral
> protection. Circa 1965 (at least that's the date on my manual).
>
> --Chuck
>
4:46 a.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
On Thu, 5 May 2016, Dave Wade wrote:
I thought we were talking Mini Computers. The
Ferranti/Manchester Atlas had
virtual memory of a sort which provided protection, and indeed IBM bought
the Virtual Memory patents from Manchester University. I gather the PDP/11
received Memory Mapping boards early in its life, didn't these offer some
protection?
Yes, originally I had Minis in mind, and yes - this term is not well
defined. Anyhow I learned a lot from all the answers as many differnet
machines have been covered in contrast to the normally DEC biased
traffic here. What still is a quite interesting question for me is,
that purely "mapped memory" or "virtual memory" does not match the
answerr to my question as it still lacks the IO aspect and I'd call
it only memory protection if there is a method to prevent ordinary
programs from modifying the memory map.
So from my point of view the HP2100 matches the criteria of my initial
question, the Rolm APM does but probabaly some others mentioned for their
virtual memory do not. E.g. the DG mapped memory (Nova3 or Eclipse)
lacks the "user prohibited to modify map" feature I think.
And for sure: Atlas was a very interesting machine with great capa-
bilities for its days!
Erik.
-----Original
Message-----
From: cctalk [mailto:cctalk-bounces at classiccmp.org] On Behalf Of Curious
Marc
Sent: 05 May 2016 01:07
To: General Discussion: On-Topic and Off-Topic Posts
<cctalk at
classiccmp.org>
Subject: Re: AW: When did Memory- and IO
Protection Emerge (Esp. in
Minis)?
In the not mini but very maxi category, I just learned that IBM
implemented
memory protection as an RPQ (customer feature) at
the request of the MIT
folks that built the first IBM time sharing system (CTSS, the predecessor
of
Multics), on their IBM 7094. Around 1963, unless
it was already
implemented
in the IBM 7090 which would have been 1961. At
least that's my cursory
understanding of it from www.multicians.org/thvv/7094.html . I was very
surprised it was that early!
Marc
Sent from my iPad
> On May 4, 2016, at 3:02 PM, Chuck Guzis <cclist at sydex.com> wrote:
>
> Is the CDC 1700 considered to be in the family of "minicomputers"?
> (i.e. was the word invented before then?).
>
> If so, the 1700 had a rather elaborate system of memory and peripheral
> protection. Circa 1965 (at least that's the date on my manual).
>
> --Chuck
>
4 May
4 May
8:07 p.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
Is the CDC 1700 considered to be in the family of
"minicomputers"?
(i.e. was the word invented before then?).
Though functionally it sort of had the minicomputer nature, it
was physically a bit large for that term ? would have been
called a "process control" computer. I also don't think I heard the
word "minicomputer" until a couple of years after I first saw a CDC 1700.
10:33 p.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
On 05/04/2016 05:07 PM, ANDY HOLT wrote:
Well, I don't know. By the time the Cyber 18 came out, it was a 120 VAC
powered unit that a strongish person could lift off the floor (about 90
lbs)--and functionally pretty much the same machine as the original
1700, just implemented with more advanced technology.
If that's not a minicomputer, I don't know what is.
We used them as data concentrators hooked to leased lines, card readers
and punches and various other peripherals.
If the 1700 isn't a minicomputer, you'll have to correct the Wikipedia
article.
--Chuck
Is the CDC 1700 considered to be in the family of
"minicomputers"?
(i.e. was the word invented before then?).
Though functionally it sort of had the minicomputer nature, it was
physically a bit large for that term ? would have been called a
"process control" computer. I also don't think I heard the word
"minicomputer" until a couple of years after I first saw a CDC 1700.
11:24 p.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
Could someone with access to the OED please check up the first use of the term
"minicomputer"
I strongly suspect it was around the time that the PDP11/20 came out or slightly later.
The IBM 1130 and 1800 were comparable to the /original/ CDC 1700, were similarly launched
in the mid 60s,
but similarly they were not /at that time/ referred to as minis.
In retrospect we might well call these minicomputers but that is not the question as
stated.
----- Original Message -----
From: "Chuck Guzis" <cclist at sydex.com>
To: "General Discussion: On-Topic and Off-Topic Posts" <cctalk at
classiccmp.org>
Sent: Thursday, 5 May, 2016 3:33:03 AM
Subject: Re: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
On 05/04/2016 05:07 PM, ANDY HOLT wrote:
Well, I don't know. By the time the Cyber 18 came out, it was a 120 VAC
powered unit that a strongish person could lift off the floor (about 90
lbs)--and functionally pretty much the same machine as the original
1700, just implemented with more advanced technology.
If that's not a minicomputer, I don't know what is.
We used them as data concentrators hooked to leased lines, card readers
and punches and various other peripherals.
If the 1700 isn't a minicomputer, you'll have to correct the Wikipedia
article.
--Chuck
Is the CDC 1700 considered to be in the family of
"minicomputers"?
(i.e. was the word invented before then?).
Though functionally it sort of had the minicomputer nature, it was
physically a bit large for that term ? would have been called a
"process control" computer. I also don't think I heard the word
"minicomputer" until a couple of years after I first saw a CDC 1700.
5 May
5 May
1:07 a.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
On 05/04/2016 08:24 PM, ANDY HOLT wrote:
Could someone with access to the OED please check up
the first use of
the term "minicomputer" I strongly suspect it was around the time
that the PDP11/20 came out or slightly later. The IBM 1130 and 1800
were comparable to the /original/ CDC 1700, were similarly launched
in the mid 60s, but similarly they were not /at that time/ referred
to as minis.
Merriam Webster and the OED gives the first published use in 1967
http://www.wordorigins.org/index.php/site/comments/1967_words/
The 1700 would fit into that in at least one of its incarnations. The
160A would not, as it went out of production in 1965, for example.
Sometimes words can be misleading. The word "microcomputer" when
initially used, did not mean what we think of today--that of a certain
level of integration of an IC that includes most of the component parts
of a computer.
AES used this same term much earlier for "a computer than can be
microprogrammed" as implemented in a cage of of cards.
--Chuck
7 May
7 May
6:18 p.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
On 05/05/16 06:07, Chuck Guzis wrote:
On 05/04/2016 08:24 PM, ANDY HOLT wrote:
The OED cites:
1967. Datamation. Jan 1985/1.
The quote is roughly '"an overcoat-pocket-sized" aerospace computer,
the 449...'
The OED does point out that the usage of the term has shifted over the
years.
Interestingly (well, interesting to me at least) the earliest OED
citation from Datamation is "personal computer" from an advert in 1959
(May-Jun 29).
There's an earlier citation for "personal computer" from 1954.
Antonio
--
Antonio Carlini
arcarlini at iee.org
Could someone with access to the OED please check
up the first use of
the term "minicomputer" I strongly suspect it was around the time
that the PDP11/20 came out or slightly later. The IBM 1130 and 1800
were comparable to the /original/ CDC 1700, were similarly launched
in the mid 60s, but similarly they were not /at that time/ referred
to as minis.
Merriam Webster and the OED gives the first published use in 1967
http://www.wordorigins.org/index.php/site/comments/1967_words/
5 May
5 May
1:10 a.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
For the fun of the argument: I was privileged enough to see Carl's IBM 1130, and to my
newbie eye, it may justifiably earn the title of "small" computer, when compared
to its brethren of the time. But it would never occur to me to call it a mini! It's
quite a biggie computer actually. Heavy stuff, forklift or winch needed to put it safely
in the truck as I recall. Then I thought our IBM 1401 was big. That's when more
knowledgeable people pointed me to the IBM 7090. Now that's *really* big. And then you
have SAGE. Now that's huge. Or insane, depending on your engineering point of view
:-).
Marc
Sent from my iPad
On May 4, 2016, at 8:24 PM, ANDY HOLT <andy.holt at
tesco.net> wrote:
Could someone with access to the OED please check up the first use of the term
"minicomputer"
I strongly suspect it was around the time that the PDP11/20 came out or slightly later.
The IBM 1130 and 1800 were comparable to the /original/ CDC 1700, were similarly launched
in the mid 60s,
but similarly they were not /at that time/ referred to as minis.
In retrospect we might well call these minicomputers but that is not the question as
stated.
----- Original Message -----
From: "Chuck Guzis" <cclist at sydex.com>
To: "General Discussion: On-Topic and Off-Topic Posts" <cctalk at
classiccmp.org>
Sent: Thursday, 5 May, 2016 3:33:03 AM
Subject: Re: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
On 05/04/2016 05:07 PM, ANDY HOLT wrote:
Well, I don't know. By the time the Cyber 18 came out, it was a 120 VAC
powered unit that a strongish person could lift off the floor (about 90
lbs)--and functionally pretty much the same machine as the original
1700, just implemented with more advanced technology.
If that's not a minicomputer, I don't know what is.
We used them as data concentrators hooked to leased lines, card readers
and punches and various other peripherals.
If the 1700 isn't a minicomputer, you'll have to correct the Wikipedia
article.
--Chuck
Is the CDC 1700 considered to be in the family of
"minicomputers"?
(i.e. was the word invented before then?).
Though functionally it sort of had the minicomputer nature, it was
physically a bit large for that term ? would have been called a
"process control" computer. I also don't think I heard the word
"minicomputer" until a couple of years after I first saw a CDC 1700.
1:29 a.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
On 05/04/2016 10:10 PM, Curious Marc wrote:
For the fun of the argument: I was privileged enough
to see Carl's
IBM 1130, and to my newbie eye, it may justifiably earn the title of
"small" computer, when compared to its brethren of the time. But it
would never occur to me to call it a mini! It's quite a biggie
computer actually. Heavy stuff, forklift or winch needed to put it
safely in the truck as I recall. Then I thought our IBM 1401 was big.
That's when more knowledgeable people pointed me to the IBM 7090. Now
that's *really* big. And then you have SAGE. Now that's huge. Or
insane, depending on your engineering point of view :-).
On the other hand, the PB250 was contained in a single 5' rack (table
model), ran off of a single 15A 120V circuit and weighed a bit over 100
lbs. I'd call it a minicomputer if it weren't for the fact that it was
brought out around 1961. 22 bit words. Up to about 16KW in the box;
magnetostrictive delay line memory, bit-serial ALU. IIRC, lotsa diodes,
but comparatively few (ca. 300-400) transistors.
--Chuck
1:06 p.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
On 05/05/2016 12:29 AM, Chuck Guzis wrote:
On the other hand, the PB250 was contained in a single 5' rack (table
model), ran off of a single 15A 120V circuit and weighed a bit over 100
lbs. I'd call it a minicomputer if it weren't for the fact that it was
brought out around 1961. 22 bit words. Up to about 16KW in the box;
magnetostrictive delay line memory, bit-serial ALU. IIRC, lotsa diodes,
but comparatively few (ca. 300-400) transistors.
Another "mini" was the Bendix G-15, 300+ vacuum tubes, 3000
diodes and a drum memory.
Certainly no memory protection on it. It was the size of a
refrigerator.
And another, the LINC, discrete transistor machine, 2K
12-bit words and a software driven screen, so it was
actually interactive. Storage was LINCtapes, the predecessor
to DECtapes. Fit in a single 5' relay rack, with a console
on a desk.
Jon
1:19 p.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
On 05/05/2016 10:06 AM, Jon Elson wrote:
And another, the LINC, discrete transistor machine, 2K
12-bit words
and a software driven screen, so it was actually interactive. Storage
was LINCtapes, the predecessor to DECtapes. Fit in a single 5' relay
rack, with a console on a desk.
...and, of course, the tabletop Univac 422.
--Chuck
6 May
6 May
4:54 p.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
On May 5, 2016, at 10:19 AM, Chuck Guzis <cclist at
sydex.com> wrote:
...and, of course, the tabletop Univac 422.
--Chuck
I had to look it up. What a sexy small machine, with its boards showing!
http://www.computerhistory.org/collections/catalog/102633139
5 May
5 May
4:17 a.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
But an original PDP/11 is almost as massive......
-----Original Message-----
From: cctalk [mailto:cctalk-bounces at classiccmp.org] On Behalf Of Curious
Marc
Sent: 05 May 2016 06:11
To: General Discussion: On-Topic and Off-Topic Posts <cctalk at classiccmp.org>
Subject: Re: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
For the fun of the argument: I was privileged enough to see Carl's IBM 1130,
and to my newbie eye, it may justifiably earn the title of "small" computer,
when compared to its brethren of the time. But it would never occur to me to
call it a mini! It's quite a biggie computer actually. Heavy stuff, forklift or
winch
needed to put it safely in the truck as I recall. Then I thought our IBM 1401 was
big. That's when more knowledgeable people pointed me to the IBM 7090. Now
that's *really* big. And then you have SAGE. Now that's huge. Or insane,
depending on your engineering point of view :-).
Marc
Sent from my iPad
On May 4, 2016, at 8:24 PM, ANDY HOLT
<andy.holt at tesco.net> wrote:
Could someone with access to the OED please check up the first use of the
term
"minicomputer"
I strongly suspect it was around the time that
the PDP11/20 came out or
slightly later.
The IBM 1130 and 1800 were comparable to the
/original/ CDC 1700, were
similarly launched in the mid 60s, but similarly they were not /at that time/
referred to as minis.
In retrospect we might well call these minicomputers but that is not the
question
as stated.
----- Original Message -----
From: "Chuck Guzis" <cclist at sydex.com>
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk at classiccmp.org>
Sent: Thursday, 5 May, 2016 3:33:03 AM
Subject: Re: AW: When did Memory- and IO Protection Emerge (Esp. in
Minis)?
>
>> On 05/04/2016 05:07 PM, ANDY HOLT wrote:
>>
>>> Is the CDC 1700 considered to be in the family of
"minicomputers"?
>>> (i.e. was the word invented before then?).
>>
>> Though functionally it sort of had the minicomputer nature, it was
>> physically a bit large for that term ? would have been called a
>> "process control" computer. I also don't think I heard the word
>> "minicomputer" until a couple of years after I first saw a CDC 1700.
>
> Well, I don't know. By the time the Cyber 18 came out, it was a 120
> VAC powered unit that a strongish person could lift off the floor
> (about 90 lbs)--and functionally pretty much the same machine as the
> original 1700, just implemented with more advanced technology.
>
> If that's not a minicomputer, I don't know what is.
>
> We used them as data concentrators hooked to leased lines, card
> readers and punches and various other peripherals.
>
> If the 1700 isn't a minicomputer, you'll have to correct the Wikipedia
> article.
>
> --Chuck
> 
2:23 a.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
Could someone with access to the OED please check up
the first use of the term "minicomputer"
I strongly suspect it was around the time that the PDP11/20 came out or slightly later.
Ngram shows the first real use about 1967, with a peak about 1983.
--
Will
2:34 a.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
Well, I don't know. By the time the Cyber 18 came
out, it was a 120 VAC
powered unit that a strongish person could lift off the floor (about 90
lbs)--and functionally pretty much the same machine as the original
1700, just implemented with more advanced technology.
1774 was from about 1967, and basically a six foot rack.
1784 was from about 1974, and about the same volume as a PDP-8/e.
I think the Wiki article may be wrong about the date of the original
1700 systems (1704) being mid-1966. That seems late. When I get back
home, I might look at the chunks I have to see if I can get date
codes.
If that's not a minicomputer, I don't know
what is.
Wasn't minicomputer really a marketing term, anyway? Suits and all?
--
Will
1:34 p.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
On 05/04/2016 11:34 PM, William Donzelli wrote:
Wasn't minicomputer really a marketing term,
anyway? Suits and all?
Well, it was the sixties, after all. We all forget "midicomputer". :)
One thing that some may not know about the 1700 is that it had a
*per-word* protection bit as well as I/O protection on a "per device"
basis. I don't know of any other computers with that feature.
--Chuck
1:58 p.m.
On May 5, 2016, at 1:34 PM, Chuck Guzis <cclist at
sydex.com> wrote:
On 05/04/2016 11:34 PM, William Donzelli wrote:
Burroughs mainframes? While the tag bits aren't quite semantically equivalent to
protection, you get some of the same benefits.
paul
Wasn't minicomputer really a marketing term,
anyway? Suits and all?
Well, it was the sixties, after all. We all forget "midicomputer". :)
One thing that some may not know about the 1700 is that it had a
*per-word* protection bit as well as I/O protection on a "per device"
basis. I don't know of any other computers with that feature. 
3:45 p.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
Wasn't
minicomputer really a marketing term, anyway? Suits and all?
Well, it was the sixties, after all. We all forget
"midicomputer". :)
We try to.
My sexist memories of miniskirts are what lets me tolerate the marketing
silliness behind the name "minicomputer". Mmmmm...miniskirts
4:35 a.m.
New subject: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
On Wed, 4 May 2016, Gottfried Specht wrote:
I'm not sure whether it qualifies for your full
list, but the HP2100A (that came out in 1971) had a "Memory Protect" hardware
that
Hi Gottfried,
thanks for the excellent answer - yes I think this is exactly what
matches my specification! Thanks.
It is really astonishing how many people know a lot on various machines
which is really great. I suspected that HP had something, too.
Fence Register: Set under program control; memory
below fence is protected.
This is a clever and somewhat outstanding feature - most others use
protection on basis of blocks ar abuse the virtual memory for the
purpose ;-)
Best regards,
Erik.
-----Urspr?ngliche Nachricht-----
Von: cctalk [mailto:cctalk-bounces at classiccmp.org] Im Auftrag von Erik Baigar
Gesendet: Dienstag, 3. Mai 2016 17:53
An: cctalk at classiccmp.org
Betreff: When did Memory- and IO Protection Emerge (Esp. in Minis)?
Dear Experts,
during discussing the Rolms I came accross the following question:
What was the first (Minicomputer) architecture which offered
memory- and IO protection? I'd define the minimum requirements as:
- Existence of a superuser mode (Rolm calls this Executive mode)
- Existence of a user mode (With at least two users, Rolm offers 4)
- In superuser mode, IO and memory protection for each user can be
set up individually.
- Any access violation is trapped and handeled by superuser code.
- Of course commands for mode switching and setting up the
memory and IO ranges must exist.
I have got a real machine (Rolm 1602) having this implemented and dating from 1975. A
document on this "Access Protection Module" as Rolm calls it also is dated 1975.
It consists of a microcode module which realizes an extension of the 16 bit Nova
instruction set and an additinoal CPU module, taking care of the new modes and supervising
the IO- and memory accesses.
My question is not regarding virtual memory memory, but regarding protection (IO and
memory) to ensure capsulation of indivitual processes - not necessarily for multi user
environments but e.g.
for safety critical applications...
Probably OS/2 in 1987 was one of the first home computer OSes to support memory
protection (how about IO protection?), BSD on some Digital PDP-* was earlier (1977?) but
still after the 1602.
Any hints out there on other "Mini" architectures of that era having someting
similar?
Erik.
2:55 p.m.
New subject: AW: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
Thanks, Erik.
How do I remember this ca. 40 years later?
Well, while servicing these systems they would frequently stop with a "Memory Protect
Error" (various Operating Systems).
Guess what the intuitive action was: Replace the "Memory Protect Board" - which
n e v e r fixed the problem.
So digging into the technology it became clear, that the Memory Protect Board in these
cases had only fulfilled its duty: protect the memory below the fence register from some
other piece of hardware (usually a processor or DMA-board) running havoc in memory. That
learning stuck ...
Kind regards,
Gottfried
_____
Gottfried Specht | Gottfried at specht-online.com | +49 211 151695?+49 151 2911 2915
-----Urspr?ngliche Nachricht-----
Von: cctalk [mailto:cctalk-bounces at classiccmp.org] Im Auftrag von Erik Baigar
Gesendet: Donnerstag, 5. Mai 2016 10:36
An: General Discussion: On-Topic and Off-Topic Posts
Betreff: Re: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
On Wed, 4 May 2016, Gottfried Specht wrote:
I'm not sure whether it qualifies for your full
list, but the HP2100A
(that came out in 1971) had a "Memory Protect" hardware that
Hi Gottfried,
thanks for the excellent answer - yes I think this is exactly what matches my
specification! Thanks.
It is really astonishing how many people know a lot on various machines which is really
great. I suspected that HP had something, too.
Fence Register: Set under program control; memory
below fence is protected.
This is a clever and somewhat outstanding feature - most others use protection on basis of
blocks ar abuse the virtual memory for the purpose ;-)
Best regards,
Erik.
-----Urspr?ngliche Nachricht-----
Von: cctalk [mailto:cctalk-bounces at classiccmp.org] Im Auftrag von Erik
Baigar
Gesendet: Dienstag, 3. Mai 2016 17:53
An: cctalk at classiccmp.org
Betreff: When did Memory- and IO Protection Emerge (Esp. in Minis)?
Dear Experts,
during discussing the Rolms I came accross the following question:
What was the first (Minicomputer) architecture which offered
memory- and IO protection? I'd define the minimum requirements as:
- Existence of a superuser mode (Rolm calls this Executive mode)
- Existence of a user mode (With at least two users, Rolm offers 4)
- In superuser mode, IO and memory protection for each user can be
set up individually.
- Any access violation is trapped and handeled by superuser code.
- Of course commands for mode switching and setting up the
memory and IO ranges must exist.
I have got a real machine (Rolm 1602) having this implemented and dating from 1975. A
document on this "Access Protection Module" as Rolm calls it also is dated 1975.
It consists of a microcode module which realizes an extension of the 16 bit Nova
instruction set and an additinoal CPU module, taking care of the new modes and supervising
the IO- and memory accesses.
My question is not regarding virtual memory memory, but regarding protection (IO and
memory) to ensure capsulation of indivitual processes - not necessarily for multi user
environments but e.g.
for safety critical applications...
Probably OS/2 in 1987 was one of the first home computer OSes to support memory
protection (how about IO protection?), BSD on some Digital PDP-* was earlier (1977?) but
still after the 1602.
Any hints out there on other "Mini" architectures of that era having someting
similar?
Erik.
3:50 p.m.
New subject: AW: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
On 2016-05-05 2:55 PM, Gottfried Specht wrote:
Thanks, Erik.
How do I remember this ca. 40 years later?
Well, while servicing these systems they would frequently stop with a "Memory
Protect Error" (various Operating Systems).
Guess what the intuitive action was: Replace the "Memory Protect Board" - which
n e v e r fixed the problem.
So digging into the technology it became clear, that the Memory Protect Board in these
cases had only fulfilled its duty: protect the memory below the fence register from some
other piece of hardware (usually a processor or DMA-board) running havoc in memory. That
learning stuck ...
Well, at least the directive wasn't "remove the Memory Protect board"? :-)
--Toby
Kind regards,
Gottfried
_____
Gottfried Specht | Gottfried at specht-online.com | +49 211 151695?+49 151 2911 2915
-----Urspr?ngliche Nachricht-----
Von: cctalk [mailto:cctalk-bounces at classiccmp.org] Im Auftrag von Erik Baigar
Gesendet: Donnerstag, 5. Mai 2016 10:36
An: General Discussion: On-Topic and Off-Topic Posts
Betreff: Re: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
On Wed, 4 May 2016, Gottfried Specht wrote:
I'm not sure whether it qualifies for your
full list, but the HP2100A
(that came out in 1971) had a "Memory Protect" hardware that
...
6 May
6 May
9:45 a.m.
New subject: AW: AW: When did Memory- and IO Protection Emerge (Esp. in Minis)?
On Thu, 5 May 2016, Gottfried Specht wrote:
How do I remember this ca. 40 years later?
[snip]
Well, while servicing these systems they would
frequently stop with a
"Memory Protect Error" (various Operating Systems). Guess what the
intuitive action was: Replace the "Memory Protect Board" - which n e v e
r fixed the problem. So digging into the technology it became clear,
that the Memory Protect Board in these cases had only fulfilled its
duty: protect the memory below the fence register from some other piece
of hardware (usually a processor or DMA-board) running havoc in memory.
Essentially the board did what it was supposed to do! That is exactly
what the APM is good for in my Rolm 1602: As these machines where
used in applcations where errors in hardware or software running
havoc would have resulted in really severe problems these where
a good idea ;-)
Have a nice weekend,
Erik.
On Wed, 4 May 2016, Gottfried Specht wrote:
I'm not sure whether it qualifies for your
full list, but the HP2100A
(that came out in 1971) had a "Memory Protect" hardware that
Hi Gottfried,
thanks for the excellent answer - yes I think this is exactly what matches my
specification! Thanks.
It is really astonishing how many people know a lot on various machines which is really
great. I suspected that HP had something, too.
Fence Register: Set under program control; memory
below fence is protected.
This is a clever and somewhat outstanding feature - most others use protection on basis
of blocks ar abuse the virtual memory for the purpose ;-)
Best regards,
Erik.
-----Urspr?ngliche Nachricht-----
Von: cctalk [mailto:cctalk-bounces at classiccmp.org] Im Auftrag von Erik
Baigar
Gesendet: Dienstag, 3. Mai 2016 17:53
An: cctalk at classiccmp.org
Betreff: When did Memory- and IO Protection Emerge (Esp. in Minis)?
Dear Experts,
during discussing the Rolms I came accross the following question:
What was the first (Minicomputer) architecture which offered
memory- and IO protection? I'd define the minimum requirements as:
- Existence of a superuser mode (Rolm calls this Executive mode)
- Existence of a user mode (With at least two users, Rolm offers 4)
- In superuser mode, IO and memory protection for each user can be
set up individually.
- Any access violation is trapped and handeled by superuser code.
- Of course commands for mode switching and setting up the
memory and IO ranges must exist.
I have got a real machine (Rolm 1602) having this implemented and dating from 1975. A
document on this "Access Protection Module" as Rolm calls it also is dated 1975.
It consists of a microcode module which realizes an extension of the 16 bit Nova
instruction set and an additinoal CPU module, taking care of the new modes and supervising
the IO- and memory accesses.
My question is not regarding virtual memory memory, but regarding protection (IO and
memory) to ensure capsulation of indivitual processes - not necessarily for multi user
environments but e.g.
for safety critical applications...
Probably OS/2 in 1987 was one of the first home computer OSes to support memory
protection (how about IO protection?), BSD on some Digital PDP-* was earlier (1977?) but
still after the 1602.
Any hints out there on other "Mini" architectures of that era having someting
similar?
Erik.
5 May
5 May
4:41 a.m.
2016-05-03 17:52 GMT+02:00 Erik Baigar <erik at baigar.de>:
Dear Experts,
during discussing the Rolms I came accross the following question:
What was the first (Minicomputer) architecture which offered
memory- and IO protection? I'd define the minimum requirements as:
- Existence of a superuser mode (Rolm calls this Executive mode)
- Existence of a user mode (With at least two users, Rolm offers 4)
- In superuser mode, IO and memory protection for each user can be
set up individually.
- Any access violation is trapped and handeled by superuser code.
- Of course commands for mode switching and setting up the
memory and IO ranges must exist.
I have got a real machine (Rolm 1602) having this implemented
and dating from 1975. A document on this "Access Protection Module" as
Rolm calls it also is dated 1975. It consists of a microcode module
which realizes an extension of the 16 bit Nova instruction set and an
additinoal CPU module, taking care of the new modes and supervising
the IO- and memory accesses.
My question is not regarding virtual memory memory, but regarding
protection (IO and memory) to ensure capsulation of indivitual
processes - not necessarily for multi user environments but e.g.
for safety critical applications...
Probably OS/2 in 1987 was one of the first home computer OSes to
support memory protection (how about IO protection?), BSD on some
Digital PDP-* was earlier (1977?) but still after the 1602.
Any hints out there on other "Mini" architectures of that era having
someting similar?
Erik.
What about the Norsk Data series of machines, NORD-1, NORD-10 etc.
The NORD-10 had memory protection and paging. Circa 1973. According to the
wiki page the NORD-1 had an option to provide virtual memory. The wiki page
claim the NORD-1 to be the first mini to have virtual memory (1969). I
cannot really tell if this is true. https://en.wikipedia.org/wiki/Nord-1
/Mattis
6 May
6 May
10:29 a.m.
On 5 May 2016 at 10:41, Mattis Lind <mattislind at gmail.com> wrote:
What about the Norsk Data series of machines, NORD-1,
NORD-10 etc.
The NORD-10 had memory protection and paging. Circa 1973. According to the
wiki page the NORD-1 had an option to provide virtual memory. The wiki page
claim the NORD-1 to be the first mini to have virtual memory (1969). I
cannot really tell if this is true. https://en.wikipedia.org/wiki/Nord-1
/Mattis
The NORD-1 (developed 1967, sold from 1968) was a real-time 16-bit
minicomputer with up to 64kw of core memory. The NORD-1 reference
manual describes the memory protection system this way:
"Both real-time and background programs can be run concurrently in a
NORD-1 system since the real-time program can be protected against
alteration. The optional features guarantees that protected areas of
memory cannot be written into. The protection feature also prevents
the execution of instructions that could initiate I/O transfers or
change the status of the protection system."
I believe this refers to, or is related to the optional virtual memory
system. And more so this passage:
"The NORD-1 dynamic core allocation system (paging system) is an
automatic address interpretation system which allows programs to be
written for 64K virtual core, with only parts of the program in
physical core at a given time[..] The page table is kept in core. For
each memory reference there is an automatic hardware address
translation via an automatic page table reference.[..]"
And it goes on to describing page fault interrupts, and privileged instructions.
But although it had privileged and non-privileged instructions from
the beginning, the NORD-1 did not acquire a time-sharing operating
system until after mid-1971, and that was only a proof-of-concept
version. It was announced in 1972 and became a product in 1973. The
user- and superuser separation was implemented using the already
existing instruction type separation, as mentioned. In 1973 the
NORD-10 appeared as well, which fully covers the OP's requirements.
The NORD-1 of 1968, or NORD-1 with VM of 1969 only partly covers it,
as the time sharing operating system was not yet in place at that
time.
10:59 a.m.
On Fri, 6 May 2016, Tor Arntsen wrote:
On 5 May 2016 at 10:41, Mattis Lind <mattislind at
gmail.com> wrote:
[snip]
minicomputer with up to 64kw of core memory. The
NORD-1 reference
[snip]
alteration. The optional features guarantees that
protected areas of
memory cannot be written into. The protection feature also prevents
the execution of instructions that could initiate I/O transfers or
change the status of the protection system."
OK, this machine also matches the requirements of my initial
question - great. So there quite some machine offering such
features well before 1975...
Thanks for citing from the reference manual!
Erik.
3070
days inactive
3074
days old
50 comments
19 participants
participants (19)
-
a.carlini＠ntlworld.com -
aek＠bitsavers.org -
andy.holt＠tesco.net -
cclist＠sydex.com -
cisin＠xenosoft.com -
curiousmarc3＠gmail.com -
dave.g4ugm＠gmail.com -
elson＠pico-systems.com -
erik＠baigar.de -
Gottfried.Specht＠t-online.de -
jwest＠classiccmp.org -
kspt.tor＠gmail.com -
macro＠linux-mips.org -
mattislind＠gmail.com -
mhs.stein＠gmail.com -
paulkoning＠comcast.net -
radiotest＠juno.com -
toby＠telegraphics.com.au -
wdonzelli＠gmail.com