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.
-----Ursprüngliche Nachricht-----
Von: cctalk [mailto:cctalk-boun...@classiccmp.org] Im Auftrag von Erik
Baigar
Gesendet: Dienstag, 3. Mai 2016 17:53
An: cctalk@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.
