3 Jul
2022
3 Jul
'22
2:39 p.m.
On Sun, Jul 03, 2022 at 05:55:15PM +1000, steve jenkin wrote:
Yes, and they all run Linux or some tiny OS. Has anyone ported v7 to
any of these devices and seen it take off? Of course not, it doesn't
have TCP/IP.
On 3 Jul 2022, at 12:27, Larry McVoy
<lm(a)mcvoy.com> wrote:
I love the early Unix releases because they were so simple, processors
were simple then as well.
Bell???s Observation on Computer Classes has brought surprises
- we???ve had some very popular new devices appear at the bottom end of the market and
sell in the billions. 
3 Jul
3 Jul
2:59 p.m.
New subject: Thoughts on Licenses
On Sun, Jul 3, 2022 at 10:39 AM Larry McVoy <lm(a)mcvoy.com> wrote:
Has anyone ported v7 to any of these devices and seen
it take off?
I agree with your observation BTW, but I will point out that every server
based on Intel Si runs a V7 flavor called Minux in the IPMI node.
WRT to Licenses - it does matter for a practical purpose. As Larry points
out no one cares ... (until they do).
Without going into specifics (this is a real example BTW), let's take a
famous Science and Technology Museum in the EU that wants to use simh to
put up a demo of traditional computing (in this case it's different from
but let's say - using a PiDP-8 or PiDP-11 like display), they would like to
know that it's a legitimate thing to do so.
Lawyers are unlikely to look at the file called Metadata -- as I said, just
last week, I would be easily able to point some of them to the top of
Warren's tree for the Caldera license. The more licenses we have together
in one place will make folks like that a lot more comfortable with
proceeding. Again keeping the specific licenses with each release is
probably a good thing (I would suggest links the master copy), just make it
100% clear we are *trying to do the right thing*.
ᐧ
3:30 p.m.
New subject: Thoughts on Licenses
On Jul 3, 2022, at 9:59 AM, Clem Cole <clemc(a)ccc.com> wrote:
I agree with your observation BTW, but I will point out that every server based on Intel
Si runs a V7 flavor called Minux in the IPMI node.
You may be thinking of MINIX 1. It was a from-scratch implementation that was syscall
compatible with V7 but IIRC it didn't have any sort of memory protection as it was
designed to run on 8088. What runs on the Intel Management Engine is MINIX 3, a µkernel
design with NetBSD userland. As Prof. Tanenbaum said, "MINIX 1 and MINIX 3 are
related in the same way as Windows 3.1 and Windows XP are: same first name".
4:26 p.m.
New subject: Thoughts on Licenses
On Sun, Jul 3, 2022 at 11:30 AM Bakul Shah <bakul(a)iitbombay.org> wrote:
You may be thinking of MINIX 1. It was a from-scratch
implementation that
was syscall compatible with V7 but IIRC it didn't have any sort of memory
protection as it was designed to run on 8088.
Minux and specifically M1 was and always has been, a uK. And yes, M1 does
not need an MMU - since it was designed to run on an 8088. IIRC this was
Linus' original objection when he wanted to run on his 386-based PC (Wyse
32:16 box, IIRC). The key was Andy wanted to teach his students about V7
without running afoul of the AT&T license as Lions had with V6.
What runs on the Intel Management Engine
It's called the Intelligent Platform Management Interface
<https://streaklinks.com/BHBsBCyMjMy7xGTaiw4nohuO/https%3A%2F%2Fwww.intel.com%2Fcontent%2Fwww%2Fus%2Fen%2Fproducts%2Fdocs%2Fservers%2Fipmi%2Fipmi-second-gen-interface-spec-v2-rev1-1.html>
- *a.k.a.* IPMI
is MINIX 3, ... with NetBSD userland.
Actually, if you want to pick nits, neither statement is correct (remember
for whom I work).
MINIX 1 and MINIX 3 are related
That's because M3 added the MMU support that M1 lacked. But there is
nothing in M3 that IPMI is using other than it is the current version from
Andy's team. What IPMI has as an underlying uK is heavily hacked and is a
'derivative work' - the local uk is basically providing V7 interfaces to
some special programs.
It made little sense to recreate something for the platform engine, and
Minux was picked because it was smaller than any of the *BSDs and was not
GPL'ed so Intel IP was still protected.
ᐧ
ᐧ
4:55 p.m.
New subject: Thoughts on Licenses
On Sun, Jul 3, 2022 at 9:27 AM Clem Cole <clemc(a)ccc.com> wrote:
On Sun, Jul 3, 2022 at 11:30 AM Bakul Shah <bakul(a)iitbombay.org> wrote:
IPMI is an entirely different subsystem from the ME.
ME runs on one of at least two superfluous x86 cores within the CPU die of
intel CPUs for a while that can run Intel and vendor supervisors. When
this blew up in Intel’s face several years ago Minix 3 was the latest
incarnation.
https://github.com/hardenedlinux/firmware-anatomy/blob/master/hack_ME/me_in…
IPMI is a sideband and could be anything including embedded Linux on a
dedicated ARM SoC which is common in large scale installations.
is MINIX 3, ... with NetBSD userland.
You may be thinking of MINIX 1. It was a
from-scratch implementation that
was syscall compatible with V7 but IIRC it didn't have any sort of memory
protection as it was designed to run on 8088.
Minux and specifically M1 was and always has been, a uK. And yes, M1 does
not need an MMU - since it was designed to run on an 8088. IIRC this was
Linus' original objection when he wanted to run on his 386-based PC (Wyse
32:16 box, IIRC). The key was Andy wanted to teach his students about V7
without running afoul of the AT&T license as Lions had with V6.
What runs on the Intel Management Engine
It's called the Intelligent Platform Management Interface
<https://streaklinks.com/BHBsBCyMjMy7xGTaiw4nohuO/https%3A%2F%2Fwww.intel.com%2Fcontent%2Fwww%2Fus%2Fen%2Fproducts%2Fdocs%2Fservers%2Fipmi%2Fipmi-second-gen-interface-spec-v2-rev1-1.html>
- *a.k.a.* IPMI
Actually, if you want to pick nits, neither statement is correct (remember
for whom I work).
Maybe you are talking about something else. Minix 3 (Andy’s last OS) does
indeed match this description, uK with NetBSD user.
MINIX 1 and MINIX 3 are related
That's because M3 added the MMU support that M1
lacked. But there is
nothing in M3 that IPMI is using other than it is the current version from
Andy's team. What IPMI has as an underlying uK is heavily hacked and is a
'derivative work' - the local uk is basically providing V7 interfaces to
some special programs.
It made little sense to recreate something for the platform engine, and
Minux was picked because it was smaller than any of the *BSDs and was not
GPL'ed so Intel IP was still protected.
ᐧ
ᐧ
5:15 p.m.
New subject: Thoughts on Licenses
On Sun, Jul 3, 2022 at 12:55 PM Kevin Bowling <kevin.bowling(a)kev009.com>
wrote:
Maybe you are talking about something else. Minix 3
(Andy’s last OS) does
indeed match this description, uK with NetBSD user.
Sigh .. I did not say it was not.
ᐧ
5:48 p.m.
New subject: Thoughts on Licenses
On Jul 3, 2022, at 11:26 AM, Clem Cole <clemc(a)ccc.com> wrote:
On Sun, Jul 3, 2022 at 11:30 AM Bakul Shah <bakul(a)iitbombay.org
<mailto:bakul@iitbombay.org>> wrote:
You may be thinking of MINIX 1. It was a from-scratch implementation that was syscall
compatible with V7 but IIRC it didn't have any sort of memory protection as it was
designed to run on 8088.
Minux and specifically M1 was and always has been, a uK. And yes, M1 does not need an
MMU - since it was designed to run on an 8088. IIRC this was Linus' original
objection when he wanted to run on his 386-based PC (Wyse 32:16 box, IIRC). The key was
Andy wanted to teach his students about V7 without running afoul of the AT&T license
as Lions had with V6.
Er.. a "microkernel" without an MMU is basically nothing more than a thread
switcher (not unlike a variety of "realtime" embedded kernels like threadX and
what not).
What runs on the Intel Management Engine
It's called the Intelligent Platform Management Interface
<https://streaklinks.com/BHBsBCyMjMy7xGTaiw4nohuO/https%3A%2F%2Fwww.intel.com%2Fcontent%2Fwww%2Fus%2Fen%2Fproducts%2Fdocs%2Fservers%2Fipmi%2Fipmi-second-gen-interface-spec-v2-rev1-1.html>
- a.k.a. IPMI
is MINIX 3, ... with NetBSD userland.
Actually, if you want to pick nits, neither statement is correct (remember for whom I
work).
Not sure which statements you are talking about that are incorrect.
Minix3 running on ME was in the news a few years back. See for instance:
https://itsfoss.com/fact-intel-minix-case/ -- Websearch reveals many articles on IME +
MINIX, hardly any to IPMI + MINIX.
MINIX 3 + NetBSD userland is pretty much what minix3.org website says!
[And no, I was not aware of who you work for.]
In any case, sounds like you were talking about something not (well)known outside of
Intel.
MINIX 1 and MINIX 3 are related
That's because M3 added the MMU support that M1 lacked. But there is nothing in M3
that IPMI is using other than it is the current version from Andy's team. What IPMI
has as an underlying uK is heavily hacked and is a 'derivative work' - the local
uk is basically providing V7 interfaces to some special programs.
I think Tanenbaum's point was that MINIX3 is nothing like MINIX1 except in name.
It made little sense to recreate something for the
platform engine, and Minux was picked because it was smaller than any of the *BSDs and was
not GPL'ed so Intel IP was still protected.
No argument here :-)
ᐧ
ᐧ
7:48 p.m.
New subject: Thoughts on Licenses
On Sun, Jul 3, 2022 at 1:48 PM Bakul Shah <bakul(a)iitbombay.org> wrote:
Er.. a "microkernel" without an MMU is
basically nothing more than a
thread switcher (not unlike a variety of "realtime" embedded kernels like
threadX and what not).
Argue with Andy and not me and read his book. Andy calls it a
message-based uK. Given how they structured it. I would agree.
I think Tanenbaum's point was that MINIX3 is *nothing* like MINIX1 except
in name.
Having talked to Andy about this in person, as well as looking at the code,
I think I differ with your interpretation
yes, M3 supports a ton of things M1 did not. But the core API and KPI are
supersets.
I also know a bit about what Intel uses for the IPMI support as part of my
$ day job. Let's just say this is a great deal that is known outside of
Intel and a good bit that is not and/or misunderstood.
My point that started this rat hole was that Larry made a comment about V7
having little value. I know for a fact Larry's observation was not true.
And how we use the core Minix subsystem (*as a basic V7 platform for a
single custom application* that allows us to manage the server - as I said
to Larry think the LS1-11 on the Vax and the 11/40 on the KL processors)
and offered it as a counter-example. It could have been almost anything
that you called a thread switcher. A simple V7 was 'good enough' and Minux
supplied that for the team.
In fact, I can think of other applications where V6 or V7 is more than
enough for a lot of the tasks.
ᐧ
9:23 p.m.
New subject: Thoughts on Licenses
On Sun, Jul 03, 2022 at 03:48:33PM -0400, Clem Cole wrote:
single custom application* that allows us to manage
the server - as I said
to Larry think the LS1-11 on the Vax and the 11/40 on the KL processors)
I have a MicroVAXII with a KMV1A. Are you refering to such a card with
the "LSI-11 on the Vax"? Does this mean there is a V7 running on it?
I really don't know much about that card. Might have been used for I/O with
an IBM 3090 in this computer.
Matthias
--
When You Find Out Your Normal Daily Lifestyle Is Called Quarantine
10:40 p.m.
New subject: Thoughts on Licenses
Sorry for some of us old guys vax = 780
ᐧ
On Sun, Jul 3, 2022 at 5:30 PM Matthias Bruestle <m(a)mbsks.franken.de> wrote:
On Sun, Jul 03, 2022 at 03:48:33PM -0400, Clem Cole
wrote:
single custom application* that allows us to
manage the server - as I
said
to Larry think the LS1-11 on the Vax and the
11/40 on the KL processors)
I have a MicroVAXII with a KMV1A. Are you refering to such a card with
the "LSI-11 on the Vax"? Does this mean there is a V7 running on it?
I really don't know much about that card. Might have been used for I/O with
an IBM 3090 in this computer.
Matthias
--
When You Find Out Your Normal Daily Lifestyle Is Called Quarantine
4 Jul
4 Jul
3:01 a.m.
New subject: Thoughts on Licenses
On July 3, 2022 4:40:18 p.m. MDT, Clem Cole <clemc(a)ccc.com> wrote:
Sorry for some of us old guys vax = 780
Did you, like me, wonder why Trump wanted
exactly an original-VAX-model-number of
phony votes in Georgia?
Maybe someone told him that the internal
DEC name for VAX/VMS, at least during early
development, was Starlet, and he wanted to
grab it by the SYS$SPAWN.
12:33 p.m.
New subject: Thoughts on Licenses
On 7/3/22, Matthias Bruestle <m(a)mbsks.franken.de> wrote:
On Sun, Jul 03, 2022 at 03:48:33PM -0400, Clem Cole
wrote:
Clem is referring to the VAX-11/780 (and its follow-ons the 782 and
785), which used a LSI-11 processor (PDP-11/03) to load its microcode
on power-on. IIRC the LSI-11 front end also ran the system console
device (originally a LA-36 DECwriter terminal).
The third generation PDP-10 CPU (the KL10) similarly had a PDP-11/40
front end that did the bootstrap loading and also acted as a
peripheral device controller.
-Paul W.
single custom application* that allows us to
manage the server - as I
said
to Larry think the LS1-11 on the Vax and the 11/40 on the KL processors)
I have a MicroVAXII with a KMV1A. Are you refering to such a card with
the "LSI-11 on the Vax"? Does this mean there is a V7 running on it?
I really don't know much about that card. Might have been used for I/O with
an IBM 3090 in this computer.
672
days inactive
673
days old
11 comments
7 participants
participants (7)
-
Bakul Shah -
Clem Cole -
Kevin Bowling -
Larry McVoy -
Matthias Bruestle -
Norman Wilson -
Paul Winalski