Here's one to stump you!
A Halliwell
Sinclair QL?
<What's a QL? | I hear a chorus of cries>
--
______________________________________________________________________________
| |What to do if you find yourself stuck in a crackin |
|u5...@cc.keele.ac.uk |the ground beneath a giant bolder, which you can't |
| |move, with no hope of rescue. |
|Andrew Halliwell |Consider how lucky you are that life has been good |
|Principal in:- |to you so far... |
|Comp Sci & Visual Arts | -The BOOK, Hitchhikers guide to the galaxy. |
------------------------------------------------------------------------------
David Carter
: Sinclair QL?
Sure it could be ported to the QL. If you could find one with a 68020 + MMU
or 68030 + 4MBytes RAM. Hmm.
A Halliwell
: : Is linux 'C' source code available so that it could be ported over to the
: : Sinclair QL?
: Sure it could be ported to the QL. If you could find one with a 68020 + MMU
: or 68030 + 4MBytes RAM. Hmm.
Damn. I can't Afford a QXL card and PC. (68030 + 8 Megs)
I can't even afford a Super Gold Card (68020 + 4 Megs)
Oh well, never mind......
Torsten Scherer
|> Is linux 'C' source code available
Dumb question, of course it is. :)
|> so that it could be ported over to the Sinclair QL? <What's a QL? | I hear
|> a chorus of cries>
Well, the fact that PC-Linux needs at least a 386 isn't included in the name,
and for us (the m68k community) the fact that Linux-m68k needs at least a
68020 + external PMMU isn't included in the name.
The QL I know has got a plain 68000 and you can therefore *never* run Linux
on it - and I've never heard of never versions or upgrades for that machine.
If there is at least a 68020 + external PMMU there's a chance - if YOU do all
the driver stuff...
ciao,
TeSche
Thomas Roehr
There is a Minux 68K port out that would run on a 68000 system. It is an
earlier Unix work-alike.
Torsten Scherer (itsc...@techfak.uni-bielefeld.de) wrote:
: u5...@cc.keele.ac.uk (A Halliwell) writes:
:
: |> Is linux 'C' source code available
:
: Dumb question, of course it is. :)
:
: |> so that it could be ported over to the Sinclair QL? <What's a QL? | I hear
: |> a chorus of cries>
:
: Well, the fact that PC-Linux needs at least a 386 isn't included in the name,
: and for us (the m68k community) the fact that Linux-m68k needs at least a
: 68020 + external PMMU isn't included in the name.
:
: The QL I know has got a plain 68000 and you can therefore *never* run Linux
Not true, one would have to write the mmu handling into the kernel. No one has
done that yet, apparently for Linux. There are a huge quantity of machines
that ran Unix on the 68000 and 68010 cpu. Actually, Unix got a real good
market period with the 68000 chip. It was the first generally available 32bit
CPU.
: on it - and I've never heard of never versions or upgrades for that machine.
: If there is at least a 68020 + external PMMU there's a chance - if YOU do all
: the driver stuff...
:
: ciao,
: TeSche
Tom Roehr
tro...@zei.com
EMACS TNA100:1
)Is linux 'C' source code available so that it could be ported over to the
)Sinclair QL?
)<What's a QL? | I hear a chorus of cries>
Aaah. Nostalgia! Sinclair! Saving on tapes. Mom and dad always missed
the headlines on the news. Got myself a TV on X-mas. Sinclair
Basic. :)
--
I love civilisation, | The number of the | You may not add me to a
it's people I can't stand. | beast is not 666. | commercial mailing list
_ _ ___ ___ _____ _ _ _ | The number is 95, | or send me commercial
|_`'_|/_-_\|_-_<`-_-'|_||_.`_| | and it is awake! | advertising without my
I came, I read, I unsubscribed. `-----------------' consent, dammit.
Michael Engel
: There is a Minux 68K port out that would run on a 68000 system. It is an
: earlier Unix work-alike.
(That's Minix, of course ... just nitpicking :)) You can find several
versions of Minix for 68k machines on ftp.cs.vu.nl in /pub/minix/
[mac,atari,amiga]
: Not true, one would have to write the mmu handling into the kernel. No one has
: done that yet, apparently for Linux. There are a huge quantity of machines
: that ran Unix on the 68000 and 68010 cpu. Actually, Unix got a real good
: market period with the 68000 chip. It was the first generally available 32bit
: CPU.
Oops, don't forget that there also existed a MMU for the 68000 and 68010.
Although I'm sure few people have ever heard of the M68451 (hmmm, even I
am no longer sure about the right number ... but that should be 68451) !
I have an ancient Unix system made by ProComp in Switzerland that has a
68000 CPU, 4 MB RAM and that 68451 MMU. Unfortunately, I didn't get a hard disk
or system media for this oldtimer :( It ran a port of 7th edition Unix as well
as some System III variant. Would surely be nice to see Linux running on
that machine ;) but I should better go and work on the MVME147 port ...
regards,
Michael Engel (en...@numerik.fb6.uni-siegen.de)
Thomas Roehr
: Thomas Roehr (tro...@ritz.mordor.com) wrote:
:
: : Not true, one would have to write the mmu handling into the kernel. No one has
: : done that yet, apparently for Linux. There are a huge quantity of machines
: : that ran Unix on the 68000 and 68010 cpu. Actually, Unix got a real good
: : market period with the 68000 chip. It was the first generally available 32bit
: : CPU.
:
: Oops, don't forget that there also existed a MMU for the 68000 and 68010.
: Although I'm sure few people have ever heard of the M68451 (hmmm, even I
: am no longer sure about the right number ... but that should be 68451) !
<deleted>
They did not have a mmu on the PDP7 that Unix was developed on, and Bell
Labs did a port to an 8086 Box. I cut my Unix teeth on a AT&T 7300 which
was a 68010 based machine. That machine ran SYS 5 R2 in 512K of Ram and
a 10 Meg Disk in minimal configuration. I also believe some people are
trying to port Linux over onto that machine. I don't remember any hardware
MMU on it, but I will check.
Later,
Tom Roehr
tro...@zei.com
Todd Fries
: u5...@cc.keele.ac.uk (A Halliwell) writes:
: Dumb question, of course it is. :)
latest is currently:
ftp://linux.ucs.indiana.edu/pub/linux/kernel/v1.3/linux-1.3.54.tar.gz
: |> so that it could be ported over to the Sinclair QL? <What's a QL? | I hear
: |> a chorus of cries>
: Well, the fact that PC-Linux needs at least a 386 isn't included in the name,
: and for us (the m68k community) the fact that Linux-m68k needs at least a
: 68020 + external PMMU isn't included in the name.
Don't count your chickens before they are hatched! Anybody who takes a gander
at ftp://tsx-11.mit.edu/pub/linux/ALPHA/linux-16 might deduce that a port is
underway to enable 286's and 8088's /8086's to run (imho) one of the most
challenging ports of Linux...
: The QL I know has got a plain 68000 and you can therefore *never* run Linux
: on it - and I've never heard of never versions or upgrades for that machine.
'never' never did anything. I might suggest it would be a very difficult task
but I don't think it would be totally out of the question..
...but who said I was an expert? This is all only my opinion!
--
<a href=mailto:tfr...@umr.edu>Todd T. Fries</a>
<a href=http://www.cs.umr.edu/~tfries/>My Homepage!</a>
Michael Engel
: Michael Engel (en...@numerik.fb6.uni-siegen.de) wrote:
: : Thomas Roehr (tro...@ritz.mordor.com) wrote:
: :
: : Although I'm sure few people have ever heard of the M68451 (hmmm, even I
: : am no longer sure about the right number ... but that should be 68451) !
: :
: <deleted>
: They did not have a mmu on the PDP7 that Unix was developed on, and Bell
: Labs did a port to an 8086 Box. I cut my Unix teeth on a AT&T 7300 which
: was a 68010 based machine. That machine ran SYS 5 R2 in 512K of Ram and
: a 10 Meg Disk in minimal configuration. I also believe some people are
: trying to port Linux over onto that machine. I don't remember any hardware
: MMU on it, but I will check.
Hmmm, I once saw a version of Xenix for the intel 8086. I still have my 80286
running Xenix here ! Not very useful, but it's good to see a kernel that
gives you quite a lot of functionality in about 120 kB ! :) Xenix/286 will
run fine with 640 kB RAM and a 20-40 MB hard disk.
There also is a Unix 7th Edition-like OS called UZI that runs on
a Z80 in 64kB RAM ... unfortunately, I don't own a CP/M machine any longer
to give it a try.
I seem to remember that the 7300 (was this the machine also known as the 3B2 ?)
had a second 68010 CPU to handle virtual memory requests ? I may be completely
wrong about that, but I thought I read that somewhere...
Btw, does anyone intend to port Linux over to the TI1500 ? That's a really
nice 680[234]0+NuBus based multiprocessor system with "intelligent" I/O
controllers (every controller board has at least a 10 MHz 68010 on it ...).
I wish I had such a machine, but these are hard to find here in Germany :(
regards,
Michael Engel (en...@numerik.fb6.uni-siegen.de)
Thomas J. Trebisky
>: : Oops, don't forget that there also existed a MMU for the 68000 and 68010.
>: : Although I'm sure few people have ever heard of the M68451 (hmmm, even I
>: : am no longer sure about the right number ... but that should be 68451) !
Sounds like the part all right. It is a strange mmu compared to what
people usually think of running unix on (altho System V was made to work on
a 68010/68451 combo in at least one case that I know of). The 68451 supports
a segmented memory model rather than the paged model that so many of us are
used to. It could perhaps be made to support virtual memory, but it would
be a significantly different scheme that the paged model that is all but
ubiquitous these days. It would run a non-VM unix like edition7 quite
nicely though.
>: I cut my Unix teeth on a AT&T 7300 which
>: was a 68010 based machine. That machine ran SYS 5 R2 in 512K of Ram and
>: a 10 Meg Disk in minimal configuration. I also believe some people are
>: trying to port Linux over onto that machine. I don't remember any hardware
>: MMU on it, but I will check.
>I seem to remember that the 7300 (was this the machine also known as the 3B2 ?)
>had a second 68010 CPU to handle virtual memory requests ? I may be completely
>wrong about that, but I thought I read that somewhere...
The 7300 is also known as the 3b1, the 3b2 is a 286 based machine I am pretty
sure. The 7300/3b1 I am quite sure about -- it is a 68010 with a custom mmu.
It supports 4M of physical memory max and has a 4M virtual address space.
0.5 meg of this VM space is dedicated to the kernel and the remaining 3.5M
is the per-process VM space. The System Vr2 port to the 3b1 stole an additional
0.5M of this for use by the kernel, so the per-process limit was restricted
by software to 3.0M
At one time I was contemplating a BSD port to the 3b1/7300 and have written
some code that would be useful for anyone wanting to work on this - in
particular a standalone HD formatter and standalone drivers for the serial
port, hard drive and floppy controllers.
Tom
--
Tom Trebisky Steward Observatory
ttre...@as.arizona.edu University of Arizona
(602) 621-5135 Tucson, Arizona 85721
Michael Engel
: At various times Michael Engel or Thomas Roehr wrote:
[... 68451 MMU ...]
: Sounds like the part all right. It is a strange mmu compared to what
: people usually think of running unix on (altho System V was made to work on
: a 68010/68451 combo in at least one case that I know of). The 68451 supports
: a segmented memory model rather than the paged model that so many of us are
: used to. It could perhaps be made to support virtual memory, but it would
: be a significantly different scheme that the paged model that is all but
: ubiquitous these days. It would run a non-VM unix like edition7 quite
: nicely though.
Do you have any data sheets on the 68451 ? I wasn't able to get some from
Motorola (no wonder ...). Just curious ...
[ ... AT&T 7300 ... ]
: >I seem to remember that the 7300 (was this the machine also known as the 3B2
: >had a second 68010 CPU to handle virtual memory requests? I may be completely
: >wrong about that, but I thought I read that somewhere...
: The 7300 is also known as the 3b1, the 3b2 is a 286 based machine I am pretty
: sure. The 7300/3b1 I am quite sure about -- it is a 68010 with a custom mmu.
: It supports 4M of physical memory max and has a 4M virtual address space.
: 0.5 meg of this VM space is dedicated to the kernel and the remaining 3.5M
: is the per-process VM space.The System Vr2 port to the 3b1 stole an additional
: 0.5M of this for use by the kernel, so the per-process limit was restricted
: by software to 3.0M
Ok, I seem to have too much net.rumour confused :) My memories about old
machines get a little fuzzy since they don't feed alt.folklore.computers
here any more :( (our whole University has _one_ 128 kBit internet conection,
so I understand why they abandoned the alt.* news hierarchy ...)
: At one time I was contemplating a BSD port to the 3b1/7300 and have written
: some code that would be useful for anyone wanting to work on this - in
: particular a standalone HD formatter and standalone drivers for the serial
: port, hard drive and floppy controllers.
Well, I don't have a 3B1 :( I don't think these machines ever appeared here in
Germany ... But if you find someone, I'd like to offer you space on our ftp
server to put your code on. Seems we're getting an ftp server for Long Forgotten
Machines (TM)... TI1500, UZI on Z280, NetBSD on uVAX .............
Michael Engel (en...@numerik.fb6.uni-siegen.de)
Ross Alexander
> The 7300 is also known as the 3b1, the 3b2 is a 286 based machine I
> am pretty sure. The 7300/3b1 I am quite sure about -- it is a 68010
> with a custom mmu.
The 3b2 series machines are built around variants of the WE32000. For
instance, a 3b2/1000-70 uses the WE32200 chipset (CPU, MAU, and FPU).
regards,
Ross
--
Ross Alexander, ve6pdq -- (403) 675 6311 -- r...@cs.athabascau.ca
Thomas J. Trebisky
>Motorola (no wonder ...). Just curious ...
Yes, I do. Holding it right here. Amusingly, it is labeled
"Advance Information" and is dated April, 1983. How time does fly.
If you want me to look things up in it for you, let me know.
I have a stack of VME10 motherboards (this was a desktop machine made
by/for Motorola that had a 68010 and one or more 68451), and did run
System V rev 2, thus my interest in this chip.
>[ ... AT&T 7300 ... ]
>Ok, I seem to have too much net.rumour confused :)
The 7300 was the first working unix box at my own home and finds a place
in my heart. Many folks laught at it, but it is always a reference point
to me that System Vr2 would run on a hardware configuration with 0.5M ram
and a 20M hard disk. Sure it might not run well (and you wouldn't have
online man pages), but it actually works, and with virtual memory too
(and you would need it).
>: At one time I was contemplating a BSD port to the 3b1/7300 and have written
>: some code that would be useful for anyone wanting to work on this - in
>: particular a standalone HD formatter and standalone drivers for the serial
>: port, hard drive and floppy controllers.
>
>Well, I don't have a 3B1 :(
>But if you find someone, I'd like to offer you space on our ftp
>server to put your code on.
>Seems we're getting an ftp server for Long Forgotten
>Machines (TM)... TI1500, UZI on Z280, NetBSD on uVAX .............
Cool. I don't think any of my code has made it outside of my own little
world. A lot of what I had done is actually for the Miniframe
(aka Motorola 6300, NOT the ATT 6300 which is a x86 box). The miniframe
is a sister machine to the 7300, same mmu and much the same i/o structure.
The 7300 has a memory mapped video. The miniframe does not, but has a
sort of network port (2Mb/sec SDLC/HDLC) and a second hard drive.
Great fun to write code for these old machines!
Bear
[snip]
>: Not true, one would have to write the mmu handling into the kernel. No one has
>: done that yet, apparently for Linux. There are a huge quantity of machines
>: that ran Unix on the 68000 and 68010 cpu. Actually, Unix got a real good
>: market period with the 68000 chip. It was the first generally available 32bit
>: CPU.
>
>Although I'm sure few people have ever heard of the M68451 (hmmm, even I
>am no longer sure about the right number ... but that should be 68451) !
>
>68000 CPU, 4 MB RAM and that 68451 MMU. Unfortunately, I didn't get a hard disk
>or system media for this oldtimer :( It ran a port of 7th edition Unix as well
>as some System III variant. Would surely be nice to see Linux running on
>that machine ;) but I should better go and work on the MVME147 port ...
I recall reading in an Amiga magazine, some time ago, that in the early
days of the Amiga, Commodore had their own version of Unix available.
IIRC, it died 'because the standard Amiga OS was so good anyway', or
words to that effect.
Is my memory going or is there any truth to this?
--
Just my two sestertii worth... :-)
Phil Pennock ; be...@dcs.warwick.ac.uk ; GCS d- H+ s+:+ g-(+) p6+ !au0 a20 w+++
v+@ C++(++++) UL++++/S++ P++@ L++ 3 E+(++) W(+) N++ !K w--- O+ M V-- !PS PE+ Y+
PGP- t- 5++ X+ R G' tv- b++>+++ DI(+) D+ G++ e+ u+ h++ f !r n+(-)@ !y+
Torsten Scherer
|> Don't count your chickens before they are hatched! Anybody who takes a
|> gander at ftp://tsx-11.mit.edu/pub/linux/ALPHA/linux-16 might deduce that
|> a port is underway to enable 286's and 8088's /8086's to run (imho) one of
|> the most challenging ports of Linux...
IMHO one of the best suggestions how to waste time - taken this for true,
which I'm not even interested to check.
|> 'never' never did anything. I might suggest it would be a very difficult
|> task but I don't think it would be totally out of the question..
Well, if you want it this way: All of them are turing-machines, so Linux
could in principle be run on *every* CPU ever existed (theorem). ;)
|> ...but who said I was an expert? This is all only my opinion!
So is mine, but I'm very suspicious about those rumors. Fact is all this
stuff doesn't exist *yet*, so my original answer will stay right for a couple
of time.
ciao,
TeSche