Message-ID: <3808CDC7.210BC969@simul8.demon.co.uk>
Date: Sat, 16 Oct 1999 20:11:03 +0100
From: James Lothian <simul8@simul8.demon.co.uk>
MIME-Version: 1.0
To: pups@minnie.cs.adfa.edu.au
Subject: Re: vtserver
References: <199910141836.LAA21478@moe.2bsd.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-pups@minnie.cs.adfa.edu.au
Precedence: bulk

Hmm.. I seem to remember, from when I was thinking about rolling my own 
11 OS a few years ago, that the /34 differs from most of the other
mid-range
11s in automatically restoring the CPU registers on a page fault. I
think I
picked this up from the differences table in the /04, /34 & /60 CPU
handbook.
(This was unfortunately no use at all to me, as I've got a /40 not a
/34.) 

Of course, that was a while ago and I might be wrong.

James

"Steven M. Schultz" wrote:
> 
> Hi -
> 
> > From: Wilko Bulte <wilko@yedi.iaf.nl>
> 
>         I will be doing some more research on this when I get home from
>         work tonight.
> 
> > I once had Ultrix-11 3.1 running on a dual RK05 11/34. What I'd call a
> > very minimal system ;-) But it ran
> 
>         That is because DEC put the extra effort into supporting non-split I/D
>         machines.  The "stock" V7 really wanted a 11/70.  In fact there was a
>         chapter in the back of one of the manuals/books detailing what it took
>         to get V7 running on an 11/40 (it was a non-trivial project).
> 
>         Several things conspire against V7 and later on 11/34 (or 35, 40, 60,
>         etc).  The two most notable ones are the limited address space,
>         everything (drivers, data structures, general kernel code) must fit
>         in 56kb instead of 120kb - (8kb reserved for the I/O page) and lack
>         of instruction restart on MMU faults.
> 
>         I'll take a look at the V7 layout later but my memory is that it
>         wanted an 11/70.
> 
>         Steven Schultz