Re: Hardware and 8MHz 68000 machines

[Stephen Usher <Stephen.Usher@earth.ox.ac.uk>]

>> I admit this would break programs which were not
>> virtual-memory aware which go straight for the hardware.
>
>Which in some cases is the only option (STE sound hardware for example)!

Which is why I would suggest that TOS-domain processes should be assumed to
be non-relocatable, provided with only contiguous memory and live in
physical address space. Of course, all the normal hardware would appear in
its physical address space whatever.

>> The main problem I see is catering for the ST{F,FM,M} series machines, those
>> with mere 68000's. The only way I can see a non-blocking fork being
>> implemented is via shadowing, as used in Minix.. and does that slow the
>> machine down or what?
>
>Sorry, but if we go on wanting to cater for 68000 machines we're never going
>to get something which matches the standards of today.  I mean think about
>it.

Hmm... but the majority of Atari users are still running 68000 based
machines. This means that we MUST cater for them if at all possible. Hence,
we should allow a non-blocking fork on the 68000 machines too.. this doesn't
mean that it will have to be done in the same way on 68030 based machines.
These machines could easily have swapping and demand paging added once
virtual addressing is implemented.

>
>regards,
>-- 
>Annius V. Groenink | E-mail: avg@cwi.nl      |  Private & ZFC:
>CWI, Kruislaan 413 | Office: M233            |  P.O. Box 799
>1098 SJ Amsterdam  | Ext:    4077            |  NL 3500 AT Utrecht
>Netherland         | Phone:  +31 20 592 4077 |  Phone: +31 30 803740

Steve

-- 
---------------------------------------------------------------------------
Computer Systems Administrator, Dept. of Earth Sciences, Oxford University.
E-Mail: steve@uk.ac.ox.earth (JANET) steve@earth.ox.ac.uk (Internet).
Tel:- Oxford (0865) 282110 (UK) or +44 865 282110 (International).