Path: utzoo!attcan!uunet!tut.cis.ohio-state.edu!ucsd!usc!apple!altos!altos86!jerry
From: jerry@altos86.Altos.COM (Jerry Gardner)
Newsgroups: comp.arch
Subject: Re: Xenix address space (Was: Adam Smith's Hand)
Message-ID: <74@altos86.Altos.COM>
Date: 25 May 90 16:54:48 GMT
References: <4698@uceng.UC.EDU> <293@cvbnetPrime.COM> <_UG3ERF@xds13.ferranti.com> <60382@sgi.sgi.com> <444@dg.dg.com> <9AJ3FT5@xds13.ferranti.com> <1211.26559700@gp.govt.nz> <1990May19.230135.16025@utzoo.uucp> <1217.2656b909@gp.govt.nz>
Reply-To: jerry@altos86.UUCP (Jerry Gardner)
Organization: Altos Computer Systems, San Jose, CA
Lines: 29

In article <1217.2656b909@gp.govt.nz> don@gp.govt.nz (Don Stokes) writes:

>The Amiga and other small system (ie no memory management) OSs do similar
>things, but have the rather severe limitation that all code running on 
>them must be "well behaved".  This is generally fine for embedded systems 
>etc, but hopeless for any serious timesharing, especially if the 

>I suppose it comes back to the old question of "what is multi-user".  It 
>could be argued that raw Unix, without any real schemes for controlling 


A company that I worked for at one time, Qantel, sells large, multi-user
business computer systems based on an architecture that has no memory 
protection at all.  And since there is no clock interrupt, the operating
system is not preemptive--tasks have to execute a task break system call
to return control to the operating system.

These systems support, in some cases, over 64 users--but at a price. Any
program that has not been extensively tested has a good chance of bringing
the machine to a grinding halt, and since there is no memory protection,
security is virtually nonexistant.



-- 
Jerry Gardner, NJ6A					Altos Computer Systems
UUCP: {sun|pyramid|sco|amdahl|uunet}!altos86!jerry	2641 Orchard Parkway
Internet: jerry@altos.com				San Jose, CA  95134
Guns don't kill people, bullets do.                     (408) 432-6200