Path: utzoo!bnr-vpa!bnr-fos!bnr-public!schow
From: schow@bnr-public.uucp (Stanley Chow)
Newsgroups: comp.sys.amiga
Subject: Re: Benchmarks (was: Re: A2620)
Message-ID: <612@bnr-fos.UUCP>
Date: 16 Jun 89 19:07:29 GMT
References: <600@bnr-fos.UUCP> <14519@watdragon.waterloo.edu>
Sender: news@bnr-fos.UUCP
Reply-To: schow%BNR.CA.bitnet@relay.cs.net (Stanley Chow)
Distribution: na
Organization: Bell-Northern Research, Ottawa, Canada
Lines: 40
Summary:
Followup-To:
Keywords:

In article <14519@watdragon.waterloo.edu> sjorr@rose.waterloo.edu (Stephen Orr) writes:
>In article <600@bnr-fos.UUCP> schow%BNR.CA.bitnet@relay.cs.net (Stanley Chow) writes:
>>I had always thought the same thing. Then I started to work with the latest
>>processor chips from both companies. I think Intel has learnt to make their
>>chips more friendly to programmers (at least to OS types). It is not clear
>>to me if Motorola cares about friendliness to either EE or programmers.
>
>    Just in case your not looking (and I know this is a detail) I'd much rather
>program a machine like the 030 when writing an OS, it has linked list assembly
>commands, (not Macros!). Now what else could you do to make programing an OS
>friendly?
>

Hmm, the only instructions on the 68K family that can be remotely associated with
linked list handling are the CAS & CAS2 instructions. You are telling me these
are good? Have you tried to write a two way linked list insertion routine? In a
multi-tasking environment? In a multi-processor environment? I will bet you dollars
to doughnuts that I can trap to supervisor level, inhibit interrupts, do the
fiddling and return before any user level instruction seqeuence is done.

As to your question of what other things make OS programming easier: most of OS
programming problems are dealing with interrupts, providing process communications,
trap/guru handling, debuging tools, ... Linked list handling is not often thought
of as a major OS problem. Cache flushing and address mapping with process swapping
and memory protection usually do present problems.


>    By the way, I'm both a programmer and an EE, and I've worked with both, and
>while I don't think I'll ever program a Mac, I'll take a 68000 series over any
>Intel product I've used so far...
>

Perhaps you have not used all the Intel products. And perhaps you have not done
a real OS for multi-processors with the latest chips.


Stanley Chow        BitNet:  schow@BNR.CA
BNR		    UUCP:    ..!psuvax1!BNR.CA.bitnet!schow
(613) 763-2831		     ..!utgpu!bnr-vpa!bnr-fos!schow%bnr-public
Me? Represent other people? Don't make them laugh so hard.