Path: utzoo!attcan!uunet!mcvax!hp4nl!targon!andre
From: andre@targon.UUCP (andre)
Newsgroups: comp.lang.c
Subject: Re: Another silly question
Message-ID: <549@targon.UUCP>
Date: 22 May 89 15:47:13 GMT
References: <17812@cup.portal.com> <2336@Portia.Stanford.EDU> <25671@amdcad.AMD.COM> <1176@mcrware.UUCP> <18560@cup.portal.com>
Reply-To: andre@targon.UUCP (andre)
Organization: Nixdorf Computer BV., OSP, P.O. Box 29,Vianen, The Netherlands
Lines: 23

In article <18560@cup.portal.com> Tim_CDC_Roberts@cup.portal.com writes:
>In <1176@mcrware.UUCP>, jejones@mcrware.UUCP (James Jones) writes:
>[pointer story]
>I disagree with this!  I assert that EVEN if the intermediate result
>goes negative, the final value will be correct, even on segmented 
>architectures.
Don't underestimate the intel approach to computing :-)
I have it on good authority that on the 386,
((adress) 0x0010 - 0x0100) + 0x0100 != 0x0010
but instead it winds up somewhere at the top of memory :-(.

>Yes, the intermediate value is not a valid address, but I don't think that's 
>important.
If the intermediate result would be put in an address register (on the '386)
(where else does an address even a bogus one belong else ?)
you will get a trap from the processors 'MMU'.


-- 
~----~ |m    AAA         DDDD  It's not the kill, but the thrill of the chase.
~|d1|~@--   AA AAvv   vvDD  DD        Segment registers are for worms.
~----~  &  AAAAAAAvv vvDD  DD
~~~~~~ -- AAA   AAAvvvDDDDDD        Andre van Dalen, uunet!mcvax!targon!andre