Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxt!houxm!whuxl!whuxlm!akgua!gatech!ut-sally!topaz!harvard!h-sc1!breuel
From: breuel@h-sc1.UUCP
Newsgroups: net.micro.amiga
Subject: Re: What's BCPL?
Message-ID: <907@h-sc1.UUCP>
Date: Tue, 4-Feb-86 07:48:59 EST
Article-I.D.: h-sc1.907
Posted: Tue Feb  4 07:48:59 1986
Date-Received: Thu, 6-Feb-86 20:37:46 EST
References: <410@wlbr.UUCP> <148600045@uiucuxc>
Organization: Harvard Univ. Science Center
Lines: 19

> now, can somebody explain to me the utility of these #$%^@ shifted pointers?
> i can see how they would serve to enforce boundary alignment of certain
> structures in memory, but does this justify the contortions required?
> or is it some kind of vestige of pdp11's with 18-bit(?) physical addresses
> and 16-bit words?

The reason is very simple: since the language does not have data types,
this is the only way that pointer arithmetic can work in a reasonably
portable manner (i.e. without having to clutter your code with constants
or special functions to do arithmetic). Furthermore, this happens to
work extremely well on architectures on which the smallest addressable
unit of memory is the word.

						Thomas.

PS: needless to say, it has nothing whatsoever to do with PDP's in
particular. You might, at most, say that it has something to do with
DEC 10's, since they are in some sense word addressable, but
whether this had any influence on the development of BCPL, I don't know.