NetNews Offline 2/NetNews Offline Volume 2.iso/news/comp/lang/c-part1/3489

Internet Message Format | 1996-08-05 | 1.9 KB

Path: Rezonet.net!news
From: [e-mail] (Ray Dunn)
Newsgroups: comp.lang.c
Subject: Re: quick decision: is n a power of 2?
Date: 29 Jan 1996 14:03:14 GMT
Organization: Ultimate Technographics Inc.
Message-ID: <4eik32$[e-mail]>
References: <[e-mail]> <[e-mail]> <4e6rs0$[e-mail]> <[e-mail]> <[e-mail]>
NNTP-Posting-Host: 204.19.230.7
Mime-Version: 1.0
Content-Type: Text/Plain; charset=US-ASCII
X-Newsreader: WinVN 0.99.7

In referenced article, Lawrence Kirby says...
>Two's complement is a representation for signed integers so
>unsigned arithmetic clearly doesn't use it. However the nice thing
>about 2's complement arithmetic is that in most instances the
>underlying bit manipulations are (or can be for a suitable definition
>of overflow behaviour) the same as those for unsigned arithmetic (so
>at the archetecture level a single form of, e.g. an ADD instruction,
>can deal with both).
>
>So the expression (-5) and (-5U) are each evaluated by its own set of
>rules but where signed integers are represented in 2's complement the
>2 results will have the same underlying bit pattern.

So what would you call this operation that is being done on the
unsigned int?

I really think you're nit-picking unnecessarily here, so let me
nit-pick back... (:-)

Dan Pop's assertion that unary minus does a two's complement negate on
unsigned types in 'C' was a perfectly reasonable statement.

A two's complement negate is a bitwise operation on a set of bits, just
like a one's complement negate is.  The signedness of that set of bits
is irrelvant to the operation, like any other, it is only the
subsequent interpretation of the sign bit that matters.
--
Ray Dunn (opinions are my own) | Phone: [phone]
Montreal                       | Phax : [phone]
[e-mail]          | Home : [phone]