NetNews Offline 2/NetNews Offline Volume 2.iso/news/comp/lang/c++-part1/7648

Internet Message Format | 1996-08-05 | 2.6 KB

Path: mail2news.demon.co.uk!genesis.demon.co.uk
From: Lawrence Kirby <[e-mail]>
Newsgroups: comp.arch.arithmetic,comp.lang.c,comp.lang.c++
Subject: Re: Access carry flag from C
Date: Sat, 24 Feb 96 17:37:59 GMT
Organization: none
Message-ID: <[e-mail]>
References: <[e-mail]> <[e-mail]> <[e-mail]> <danpop.824859220@rscernix> <[e-mail]> <[e-mail]> <[e-mail]>
Reply-To: [e-mail]
X-NNTP-Posting-Host: genesis.demon.co.uk
X-Newsreader: Demon Internet Simple News v1.27
X-Mail2News-Path: genesis.demon.co.uk

In article <[e-mail]> [e-mail] "Franz Korntner" writes:

>Lawrence Kirby wrote:
>
>> No. The value defined for INT_MAX in limits.h is the maximum value an int
>> can hold *on that particular implementation*. The standard requires that
>> INT_MAX be at least 32767 on all implementations (but it can be and
>> often is higher). So on any implementation:
>>
>>   int x = INT_MAX+1;
>>
>> is a guaranteed overflow.
>
>Wrong. The limits are minima. The above operation does not have to overflow!

5.2.4.2.1

"- maximum value for an object of type int
   INT_MAX                              +32767"

INT_MAX is a maximum (as its name suggests). +32767 is a minimum and is the
smallest value an implementation is allowed to define INT_MAX to be. Hence
int's on any implementation are resuires to be able to represent integers
up to at least 32767. See Tanmoy's post for a more detailed (and correct)
explanation.

>> >and what I really miss in the
>> >standard and/or implementation is a inline function 'maxlimitof(int)',
>> >analogical to 'sizeof(int)'.
>>
>> That is precisely what INT_MAX gives you.
>
>INT_MAX is a constant. On some platforms you can force GCC to assume int's
>to be 16 bit values. What about INT_MAX then?!?

To remain conforming gcc would have to ensure that INT_MAX is defined to
a value consistent with how an integer is represented for that particular
compilation. It can do that easily by using conditional preprocessing within
limits.h. Whether it does that in practice is another matter - as soon as
you start playing around with type representations you will hit trouble
if every translation unit doesn't use the same one, notably the standard
library. Most likely you simply don't get a conforming compiler if you
try to do this, notably where gcc is used with the standard library supplied
by the system.

--
-----------------------------------------
Lawrence Kirby | [e-mail]
Wilts, England | [e-mail]
-----------------------------------------