Path: csiph.com!eternal-september.org!feeder.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: Ben Bacarisse <ben.usenet@bsb.me.uk>
Newsgroups: comp.lang.c
Subject: Re: A glitch regarding the usual malloc() idiom
Date: Thu, 28 Jul 2016 14:51:21 +0100
Organization: A noiseless patient Spider
Lines: 40
Message-ID: <87zip1yip2.fsf@bsb.me.uk>
References: <lnr3apmfbl.fsf@kst-u.example.com> <f21ce6b1-3be2-422e-8511-c84d5b13e757@googlegroups.com> <5793c8e5.2941812@news.xs4all.nl> <b80e8423-7bb0-432b-894a-8cd6bb038ea6@googlegroups.com> <nn0p0u$jgm$1@jstuckle.eternal-september.org> <87oa5n97rz.fsf@bsb.me.uk> <nn2lht$lsh$1@jstuckle.eternal-september.org> <87wpkar54r.fsf@bsb.me.uk> <nn3odc$6ca$5@jstuckle.eternal-september.org> <87d1m1r99c.fsf@bsb.me.uk> <nn5f61$310$1@jstuckle.eternal-september.org> <871t2hqn06.fsf@bsb.me.uk> <nn834o$elo$1@jstuckle.eternal-september.org> <87fuqwp5fc.fsf@bsb.me.uk> <nn8dt4$o57$1@jstuckle.eternal-september.org> <2c152373-28b9-49dc-b221-d24dcb9ee08c@googlegroups.com> <nna1vh$9gb$1@dont-email.me> <nna9ot$4ji$1@dont-email.me> <nnagks$u8l$1@dont-email.me> <8760rrf6q8.fsf@bsb.me.uk> <nnbd27$c7d$1@dont-email.me>
Mime-Version: 1.0
Content-Type: text/plain
Injection-Info: mx02.eternal-september.org; posting-host="017616aa25f81ec581c44d76d61ba2f3"; logging-data="30829"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1/BwjINWPTTTyyPxc1JYbSwacGQcQ8KH5Q="
Cancel-Lock: sha1:LODD5zZs3TWfkqVZwK9V19TQcgU= sha1:3Fz7XtXFcBvvsMt1Vy+p20DjnIc=
X-BSB-Auth: 1.14e4a68620d01ddbf510.20160728145121BST.87zip1yip2.fsf@bsb.me.uk
Xref: csiph.com comp.lang.c:91231

BartC <bc@freeuk.com> writes:

> On 27/07/2016 16:20, Ben Bacarisse wrote:
>> David Brown <david.brown@hesbynett.no> writes:
>>
>>> On 27/07/16 14:34, Richard Heathfield wrote:
>>>> On 27/07/16 11:21, BartC wrote:
>>>
>>>>
>>>>> (BTW why is it only C that is obsessed with padding bits? They don't
>>>>> seem to crop up with any other language. Not even assembly.)
>>>>
>>>
>>> Assembly can have padding bits too - it's just assembly on the
>>> processors /you/ are familiar with that don't have them.  To be fair,
>>> few modern cpus need any sort of padding unless you are trying to use
>>> bitfields.  But they do turn up in odd processor architectures,
>>> including some mainframes or older machines.  So C allows the
>>> possibility of padding bits for much the same reasons it allows signed
>>> integers in formats other than two's complement.
>>
>> There's another more direct use of padding bits: _Bool.  In most C
>> implementations _Bool has one value bit and 7 padding bits.
>
> Are the really padding bits, or does it just restrict the value that
> can be stored in an 8-bit type to 0 or 1 instead of 0 to 255?

The bits in an integer type's representation are either a sign bit,
values bits or padding bits.

> Because I could easily store data in those extra 7 bits in all six C
> compilers I tried (by casting _Bool* to char*).

Yes, that's common.  Exactly what happens then is then up to the
implementation, but it's got carte blanche since almost everything to do
with padding bits is undefined.

<snip>
-- 
Ben.