Path: csiph.com!eternal-september.org!feeder.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: Keith Thompson <kst-u@mib.org>
Newsgroups: comp.lang.c
Subject: Re: usecases of void foo(char **b) {}
Date: Tue, 08 Nov 2016 08:06:40 -0800
Organization: None to speak of
Lines: 21
Message-ID: <ln1symrm5r.fsf@kst-u.example.com>
References: <f04d8120-e4c9-464b-923c-7dc262da5b5b@googlegroups.com> <877f8hgb0p.fsf@bsb.me.uk> <nvl38b$lpf$1@dont-email.me> <87eg2pedkx.fsf@bsb.me.uk> <nvlrks$e8i$1@dont-email.me> <87wpghcu23.fsf@bsb.me.uk> <nvm3i5$4tu$1@dont-email.me> <877f8gx44n.fsf@bsb.me.uk> <nvn53j$hig$1@dont-email.me> <nvn908$tu2$1@dont-email.me> <nvne52$fri$1@dont-email.me> <nvnfs9$m6u$1@dont-email.me> <lnh97ktjns.fsf@kst-u.example.com> <878tswb89s.fsf@bsb.me.uk> <lnshr4s1or.fsf@kst-u.example.com> <87wpgg9l8k.fsf@bsb.me.uk> <lnfun4rqax.fsf@kst-u.example.com> <87a8dba7ab.fsf@bsb.me.uk> <1a898248-9e83-4ec5-960a-23fa13461a06@googlegroups.com> <nvqtim$r67$1@dont-email.me> <305bbfee-a379-4eef-a581-175f983bb589@googlegroups.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Injection-Info: mx02.eternal-september.org; posting-host="e43f89eff67778d2e2548cd041b513fe"; logging-data="15168"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX19DpolAjHH4u5ojQMrDRFpM"
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.1 (gnu/linux)
Cancel-Lock: sha1:oZ/Rk6z+V8SWC7RTuGk2J+2+Ny8= sha1:mstCsIDXMw3v+iUuFkaOm64VFCg=
Xref: csiph.com comp.lang.c:96093

supercat@casperkitty.com writes:
[...]
> The Standard would allow for arrays with an even number of elements to
> be placed in structures using an alignment coarser than the element type,
> and on some hardware platforms (especially those which do not have octet-
> addressable memory but emulate an octet "char" type) there may be benefits
> to doing so.

The standard allows for any struct member other than the first to be
placed at any alignment the compiler chooses, as long as it satisfies
the alignment requirements of the member's type.  In
    struct { char c0; char c1; };
the compiler could legally place c1 at offset 137.

[snip]

-- 
Keith Thompson (The_Other_Keith) kst-u@mib.org  <http://www.ghoti.net/~kst>
Working, but not speaking, for JetHead Development, Inc.
"We must do something.  This is something.  Therefore, we must do this."
    -- Antony Jay and Jonathan Lynn, "Yes Minister"