delorie.com/archives/browse.cgi | search |
X-Authentication-Warning: | delorie.com: mail set sender to djgpp-workers-bounces using -f |
X-Recipient: | djgpp-workers AT delorie DOT com |
DKIM-Signature: | v=1; a=rsa-sha256; c=relaxed/relaxed; |
d=gmail.com; s=20120113; | |
h=mime-version:in-reply-to:references:date:message-id:subject:from:to | |
:content-type; | |
bh=RHFJ5VyUm2Hv/cqnQ8OmIW15T2SFZca4iJHpaJPzEw0=; | |
b=pwoaI2BasZP1zP5hFpf0llOR0RibS5CLgwf0HWCbFxuKLzvmOCf5gjnIvCRSoyTvpO | |
5r3/U/pqR8aTJgERxLDo9iZj1KRQHfeNUTMdejluf0jQ19jr/mlM34Dhs2jLfjGaov3b | |
Dy6kIUjkXiIZZ+NfO+WM75tYf/uuNm+PD4WpvZy7A+5zMB9slx4DRm9FDfEVd4IDz7Fo | |
Ee+Q1PxlaMdsRG/LxuJ6QRO3NgJ6NwvGKC10U0GfzInys3y+gqqKwoNzYjtpLycESnn1 | |
RsX1SgGuoLWvnZXOd8EtDTSESeESktC8ao31y1GCKNY+Ul5jmSmAlka4h5yRT41TJzLv | |
t+ag== | |
MIME-Version: | 1.0 |
In-Reply-To: | <CAA-ihx-+juCuiapWVXVbJK5gWA-5NOd-idVGTXH93DNtaGN=Xg@mail.gmail.com> |
References: | <CAA-ihx-+juCuiapWVXVbJK5gWA-5NOd-idVGTXH93DNtaGN=Xg AT mail DOT gmail DOT com> |
Date: | Sun, 2 Sep 2012 14:35:22 -0500 |
Message-ID: | <CAA-ihx9e5upk6kwNBoufOwBFweJ6-wTODyy8G9N-ftRgDfm_Yg@mail.gmail.com> |
Subject: | Re: memalign() -- backwards to Linux |
From: | Rugxulo <rugxulo AT gmail DOT com> |
To: | djgpp-workers AT delorie DOT com |
Reply-To: | djgpp-workers AT delorie DOT com |
Errors-To: | nobody AT delorie DOT com |
X-Mailing-List: | djgpp-workers AT delorie DOT com |
X-Unsubscribes-To: | listserv AT delorie DOT com |
Hi again, On Tue, Aug 28, 2012 at 6:13 PM, Rugxulo <rugxulo AT gmail DOT com> wrote: > > I hate to mention this, I know it's not that important ... > > I read a while back that DJGPP's memalign() is backwards to Linux > > void *memalign(size_t size, size_t alignment); // DJGPP > void *memalign(size_t alignment, size_t size); // Linux > > ANSI/ISO C No > POSIX No > > Even if you don't find this that big a deal, wouldn't it be better to > use #warning or similar on this (and other things like fork) so that > nobody gets the wrong idea??? Apparently the Linux version is in malloc.h while ours is in stdlib.h (probably a bad place, IMHO). But how would you (only) warn (at compile time) when including stdlib.h without being annoying to everybody? Well, I (weakly) propose moving it out of stdlib and putting it exclusively in malloc. (And presumably fixing it to match Linux, of course.) It seems 2.04's malloc.h is pure boilerplate, does nothing but include stdlib.h anyways, so it wouldn't hurt much there. http://www.delorie.com/bin/cvsweb.cgi/djgpp/include/malloc.h Then people would have to know to include malloc.h anyways. Actually, if you fix it to match Linux, it probably doesn't need a #warning anyways! (I know #warning isn't ANSI, but GCC does support it. I guess?? you could instead do #pragma warning , but I don't know offhand if that works with GCC.)
webmaster | delorie software privacy |
Copyright © 2019 by DJ Delorie | Updated Jul 2019 |