From owner-freebsd-questions Thu Jan 13 12:37:53 2000 Delivered-To: freebsd-questions@freebsd.org Received: from speedbuggy.telerama.com (speedbuggy.telerama.com [205.201.1.216]) by hub.freebsd.org (Postfix) with SMTP id 39C1015182 for ; Thu, 13 Jan 2000 12:37:48 -0800 (PST) (envelope-from evs@telerama.com) Received: (qmail 11098 invoked by uid 65534); 13 Jan 2000 20:37:43 -0000 Message-ID: <20000113203743.11097.qmail@speedbuggy.telerama.com> Date: 13 Jan 2000 15:37:43 -0500 From: evs@telerama.com To: keramida@ceid.upatras.gr, Giorgos Keramidas Reply-To: evs@telerama.com Cc: Mikhail Evstiounin , freebsd-questions@freebsd.org References: <00a201bf5d70$fb0cbea0$fc353018@evstiouninadelphia.net.pit.adelphia.net> <20000113143619.A2590@hades.hell.gr> In-Reply-To: <20000113143619.A2590@hades.hell.gr> MIME-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit X-Mailer: IMP/PHP3 Imap webMail Program 2.0.9 X-Originating-IP: 141.202.248.55 Subject: Re: Giving a sighandler more information Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Quoting Giorgos Keramidas : > On Wed, Jan 12, 2000 at 09:50:41PM -0500, Mikhail Evstiounin wrote: > > > > -----Original Message----- > > From: Oliver Fromme > > To: freebsd-questions@FreeBSD.ORG > > Date: Wednesday, January 12, 2000 9:37 PM > > Subject: Re: Giving a sighandler more information > > > > > > In fact in most Unix incarnations that I've worked with so far, the > size of `int' is the same as the size of `long'. However, `long' has > the size of the longest value that can be stored in a word on a given > architecture, and this is usually just the right size for storing > pointers ;) Yeap, 100% agreed, I installed gcc on all platforms I worked, excep those that have GCC as a nativre compiler (FreeBSD, Linux, OpenBSD, NetBSD) - HP-UX 9.x, 10.x, 11.x; IBM AIX 3.xx, 4.xx, Sequent Dynix/ptx - numerous versions. But we always implemented int as a longest value that could be stored in a word and long usually made longer than int (not always) - reason beind it if somebody wants to use long - compiler should give thos possibility to user, even long can work a little bit slower. > > > Sorry, guys, but it requirenments of ANSI that sizeof(void*) == > > sizeof(int). Don't get me wrong, I like Alpha, but alpha compiler is > > not 100% ANSI compatible. If you take a look at HP aCC and 64 bit > > architecture - you will see the same. So, let me rephrase it - in > > ANSI world you could do it. and question does it mean that alpha > > integer is 32 bit, or pointer has more than 64? I used GNU C a couple > > of times there and everything was OK. Is it a special compiler? > > GNU cc is a different compiler than those usually shipped from the so > called `vendors'. > Yeap, and gcc team never claimed like many vendors, that gcc is ANSI copatible. > It is not 100% ANSI compliant either, but those few features that are > not ANSI-safe can be turned off with an option. To qquote the GCC > manuals, one can use -ansi to "turn off certain features of GNU C that > are incompatible with ANSI C, such as `asm', `inline' and `typeof' > keywords". > > For even stricter ANSI support, the use of -pedantic will "cause > non-ANSI programs to be rejected gratuitously". > > -- > Giorgos Keramidas, < keramida @ ceid . upatras . gr > > "What we have to learn to do, we learn by doing." [Aristotle] > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-questions" in the body of the message > Sincerely yours Mikhail Evstiounin To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message