Date: Thu, 06 Aug 1998 20:49:05 -0700 From: John Polstra <jdp@polstra.com> To: Terry Lambert <tlambert@primenet.com> Cc: Nicolas.Souchu@prism.uvsq.fr, hackers@FreeBSD.ORG Subject: Re: C and static initialization with unions Message-ID: <199808070349.UAA29774@austin.polstra.com> In-Reply-To: Your message of "Fri, 07 Aug 1998 01:36:31 -0000." <199808070136.SAA24978@usr06.primenet.com>
next in thread | previous in thread | raw e-mail | index | archive | help
> > GCC extensions are occasionally used in FreeBSD. However, they > > are virtually always optional and for the purpose of (a) better > > efficiency, or (b) better compiler warnings. When GCC extensions are > > used, they are protected by "#ifdef __GNUC__", and an alternative > > portable implementation is provided in the "#else" clause. For case > > (a) the alternative implementation is less efficient; for case (b), > > it's a no-op. See <sys/cdefs.h> for some examples of (b). > > This is not quite correct. The <sys/cdefs.h> defined __P(), for > example, and this appears to have been deprecated despite its > ability to ensure portability; which is really bizarre, because > "const", "volatile", and __CONCAT() all haven't, and it the same > portability barrier that all 4 address (non-ANSI K&R compilers). I was talking about GCC extensions (extra-ANSI), and you are talking about K&R vs. ANSI. Those are different topics. -- John Polstra jdp@polstra.com John D. Polstra & Co., Inc. Seattle, Washington USA "Self-knowledge is always bad news." -- John Barth To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199808070349.UAA29774>