Date: Sun, 10 Dec 2000 19:01:48 +0200 From: Mark Murray <mark@grondar.za> To: Garrett Wollman <wollman@khavrinen.lcs.mit.edu> Cc: Jeroen Ruigrok van der Werven <asmodai@FreeBSD.org>, John Baldwin <jhb@FreeBSD.org>, cvs-all@FreeBSD.org, cvs-committers@FreeBSD.org Subject: Re: cvs commit: src/usr.bin/tail forward.c read.c reverse.c tail Message-ID: <200012101702.eBAH22I29827@gratis.grondar.za> In-Reply-To: <200012101643.LAA29818@khavrinen.lcs.mit.edu> ; from Garrett Wollman <wollman@khavrinen.lcs.mit.edu> "Sun, 10 Dec 2000 11:43:33 EST." References: <200012101643.LAA29818@khavrinen.lcs.mit.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
> Actually, the `register' keyword is *not* ignored by compilers, and > *does* serve a purpose (which may or may not turn out to be useful). > Variables with the `register' storage class may not have their > addresses taken, and the compiler is required to emit a diagnostic if > you try. The C compiler can not store a variable in a register across > a function call if its address is ever taken. (This is important for > SPARC processors, but less so on ia32 since there are only two > call-preserved registers.) Wrong. As of ANSI, "register" variables _may_ have their addresses taken, and the "register" keyword is officially a *hint*, not an instruction. M -- Mark Murray Join the anti-SPAM movement: http://www.cauce.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200012101702.eBAH22I29827>