Date: Thu, 09 Jan 1997 10:24:33 -0700 From: Warner Losh <imp@village.org> To: Chuck Robey <chuckr@glue.umd.edu> Cc: ports@freebsd.org Subject: Re: Niklas Hallqvist: archivers/hpack.non-usa.only Message-ID: <E0viODS-0005Ep-00@rover.village.org> In-Reply-To: Your message of "Thu, 09 Jan 1997 12:10:55 EST." <Pine.OSF.3.95.970109120824.7826A-100000@packet.eng.umd.edu> References: <Pine.OSF.3.95.970109120824.7826A-100000@packet.eng.umd.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
In message <Pine.OSF.3.95.970109120824.7826A-100000@packet.eng.umd.edu> Chuck Robey writes: : I don't know ... does sys/param.h exist on _every_ last system that has : unix or __unix__ defined? If not, the code above is a fatal error, and no : good. If it's guaranteed true, it's fine, and I'd use it. It does seem to exist on almost every unix system today. I'm in the process of looking for examples that aren't true. I'd say that at least 99.999% of all systems that define unix or __unix__ have sys/param.h based on my porting experiences. The Lions book, for example, has a file called unix/param.h, which matches closely the stuff found in sys/param.h, kinda sorta. So there is precidence for this sort of stuff going back a long ways (to V6, for those that don't know the Lions book). : Can you make that guarantee? If not, did you see the thing about using : .if exists ? Nearly. And it would be easy to add exception code for those few systems that it doesn't exist on, if any. Almost nobody else in the unix world has bmake on their system, so that solution is much less portible by quite a bit. Warner
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?E0viODS-0005Ep-00>