Date: Thu, 23 Feb 95 9:22:17 MST From: terry@cs.weber.edu (Terry Lambert) To: nate@trout.sri.MT.net (Nate Williams) Cc: phk@ref.tfs.com, wollman@halloran-eldar.lcs.mit.edu, current@freefall.cdrom.com Subject: Re: TRUE and FALSE Message-ID: <9502231622.AA02944@cs.weber.edu> In-Reply-To: <199502222345.QAA15987@trout.sri.MT.net> from "Nate Williams" at Feb 22, 95 04:45:52 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> At what gain are we doing this? I believe it's a noble gain to have the > source tree compile w/out reference to /usr/include, but what does it > gain us? The only thing I can see where it's a big deal is building a > brand-new $(DESTDIR) tree. Other than that, most of the time I *want* > to use the files in /usr/include and NOT those in /usr/src (speaking as > a user-land kind of guy). 1) The ability to build multiple source tree instances on a single box prepatory to doing a distribution. 2) The ability to rebuild everything then chroot to test it prior to actually installing it. 3) The ability (eventually) to rebuild with a relocated source tree. For instance, off a remote mount or a cdrom. 4) The ability (eventually) to do a fully hosted crosscompile. This one is tricky, since you can't use any of the tools that result from the build in the build. This is an eventual goal since the tools-used-during-build reference and the divisions between system dependent code aren't there in FreeBSD's source tree. That's what it buys us. Terry Lambert terry@cs.weber.edu --- Any opinions in this posting are my own and not those of my present or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?9502231622.AA02944>