Date: Fri, 9 Mar 2001 18:45:55 +1100 (EST) From: Bruce Evans <bde@zeta.org.au> To: Kris Kennaway <kris@obsecurity.org> Cc: Warner Losh <imp@village.org>, "Rodney W. Grimes" <freebsd@gndrsh.dnsmgr.net>, "Matthew D. Fuller" <fullermd@futuresouth.com>, arch@FreeBSD.ORG Subject: Re: Breaking up make.conf Message-ID: <Pine.BSF.4.21.0103091826260.484-100000@besplex.bde.org> In-Reply-To: <20010308201422.A94052@mollari.cthul.hu>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, 8 Mar 2001, Kris Kennaway wrote: > On Thu, Mar 08, 2001 at 08:49:16PM -0700, Warner Losh wrote: > Anyway, the only real question here is whether to go for the iterative > .for search, or a recursive include. Both can give the same > behaviour, so it comes down to which is more efficient. Except the .for search is easier to misconfigure and harder to debug (where is your make.conf today?). The search is bad enough in the limited scope of kmod.mk. > > Something like the following patch, included after my sig. Note, I'd > > also go for nuking bsd.own.mk and bsd.cpu.mk and moving them into > > etc/defaults/make.conf so we stop polluting the global make space with > > them. .for loops are hard to terminate in make, so I didn't try. > > You can't put bsd.cpu.mk in /etc/defaults/make.conf or I would have > done this from the beginning. It has to be included AFTER > /etc/make.conf, because thats where CPUTYPE is set. This is a general problem. Parts of make.conf would work better if it were included before sys.mk. E.g., it should used "?=" to set most variables so that it doesn't clobber any previous setting, but it can't do that since sys.mk has alreading provided a previous setting (only a default setting, but make.conf doesn't know that). Bruce To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.21.0103091826260.484-100000>