Date: Tue, 8 May 2001 19:14:44 -0700 From: Eric Melville <eric@FreeBSD.org> To: Mikhail Teterin <mi@misha.privatelabs.com> Cc: mi@aldan.algebra.com, knu@iDaemons.org, will@physics.purdue.edu, sobomax@FreeBSD.ORG, kris@obsecurity.org, kris@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG Subject: Re: port policies Message-ID: <20010508191444.A75739@FreeBSD.org> In-Reply-To: <200105082029.f48KTtC38076@misha.privatelabs.com>; from mi@misha.privatelabs.com on Tue, May 08, 2001 at 04:29:54PM -0400 References: <20010508122714.B48723@FreeBSD.org> <200105082029.f48KTtC38076@misha.privatelabs.com>
next in thread | previous in thread | raw e-mail | index | archive | help
> > Don't do things that the user didn't ask to be done. > > > > If the user wants more excitment, we provide facilities that make it > > easy to do so. > > The user did not ask for ``-O -pipe'' in the CFLAGS either. -pipe, > consumes more memory, while -O makes compiling longer. Should the > default CFLAGS be empty too? By your logic, they should... We're getting a bit off on a tangent here. I suppose that previously I should have mentioned something about consistancy with procedure that is documented to be safe. Using -j(>1) is not encouraged by the documentation, nor is it default policy for building world. We ought to be following our own documented policies, and ports are to be considered more fragile than world. It follows that both the documentation and the world build should sport -j(>1) before ports do. If you were to change the documentation to encourage this, and set -j(>1) in the default make.conf, I'm willing to bet that a much larger number of people would be raining on your parallel-building parade. If not, then I'm wrong, and perhaps we should be using -j(>1) all over the place. 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?20010508191444.A75739>