Date: Sat, 26 Jul 2003 00:14:08 -0600 From: Scott Long <scottl@freebsd.org> To: src-committers@FreeBSD.org Cc: cvs-all@FreeBSD.org Subject: Re: cvs commit: src/release Makefile Message-ID: <3F221C30.4050003@freebsd.org> In-Reply-To: <200307260557.h6Q5vrm6002959@repoman.freebsd.org> References: <200307260557.h6Q5vrm6002959@repoman.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Scott Long wrote: > scottl 2003/07/25 22:57:53 PDT > > FreeBSD src repository > > Modified files: > release Makefile > Log: > Fix the release build with a forgotten '-f'. > > Revision Changes Path > 1.800 +3 -3 src/release/Makefile > Instead of putting a snide remark into commit log about this, I'll take a different approach. For the love of god, people, the minimum requirement for checking something in to our repo is that it *COMPILES*! Things have gotten far too sloppy, and it only wastes developer and user time! What good are the daily release snapshots if they don't build for days or weeks on end? What good is it to tell our userbase to track the tree with cvs and cvsup when the tree won't build? Not long ago we used to require that any commit to the kernel be tested before-hand by building LINT, and that any commit to the rest of the tree be tested by building world. That was back in the days when building world took 2-3 hours on fast hardware. The project has resources available to it now that can build world in less than an hour and build a release in less than two. There is no excuse for not testing before committing. That said, I'm pleased with the work that has gone on this summer and appreciate all of the hard effort. The schedules for 4.9 and 5.2 will be published shortly, both of which have the potential to be important and exciting releases =-) Scott
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3F221C30.4050003>