From owner-cvs-all@FreeBSD.ORG Fri Jun 17 08:07:17 2011 Return-Path: Delivered-To: cvs-all@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 9E7151065674; Fri, 17 Jun 2011 08:07:17 +0000 (UTC) (envelope-from itetcu@FreeBSD.org) Received: from worf.ds9.tecnik93.com (worf.ds9.tecnik93.com [81.196.207.130]) by mx1.freebsd.org (Postfix) with ESMTP id 191768FC0A; Fri, 17 Jun 2011 08:07:16 +0000 (UTC) Received: from User-PC (unknown [81.181.146.246]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by worf.ds9.tecnik93.com (Postfix) with ESMTPSA id 867DC22C55AC; Fri, 17 Jun 2011 11:07:15 +0300 (EEST) Date: Fri, 17 Jun 2011 11:07:11 +0300 From: Ion-Mihai Tetcu To: Stephen Montgomery-Smith Message-Id: <20110617110711.3c4f8e3a.itetcu@FreeBSD.org> In-Reply-To: <20110616234556.GB19139@magic.hamla.org> References: <201106162238.p5GMcfrV085514@repoman.freebsd.org> <20110616231146.GA19139@magic.hamla.org> <4DFA940C.2030909@missouri.edu> <20110616234556.GB19139@magic.hamla.org> X-Mailer: Sylpheed 3.1.1 (GTK+ 2.10.14; i686-pc-mingw32) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: Maho NAKATA , Sahil Tandon , Stephen Montgomery-Smith , "cvs-all@FreeBSD.org" , "ports-committers@FreeBSD.org" , gabor@FreeBSD.org, "cvs-ports@FreeBSD.org" Subject: Re: cvs commit: ports/astro/orsa Makefile X-BeenThere: cvs-all@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: **OBSOLETE** CVS commit messages for the entire tree List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 17 Jun 2011 08:07:17 -0000 On Thu, 16 Jun 2011 19:45:57 -0400 Sahil Tandon wrote: >> Should one wait a day or so before committing the bumps to avoid >> certain "race" conditions with tinderbox, where it might build the port >> requiring the dependent before rebuilding the dependent? No, there is no such race, on regular tindy, QAT (soon to be back) or pointyhat. You should commit this kind of updates _exactly_after_ you did the update of the dependency. Waiting hours / days / whatever gets you pointyhats and agry mails from users unable to run their ports after your update of the dependency. Please talk about this with your mentors (cc'ed). > I believe shlib-related PORTREVISION bumps should be committed right > away. Yes. Not the same commit, but exactly after, one commit for all afftected ports. Finding affected ports: - ports that list the lib in question in LIB_DEPENDS - if the dependency has one of the USE_XXX, all the ports that have that USE_XXX defined (but take a look at what that USE_XXX actually does, there might be exceotions) To be safe: grep -R libsomthing /usr/ports This is actually documented in PH. Please take some time to read it all, even the sections that are not in your usual area of interest. HTH, -- Ion-Mihai Tetcu