From owner-freebsd-ports@FreeBSD.ORG Tue Jan 13 14:55:06 2015 Return-Path: Delivered-To: freebsd-ports@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [8.8.178.115]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTPS id B9E37542 for ; Tue, 13 Jan 2015 14:55:06 +0000 (UTC) Received: from be-well.ilk.org (be-well.ilk.org [23.30.133.173]) by mx1.freebsd.org (Postfix) with ESMTP id 8F54FA91 for ; Tue, 13 Jan 2015 14:55:06 +0000 (UTC) Received: from lowell-desk.lan (lowell-desk.lan [172.30.250.41]) by be-well.ilk.org (Postfix) with ESMTP id B14F033C1D for ; Tue, 13 Jan 2015 09:54:54 -0500 (EST) Received: by lowell-desk.lan (Postfix, from userid 1147) id D98D13983C; Tue, 13 Jan 2015 09:54:53 -0500 (EST) From: Lowell Gilbert To: freebsd-ports@freebsd.org Subject: Re: poudriere: reduce the number of rebuilt packages? References: <54A67B1A.5060007@gmx.net> <54A97748.9000401@gmx.net> <54B3FD78.5060404@gmx.net> <54B40822.9000702@gmx.net> <23A49BD336E4EC29F493416A@ogg.in.absolight.net> <54B4372C.9040803@siol.net> <54B4CBF6.70001@FreeBSD.org> Date: Tue, 13 Jan 2015 09:54:53 -0500 In-Reply-To: <54B4CBF6.70001@FreeBSD.org> (Matthew Seaman's message of "Tue, 13 Jan 2015 07:40:38 +0000") Message-ID: <44vbkabuiq.fsf@lowell-desk.lan> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/24.4 (berkeley-unix) MIME-Version: 1.0 Content-Type: text/plain X-BeenThere: freebsd-ports@freebsd.org X-Mailman-Version: 2.1.18-1 Precedence: list List-Id: Porting software to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 13 Jan 2015 14:55:06 -0000 Matthew Seaman writes: > poudriere only knows that the dependency changed. In effect, to find > out if the package of interest would be changed because of that, it has > no other recourse than to build the package. Now, if you can come up > with some heuristics whereby you can examine the changes to a port and > determine that they will not cause significant downstream changes, and > do that reliably and faster than just rebuilding the package, then I'm > sure the poudriere developers would be eager to incorporate them. > > Failing that, poudriere re-building everything that might be affected is > the sensible choice. If I know that only the actually changed ports need to be rebuilt, I go into my jail but instead of running poudriere, I use "portmaster -g". Unfortunately, it's really easy to be wrong about "knowing" that.