Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262787AbVCER7D (ORCPT ); Sat, 5 Mar 2005 12:59:03 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261688AbVCER4f (ORCPT ); Sat, 5 Mar 2005 12:56:35 -0500 Received: from parcelfarce.linux.theplanet.co.uk ([195.92.249.252]:31456 "EHLO parcelfarce.linux.theplanet.co.uk") by vger.kernel.org with ESMTP id S261694AbVCERuV (ORCPT ); Sat, 5 Mar 2005 12:50:21 -0500 Message-ID: <4229F14A.8030109@pobox.com> Date: Sat, 05 Mar 2005 12:50:02 -0500 From: Jeff Garzik User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.3) Gecko/20040922 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Linus Torvalds CC: Russell King , Greg KH , Andrew Morton , linux-kernel@vger.kernel.org, chrisw@osdl.org Subject: Re: Linux 2.6.11.1 References: <20050304175302.GA29289@kroah.com> <20050304124431.676fd7cf.akpm@osdl.org> <20050304205842.GA32232@kroah.com> <20050304131537.7039ca10.akpm@osdl.org> <20050304135933.3a325efc.akpm@osdl.org> <20050304220518.GC1201@kroah.com> <20050305095139.A26541@flint.arm.linux.org.uk> <4229EA0A.8010608@pobox.com> In-Reply-To: Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1753 Lines: 46 Linus Torvalds wrote: > > On Sat, 5 Mar 2005, Jeff Garzik wrote: > >>Yup, BK could definitely handle that... > > > However, it's also true that the thing BK is _worst_ at is cherry-picking > things, and having a collection of stuff where somebody may end up vetoing > one patch and saying "remove that one". In general, I agree. Andrew and I mentioned this to BitMover recently [though its certainly not a new comment], when they asked us why I had to occasionally blow away the netdev-2.6 tree, and reconstitute it from scratch. > I love BK, but what BK does well is merging and maintaining trees full of > good stuff. What BK sucks at is experimental stuff where you don't know > whether something should be eventually used or not. I use BitKeeper to maintain such a tree, "libata-dev". Most stuff in there will go upstream. Some stuff may never go upstream. Some stuff needs to simmer for a while before going upstream. So "change streams" get divided up locally: [jgarzik@pretzel libata-dev]$ ls -FC adma/ atapi-enable/ janitor/ remove-one-fix/ adma-mwi/ bridge-detect/ passthru/ sata-sil-irq/ ahci-msi/ chs-support/ pdc2027x/ tf-cleanup/ ahci-tf-read/ ioctl-get-identity/ pdc20619/ via-6421/ iomap/ promise-sata-pata/ and then I cherrypick from that. netdev-2.6 queue is maintained the same way. It's simply a merge tree composed of 40+ individual trees, all merged together. Jeff - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/