Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1750716AbWIAR4V (ORCPT ); Fri, 1 Sep 2006 13:56:21 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1750707AbWIAR4V (ORCPT ); Fri, 1 Sep 2006 13:56:21 -0400 Received: from einhorn.in-berlin.de ([192.109.42.8]:1676 "EHLO einhorn.in-berlin.de") by vger.kernel.org with ESMTP id S1750701AbWIAR4U (ORCPT ); Fri, 1 Sep 2006 13:56:20 -0400 X-Envelope-From: stefanr@s5r6.in-berlin.de Message-ID: <44F8732B.8080102@s5r6.in-berlin.de> Date: Fri, 01 Sep 2006 19:51:39 +0200 From: Stefan Richter User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8.0.5) Gecko/20060720 SeaMonkey/1.0.3 MIME-Version: 1.0 To: Adrian Bunk CC: =?ISO-8859-1?Q?J=F6rn_Engel?= , David Woodhouse , "Randy.Dunlap" , Roman Zippel , linux-kernel@vger.kernel.org, Christoph Hellwig , David Howells , linux-fsdevel@vger.kernel.org Subject: Re: [PATCH 17/17] BLOCK: Make it possible to disable the block layer [try #2] References: <44F5DA00.8050909@s5r6.in-berlin.de> <20060830214356.GO18276@stusta.de> <1157069717.2347.13.camel@shinybook.infradead.org> <20060831174852.18efec7e.rdunlap@xenotime.net> <1157074048.2347.24.camel@shinybook.infradead.org> <20060901134425.GA32440@wohnheim.fh-wedel.de> <44F85267.1000607@s5r6.in-berlin.de> <20060901161920.GB32440@wohnheim.fh-wedel.de> <20060901163403.GC18276@stusta.de> In-Reply-To: <20060901163403.GC18276@stusta.de> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1983 Lines: 47 Adrian Bunk wrote: > On Fri, Sep 01, 2006 at 06:19:20PM +0200, J?rn Engel wrote: [...] >> Assuming that select gets removed in the process, and >> concentrating on oldconfig, would it be enough to have something like >> this in the .config? >> >> # CONFIG_USB_STORAGE has unmet dependencies: CONFIG_SCSI, CONFIG_BLOCK >> >> Now people looking for usb mass storage can find the option without >> grepping through Kconfig files, but also every single driver for every >> single disabled subsystem shows up. Might be a bit too much. This comment or similar things are apparently not necessary _within subsystems_, just across subsystems, i.e. where the hierarchy of subdirectories and files does not match the hierarchy of dependencies. > Common use case: > A driver was changed to use FW_LOADER. > The .config for the old kernel contains CONFIG_FW_LOADER=n. > The user runs "make oldconfig" with the old .config in the new kernel. > > How do you plan to handle this reasonably without select? "make oldconfig" could ask questions when it sees need to disable formerly enabled options. In general I think: As long as we talk about the various prefab UIs to manipulate .config (i.e. "make {allyes, allmod, allno, def, g, menu, old, rand, silentold, update-po-, x, ''}config"), there may be ways to implement modes of operation which do what people expect from 'select' but with 'depends on' alone. To ensure that no user group is discriminated in the process, committees could be formed. (<- attempt on irony) It will get difficult to entirely please users who don't use these interfaces to .config. But it seems these users are better off without 'select'. -- Stefan Richter -=====-=-==- =--= ----= http://arcgraph.de/sr/ - 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/