From mboxrd@z Thu Jan 1 00:00:00 1970 From: Chris Ball Subject: Re: [PATCHv4] MMC: MMC boot partitions support. Date: Wed, 30 Mar 2011 18:46:48 -0400 Message-ID: References: <1300533491-2378-2-git-send-email-andreiw@motorola.com> <1300828300-8677-2-git-send-email-andreiw@motorola.com> <201103301403.20779.arnd@arndb.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from void.printf.net ([89.145.121.20]:52829 "EHLO void.printf.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753491Ab1C3Wkx (ORCPT ); Wed, 30 Mar 2011 18:40:53 -0400 In-Reply-To: (Chris Ball's message of "Wed, 30 Mar 2011 18:43:21 -0400") Sender: linux-mmc-owner@vger.kernel.org List-Id: linux-mmc@vger.kernel.org To: Andrei Warkentin Cc: Arnd Bergmann , linux-mmc@vger.kernel.org Hi, On Wed, Mar 30 2011, Chris Ball wrote: > Hi Andrei, > > On Wed, Mar 30 2011, Andrei Warkentin wrote: >> Unfortunately yes, although I am open for suggestions. An idea I >> considered is to add a postfix for device partitions. So, for card 0, >> the user area will remain mmcblk0, while boot partitions will be >> mmcblk0b0 and mmcblk0b1, generic partitions will be mmcblk0g0, and so >> on. > > Like Arnd, my main concern was around naming, breaking boot setups, and > freaking people out by telling them that their cards have block devices > that they can't usually see and not making it obvious where they came > from. :) > > Your "b/g" postfix solution appears to fix all of these, so I like it. > We still have to think about where we expect users wondering what the > "b" and "g" mean to go in order to find out.. Oh, one more thing -- if we go with the b/g postfix, what's the intuition for requiring a Kconfig entry to turn this on? Is there any good argument against just turning it on for everyone, as long as it's not affecting mmcblk*p* naming? Thanks, - Chris. -- Chris Ball One Laptop Per Child