From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932366AbWDKI5O (ORCPT ); Tue, 11 Apr 2006 04:57:14 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S932357AbWDKI5O (ORCPT ); Tue, 11 Apr 2006 04:57:14 -0400 Received: from hobbit.corpit.ru ([81.13.94.6]:38748 "EHLO hobbit.corpit.ru") by vger.kernel.org with ESMTP id S932366AbWDKI5N (ORCPT ); Tue, 11 Apr 2006 04:57:13 -0400 Message-ID: <443B6F66.3020302@tls.msk.ru> Date: Tue, 11 Apr 2006 12:57:10 +0400 From: Michael Tokarev User-Agent: Mail/News 1.5 (X11/20060318) MIME-Version: 1.0 To: Kris Shannon CC: Linux Kernel Mailing List Subject: Re: Separate Initramfs dependency on initrd (and therefore ramdisks) References: <5d4799ae0602170820j33795815u7104bd41c7fe7e67@mail.gmail.com> <5d4799ae0604101855o72f01453l438d0d4d628bbb7@mail.gmail.com> In-Reply-To: <5d4799ae0604101855o72f01453l438d0d4d628bbb7@mail.gmail.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Kris Shannon wrote: > A number of distributions (most importantly for me - Debian) use the > initrd as initramfs facility. I assumed that the passing of the data > block would be independent of ram disks seeing as not using a ram > disk was one of the major reasons for initramfs, but it seems that > you need CONFIG_BLK_DEV_INITRD=y which depends on > CONFIG_BLK_DEV_RAM=y > > Would a patch separating out the init image handling from the initrd > handling be welcome and if so should the resulting init image code > be dependant on a CONFIG variable or always on (like initramfs is now) > > The only reference to this I found in the archives was: > > http://www.uwsg.indiana.edu/hypermail/linux/kernel/0508.1/0097.html http://www.ussg.iu.edu/hypermail/linux/kernel/0603.1/0969.html /mjt