From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757484Ab2GNAtw (ORCPT ); Fri, 13 Jul 2012 20:49:52 -0400 Received: from zfrontend1.aha.ru ([195.2.83.147]:47942 "EHLO aha.ru" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1757297Ab2GNAtu (ORCPT ); Fri, 13 Jul 2012 20:49:50 -0400 From: "werner" Subject: To: linux-kernel@vger.kernel.org, mjt@tls.msk.ru X-Mailer: CommuniGate Pro WebUser Interface v.4.3.12 Date: Fri, 13 Jul 2012 20:49:45 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8"; format="flowed" Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Since years I use the same routine for compile / link / package the kernel. Until now, rdev gave and wrote always the boot device in the form /dev/... , not as the bios number. This problem starts now only. So I suppose there was anything changed, either on the kernel or in one of the kernel-near basic routines (such as, rdev itself) I agree that the bios device numbers are not good for identify installations or devices, as they can change, if one change the BIOS boot order, or if one only change via F12 the actual boot order, or even if one stick in an usb key before booting; specially it changes after one install Linux from an USB key. Just because of this I use the /dev/... names for lilo and for the kernel command line. rdev may be obsolete, but it's one of the oldest functions and used in many install systems, so that for compatibility it should be kept, nor be changed. --- Professional hosting for everyone - http://www.host.ru