From mboxrd@z Thu Jan 1 00:00:00 1970 From: Declan Moriarty Subject: Re: IDE Problem with old Gateway laptop -SOLVED Date: Mon, 05 Nov 2007 15:23:10 +0000 Message-ID: <1194276190.2633.137.camel@localhost.localdomain> References: <884696.96067.qm@web54404.mail.yahoo.com> <1193830965.2347.66.camel@turtle-25895.iol.ie> <20071031175812.7dc9cc8f@the-village.bc.nu> Reply-To: declan.moriarty@gmail.com Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: Received: from mail2.mail.iol.ie ([193.95.141.54]:37193 "EHLO mail2.mail.iol.ie" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751550AbXKEPX3 (ORCPT ); Mon, 5 Nov 2007 10:23:29 -0500 In-Reply-To: <20071031175812.7dc9cc8f@the-village.bc.nu> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Alan Cox , Kevin Day Cc: Vlad Codrea , linux-ide On Wed, 2007-10-31 at 17:58 +0000, Alan Cox wrote: > > kjournald starting Commit interval 5 seconds > > VFS: Mounted root (ext3 filesystem) readonly > > Freeing unused kernel memory: 168k freed > > (keyboard still echoed onscreen, disk parked) > > > > Adding init=/bin/bash is no better. > > So its the user space which is broken. Make sure the user space has the > right /dev etc - and is for the right CPU - a 200MHz pentium means you > need i386 or i586 binaries. i686/athlon binaries would produce the effect > you report. Remember this? The problem appears to have been uClibc-0.9.29. Dropping the system to uClibc-0.9.28.3 and building the minimum as a proof of concept, I'm running with init=/bin/bash. This is using the same 2 kernels, one compiled under uClibc-0.9.29, and one under glibc (not installed there) which seems to indicate that the kernel is fairly libc independent. I have both systems on that laptop currently. Hda3(with uClibc-0.9.28.3) boots whereas the same system on hda4 (uClibc-0.9.29)does not. It's actually quite audible whether it works or not, as you barely hear the hard disk on the dodgy one at all, while the good one gives it a thrashing before coming up with. It's either that libc or some i686 code sneaked through to the old pentium. If I knew a handy way to check that, I'd do for it. -- With Best Regards, Declan Moriarty