From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1750968AbVJXWJM (ORCPT ); Mon, 24 Oct 2005 18:09:12 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751118AbVJXWJM (ORCPT ); Mon, 24 Oct 2005 18:09:12 -0400 Received: from gprs189-60.eurotel.cz ([160.218.189.60]:24728 "EHLO amd.ucw.cz") by vger.kernel.org with ESMTP id S1750968AbVJXWJL (ORCPT ); Mon, 24 Oct 2005 18:09:11 -0400 Date: Tue, 25 Oct 2005 00:09:01 +0200 From: Pavel Machek To: Richard Purdie Cc: kernel list Subject: Re: Sharp brick c-3000 Message-ID: <20051024220901.GD10057@elf.ucw.cz> References: <20051024211632.GA7127@elf.ucw.cz> <1130191145.8345.185.camel@localhost.localdomain> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1130191145.8345.185.camel@localhost.localdomain> X-Warning: Reading this can be dangerous to your mental health. User-Agent: Mutt/1.5.9i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hi! > > I'm not sure what I did wrong... I can still access "flasher" menu, by > > holding down okay after reset, I press 4, 2, Y after that, screen just > > goes black, but I do not get "update screen" after that. SD inserted > > or not, no change. I'll play a bit more. > > Ah, I can guess :-(. Have a look at this: > http://www.h5.dion.ne.jp/~rimemoon/zaurus/pic/nandmap.jpg (from > http://www.h5.dion.ne.jp/~rimemoon/zaurus/memo_006.htm) > > updater.sh writes the kernel into the space marked "2nd Kernel" - 1264kb > long. I'd guess you wrote a kernel larger than this, you overwrite > the Last kernel I flashed was 1285752 bytes big... > initrd.gz which is used to run updater.sh. We really need to add a size > check to the OE copy of updater.sh. Anyhow, the damage has been done and > that update route is now broken. I should have warned about this > although I assumed Sharp might have fixed it on newer devices :-/. Heh, apparently they did not :-(. Or I have some other problem. > Your other alternative is to perform a NAND Restore, if you can find a > NAND backup for the C3000. The D+M menu has an option for this and I'm > sure it will also be documented on the web. When in this state I > couldn't get that to work on my c760 though it could have been the image > I was trying to restore it with or my CF card... I can't seem to enter D+M menu. Instructions say "wait for up-to an hour", so I guess that's what I should do. > Incidentally OE has safeguards that warn you if the kernel is too big... I should really learn how to use that framework... Pavel -- Thanks, Sharp!