From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from fed1rmmtao04.cox.net (fed1rmmtao04.cox.net [68.230.241.35]) by ozlabs.org (Postfix) with ESMTP id 70B5068696 for ; Tue, 8 Nov 2005 03:31:52 +1100 (EST) Date: Mon, 7 Nov 2005 09:31:44 -0700 From: Tom Rini To: David Woodhouse Message-ID: <20051107163144.GF3839@smtp.west.cox.net> References: <1131357857.27347.28.camel@baythorne.infradead.org> <20051107160939.GD3839@smtp.west.cox.net> <1131380422.27347.72.camel@baythorne.infradead.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <1131380422.27347.72.camel@baythorne.infradead.org> Cc: linuxppc-dev@ozlabs.org Subject: Re: [PATCH] Fix 8250 probe on ppc32 List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Mon, Nov 07, 2005 at 04:20:22PM +0000, David Woodhouse wrote: > On Mon, 2005-11-07 at 09:09 -0700, Tom Rini wrote: > > That depends on what we're going to expect out of arch/powerpc/boot, > > which I'm talking with Paul about. > > I'd be quite happy to see the bootloader code kept separate from the > kernel in future. > > Fedora's ppc64-utils package currently has a 'mkzimage' script and a > bootloader stub based on what's in arch/ppc64/kernel, but modified to be > capable of loading 32-bit kernels too. We can quite happily put together > the stub, the vmlinux and the initrd after the fact. I've always found it quite handy, and thought it was a good new user experiance that they could just build and boot the kernel with just the kernel tarball around. If everything going forward ships from the hw vendor with U-Boot, that's less of an issue, but I don't know how likely !Freescale is to change. > I don't really see why any of the other stubs necessarily need to live > with the kernel either. Avoiding duplication of hardware defines (serial, whack-a-device, etc) has been part of it. -- Tom Rini http://gate.crashing.org/~trini/