On Thu, Sep 20, 2001 at 05:43:10PM -0400, Dan Malek wrote: > Tom Rini wrote: > > > .... In some parts of the PPC world, we can't/don't for various > > reasons (technical and/or ingrained). > > I can't think of any that don't......Help me remember :-).
heh. Well, chrp and pmac really should use yaboot or something similar. I'd go as far as to say on a machine which reasonably could be used for desktop (and thus the choice of multiple kernels) you want something to find where they are and let you pick one. > > .... For example, I don't see a problem > > with something which takes a vmlinux.gz and boots it. > > First of all, someone has to uncompress it. Then someone has > to handle any initrd loading. Then someone has to handle command line > and other parameter passing. Then..... In other words, do all of the stuff the in-kernel 'bootloader' does. I don't think this is a problem in a stable line. This stuff shouldn't change in a way that's not backwards compat. We've diddled this a bit along the way in 2.2 and 2.4 I'm sure, but we shouldn't. > > .... But we should define what we're going to do before we do > > it so that yaboot/ppcboot and apus can figure out how much re-working > > they'll need and whatnot.. > > Well, I trusted Ben to provide some of this information. Mind you, > it was only a few lines on IRC, but it started with moving all of > the cputable initialization stuff into the bootloaders......and went > downhill from there :-). heh. Well, I do think we should outline what we're going to attempt to do a bit and then throw out some code that does it. There's other people who for some reason or another want to have a bootloader external to the tree, and I think that's a good idea. I'd go as far as yanking our in-kernel crap out and making it it's own package once we got things settled down perhaps... But that's just my opinion. :) -- Tom Rini (TR1265) http://gate.crashing.org/~trini/ ** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/
