LWN.net Logo

Re: [RFC, Announce] Unified x86 architecture, arch/x86

From:  Linus Torvalds <torvalds-AT-linux-foundation.org>
To:  Jeff Garzik <jeff-AT-garzik.org>
Subject:  Re: [RFC, Announce] Unified x86 architecture, arch/x86
Date:  Fri, 20 Jul 2007 15:51:51 -0700 (PDT)
Message-ID:  <alpine.LFD.0.999.0707201546140.27249@woody.linux-foundation.org>
Cc:  Ingo Molnar <mingo-AT-elte.hu>, Thomas Gleixner <tglx-AT-linutronix.de>, LKML <linux-kernel-AT-vger.kernel.org>, Andrew Morton <akpm-AT-linux-foundation.org>, Andi Kleen <ak-AT-suse.de>, Arjan van de Ven <arjan-AT-infradead.org>, Chris Wright <chrisw-AT-sous-sol.org>, Steven Rostedt <rostedt-AT-goodmis.org>
Archive-link:  Article, Thread



On Fri, 20 Jul 2007, Jeff Garzik wrote:

> Ingo Molnar wrote:
> > * Jeff Garzik <jeff@garzik.org> wrote:
> > 
> > > I agree with Andi...  it's quite nice to be able to leave some arch/i386
> > > stuff, and not carry it over to arch/x86-64.
> > 
> > we can leave those few items in arch/x86 just as much. No need to keep
> > around a legacy tree for that.
> 
> By extension it makes doing that sort of thing, in general, more difficult.
> Which is IMO not desirable.

I think it's *much* harder to carry legacy things around in an old tree 
that almost nobody even uses any more (probably not true yet, but for most 
of the main developers, I bet it will be true in a year). Especially one 
that just duplicates 99% of the stuff.

There really isn't that much legacy crud. There are things like random 
quirks, but every time I hear the (theoretical) argument about how much 
time and effort we save by having it duplicated somewhere else, I think 
about all the time we definitely waste by fixing the same bug twice (and 
worry about the cases where we don't).

			Linus


(Log in to post comments)

Copyright © 2007, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds