From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753849Ab3JKGUI (ORCPT ); Fri, 11 Oct 2013 02:20:08 -0400 Received: from mail-ee0-f45.google.com ([74.125.83.45]:46945 "EHLO mail-ee0-f45.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751762Ab3JKGUG (ORCPT ); Fri, 11 Oct 2013 02:20:06 -0400 Date: Fri, 11 Oct 2013 08:20:02 +0200 From: Ingo Molnar To: "H. Peter Anvin" Cc: Thomas Gleixner , Linus Torvalds , Linux Kernel Mailing List Subject: Re: RFD: Does CONFIG_X86_LOCAL_APIC still make sense? Message-ID: <20131011062002.GC4975@gmail.com> References: <52573279.9020404@zytor.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <52573279.9020404@zytor.com> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org * H. Peter Anvin wrote: > I'm trying to wrap my head around if there are use cases where disabling > either CONFIG_X86_UP_APIC or CONFIG_X86_LOCAL_APIC would ever make sense > anymore. > > I am guessing there are probably some small number of embedded systems > which still don't have LAPICs, but is it a significant number and does > anyone care for new kernels? We are not talking about discontinuing > support for non-APIC configurations, just the configuration option. I guess it depends on the kernel size difference. If it's more than just a few K then a patch will creep back via some CONFIG_EXPERT "reduce kernel bloat" route. The thing is, if we allow non-apic configurations then we have paid most of the modularization, #ifdef and general maintenance price already. Adding a Kconfig just makes it more testable and breakages more apparent. Unless I missed some aspects of this that would allow us to simplify code significantly that is. Thanks, Ingo