From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261658AbVGKNaP (ORCPT ); Mon, 11 Jul 2005 09:30:15 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261693AbVGKNaO (ORCPT ); Mon, 11 Jul 2005 09:30:14 -0400 Received: from fsmlabs.com ([168.103.115.128]:34780 "EHLO fsmlabs.com") by vger.kernel.org with ESMTP id S261658AbVGKNaJ (ORCPT ); Mon, 11 Jul 2005 09:30:09 -0400 Date: Mon, 11 Jul 2005 07:34:53 -0600 (MDT) From: Zwane Mwaikambo To: Andi Kleen cc: linux-kernel@vger.kernel.org Subject: Re: [RFC][PATCH] i386: Per node IDT In-Reply-To: Message-ID: References: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Sun, 11 Jul 2005, Andi Kleen wrote: > Why per node? Why not go the whole way and make it per CPU? > > I would also not define it statically, but allocate it at boot time > in node local memory. I went per node so that it would be minimal/zero impact for the no-node case, it would also simplify hotplug cpu since once a cpu in a node goes down, we still have other participating processors capable of handling its devices without having to do too much migration work. I'll definitely incorporate the node local allocations however, for some i386 systems we might be forced to stick some additional IDTs on node 0 since the IDTR will only take 32bit addresses and we could end up with only highmem on some nodes. Thanks for the feedback, Zwane