From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755262AbZARTJV (ORCPT ); Sun, 18 Jan 2009 14:09:21 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752853AbZARTJG (ORCPT ); Sun, 18 Jan 2009 14:09:06 -0500 Received: from mx3.mail.elte.hu ([157.181.1.138]:42670 "EHLO mx3.mail.elte.hu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752668AbZARTJF (ORCPT ); Sun, 18 Jan 2009 14:09:05 -0500 Date: Sun, 18 Jan 2009 20:08:49 +0100 From: Ingo Molnar To: Jack Steiner Cc: Mike Travis , LKML , "H. Peter Anvin" , Thomas Gleixner , Yinghai Lu , "Maciej W. Rozycki" Subject: Re: [PATCH] x86: put trigger in to detect mismatched apic versions. Message-ID: <20090118190849.GB858@elte.hu> References: <49704DF6.8040205@sgi.com> <20090116092540.GC4305@elte.hu> <4970C984.30202@sgi.com> <20090116223011.GB3899@elte.hu> <497116A8.5080900@sgi.com> <20090117030752.GB127262@sgi.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20090117030752.GB127262@sgi.com> User-Agent: Mutt/1.5.18 (2008-05-17) X-ELTE-VirusStatus: clean X-ELTE-SpamScore: -1.5 X-ELTE-SpamLevel: X-ELTE-SpamCheck: no X-ELTE-SpamVersion: ELTE 2.0 X-ELTE-SpamCheck-Details: score=-1.5 required=5.9 tests=BAYES_00 autolearn=no SpamAssassin version=3.2.3 -1.5 BAYES_00 BODY: Bayesian spam probability is 0 to 1% [score: 0.0000] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org * Jack Steiner wrote: > > Btw, I checked with our UV architect and the problem is that we need a > > 16 bit apic id which is what caused the MAX_APICS to be bumped to 32k. > > The lower 8 bits are the normal apic id, and the upper bit relate to > > the node. This means cpu 0 on node 0 has the same apic id as cpu 0 on > > node 1, etc. I also asked about whether we could rely on always > > having > > Not strictly true. The apicids in the ACPI tables are always globally > unique across the entire system. Because of the size of UV systems, UV > needs 16 bit apicids. This fits in the ACPI apicid id/eid fields. > > The actual processor apicid register is unfortunately only 11 bits and > there are some restrictions on the actual values loaded into the apicid > register. > > If we can put unique ids into the apicid register, we do. If we can't, > the function that reads the apicid will automatically supply the rest of > the bits. Most of the kernel is unaware that the processor apicid > register may have only a subset of the bits that are in the ACPI tables. apicid remapping is something we need/want, so we cannot remove that array. But it would be nice to offload such properties to the percpu area instead - is there any reason why that is hard? The local apic is attached to a CPU in any case. Is there some early init reason that complicates this? Ingo