From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754017Ab1HZSWY (ORCPT ); Fri, 26 Aug 2011 14:22:24 -0400 Received: from terminus.zytor.com ([198.137.202.10]:60172 "EHLO mail.zytor.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751808Ab1HZSWX (ORCPT ); Fri, 26 Aug 2011 14:22:23 -0400 Message-ID: <4E57E441.6030800@zytor.com> Date: Fri, 26 Aug 2011 11:21:53 -0700 From: "H. Peter Anvin" User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:6.0) Gecko/20110816 Thunderbird/6.0 MIME-Version: 1.0 To: Cyrill Gorcunov CC: Bjorn Helgaas , Suresh Siddha , Thomas Gleixner , Ingo Molnar , Ralf Jung , Yinghai Lu , "linux-kernel@vger.kernel.org" Subject: Re: [PATCH 1/2] x86, ioapic: Reserve only 128 bytes for IOAPICs References: <20110825230539.5355.13851.stgit@bhelgaas.mtv.corp.google.com> <1314315225.16133.30.camel@sbsiddha-desk.sc.intel.com> <20110826062224.GA3903@sun> <20110826180930.GA1930@sun> In-Reply-To: <20110826180930.GA1930@sun> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 08/26/2011 11:09 AM, Cyrill Gorcunov wrote: > > Yes, one of the spec is Intel's MP specification (as far as I remember). > Letme re-check... > > | 3.6.5 APIC Memory Mapping > | > | "Unlike the local APICs, the I/O APICs are mapped to give shared access from all > | processors, providing full symmetric I/O access. The default base address for the > | first I/O APIC is 0FEC0_0000h. Subsequent I/O APIC addresses are assigned in > | 4K increments. For example, the second I/O APIC is at 0FEC0_1000h. Non-default > | APIC base addresses can be used if the MP configuration table is provided. > | (Refer to Chapter 4.) However, the local APIC base address must be aligned > | on a 4K boundary, and the I/O APIC base address must be aligned on a 1K > | boundary." > > Ie -- 4K increment with 1K base address. If I find other sources I have in mind > I'll ping you. > OK, so that explicitly specifies a 1K alignment. The 4K bit seems to be a default policy and not a requirement. -hpa