From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S264349AbUE0Nfd (ORCPT ); Thu, 27 May 2004 09:35:33 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S264443AbUE0Nfd (ORCPT ); Thu, 27 May 2004 09:35:33 -0400 Received: from mx1.redhat.com ([66.187.233.31]:11410 "EHLO mx1.redhat.com") by vger.kernel.org with ESMTP id S264349AbUE0NfT (ORCPT ); Thu, 27 May 2004 09:35:19 -0400 Date: Thu, 27 May 2004 09:34:52 -0400 (EDT) From: Ingo Molnar X-X-Sender: mingo@devserv.devel.redhat.com To: "Maciej W. Rozycki" cc: Pavel Machek , kernel list , Andrew Morton , mingo@elte.hu Subject: Re: Cleanups for APIC In-Reply-To: Message-ID: References: <20040525124937.GA13347@elf.ucw.cz> 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 Thu, 27 May 2004, Maciej W. Rozycki wrote: > > (wrt. io_apic_sync(): i added it in 2.1.104 together with some other > > changes - i dont this it's necessary anymore - the local APICs had > > writearound erratas, but i dont remember this ever being necessary for > > IO-APICs. I'll address this in another patch.) > > Hmm, isn't that needed to make sure the iomem writeback is completed > before exiting the caller? the only thing that could happen is a POST delay in the PCI chipset - but is that really an issue? Plus we only do the io_apic_sync() for the masking, not the unmasking - so if it's needed then we dont do it consistently. Ingo