From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S265233AbUGGRCM (ORCPT ); Wed, 7 Jul 2004 13:02:12 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S265226AbUGGRCL (ORCPT ); Wed, 7 Jul 2004 13:02:11 -0400 Received: from dragnfire.mtl.istop.com ([66.11.160.179]:51142 "EHLO dsl.commfireservices.com") by vger.kernel.org with ESMTP id S265233AbUGGRCC (ORCPT ); Wed, 7 Jul 2004 13:02:02 -0400 Date: Wed, 7 Jul 2004 13:04:45 -0400 (EDT) From: Zwane Mwaikambo To: Helge Hafting Cc: Davide Rossetti , linux-kernel@vger.kernel.org Subject: Re: MSI to memory? In-Reply-To: <40EBF07B.8040003@hist.no> Message-ID: References: <200407011215.59723.bjorn.helgaas@hp.com> <20040701115339.A4265@unix-os.sc.intel.com> <40EBED33.3050707@roma1.infn.it> <40EBF07B.8040003@hist.no> 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 Wed, 7 Jul 2004, Helge Hafting wrote: > Davide Rossetti wrote: > > > Rajesh Shah wrote: > > > >> What type of usage model did you have in mind to have the > >> > >> device write to memory instead of using MSI for interrupts? > >> > >> > > for instance for a fast wake-up trick. the driver loops on a memory > > location until the MSI write access changes the memory content... > > Won't that put a bad load on the bus? Someone else might need it: > * Another cpu in a smp system > * Any device doing bus-master transfers, even in a UP system > > That polling loop had better be guaranteed to be _very_ short. mwait/monitor is best suited for such things and doesn't result in pounding on memory. Considering the box has MSI, there is a higher possibility of it also having the mwait/monitor instructions. Zwane