From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eduard - Gabriel Munteanu Subject: Re: [Qemu-devel] Re: [RFC PATCH 4/7] ide: IOMMU support Date: Thu, 15 Jul 2010 20:27:56 +0300 Message-ID: <20100715172756.GA16942@localhost> References: <1279086307-9596-1-git-send-email-eduard.munteanu@linux360.ro> <4C3E2C2E.70507@codemonkey.ws> <20100714222401.GB21126@sequoia.sous-sol.org> <201007151128.27487.paul@codesourcery.com> <20100715165214.GV14017@sequoia.sous-sol.org> <4C3F3F0D.1000103@redhat.com> <20100715171710.GZ14017@sequoia.sous-sol.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Avi Kivity , Paul Brook , Anthony Liguori , Joerg Roedel , qemu-devel@nongnu.org, kvm@vger.kernel.org To: Chris Wright Return-path: Received: from mail-bw0-f46.google.com ([209.85.214.46]:44340 "EHLO mail-bw0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933810Ab0GOR25 (ORCPT ); Thu, 15 Jul 2010 13:28:57 -0400 Received: by bwz1 with SMTP id 1so727575bwz.19 for ; Thu, 15 Jul 2010 10:28:56 -0700 (PDT) Content-Disposition: inline In-Reply-To: <20100715171710.GZ14017@sequoia.sous-sol.org> Sender: kvm-owner@vger.kernel.org List-ID: On Thu, Jul 15, 2010 at 10:17:10AM -0700, Chris Wright wrote: > * Avi Kivity (avi@redhat.com) wrote: > > > > For emulated device, it seems like we can ignore ATS completely, no? > > Not if you want to emulate an ATS capable device ;) > > Eariler upthread I said: > > IOW, if qemu ever had a device with ATS support... > > So, that should've been a much bigger _IF_ > > thanks, > -chris I think we can augment some devices with ATS capability if there are performance gains in doing so. This doesn't seem to be a detail the actual guest OS would be interested in, so we can do it even for devices that existed long before the AMD IOMMU came into existence. But I'm not really sure about this, it's just a thought. Linux seems to be issuing IOTLB invalidation commands anyway. Eduard From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from [140.186.70.92] (port=53522 helo=eggs.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1OZSF3-0000m2-Fe for qemu-devel@nongnu.org; Thu, 15 Jul 2010 13:29:02 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.69) (envelope-from ) id 1OZSEz-0007Ma-Hg for qemu-devel@nongnu.org; Thu, 15 Jul 2010 13:29:01 -0400 Received: from mail-bw0-f45.google.com ([209.85.214.45]:64212) by eggs.gnu.org with esmtp (Exim 4.69) (envelope-from ) id 1OZSEz-0007MU-BB for qemu-devel@nongnu.org; Thu, 15 Jul 2010 13:28:57 -0400 Received: by bwz19 with SMTP id 19so828308bwz.4 for ; Thu, 15 Jul 2010 10:28:56 -0700 (PDT) Sender: Eduard - Gabriel Munteanu Date: Thu, 15 Jul 2010 20:27:56 +0300 From: Eduard - Gabriel Munteanu Subject: Re: [Qemu-devel] Re: [RFC PATCH 4/7] ide: IOMMU support Message-ID: <20100715172756.GA16942@localhost> References: <1279086307-9596-1-git-send-email-eduard.munteanu@linux360.ro> <4C3E2C2E.70507@codemonkey.ws> <20100714222401.GB21126@sequoia.sous-sol.org> <201007151128.27487.paul@codesourcery.com> <20100715165214.GV14017@sequoia.sous-sol.org> <4C3F3F0D.1000103@redhat.com> <20100715171710.GZ14017@sequoia.sous-sol.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20100715171710.GZ14017@sequoia.sous-sol.org> List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Chris Wright Cc: kvm@vger.kernel.org, Joerg Roedel , qemu-devel@nongnu.org, Avi Kivity , Paul Brook On Thu, Jul 15, 2010 at 10:17:10AM -0700, Chris Wright wrote: > * Avi Kivity (avi@redhat.com) wrote: > > > > For emulated device, it seems like we can ignore ATS completely, no? > > Not if you want to emulate an ATS capable device ;) > > Eariler upthread I said: > > IOW, if qemu ever had a device with ATS support... > > So, that should've been a much bigger _IF_ > > thanks, > -chris I think we can augment some devices with ATS capability if there are performance gains in doing so. This doesn't seem to be a detail the actual guest OS would be interested in, so we can do it even for devices that existed long before the AMD IOMMU came into existence. But I'm not really sure about this, it's just a thought. Linux seems to be issuing IOTLB invalidation commands anyway. Eduard