From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([140.186.70.92]:42375) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1RFQZS-0000pG-IW for qemu-devel@nongnu.org; Sun, 16 Oct 2011 09:16:07 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1RFQZR-0001ya-JU for qemu-devel@nongnu.org; Sun, 16 Oct 2011 09:16:06 -0400 Received: from mx1.redhat.com ([209.132.183.28]:47956) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1RFQZR-0001yV-7B for qemu-devel@nongnu.org; Sun, 16 Oct 2011 09:16:05 -0400 Message-ID: <4E9AD909.1000509@redhat.com> Date: Sun, 16 Oct 2011 15:15:53 +0200 From: Avi Kivity MIME-Version: 1.0 References: <1315197304-22469-1-git-send-email-david@gibson.dropbear.id.au> <1315197304-22469-2-git-send-email-david@gibson.dropbear.id.au> <20111002102547.GC30747@redhat.com> <4E883CF4.6060606@redhat.com> <20111002105238.GE30747@redhat.com> <4E8843DB.1020404@redhat.com> <20111002111700.GF30747@redhat.com> <4E885286.30905@redhat.com> <20111002121426.GK30747@redhat.com> <4E89B5D1.4080600@us.ibm.com> <20111014021407.GB4580@truffala.fritz.box> In-Reply-To: <20111014021407.GB4580@truffala.fritz.box> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH 1/9] Add stub functions for PCI device models to do PCI DMA List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Anthony Liguori , "Michael S. Tsirkin" , qemu-devel@nongnu.org, joerg.roedel@amd.com, rth@twiddle.net, agraf@suse.de, eduard.munteanu@linux360.ro, kraxel@redhat.com On 10/14/2011 04:14 AM, David Gibson wrote: > > Virtio is a very, very special case. virtio requires coherent RAM access. > > Right. Virtio's access to memory is *not* emulated PCI DMA, it's > god-like hypervisor access to guest system memory. It should > correctly bypass any IOMMU, and so should remain as > cpu_physical_memory_rw() or the atomic accessors, rather than being > converted to this new API. virtio should definitely not bypass an iommu. A guest may assign a virtio device to nested guests, and would wish it confined by the emulated iommu. More generally, a guest sees a virtio device as just another pci device, and has no way to tell that it bypasses the iommu. -- error compiling committee.c: too many arguments to function