From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:37601) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1XhCWg-0002SU-Uv for qemu-devel@nongnu.org; Thu, 23 Oct 2014 03:09:45 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1XhCWa-0002wc-RK for qemu-devel@nongnu.org; Thu, 23 Oct 2014 03:09:38 -0400 Received: from mx1.redhat.com ([209.132.183.28]:62777) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1XhCWa-0002wO-KI for qemu-devel@nongnu.org; Thu, 23 Oct 2014 03:09:32 -0400 Date: Thu, 23 Oct 2014 10:12:51 +0300 From: "Michael S. Tsirkin" Message-ID: <20141023071251.GA6287@redhat.com> References: <1412692807-12398-1-git-send-email-cornelia.huck@de.ibm.com> <54349246.6000905@amacapital.net> <20141008110428.6fc78115.cornelia.huck@de.ibm.com> <20141022084438.GA8051@redhat.com> <5447BC84.1010402@siemens.com> <1414010056.364.20.camel@pasglop> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1414010056.364.20.camel@pasglop> Subject: Re: [Qemu-devel] [PATCH RFC 00/11] qemu: towards virtio-1 host support List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Benjamin Herrenschmidt Cc: thuth@linux.vnet.ibm.com, kvm@vger.kernel.org, Rusty Russell , Jan Kiszka , qemu-devel@nongnu.org, Andy Lutomirski , Cornelia Huck , virtualization@lists.linux-foundation.org On Thu, Oct 23, 2014 at 07:34:16AM +1100, Benjamin Herrenschmidt wrote: > On Wed, 2014-10-22 at 16:17 +0200, Jan Kiszka wrote: > > I thought about this again, and I'm not sure anymore if we can use > > ACPI > > to "black-list" the incompatible virtio devices. Reason: hotplug. To > > my > > understanding, the ACPI DRHD tables won't change during runtime when a > > device shows up or disappears. We would have to isolate virtio devices > > from the rest of the system by using separate buses for it (and avoid > > listing those in any DRHD table) and enforce that they only get > > plugged > > into those buses. I suppose that is not desirable. > > > > Maybe it's better to fix virtio /wrt IOMMUs. > > I always go back to my initial proposal which is to define that current > virtio always bypass any iommu (which is what it does really) and have > it expose via a new capability if that isn't the case. That means fixing > that Xen thingy to allow qemu to know what to expose I assume but that > seems to be the less bad approach. > > Cheers, > Ben. > OK so how does this work? If you want to run an existing guest, you use the old device. And presumably you blacklist virtio for nested virt then, unless a new capability is present? -- MST