From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([208.118.235.92]:54960) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1TsZhG-0002KY-61 for qemu-devel@nongnu.org; Tue, 08 Jan 2013 08:58:36 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1TsZhD-0004vJ-Cg for qemu-devel@nongnu.org; Tue, 08 Jan 2013 08:58:30 -0500 Received: from mx1.redhat.com ([209.132.183.28]:16318) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1TsZhD-0004vB-4c for qemu-devel@nongnu.org; Tue, 08 Jan 2013 08:58:27 -0500 Date: Tue, 8 Jan 2013 16:02:05 +0200 From: "Michael S. Tsirkin" Message-ID: <20130108140205.GB23732@redhat.com> References: <1354887155-32281-1-git-send-email-fred.konrad@greensocs.com> <20121217154508.GA28712@redhat.com> <20121218110153.GC22586@redhat.com> <50D053CC.9040203@greensocs.com> <20130107195843.GC8679@redhat.com> <50EBED66.6080606@greensocs.com> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline In-Reply-To: <50EBED66.6080606@greensocs.com> Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [RFC PATCH v6 0/6] Virtio refactoring. List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: KONRAD =?iso-8859-1?Q?Fr=E9d=E9ric?= Cc: Peter Maydell , aliguori@us.ibm.com, e.voevodin@samsung.com, mark.burton@greensocs.com, qemu-devel@nongnu.org, stefanha@redhat.com, cornelia.huck@de.ibm.com, afaerber@suse.de On Tue, Jan 08, 2013 at 10:56:54AM +0100, KONRAD Fr=E9d=E9ric wrote: > On 07/01/2013 20:58, Michael S. Tsirkin wrote: > >On Tue, Dec 18, 2012 at 12:30:20PM +0100, KONRAD Fr=E9d=E9ric wrote: > >>On 18/12/2012 12:01, Michael S. Tsirkin wrote: > >>>On Tue, Dec 18, 2012 at 10:33:37AM +0000, Peter Maydell wrote: > >>>>On 17 December 2012 15:45, Michael S. Tsirkin wrot= e: > >>>>>Is the point to allow virtio-mmio? Why can't virtio-mmio be just > >>>>>another bus, like a pci bus, and another binding, like the virtio-= pci > >>>>>binding? > >>>>(a) the current code is really not very nice because it's not > >>>>actually a proper set of QOM/qdev devices > >>>>(b) unlike PCI, you can't create sysbus devices on the > >>>>command line, because they don't correspond to a user > >>>>pluggable bit of hardware. We don't want users to have to know > >>>>an address and IRQ number for each virtio-mmio device (especially > >>>>since these are board specific); instead the board can create > >>>>and wire up transport devices wherever is suitable, and the > >>>>user just creates the backend (which is plugged into the virtio bus= ). > >>>> > >>>>-- PMM > >>>This is what I am saying: create your own bus and put > >>>your devices there. Allocate resources when you init > >>>a device. > >>> > >>>Instead you seem to want to expose a virtio device as two devices to > >>>user - if true this is not reasonable. > >>> > >>The modifications will be transparent to the user, as we will keep > >>virtio-x-pci devices. > >Then what's the point of all this? > > > >-device virtio-pci,id=3Dtransport1 -device virtio-net,bus=3Dtransport1 > > > >or > > > >-device virtio-mmio,id=3Dtransport1 -device virtio-net,bus=3Dtransport= 1 > > > >Is simply an insane way to create a network device. > > > To recap : >=20 > The idea is to have a virtio-bus between the transport device > (like pci, mmio,... ). >=20 > Then we can have a platform with several virtio-mmio and then > virtio-bus slot. >=20 > At the end user can add a virtio-device in the command line with -devic= e > parameter without recompiling the platform. That is not possible with j= ust > creating the virtio-x-mmio devices. The bus=3D option can be used to > select the bus slot, but I'm not sure it is usefull. pci uses addr option for this, I am guessing mmio can do the same. > The series keep the virtio-x-pci devices : > eg : step 11/61 for virtio-blk > So -device virtio-blk-pci or -device virtio-blk-s390 works as before. >=20 > Of course -device virtio-pci,id=3Dtransport1 -device > virtio-net,bus=3Dtransport1 is > possible but why using this command line when we could simply do : > -device virtio-net-pci ? >=20 > Fred Adding multiple ways to do one thing is a bad idea. I'm fine with modeling virtio by multiple devices internally, but exposing this to user is a mistake, we might rework this even more and command line has to be supported indefinitely.