From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1MDlIN-0000JP-5U for qemu-devel@nongnu.org; Mon, 08 Jun 2009 16:18:15 -0400 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1MDlII-0000FL-JU for qemu-devel@nongnu.org; Mon, 08 Jun 2009 16:18:14 -0400 Received: from [199.232.76.173] (port=45217 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1MDlII-0000FD-HO for qemu-devel@nongnu.org; Mon, 08 Jun 2009 16:18:10 -0400 Received: from wf-out-1314.google.com ([209.85.200.175]:20987) by monty-python.gnu.org with esmtp (Exim 4.60) (envelope-from ) id 1MDlII-0002TT-4u for qemu-devel@nongnu.org; Mon, 08 Jun 2009 16:18:10 -0400 Received: by wf-out-1314.google.com with SMTP id 26so1399611wfd.4 for ; Mon, 08 Jun 2009 13:18:08 -0700 (PDT) MIME-Version: 1.0 Sender: alex.l.williamson@gmail.com In-Reply-To: <4A2D63EC.9040003@codemonkey.ws> References: <20090605204647.3355.81929.stgit@kvm.aw> <20090605204718.3355.28647.stgit@kvm.aw> <20090606204845.GC26877@redhat.com> <7162ab20906081201y4c598899mdfd5d42c42e17038@mail.gmail.com> <4A2D63EC.9040003@codemonkey.ws> Date: Mon, 8 Jun 2009 14:18:07 -0600 Message-ID: <7162ab20906081318g730eaf82pb526913b8b4c5d30@mail.gmail.com> Subject: Re: [Qemu-devel] [PATCH 6/7] virtio-net: Add new RX filter controls From: Alex Williamson Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Anthony Liguori Cc: qemu-devel@nongnu.org, "Michael S. Tsirkin" On Mon, Jun 8, 2009 at 1:18 PM, Anthony Liguori wrot= e: > Alex Williamson wrote: >> >> e1000 also allows the driver to selectively enable/disable RX of >> packets to the broadcast address. =C2=A0This is replicated with the >> all/no-bcast options. =C2=A0Finally, there may be cases where we want to >> receive only unicast or only multicast address for special purpose >> network devices. =C2=A0This is provided by the nouni and nomulti options= . >> A proprietary guest know as DMX intends to make use of these extra >> modes. =C2=A0Are there any other interesting, useful and lightweight pac= ket >> filters we could implement? =C2=A0Thanks, >> > > I've been thinking about whether doing VLAN filtering/tagging within QEMU > would make sense. =C2=A0It could potentially simplify bridge setups treme= ndously. > =C2=A0Today, if you want to isolate VMs on separate vlans, it involves cr= eating > multiple bridges which gets ugly quickly. IIRC, you have to be careful that the host NIC doesn't strip the VLAN tag itself, which means you want the VLAN guests on a non-VLAN bridge. It's all rather confusing and I wouldn't be surprised if there's some dependency on how much offloading the host NIC does. We do have VLAN filtering in virtio-net today, but of course it would be better if it was done at a generic network level in QEMU or pushed deeper into the host. Tagging doesn't make much sense at the level we're doing it now. Alex