From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1MOAoS-00082U-7W for qemu-devel@nongnu.org; Tue, 07 Jul 2009 09:34:24 -0400 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1MOAoN-00081u-Ls for qemu-devel@nongnu.org; Tue, 07 Jul 2009 09:34:23 -0400 Received: from [199.232.76.173] (port=38913 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1MOAoN-00081r-Es for qemu-devel@nongnu.org; Tue, 07 Jul 2009 09:34:19 -0400 Received: from fwil.voltaire.com ([193.47.165.2]:20182 helo=exil.voltaire.com) by monty-python.gnu.org with esmtp (Exim 4.60) (envelope-from ) id 1MOAoN-0000m3-08 for qemu-devel@nongnu.org; Tue, 07 Jul 2009 09:34:19 -0400 Message-ID: <4A534EC4.5030209@voltaire.com> Date: Tue, 07 Jul 2009 16:33:56 +0300 From: Or Gerlitz MIME-Version: 1.0 Subject: Re: [Qemu-devel] [PATCH] net: add raw backend References: <20090701162115.GA4555@shareable.org> <4A4CA747.1050509@Voltaire.com> <20090703023911.GD938@shareable.org> In-Reply-To: <20090703023911.GD938@shareable.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Jamie Lokier , Anthony Liguori Cc: Herbert Xu , Jan Kiszka , qemu-devel@nongnu.org Jamie Lokier wrote: > No, I meant "-net pcap", whose patch is announced at > http://lists.freebsd.org/pipermail/freebsd-emulation/2007-February/003108.html > and found here: http://people.freebsd.org/~jkim/patch-qemu-pcap.diff I saw it mention on this list in the last few months. It sends and receives packets over a host network interface using libpcap. okay, looking a bit on the archives I realized that the -net pcap backend was suggested on the qemu-devel list at least twice, on 2007 (your pointer) and recently, e.g @ http://lists.gnu.org/archive/html/qemu-devel/2009-03/msg00895.html and the mail threads that followed from March, April and May 2009. Under Linux the pcap library uses a packet socket, so basically it could make some sense to go through libpcap and not directly to sockets, but there are also some disadvantages which will not less qemu implement some related optimizations which are not integrated into libpcap. Now, before going into libpcap vs. packet socket, I'd be happy if you or Antony can help me understand the comments that with this approach guest <--> host communication is impossible. AFAIK, the only packet sent by Qemu is gratuitous ARP after migration, but it doesn't fall into guest <--> host communication over the NIC/vlan/back-end gang, so I don't see what is this traffic which is impossible with the bridge-less approach. Also, if for some reason one need to communicate from the guest to the host the -net raw can instructed to run over a veth Linux interface couple which are connected to a bridge. Or.