From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:44362) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1c9P8y-0007p7-Nt for qemu-devel@nongnu.org; Tue, 22 Nov 2016 23:26:49 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1c9P8t-0002dY-PA for qemu-devel@nongnu.org; Tue, 22 Nov 2016 23:26:48 -0500 Received: from mx1.redhat.com ([209.132.183.28]:48404) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1c9P8t-0002dA-Ji for qemu-devel@nongnu.org; Tue, 22 Nov 2016 23:26:43 -0500 Date: Wed, 23 Nov 2016 06:26:41 +0200 From: "Michael S. Tsirkin" Message-ID: <20161123062418-mutt-send-email-mst@kernel.org> References: <0105e7e3-6003-04c8-483d-30ed1208e5fc@redhat.com> <20161121182040-mutt-send-email-mst@kernel.org> <38ee6d4c-ff87-d8cc-1a1d-f45b948dcdb3@redhat.com> <796ff318-c3e9-e885-f345-7f6e48d3bb8b@redhat.com> <20161122163413-mutt-send-email-mst@kernel.org> <2d971a18-4253-1bac-4098-6879c9ccc01f@redhat.com> <20161122223744-mutt-send-email-mst@kernel.org> <5cbd7441-a1d1-7e34-56ff-33c2305736d4@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <5cbd7441-a1d1-7e34-56ff-33c2305736d4@redhat.com> Subject: Re: [Qemu-devel] [RFC v2 0/3] virtio-net: Add support to MTU feature List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Jason Wang Cc: Maxime Coquelin , yuanhan.liu@linux.intel.com, pbonzini@redhat.com, Aaron Conole , qemu-devel@nongnu.org On Wed, Nov 23, 2016 at 11:42:52AM +0800, Jason Wang wrote: > > > > > > > > Seems to me like an easy way to get out of sync. > > > > > > > > > > > >If we send it to the backend, that has a chance to check > > > > > >mtu and disconnect on error. > > > > > > > >For vhost-user backend, we can send it the MTU value with a > > > >vhost-user protocol feature. > > > > > > > >For tun/macvtap, how do you do without adding a new ioctl ? > > Have management configure same mtu on the backend and in qemu. > > > > > > Then why not do same for vhost-user (instead of using two different > methods)? That's what I'm saying. If backend supports that, we can also check the mtu in some way to make sure it matches. -- MST