From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Michael S. Tsirkin" Subject: Re: [PATCH 2/2] vhost_net: a kernel-level virtio server Date: Thu, 13 Aug 2009 11:45:13 +0300 Message-ID: <20090813084512.GA4401@redhat.com> References: <20090810185340.GC13924@redhat.com> <200908121903.22325.arnd@arndb.de> <20090812172141.GA29966@redhat.com> <4A83167E.2080701@codemonkey.ws> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Arnd Bergmann , virtualization@lists.linux-foundation.org, "Ira W. Snyder" , netdev@vger.kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org To: Anthony Liguori Return-path: Content-Disposition: inline In-Reply-To: <4A83167E.2080701@codemonkey.ws> Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On Wed, Aug 12, 2009 at 02:22:38PM -0500, Anthony Liguori wrote: > Michael S. Tsirkin wrote: >>> >>> We discussed this before, and I still think this could be directly derived >>> from struct virtqueue, in the same way that vring_virtqueue is derived from >>> struct virtqueue. >>> >> >> I prefer keeping it simple. Much of abstraction in virtio is due to the >> fact that it needs to work on top of different hardware emulations: >> lguest,kvm, possibly others in the future. vhost is always working on >> real hardware, using eventfd as the interface, so it does not need that. >> > > Actually, vhost may not always be limited to real hardware. Yes, any ethernet device will do. What I mean is that vhost does not deal with emulation at all. All setup is done in userspace. > > We may on day use vhost as the basis of a driver domain. There's quite > a lot of interest in this for networking. You can use veth for this. This works today. > At any rate, I'd like to see performance results before we consider > trying to reuse virtio code. > > Regards, > > Anthony Liguori