On Wed, Dec 4, 2013 at 2:56 PM, Stefan Hajnoczi <stefa...@gmail.com> wrote:
> On Fri, Nov 29, 2013 at 08:52:21PM +0100, Antonios Motakis wrote: > > In this patch series we would like to introduce our approach for putting > a > > virtio-net backend in an external userspace process. Our eventual target > is to > > run the network backend in the Snabbswitch ethernet switch, while > receiving > > traffic from a guest inside QEMU/KVM which runs an unmodified virtio-net > > implementation. > > > > For this, we are working into extending vhost to allow equivalent > functionality > > for userspace. Vhost already passes control of the data plane of > virtio-net to > > the host kernel; we want to realize a similar model, but for userspace. > > > > In this patch series the concept of a vhost-backend is introduced. > > > > We define two vhost backend types - vhost-kernel and vhost-user. The > former is > > the interface to the current kernel module implementation. Its control > plane is > > ioctl based. The data plane is the kernel directly accessing the QEMU > allocated, > > guest memory. > > > > In the new vhost-user backend, the control plane is based on > communication > > between QEMU and another userspace process using a unix domain socket. > This > > allows to implement a virtio backend for a guest running in QEMU, inside > the > > other userspace process. > > One thing that came to mind when reading the patches is that you are > implementing the vhost interface pretty much exactly as-is. Did you > look at FUSE's character devices in userspace (CUSE)? IIRC even ioctl > is supported so you might be able to skip the userspace backend entirely > if you mimic vhost_net.ko's ioctl interface. > > Then all that's needed is some configuration/startup code to use shared > memory and pass the eventfds. > > Stefan > Using UNIX domain sockets we can easily exchange the file descriptors we need, I don't know if this is the case with CUSE. Even though we reimplement ioctls via the UNIX domain socket, those are simple enough so I am also not sure CUSE would actually make things simpler. Even so, the vhost server is not really a device to be represented as a character device. I don't know if it is the right approach to do it like that. Antonios
