Re: vhost-user (virtio-fs) migration: back end state

[Stefan Hajnoczi]

On Wed, 8 Feb 2023 at 10:58, Hanna Czenczek <hreitz@redhat.com> wrote:
>
> On 08.02.23 15:32, Stefan Hajnoczi wrote:
> > On Tue, 7 Feb 2023 at 07:29, Stefan Hajnoczi <stefanha@gmail.com> wrote:
> >> On Tue, 7 Feb 2023 at 04:08, Hanna Czenczek <hreitz@redhat.com> wrote:
> >>> On 06.02.23 17:27, Stefan Hajnoczi wrote:
> >>>> On Mon, 6 Feb 2023 at 07:36, Hanna Czenczek <hreitz@redhat.com> wrote:
> >>>> Should we schedule a call with Jason, Michael, Juan, David, etc to
> >>>> discuss further? That way there's less chance of spending weeks
> >>>> working on something only to be asked to change the approach later.
> >>> Sure, sounds good!  I’ve taken a look into what state we’ll need to
> >>> migrate already, but I’ll take a more detailed look now so that it’s
> >>> clear what our requirements are.
> > Hi Hanna,
> > The next step is getting agreement on how the vhost-user device state
> > interface will work. Do you want to draft the new vhost-user protocol
> > messages and put together slides for discussion with Michael, Jason,
> > Juan, and David in the next KVM call? That might be the best way to
> > get agreement. Doing it via email is possible too but I guess it will
> > take longer.
> >
> > If you don't want to design the vhost-user protocol changes yourself
> > then someone on this email thread can help with that.
>
> I’ll need to talk about the whole thing to Stefano and German first
> (we’re collaborating on virtio-fs migration, looking at different
> aspects of it).  Also, I think I’ll want to look into the code a bit
> first and fiddle around to get a working prototype so I get an idea of
> what might be feasible at all.  I wouldn’t want to propose something
> that actually can’t work when I try to make it work in practice. O:)

Okay.

The specifics of the virtiofs device state don't need much discussion
here. Only the vhost-user protocol changes need agreement because they
are generic infrastructure that will affect other vhost-user device
types, vDPA, and probably core VIRTIO. So don't worry about defining
what each piece of virtiofs state needs to be at this point, I guess
you only need a rough idea in order to come up with the vhost-user
protocol changes.

Stefan