[Qemu-devel] Re: The State of the SaveVM format

["Michael S. Tsirkin" <mst@redhat.com>]

On Thu, Sep 10, 2009 at 02:55:07PM +0200, Juan Quintela wrote:
> "Michael S. Tsirkin" <mst@redhat.com> wrote:
> > On Wed, Sep 09, 2009 at 09:02:07PM -0500, Anthony Liguori wrote:
> >>>   If we complicate vmstate now to shoehorn pre-vmstate
> >>> formats into vmstate, that ancient history will continue to haunt us.
> >>> Complicating a program is far easier than the other direction.
> >>>   
> >>
> >> Let's take it one step at a time.  There is an awful lot of areas where  
> >> we can support older versions without adding complications.  Let's  
> >> approach the complicated ones one at a time.
> >
> > I'm not sure I understand this talk about "pre-vmstate formats".
> > I thought vmstate patches were, at least for the most part, trying
> > to reimplement existing format with the table-driver design?
> >
> > If that's not so and we are changing the format now, is it too late to
> > consider some standard serialization format rather than rolling our own?
> 
> We are using previous format.  At some point we should move to other
> format.  When/what is still not decided.  Each time at its time.

Ah, that's what I thought.

> Once
> we have everything using vmstate, we have a declarative description of
> the state.  Going for tables with names + types to any format is just an
> exercise of walking the tables and writing a pretty-printer and a parser.
> 
> Later, Juan.

Good point. But if we do intend to switch formats because of vmstate and
separately switch to a standard format, it might be easier on users if
we do a single switch, in the same release cycle.

-- 
MST