On Thu, Dec 31, 2009 at 08:16:27AM -0600, Anthony Liguori wrote: > On 12/30/2009 06:23 PM, Gleb Natapov wrote: > >I've got you point. When VMState will have this cool powers you are talking > >about I'll use it. I don't see the point of bragging about something > >that doesn't exists. > > It *does* exist. The ioapic has already been converted to VMState. > It *does not*, VMstate has nothing to do with displaying device state as of today. And implementing of general "device state printing facility" that dumps binary data was possible event before VMstate: just call device's save function with fake QEMUFile and dump it in binary. This haven't stopped addition of info functions that understand device internals and print relevant information in human readable form. This has nothing to do with VMState.
-- Gleb.