On Thu, Dec 31, 2009 at 01:05:18PM -0600, Anthony Liguori wrote:
> On 12/31/2009 09:42 AM, Gleb Natapov wrote:
> >On Thu, Dec 31, 2009 at 08:20:06AM -0600, Anthony Liguori wrote:
> >>On 12/30/2009 06:20 PM, Gleb Natapov wrote:
> >>>I included only the state I need for debugging. I don't what to see
> >>>complete state. It will just clatter important info.
> >>
> >>Someone may find that full state useful though. There's no good
> >>reason to not show the entire device's state.
> >>
> >There is no good reason to not show the entire device's state in
> >addition to nicely formated most useful part of it. Here I fixed it for
> >you.
>
>
> Have you even attempted to look at what the generic implementation
> would be? ioapic has three vmstate fields. If you decode ioredtbl
> into six subfields, then you're only adding two addition fields to
> be printed.
>
I you seriously suggesting that we should tie the way device state is
migrated to the way device information is printed?
> I can't believe that having those two extra fields is really going
> to make it any more difficult to debug.
>
Can't parse that.
> And being able to dump any device state is certainly going to be
> useful in the future.
>
No arguing with that.
--
Gleb.
