Am 11.01.2011 09:53, Gerd Hoffmann wrote:
>   Hi,
> 
>> Actually, there is already a channel to pass pointers to qdev devices:
>> the pointer property hack. I'm not sure we should contribute to its user
>> base or take the chance for a cleanup, but we are not alone with this
>> requirement. Point below remains valid, though.
> 
> It is considered bad/hackish style as you can't create that kind of
> devices using the -device command line switch (or from a machine
> description config file some day in the future).

That kind of instantiation wouldn't be possible for device models that
require someone actively passing kvm_state to them...

>  So we should not add
> more uses of this, especially not in patches which are supposed to
> cleanup things ;)

You won't see me disagree.

Jan