Am 13.03.2012 13:20, schrieb Paolo Bonzini:
> Il 13/03/2012 13:13, Andreas Färber ha scritto:
>>> It will be easier to generalize later qdev code and not make special
>>> case when adding cpus.
>>
>> I never heard anyone wanting to generalize reset so far. I don't think
>> it belongs into Object at least. Maybe DeviceState. Anthony? Paolo?
>
> I believe long term we want CPUs to become a DeviceState. For now, I
> think Andreas's prototype is fine.
I have prepared $(qom-obj-twice-y) to allow for:
#ifdef CONFIG_SOFTMMU
.parent = TYPE_DEVICE, // or TYPE_SYS_BUS_DEVICE
#else
.parent = TYPE_OBJECT,
#endif
So far it was not needed. :)
> Methods should not take a superclass
> argument in general.
So to clarify, this is pro CPUState?
>> This series is taking much too long to move forward (the QOM "steam"
>> seems to be gone?) and I'm worried that introducing much more basic
>> infrastructure will make review and applying even slower, cf.
>> object_class_foreach_ordered()/_get_list().
>
> Agreed, this series looks more or less good (and mostly mechanical
> anyway).
Thanks.
> Is it an RFC or what? :) I wonder if reviewers are put off by
> the subject.
The implied RFC is, are we okay with reusing "CPUState" this way? Or
does someone - last call! - have a better identifier name?
Getting this series merged either means coordinating the PULL with a
maintainer so that no merge conflicts arise in-flight, or having the
maintainer re-run the commit-creating script himself.
Andreas
--
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer; HRB 16746 AG Nürnberg
