Am 28.08.2013 12:31, schrieb Alexey Kardashevskiy:
> On 08/27/2013 12:32 AM, Andreas Färber wrote:
>> Am 26.08.2013 15:04, schrieb Alexander Graf:
>>>
>>> On 19.08.2013, at 06:06, Alexey Kardashevskiy wrote:
>>>
>>>> diff --git a/target-ppc/translate_init.c b/target-ppc/translate_init.c
>>>> index 13b290c..dfe398d 100644
>>>> --- a/target-ppc/translate_init.c
>>>> +++ b/target-ppc/translate_init.c
>>>> @@ -3104,7 +3104,6 @@ static int check_pow_hid0_74xx (CPUPPCState *env)
>>>> glue(glue(ppc_, _name), _cpu_family_type_info) = {
>>>> \
>>>> .name = stringify(_name) "-family-" TYPE_POWERPC_CPU,
>>>> \
>>>> .parent = TYPE_POWERPC_CPU,
>>>> \
>>>> - .abstract = true,
>>>> \
>>>> .class_init = glue(glue(ppc_, _name), _cpu_family_class_init),
>>>> \
>>>> };
>>>> \
>>>>
>>>> \
>>
>> This should no longer be necessary (once the fuzzy search is implemented
>> in KVM host type registration code path) and trivially solves the -cpu ?
>> / QMP aspect Alex mentioned above.
>
>
> I do not really understand this part. Will we still need/want
> .abstract==true?
No, we don't need/want it since the host CPU type is non-abstract
already and can inherit from abstract types without problem.
Andreas
> I'll post yet another version of my patch in next 3
> minutes, there I skip PPC CPU classes which parent is TYPE_POWERPC_CPU,
> pretty trivial already :)
--
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer; HRB 16746 AG Nürnberg
