Paolo Bonzini <[email protected]> writes:
> On 13/03/21 10:35, Markus Armbruster wrote:
>>> + object_option_foreach_add(object_create_late, &error_fatal);
>> Likewise.
>>
>> Before and after the patch, object_create_late(opt) ==
>> !object_create_early(opt). Your patch provides us with an opportunity
>> to assert this: object_opts must be empty now. Suggestion, not demand.
>>
>
> Nice idea, I'll include it.
Possible alternative:
static bool object_create_late(const char *type)
{
return true;
}
Your choice.
