Re: [PATCH v6 2/9] virtio-pci:decouple virtqueue from interrupt setting process

[Jason Wang <jasowang@redhat.com>]

在 2021/4/27 下午2:40, Jason Wang 写道:
>>   static int virtio_pci_vq_vector_unmask(VirtIOPCIProxy *proxy,
>>                                          unsigned int queue_no,
>>                                          unsigned int vector,
>> -                                       MSIMessage msg)
>> +                                       MSIMessage msg,
>> +                                        EventNotifier *n)
>
>
> So you switch to use EventNotifier but keep using queue_no/vector, 
> this looks kind of duplication.
>
> If we can keep queue_no or virtio_queue_get_guest_notifier working as 
> in the past, I don't see any reason for this effort.


Ok I see it is because we need to support config interrupt.

But using queue_no/vector/n at the same time is a hint that the decouple 
is not done completely.

Thanks


>
> Thanks