Re: [PATCH 01/11] qdev/qbus: add hidden device support

[Jens Freimann <jfreimann@redhat.com>]

On Mon, Oct 21, 2019 at 01:53:58PM +0100, Peter Maydell wrote:
>On Fri, 18 Oct 2019 at 21:22, Jens Freimann <jfreimann@redhat.com> wrote:
>>
>> This adds support for hiding a device to the qbus and qdev APIs.  The
>> first user of this will be the virtio-net failover feature but the API
>> introduced with this patch could be used to implement other features as
>> well, for example hiding pci devices when a pci bus is powered off.
>
>In what sense is a hidden device hidden? Is it hidden from
>the guest (in what way) ? Is it hidden from the QMP/HMP monitor
>commands? Is it hidden from QEMU functions that iterate through
>the qbus, or that walk the QOM tree ? What does a hidden device
>have to do to be successfully hidden ? Is it completely disconnected
>from the rest of the simulation, or does it itself have to avoid
>doing things like asserting IRQ lines ?
>
>Expanding the DeviceClass doc comment might be a good place to answer
>questions like the above and generally describe the 'hidden device'
>mechanism.

ok, I will add to the DeviceClass comment. 

>
>>  };
>>
>> @@ -451,4 +457,6 @@ static inline bool qbus_is_hotpluggable(BusState *bus)
>>  void device_listener_register(DeviceListener *listener);
>>  void device_listener_unregister(DeviceListener *listener);
>>
>> +bool qdev_should_hide_device(QemuOpts *opts);
>
>New globally visible functions should always have doc-comment
>format documentation comments, please.

Will add this too. Thanks for the review!

regards,
Jens