Re: [Qemu-devel] [PATCH 22/28] virtio-serial: convert to QEMU Object Model

[Anthony Liguori <aliguori@us.ibm.com>]

On 01/24/2012 01:59 PM, Amit Shah wrote:
> On (Tue) 24 Jan 2012 [13:33:14], Anthony Liguori wrote:
>
>> diff --git a/hw/virtio-serial-bus.c b/hw/virtio-serial-bus.c
>> index 982ffbf..edf54d7 100644
>> --- a/hw/virtio-serial-bus.c
>> +++ b/hw/virtio-serial-bus.c
>
>
>> @@ -157,7 +157,7 @@ static void do_flush_queued_data(VirtIOSerialPort *port, VirtQueue *vq,
>>               ssize_t ret;
>>
>>               buf_size = port->elem.out_sg[i].iov_len - port->iov_offset;
>> -            ret = info->have_data(port,
>> +            ret = vsc->have_data(port,
>>                                     port->elem.out_sg[i].iov_base
>>                                     + port->iov_offset,
>>                                     buf_size);
>
> Would be nice if everything was re-indented.

That'll need to be a follow-on.  This was mostly scripted.

>
>> @@ -176,7 +176,7 @@ static void do_flush_queued_data(VirtIOSerialPort *port, VirtQueue *vq,
>>                    * 1: chardevs can notify frondends
>>                    * 2: the guest driver does not spin in these cases
>>                    */
>> -                if (!info->is_console) {
>> +                if (vsc->is_console) {
>>                       virtio_serial_throttle_port(port, true);
>>                   }
>>                   port->iov_idx = i;
>
> This change is wrong, reverses the if condition.  Was this done
> manually or via a script?  If the latter, it would be worthwhile to
> check all the conversions.

This was a manual rebase conflict resolution change on my part.  I'll fix.

>
>> -void virtio_serial_port_qdev_register(VirtIOSerialPortInfo *info)
>> +void virtio_serial_port_qdev_register(DeviceInfo *info)
>
> Should the fn be renamed (replace qdev)?

It'll go away entirely in 3/3 which I'll post in a couple days.

Regards,

Anthony Liguori

>
> 		Amit
>