Re: [Qemu-devel] [PATCH 08/11] QMP: Asynchronous messages enable/disable support

[Anthony Liguori <aliguori@linux.vnet.ibm.com>]

On 01/22/2010 02:09 PM, Luiz Capitulino wrote:
> On Fri, 22 Jan 2010 12:05:19 -0600
> Anthony Liguori<anthony@codemonkey.ws>  wrote:
>
>    
>> On 01/21/2010 03:09 PM, Luiz Capitulino wrote:
>>      
>>> This commit disables asynchronous messages by default and
>>> introduces two new QMP commands: async_msg_enable and
>>> async_msg_disable.
>>>
>>> Each QMP Monitor has its own set of asynchronous messages,
>>> so for example, if QEMU is run with two QMP Monitors async
>>> messages setup in one of them doesn't affect the other.
>>>
>>> To implement this design a bitmap is introduced to the
>>> Monitor struct, each async message is represented by one bit.
>>>
>>> Signed-off-by: Luiz Capitulino<lcapitulino@redhat.com>
>>>
>>>        
>> Ah, I see I was a little confused.
>>
>> I'd suggest making async message masking an independent mechanism.
>> Capabilities should strictly deal with protocol changes, not feature
>> details.
>>      
>   To summarize (after a IRC talk): async messages are considered a
> capability and should be enabled during the negotiation phase but
> the masking of particular messages are not and can be done at
> any time after the negotiation phase.
>    

Just to further clarify, the mental model I have of capabilities is that 
they are things that affect the protocol itself.  Additional features 
(new command options, new commands, new async messages) are things that 
are enabled/discovered in a different way.

Async messages themselves are a capability since it changes the 
protocol.  Timestamps would be another capability and a new data type 
(like an uint64 type) would be a new capability.

>   I'm ok with that, Markus?
>    

Regards,

Anthony Liguori