Re: [PATCH 2/4] Core support for packet switched bearer reporting

[Denis Kenzior <denkenz@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 1231 bytes --]

Hi Rémi,

On 01/10/2011 05:19 AM, Rémi Denis-Courmont wrote:
> On Friday 07 January 2011 18:25:32 ext Denis Kenzior, you wrote:
>>> @@ -2001,6 +2010,29 @@ void ofono_gprs_add_context(struct ofono_gprs
>>> *gprs,
>>>
>>>  	__ofono_atom_register(gc->atom, gprs_context_unregister);
>>>  
>>>  }
>>>
>>> +void ofono_gprs_bearer_notify(struct ofono_gprs *gprs,
>>> +				unsigned int cid, int bearer)
>>> +{
>>> +	DBusConnection *conn = ofono_dbus_get_connection();
>>> +	GSList *l;
>>> +	const char *value = packet_bearer_to_string(bearer);
>>
>> Why bother running this operation before you're sure you're going to
>> emit the signal?
> 
> Now that we're talking about network-initiated LTE bearers, I would rather be 
> safe than sorry. I wouldn't assume that the modem will signal the CPSB 
> (unchanged) value later in this case.
> 

Are we both talking about the packet_bearer_to_string call?

>> Please move this down.
> 
> To avoid any race, we have to query the existing CPSB values after we enable 
> the CPSB subscription, right? We could surely move that to a later stage, but 
> I don't see the benefit.
> 

What race?  I don't think you really read my comments...

Regards,
-Denis