Re: [PATCH v2 2/6] gatchat: Add g_at_chat_send_full.

[Denis Kenzior <denkenz@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 1445 bytes --]

Hi Andrew,

> Hi Denis,
> 
> On 10 May 2010 21:13, Denis Kenzior <denkenz@gmail.com> wrote:
> >> g_at_chat_send_full takes two additional parameters compared to
> >> g_at_chat_send.  The destroy notification is removed from
> >> g_at_chat_send and users updated.
> >> ---
> >
> > You stated earlier that the modem will return an error and not time out.
> >  If this is the case then I'm really failing to see the point of needing
> > the submit callback.
> 
> I stated that as en example of an incompliant modem.  The specs are
> not very clear here but say that we should assume the modem is dead if
> no response to the AT+CSIM is sent within a timeout of at least 5
> seconds.

The modem or the SIM?  I think the specs are talking about the SIM here and 
the modems do need to poll and detect SIM dead conditions.  If we were to 
implement a GSM soft-modem, then we'd definitely need to poll...

> 
> My thinking so far was that ofono should support an (imaginary) modem
> that's described by the specs, and on top of that add support for
> vendor specific things, but maybe that is useless in this case.  What
> do you think?

I think we should drop this capability until we actually see this 'imaginary' 
modem you describe in the wild.  Most vendors intercept everything related to 
SIM / STK and can even do hardware SIM detection.  Makes no sense for us to 
worry about it right now.

Regards,
-Denis