Re: Ofono with Sagem Hilo GPRS

[Denis Kenzior <denkenz@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 2487 bytes --]

Hi Andrew,

> Hi,
> 
> On 1 June 2010 21:11, Denis Kenzior <denkenz@gmail.com> wrote:
> > Hi Yoann,
> >
> > <snip>
> >
> >> ofonod[3221]: src/modem.c:get_modem_property() modem 0x8142058 property
> >>  status-poll-interval ofonod[3221]: > AT+CSIM=8,A0F200C0\r
> >> ofonod[3221]: < AT+CSIM=8,A0F200C0\r\r\n+CME ERROR: 3\r\n
> >
> > The culprit was sim polling which performs CSIM pdus to detect a dead
> > SIM. Your modem doesn't seem to support CSIM command, so returns an error
> > and oFono foolishly thinks the SIM has been removed.
> 
> Not if the +CME ERROR is sent withing 5 seconds.  Only if there is a 5
> second or longer lag between the AT+CSIM=... and the response (success
> or error), it decides the card is dead.

Thanks for the correction.  I was hitting a slightly different issue with HSO 
modem and atgen that was also resulting in sim removal.

Looking at the logs again, it seems that this response causes the poll to 
signal sim removal:

ofonod[3253]: > AT+CRSM=192,28618\r
ofonod[3253]: < A
ofonod[3253]: < T+CRSM=192,28618\r
ofonod[3253]: < \000
ofonod[3253]: src/message-waiting.c:mw_remove() atom: 0x9ac6c00
ofonod[3253]: src/phonebook.c:phonebook_remove() atom: 0x9ac6a98
ofonod[3253]: src/sms.c:sms_remove() atom: 0x9ac69c0
ofonod[3253]: src/ssn.c:ssn_remove() atom: 0x9ac68f0
ofonod[3253]: src/call-barring.c:call_barring_remove() atom: 0x9ac6880
ofonod[3253]: src/call-meter.c:call_meter_remove() atom: 0x9ac6780
ofonod[3253]: src/network.c:netreg_remove() atom: 0x9ac62e8
ofonod[3253]: src/call-settings.c:call_settings_remove() atom: 0x9ac64f8
ofonod[3253]: src/call-forwarding.c:call_forwarding_remove() atom: 0x9ac6418
ofonod[3253]: src/ussd.c:ussd_remove() atom: 0x9ac4e70
ofonod[3253]: src/modem.c:set_modem_property() modem 0x9ac4058 property 
status-poll-interval
ofonod[3253]: src/modem.c:unregister_property() property 0x9ac49d8

Seems we try to read EFmwis and get back a '\0' as a response.  This probably 
stalls the gatchat queue and causes the sim poll timeout to fire.  Or is there 
something else going on I'm not seeing?

There is another issue:
When we send the CPIN to unlock the PIN, the modem actually takes some time to 
initialize the SIM and our CIMI request fails.  This means that we don't 
proceed with the initialization procedure.  Disabling / Enabling the modem 
skips the PIN entry and we initialize properly.

This modem's firmware is wonderful :)

Regards,
-Denis