Re: [PATCH] telit: signal sim inserted when sim ready.

[Christopher Vogl <christopher.vogl@hale.at>]

[-- Attachment #1: Type: text/plain, Size: 4229 bytes --]

On 17/12/12 16:50, Denis Kenzior wrote:
> Hi Christopher,
>
> On 12/14/2012 08:19 AM, christopher.vogl(a)hale.at wrote:
>> From: Christopher Vogl<christopher.vogl@hale.at>
>>
>> We need to wait for #QSS: 3 until SIM inserted may be signaled or
>> subsequent commands like +CMER will report SIM BUSY which in turn
>> leads to a removal of the network registration atom. This has been
>> observed with Telit HE910.
>>
>> In case of #QSS: 1 we need to check if the SIM is locked.
>> If the SIM is unlocked we just wait for #QSS: 3,
>> if the SIM is not unlocked we signal SIM inserted to be able to
>> enter a PIN code.
>> ---
>>   plugins/telit.c |   43 +++++++++++++++++++++++++++++++++++++++++--
>>   1 files changed, 41 insertions(+), 2 deletions(-)
>>
>> diff --git a/plugins/telit.c b/plugins/telit.c
>> index fe2ccd6..cd67ef4 100644
>> --- a/plugins/telit.c
>> +++ b/plugins/telit.c
>> @@ -61,6 +61,7 @@
>>   #include "bluetooth.h"
>>
>>   static const char *none_prefix[] = { NULL };
>> +static const char *cpin_prefix[] = { "+CPIN:", NULL };
>>   static const char *rsen_prefix[]= { "#RSEN:", NULL };
>>
>>   struct telit_data {
>> @@ -211,6 +212,33 @@ static GAtChat *open_device(struct ofono_modem 
>> *modem,
>>       return chat;
>>   }
>>
>> +static void at_cpin_cb(gboolean ok, GAtResult *result, gpointer 
>> user_data)
>> +{
>> +    struct ofono_modem *modem = user_data;
>> +    struct telit_data *data = ofono_modem_get_data(modem);
>> +    GAtResultIter iter;
>> +    const char *pin_required;
>> +
>> +    if (!ok)
>> +        return;
>> +
>> +    g_at_result_iter_init(&iter, result);
>> +
>> +    if (!g_at_result_iter_next(&iter, "+CPIN:"))
>> +        return;
>> +
>> + g_at_result_iter_next_unquoted_string(&iter,&pin_required);
>> +
>> +    // No PIN required, we can wait for #QSS: 3.
>
> Nitpicking, but we don't use // style comments.
>
>> +    if (strcmp(pin_required, "READY")==0)
>> +        return;
>> +
>> +    if (data->have_sim == FALSE) {
>> +        ofono_sim_inserted_notify(data->sim, TRUE);
>> +        data->have_sim = TRUE;
>> +    }
>> +}
>> +
>>   static void switch_sim_state_status(struct ofono_modem *modem, int 
>> status)
>>   {
>>       struct telit_data *data = ofono_modem_get_data(modem);
>> @@ -226,13 +254,24 @@ static void switch_sim_state_status(struct 
>> ofono_modem *modem, int status)
>>           }
>>           break;
>>       case 1:    /* SIM inserted */
>> +        /*
>> +         * We want to call ofono_sim_inserted_notify() only if #QSS: 3
>> +         * (SIM READY), as otherwise subsequent commands like +CMER may
>> +         * report SIM BUSY.
>> +         * In case a PIN is required, we have to signal
>> +         * sim inserted on #QSS: 1 and wait in the SIM atom for #QSS: 3
>> +         * after entering the PIN.
>> +         */
>> +        g_at_chat_send(data->chat, "AT+CPIN?", cpin_prefix,
>> +                at_cpin_cb, modem, NULL);
>> +        break;
>>       case 2:    /* SIM inserted and PIN unlocked */
>> +        break;
>> +    case 3:    /* SIM inserted and READY (SMS&  Phonebook access 
>> possible) */
>>           if (data->have_sim == FALSE) {
>>               ofono_sim_inserted_notify(data->sim, TRUE);
>>               data->have_sim = TRUE;
>>           }
>> -        break;
>> -    case 3:    /* SIM inserted, SMS and phonebook ready */
>>           if (data->sms_phonebook_added == FALSE) {
>>               ofono_phonebook_create(modem, 0, "atmodem", data->chat);
>>               ofono_sms_create(modem, 0, "atmodem", data->chat);
>
> Since we are waiting for QSS:3, should phonebook and sms atom creation 
> be moved out of here and simply put into post_sim()?
>

Ok, I will put the creation of those atoms back into post_sim() so it is 
consistent with other plugins again.

I also realised that we only get #QSS: 1 iff a PIN is required, i.e. the 
modem firmware is doing the +CPIN query implicitly for us.
If there is no PIN required we always get #QSS: 2 (SIM UNLOCKED) instead 
of #QSS: 1.
Better leave the PIN querying or would you remove it?

Regards,
Christopher


--
Scanned by MailScanner.