Re: [PATCH 1/2] xmm7modem: enabling simtoolkit in modem driver

[Denis Kenzior <denkenz@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 1431 bytes --]

Hi Antara,

On 05/15/2019 06:10 AM, Antara Borwankar wrote:
> handled enabling of stk for xmm7modem vendor in atmodem
> driver.
> ---
>   drivers/atmodem/stk.c | 13 +++++++++++++
>   1 file changed, 13 insertions(+)
> 
> diff --git a/drivers/atmodem/stk.c b/drivers/atmodem/stk.c
> index f0dc5c0..43ae953 100644
> --- a/drivers/atmodem/stk.c
> +++ b/drivers/atmodem/stk.c
> @@ -191,6 +191,19 @@ static gboolean at_stk_register(gpointer user)
>   		g_at_chat_register(sd->chat, "*HCMD:", phonesim_hcmd_notify,
>   						FALSE, stk, NULL);
>   
> +	if (sd->vendor == OFONO_VENDOR_XMM) {
> +		/*	enabling stk	*/
> +		g_at_chat_send(sd->chat, "AT+CFUN=6", none_prefix,
> +						NULL, NULL, NULL);
> +		/*	Here ofono has missed stk menu proactive command
> +		 *	that comes after sim initialization only. Doing a
> +		 *	sim reset will enable the stk driver to get the
> +		 *	missed +CUSATP notifications.
> +		 */

This seems really dangerous.  STK is created in post_sim, e.g. after the 
SIM has been detected and initialized.  What happens if you reset it and 
the PIN is locked, etc?  E.g. wouldn't the SIM PIN be needed again?

Would it make more sense to perform a SIM reset as part of the .enable() 
operation?

> +		g_at_chat_send(sd->chat, "AT+CFUN=27,1", none_prefix,
> +						NULL, NULL, NULL);
> +	}
> +
>   	ofono_stk_register(stk);
>   
>   	return FALSE;
> 

Regards,
-Denis