Fwd: Cinterion plugin patch

[Aleksander Morgado]

On Sun, Nov 27, 2016 at 3:47 AM, matthew stanger <stangerm2 at gmail.com> wrote:
> attempt to be as detailed as you
>>
>>
>>  * The PDP context mapping is as suggested by Cinterion, although it looks
>> like
>>  * this isn't strictly enforced by the modem; i.e. SWWAN could work fine
>> with
>>  * any PDP context vs SWWAN interface mapping.
>
>
> It's really stupid but the reason for this mapping is for the PLS-X only,
> specifically when using Verizon Wireless as a carrier. As I interpret it
> this would also mean Verizon wouldn't be able to support dual SIM
> connections either. When I get to the dual SIM patch I'll be sure to keep it
> in mind but for now it's just an FYI.
>

Yes, we need to keep an eye on this. Plus, we should likely try to
test with other Cinterion devices providing ^SWWAN, if any.

>   /* Map PDP context from the current Bearer. USB0 -> 3rd context, USB1 ->
> 1st context.
>      * In Verizon Wireless networks PDP contexts with APN class 1, 2 and 4
> are dedicated to specific profile settings
>      * for LTE (see AT^SGAPN), meaning that <cid> 1, 2 and 4 are not allowed
> to be used for DUN, WWAN (RmNet)
>      * or IP service connections. */
>

Oh, I see what you mean... I don't think this is something we should
hack on the Cinterion plugin then... If the mapping suggested is only
due to how Verizon needs things, we should take another more generic
approach. Is there some online documentation or reference stating the
requirements from Verizon in this regard?

>>     switch (ctx->step) {
>>
>>     case CONNECT_3GPP_CONTEXT_STEP_FIRST: {
>>
>>         MMBearerIpFamily ip_family;
>
>
>
> Just curious why we moved this back to the 1st step? I thought you didn't
> want to create MBearerIpConfig before knowing if the modem is connected?

This is just validating supported IP families IIRC, not creating the
MMBearerIpConfig.


-- 
Aleksander
https://aleksander.es