MS basic connect extensions: 0d3a or 9d3a?

[Aleksander Morgado]

>> On Mon, Sep 10, 2018 at 5:36 PM, Ben Chan <benchan at chromium.org> wrote:
>> > Looks like the UUID for basic connect extension is pretty messed up.  It
>> > isn't listed in http://compliance.usb.org/mbim/.  I'm not sure what to
>> > expect. majority wins?   Worst case, we will need to carry a quirk in
>> > libmbim to treat them interchangeable, sigh :(
>> >
>> > Service UUID = 3d01dcc5-fef5-4d05-9d3a-bef7058e9aaf
>> > https://docs.microsoft.com/en-us/windows-hardware/drivers/network/mb-modem-reset-operations
>> > https://docs.microsoft.com/en-us/windows-hardware/drivers/network/mb-protocol-configuration-operations--pco-
>> >
>> > Service UUID = 3d01dcc5-fef5-4d05-0d3a-bef7058e9aaf
>> > https://docs.microsoft.com/en-us/windows-hardware/drivers/network/mb-multi-sim-operations
>> > https://docs.microsoft.com/en-us/windows-hardware/drivers/network/mb-network-blacklist-operations
>> > https://docs.microsoft.com/en-us/windows-hardware/drivers/network/mb-base-stations-information-query-support
>> > https://docs.microsoft.com/en-us/windows-hardware/drivers/network/mb-provisioned-context-operations
>> > https://docs.microsoft.com/en-us/windows-hardware/drivers/network/mb-lte-attach-operations:
>> >
>>
>> We need to support both then. The Qualcomm-based DW5821e uses 0d3a,
>> the Intel based L850-GL uses 9d3a... I'll update libmbim to
>> transparently decide which one to use the first time a command of that
>> service is requested.
>
> I'm following up on L850-GL, which we made the request to add the
> MBIM_CID_PCO support. It may be possible to address that in the
> firmware.
>
> libmbim could also do an initial device_services_list query to check
> which UUID is defined (or, in the worst case, both are defined for
> different CIDs), and then transparently select the UUID based on the
> CID.
>

That is what I was going to do, yes. I don't think we have seen both
cases at the same time, right?


-- 
Aleksander
https://aleksander.es