Re: [PATCH 1/7] extended support for LTE and NR. Some minor fixes. part 1 of 7

[Denis Kenzior <denkenz@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 1815 bytes --]

Hi Slava,

On 09/19/2018 06:14 PM, Slava Monich wrote:
> On 20/09/18 00:45, Denis Kenzior wrote:
>>> (ANY = PAP_CHAP, and don't ask me why we added new values to the 
>>> beginning of the enum - it was before we started using binary 
>>> plugins). I would be more than happy if upstream started to use the 
>>> same enum!
>>>
>>
>> That assumes that we should support your METHOD_ANY thing.  I've not 
>> heard any good arguments for that yet...
> 
> At least one Chinese modem I dealt with had AT+ EGPAU = <op>,<cid>, 
> <proto> where <proto> is 0 for PAP, 1 for CHAP, 2 for NONE and 3 for 
> PAP_CHAP. Also, Android RIL interface has value 3 reserved for PAP&CHAP 
> (see ril.h for older Androids and ApnAuthType in binder radio interface 
> for Android 8+). To me that sounds like a valid use case.

So I went in and checked (someone can correct me if I'm wrong)

- The three-four generations of Intel modems I checked do not support this
- QMI doesn't support this
- MBIM doesn't support this (but supports MSChap/MSChapV2, funnily enough)
- 3GPP doesn't support this
- Telit doesn't support this
- Model Broadband Provider Info doesn't support this

Are you sure it isn't some weird vendor feature where they try all 
possibilities one at a time?  E.g. it tries CHAP first, then tries PAP, 
then tries None?

And how do you plan on supporting this on modems that physically do not 
support such an enumeration?  What would be the exact semantics?  Is 
this modem supported by oFono upstream?

This conversation is moot until you introduce support for such a modem 
or more specific information is introduced.  And until / unless this 
happens, we're not introducing random enumerations into the D-Bus API 
that we have to live with for a long time.

Regards,
-Denis