From mboxrd@z Thu Jan 1 00:00:00 1970 Content-Type: multipart/mixed; boundary="===============4923642497780697211==" MIME-Version: 1.0 From: Denis Kenzior Subject: Re: Huawei 3131 with idVendor=12d1, idProduct=14fe Date: Wed, 13 Jan 2016 10:08:13 -0600 Message-ID: <5696766D.9060307@gmail.com> In-Reply-To: <87fuy2vco9.fsf@nemi.mork.no> List-Id: To: ofono@ofono.org --===============4923642497780697211== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Hi Bjorn, On 01/12/2016 01:08 PM, Bj=C3=B8rn Mork wrote: > matti kaasinen writes: >> 2016-01-12 18:30 GMT+02:00 Denis Kenzior : >> >>> Please note, that there are two different versions of the Huawei cards >>> supported by oFono. One uses the "gobi" driver (these are Qualcomm >>> chips). While the more traditional one uses PPP & the "huawei" driver. >> >> >> I suppose I should edit setup_gobi() and vendor_list() as before in case >> it's "gobi"? > > It's not "gobi". At least not in the firmware cfg you are using. The > 255/2/22 class/subclass/protocol indicates an NCM function with an AT > command channel transported via SEND_ENCAPSULATED_COMMAND and > GET_ENCAPSULATED_RESPONSE. The huawei_cdc_ncm driver will handle this, > exporting the AT command channel as a /dev/cdc-wdmX device. I suspected as much. Thanks for confirming that. > > The NCM network interface is managed via Huawei AT commands like e.g > AT^NDISDUP, either over the /dev/cdc-wdmX device or one of the serial > functions. Don't know if ofono supports that? If not, then I guess PPP > is your option. Looks like Marcel added support for this back in 2010: = bfd2c768cd97e92347325363b92292cd6bb27e9f. But it looks like the = 'huawei' driver was never updated to make use of it. Regards, -Denis --===============4923642497780697211==--