gobi 3000 mc8355 error: couldn't set FCC auth

ITwrx Info info at itwrx.org
Fri Aug 19 19:13:23 UTC 2016



On 08/19/2016 11:09 AM, Dan Williams wrote:
> On Thu, 2016-08-18 at 18:53 -0500, ITwrx Info wrote:
>> On 08/18/2016 11:41 AM, Dan Williams wrote:
>>> On Thu, 2016-08-18 at 10:46 -0500, ITwrx Info wrote:
>>>> On 08/18/2016 10:09 AM, Dan Williams wrote:
>>>>> On Wed, 2016-08-17 at 18:48 -0500, ITwrx Info wrote:
>>>>>> On 08/17/2016 04:26 PM, Dan Williams wrote:
>>>>>>> On Wed, 2016-08-17 at 15:37 -0500, ITwrx Info wrote:
>>>>>>>> On 08/17/2016 01:40 PM, Bjørn Mork wrote:
>>>>>>>>> ITwrx Info <info at itwrx.org> writes:
>>>>>>>>>
>>>>>>>>>> On 08/17/2016 10:08 AM, Bjørn Mork wrote:
>>>>>>>>>>> "ITwrx.org" <info at itwrx.org> writes:
>>>>>>>>>>>
>>>>>>>>>>>> my MM debug log is here:  http://pastebin.com/ev8
>>>>>>>>>>>> TGi8
>>>>>>>>>>>> J
>>>>>>>>>>> OK, so that detects the sw radio off state and runs
>>>>>>>>>>> the
>>>>>>>>>>> FCC
>>>>>>>>>>> auth logic
>>>>>>>>>>> as it is supposed to.  But the result indicates
>>>>>>>>>>> this is
>>>>>>>>>>> not
>>>>>>>>>>> the
>>>>>>>>>>> issue:
>>>>>>>>>>>
>>>>>>>>>>> ModemManager[1109]: [/dev/cdc-wdm0] Received MBIM
>>>>>>>>>>> message
>>>>>>>>>>> ModemManager[1109]: [/dev/cdc-wdm0] Received
>>>>>>>>>>> message...
>>>>>>>>>>>>>>>>> RAW:
>>>>>>>>>>>>>>>>>         length = 20
>>>>>>>>>>>>>>>>>         data   =
>>>>>>>>>>>>>>>>> 01:13:00:80:02:04:02:01:00:5F:55:07:00:
>>>>>>>>>>>>>>>>> 02:0
>>>>>>>>>>>>>>>>> 4:00
>>>>>>>>>>>>>>>>> :01:
>>>>>>>>>>>>>>>>> 00:1A:00
>>>>>>>>>>> ModemManager[1109]: [/dev/cdc-wdm0] Received
>>>>>>>>>>> message
>>>>>>>>>>> (translated)...
>>>>>>>>>>>>>>>>> QMUX:
>>>>>>>>>>>>>>>>>         length  = 19
>>>>>>>>>>>>>>>>>         flags   = 0x80
>>>>>>>>>>>>>>>>>         service = "dms"
>>>>>>>>>>>>>>>>>         client  = 4
>>>>>>>>>>>>>>>>> QMI:
>>>>>>>>>>>>>>>>>         flags       = "response"
>>>>>>>>>>>>>>>>>         transaction = 1
>>>>>>>>>>>>>>>>>         tlv_length  = 7
>>>>>>>>>>>>>>>>>         message     = "Set FCC
>>>>>>>>>>>>>>>>> Authentication"
>>>>>>>>>>>>>>>>> (0x555F)
>>>>>>>>>>>>>>>>> TLV:
>>>>>>>>>>>>>>>>>         type       = "Result" (0x02)
>>>>>>>>>>>>>>>>>         length     = 4
>>>>>>>>>>>>>>>>>         value      = 01:00:1A:00
>>>>>>>>>>>>>>>>>         translated = FAILURE: NoEffect
>>>>>>>>>>> ModemManager[1109]: error: couldn't set FCC auth:
>>>>>>>>>>> QMI
>>>>>>>>>>> protocol
>>>>>>>>>>> error (26): 'NoEffect'
>>>>>>>>>> my reading of the log is that it tried to set the fcc
>>>>>>>>>> auth
>>>>>>>>>> but
>>>>>>>>>> failed. to me that doesn't necessarily mean it's not
>>>>>>>>>> the
>>>>>>>>>> issue,
>>>>>>>>>> but i
>>>>>>>>>> don't know. that's just my reading of it from a
>>>>>>>>>> layman's
>>>>>>>>>> perspective.
>>>>>>>>> Yes, it failed, but the "NoEffect" is usually pretty
>>>>>>>>> descriptive.  FWIW,
>>>>>>>>> that's what I get on my EM7455 if I run  "Set FCC
>>>>>>>>> Authentication"
>>>>>>>>> twice:
>>>>>>>>>
>>>>>>>>>       root at miraculix:/home/bjorn# qmicli -p -d /dev/cdc-
>>>>>>>>> wdm0
>>>>>>>>> --
>>>>>>>>> device-
>>>>>>>>> open-mbim  --dms-set-fcc-authentication
>>>>>>>>>       [/dev/cdc-wdm0] Successfully set FCC
>>>>>>>>> authentication
>>>>>>>>>
>>>>>>>>>       root at miraculix:/home/bjorn# qmicli -p -d /dev/cdc-
>>>>>>>>> wdm0
>>>>>>>>> --
>>>>>>>>> device-
>>>>>>>>> open-mbim  --dms-set-fcc-authentication
>>>>>>>>>       error: couldn't set FCC authentication: QMI
>>>>>>>>> protocol
>>>>>>>>> error
>>>>>>>>> (26):
>>>>>>>>> 'NoEffect'
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> So the above indicates that your modem doesn't need
>>>>>>>>> "Set
>>>>>>>>> FCC
>>>>>>>>> Authentication".
>>>>>>>> i still don't understand how it couldn't mean it actually
>>>>>>>> failed
>>>>>>>> the
>>>>>>>> first time, but i'll take your word for it.
>>>>>>>>> There is obviously something keeping it in low-power
>>>>>>>>> mode, but it is more likely something else.
>>>>>>>> i tried powering on with "mmcli -m 3 --set-power-state-
>>>>>>>> on"
>>>>>>>> and it reports
>>>>>>>> "'error: couldn't set new power state in the modem:
>>>>>>>> 'GDBus.Error:org.freedesktop.libmbim.Error.Status.Failure
>>>>>>>> :
>>>>>>>> Failure'".
>>>>>>>> Do you know what is generally required for a modem to be
>>>>>>>> powered
>>>>>>>> on?
>>>>>>>> What all kernel modules are required for this modem, do
>>>>>>>> you
>>>>>>>> think?
>>>>>>>> maybe
>>>>>>>> i don't have the basics handled. do i need to blacklist
>>>>>>>> anything?
>>>>>>>>>>> Are you sure your modem requires FCC auth?
>>>>>>>>>> no.
>>>>>>>>>>> Are you sure there aren't
>>>>>>>>>>> enywhting else keeping it in radio off state?
>>>>>>>>>> no. there very well may be but i don't know what.
>>>>>>>>> So you haven't seen this e.g. working in Windows?
>>>>>>>> i don't use windows at all. It's a new modem, so it
>>>>>>>> should
>>>>>>>> work.
>>>>>>>> I
>>>>>>>> had
>>>>>>>> previously purchased what should have been a slightly
>>>>>>>> more
>>>>>>>> modern,
>>>>>>>> better supported modem but lenovo's uefi whitelist
>>>>>>>> suprised
>>>>>>>> me.
>>>>>>>> so i
>>>>>>>> had
>>>>>>>> a choice of two modems that would work with my carrier
>>>>>>>> and
>>>>>>>> band.
>>>>>>>> i
>>>>>>>> bought the cheaper one. For future reference, is the
>>>>>>>> em7345
>>>>>>>> known
>>>>>>>> to
>>>>>>>> work out of the box?
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>>> Typically that would a
>>>>>>>>>>> "missing" firmware image, as required by your
>>>>>>>>>>> operator
>>>>>>>>>>> settings.
>>>>>>>>>> this was the only guess i had too. i don't know what
>>>>>>>>>> firmware
>>>>>>>>>> i
>>>>>>>>>> may
>>>>>>>>>> need or where to get it or how to load it. gobi-
>>>>>>>>>> loader
>>>>>>>>>> and
>>>>>>>>>> gobi-firmware were the only things i found but they
>>>>>>>>>> appear to
>>>>>>>>>> be
>>>>>>>>>> for
>>>>>>>>>> gobi 2000 and 1000.
>>>>>>>>> Yes.  The gobi 3000 didn't normally need any firmware
>>>>>>>>> upload
>>>>>>>>> since
>>>>>>>>> it
>>>>>>>>> came with firmware on flash.  But newer modems like the
>>>>>>>>> MC73xx
>>>>>>>>> and
>>>>>>>>> MC74xx have a concept of firmware image and operator
>>>>>>>>> matching,
>>>>>>>>> where
>>>>>>>>> they will stay in low-power mode if none of the
>>>>>>>>> installed
>>>>>>>>> images
>>>>>>>>> are
>>>>>>>>> usable.
>>>>>>>>>
>>>>>>>>> I don't know if this applies to the gobi 3000
>>>>>>>>> generation,
>>>>>>>>> though.
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>>> Hmm, you don't have any AT command port enabled, do
>>>>>>>>>>> you?
>>>>>>>>>> not that i'm aware of.
>>>>>>>>>>> A look at the
>>>>>>>>>>> "AT!PCINFO?" output would be most useful...  Not to
>>>>>>>>>>> mention
>>>>>>>>>>> the
>>>>>>>>>>> "AT!PCFCCAUTH?" I've just learned about.
>>>>>>>>>>>
>>>>>>>>>>> But failing that, could you try this:
>>>>>>>>>> those commands failed to produce any results.
>>>>>>>>> You need to run them on the AT command TTY after
>>>>>>>>> entering
>>>>>>>>> the
>>>>>>>>> OEM
>>>>>>>>> password.  I.e.
>>>>>>>>>
>>>>>>>>> AT!ENTERCND="A710"
>>>>>>>>> AT!PCINFO?
>>>>>>>>> AT!PCFCCAUTH?
>>>>>>>>>
>>>>>>>>> But if you don't have an AT command port, then I guess
>>>>>>>>> we
>>>>>>>>> would
>>>>>>>>> have to
>>>>>>>>> figure out how to enable that first.
>>>>>>>> do i have to have a physical serial port or is there a
>>>>>>>> way to
>>>>>>>> send
>>>>>>>> AT
>>>>>>>> commands via the terminal using some emulation or
>>>>>>>> something?
>>>>>>> No, if you don't have an AT port exposed then you'd have to
>>>>>>> change
>>>>>>> the
>>>>>>> firmware USB mode to get one.  Try setting a different
>>>>>>> configuration on
>>>>>>> the device to get QMI mode:
>>>>>>>
>>>>>>> echo 1 > /sys/bus/usb/devices/2-2/bConfigurationValue
>>>>>>>
>>>>>>> Mine is at 2-2, yours will be different.  Run 'dmesg' and
>>>>>>> you'll
>>>>>>> see
>>>>>>> something like "cdc_mbim 2-2:2.12: cdc-wdm1: USB WDM
>>>>>>> device",
>>>>>>> take
>>>>>>> everything before the ":" (eg, 2-2) and use that.  When you
>>>>>>> do
>>>>>>> that, if
>>>>>>> the device supports it, it will automatically switch from
>>>>>>> MBIM
>>>>>>> to
>>>>>>> QMI
>>>>>>> and you'll probably get a few /dev/ttyUSBx ports.  Try them
>>>>>>> each in
>>>>>>> turn, eg "minicom -D /dev/ttyUSB2" until you find one that
>>>>>>> responds
>>>>>>> to
>>>>>>> "at!pcinfo?".
>>>>>>>
>>>>>>> For example, from a EM7355:
>>>>>>>
>>>>>>> at!pcinfo?
>>>>>>> State: LowPowerMode
>>>>>>> LPM force flags - W_DISABLE:0, User:0, Temp:0, Volt:0,
>>>>>>> BIOS:0,
>>>>>>> GOBIIM:1
>>>>>>> W_DISABLE: 0
>>>>>>> Poweroff mode: 0
>>>>>>> LPM Persistent: 0
>>>>>>>
>>>>>>> This indicates that my device is in low-power-mode, because
>>>>>>> "GOBIIM:1".
>>>>>>> When I flip the W_DISABLE switch on my dev carrier board,
>>>>>>> W_DISABLE
>>>>>>> changes to "1" too.
>>>>>>>
>>>>>>> It's likely that one of these is your problem.  Either
>>>>>>> hardware
>>>>>>> W_DISABLE pins from your Thinkpad 530, or some image
>>>>>>> management
>>>>>>> issue
>>>>>>> in the firmware.
>>>>>>>
>>>>>>> Dan
>>>>>>>
>>>>>> i followed your instructions and am able to issue AT commands
>>>>>> to
>>>>>> the
>>>>>> modem with minicom but "at!pcinfo?" returns "ERROR" then it
>>>>>> alternates
>>>>>> on it's own between "+CGREG: 1,0" and "+CREG: 1,0". some
>>>>>> commands
>>>>>> work
>>>>>> fine however. issuing "AT+CFUN=1" returns "+CME ERROR: 4", if
>>>>>> that's
>>>>>> useful. "AT+CIMI" works and "AT+CFUN?"  returns "+CFUN:
>>>>>> 0".  "AT+CPIN?"
>>>>>> returns "+CPIN: READY". "AT+CSQ" returns "+CSQ: 99,99".
>>>>> What does "ATI" return on this device?  Feel free to obscure
>>>>> the
>>>>> IMEI
>>>>> if it's printed out.
>>>>>
>>>>> Also, what does AT+CLAC and AT&V return?
>>>>>
>>>>> How about "AT!GSTATUS?"
>>>>>
>>>>> Dan
>>>>>
>>>> results here: http://pastebin.com/qpw1J8BF
>>> Ok, it looks like a much-less-customized firmware than Sierra
>>> normally
>>> has, even the 7xxx series.  How about, instead:
>>>
>>> AT^SYSINFO?
>>> AT^SYSCONFIG?
>>>
>>> If those return errors, try without the "?" too.
>>>
>>> Dan
>>>
>> i think this is new to log now that we switched to usb mode. looks
>> interesting. :)
>>
>> http://pastebin.com/TTAtwzSk
> When in QMI mode, what does:
>
> qmicli -p -d /dev/cdc-wdm0 --dms-get-operating-mode
> qmicli -p -d /dev/cdc-wdm0 --dms-get-capabilities
>
> say?
>
> Dan
>
qmicli -p -d /dev/cdc-wdm0 --dms-get-operating-mode
[/dev/cdc-wdm0] Operating mode retrieved:
     Mode: 'low-power'
     HW restricted: 'no'

qmicli -p -d /dev/cdc-wdm0 --dms-get-capabilities
[/dev/cdc-wdm0] Device capabilities retrieved:
     Max TX channel rate: '5742000'
     Max RX channel rate: '14400000'
            Data Service: 'simultaneous-cs-ps'
                     SIM: 'supported'
                Networks: 'gsm, umts'

thanks



More information about the ModemManager-devel mailing list