Sierra MC73xx working

After 3 weeks i finally have a module working!
Support was added to pfSense to support in version 2.2.3 and i bought an MC7354 from a large electronics firm. Module came in and it had a different Product ID than the module added to FreeBSD. Added was a device with VID/PID 1199:6820 and my module showed up as 1199:68C0. Would not work. Bummer. So buy another a week later. Same thing different supplier 1199:68C0. So i decided to go the hardware route and switch the module PID to 6820. Quite a learning experience.

Anybody running one of these modules???

I finally got an EM7355 working and not the two MC7354's i bought for pfSense. Seems you can only switch to PID's which are hardcoded to the module/firmware. Some MC73xx will work some won't. My suspicion is they have removed the needed PID from the newer firmwares.

Thanks to J.Porter and Netgate we have a working -currently for sale and vendor supported- LTE Module.

Here are the extended AT commands I needed. I used Putty in both Ubuntu and Windows to access the Serial Port/TTYUSB.(AT port used found in Windows Device Manager)

AT!ENTERCND="A710"
This is to enter extended commands needed for extended output

AT!UDINFO?
This pulls up the info on VID/PID

AT!UDPID=?
Here is the command to check to see if your device supports 6820 pid.
If PID is not supported it will fail showing ERROR on switch attempt.

AT!UDPID=6820
This is the command to switch to supported PID.

YOU MAY NOT BE ABLE TO SWITCH BACK. USE AT OWN RISK. YOU CAN BRICK A MODULE.

Careful with this following information. Not needed. Purely Informational.

AT!UDUSBCOMP?
The default firmware setting for your module mode is found by this command.

AT!UDUSBCOMP=?
pfSense seems to work with default #6. Notice #9 has no AT console..MBIM only. Beware.

thread Linux contributor BMORK is the most knowledgeable person on this subject.
http://lists.freedesktop.org/archives/libqmi-devel/2014-May/000788.html

-Edited for accuracy-

So I am using this EM7355 (An M.2 form factor module) In a converter board - M.2 to mini-Pcie for testing.
The newer style M.2 modules- Both wifi and cellular- use a newer Antenna connector, very similar to Ipex/u.fl but smaller yet.
Called Ipex/MHF4.

I should also note i cannot use this module in Windows anymore as I have modified the PID and driver uses that for device ID.

So in case i was not clear, The whole MC73xx family now uses the same firmware. So except for differences in the radio frequency of different regions -they are the same. Some offer "Voice over LTE" but the carriers have minimally adopted the standard. PCM audio on the modem is the difference.

Mini PCIe Modules
MC7304-Europe
MC7305-Europe-VoLTE supported
MC7330-Australia
MC7354-North America
MC7355-North America-VoLTE supported

NGFF/M.2 Modules
EM7305-Europe
EM7355-North America

Even within same models there are different SKU's

EM7355
1101637-OEM
1101766-IBM 04W3801
1101797-Dell DW5808e 2NDHX
1102173-Dell DW5808e PN10C
1101825-HP
1102091-HP
1101889-Fujistu

MC7355
1101796-Dell DW5808 1N1FY
1101888-Panasonic
1101906-OEM

MC7354
1102016-OEM

So that is what I have found. Anyone care to correct any inaccuracies.

If your thinking of buying one of these beware. The support is limited to only one PID while there are many PID's in the field.

I think I have found the problem. The one module i have that works has a firmware version that is different than the rest.

ATI command shows firmware SWI9X15C_01.03.03 r8561 -Dated 12-4-12

All the versions on the internet are SWI9X15C_05.05.16 and up.

The strange thing is that when you run AT!UDPID=? to show available PID's the 6820 PID is not shown.

PID's for the older firmware:
9041
9061
9051
9053
9054
9056
68A2

Now the PID's for the newer firmware:
9041
9054
9056
68A2
68B1
68C0

The 68A2 is the same as the MC7710 and is the PID that was assigned to the EM7355 that I switched to PID=6820 and have working.

Was really happy to find CU (terminal program) a pfSense default as well. I did not even have to set up PPP to talk to the cuaU0.1 port.

cu -l /dev/cuaU0.1

I had all kind of troubles with term, under PPP. Now i have something i can use.

Here are my screenshots for the MC7354 -with the newer firmware- and its various settings. Cannot run in pfSense as is.

Here are the some PID's found in SierraMC73xx firmware.

MC7355/MC7355 – VID=0x1199
User defined PID with AT!UDPID=xxxx
0x9041
0x9054
0x9056
0x68A2
0x68B1
0X68C0

Dell DW5808 -VID=0x413C
Factory PID=0x81A8

User defined PID with AT!UDPID=xxxx On the Dell DW5808 Module
0x9041
0x9051
0x9053
0x9054
0X9056
0x68A2
0x68B1
–--------------------------------------------------

Linux qcserial PID assignment for cross reference.

0x9041-Sierra Wireless MC7305/MC7355
0x9051-Netgear AirCard 340U Modem
0x9053-Sierra Wireless Modem
0x9054-Sierra Wireless Modem
0X9056-Netgear AirCard 341U Modem
0x68A2-MC7710 QMI Mode
0x68B1-None
–---------------------------------------------------

Current FreeBSD USBDEVS -PID Assignments-
0x9041-Not assigned
0x9051-Not Assigned
0x9053-Not Assigned
0x9054-Not Assigned
0X9056-Not Assigned
0x68A2-Not Assigned
0x68B1-Not Assigned
0x68C0-Not Assigned
–----------------------------------------------------

Product ID's I would like to be added to USBDEVS/U3G
VID:1199 PID:68C0 DEVICE_DESC: Sierra Wireless MC7355
VID:1199 PID:68A2 DEVICE_DESC: Sierra Wireless EM7355
VID:1199 PID:9041 DEVICE_DESC: Sierra Wireless MC7355
VID:413C PID:818A DEVICE_DESC: Dell DW5808 Wireless Modem (MC7355)

Here are some more informational screen shots showing firmware, PID's and USBCOMP mode. These are for the MC7355

Here are the user selectable PID's for the MC7355.

Here are the Dell DW5808 details. This is also a MC7355.
They have more PID's on their list than any other. Strange how the default PID 81A8 is not in the list. Modes are different too.

I was able to convert a Dell DW5808 to the native Sierra MC7355.

AT!UDVID=1199 and AT!UDPID=9041

Turns your rotten Dell locked module into a generic OEM module usable in any Windows machine with stock drivers.

I only tried it on one module so YMMV. It had an early firmware and i couldn't coax PID:6820 out of it…

While getting a couple of modules flashed to the latest ATT firmware for a trip to Texas and I decided to take them live. Most of my testing is done with no antennas, so last step is antennas before use. Used windows and flashed them -then fired up the Sierra Skylight Watcher. The speedometer/indicator was on LTE at first, then seem to drop to 3G HPSA+ and lots of back and forth and maybe even 2G/UMTS flashed up.. Seemed erratic. Used my new AT command skills and hit the extended commands pdf. AT!GSTATUS? gives you some info. Then found AT!BANDS, AT!GBAND and AT!DASBAND which i still don't understand the differences. I tried LTE only and it wouldn't find service. ATT uses B17 for most LTE. The fallback 3G seems to be 850GSM and 1900GSM for ATT.

I am pretty impressed with Ubuntu Modem-Manager GUI as it has a nice "Scan" feature and I see 5) T-mobile towers, and only 1ea.)VZW, Sprint and ATT tower. That makes me wonder about my service. I see 4 bars on my ATT tower. I think the module needs tuning. I even see a RF calibration recommended…Whatever that means. I know my MC7700 experience was much less painful and that module maintains LTE on the Watcher...I should probably dump my Windows platform as I have installed probably a dozen different driver packages...Super weird thing is that the module "Firmware version" seems tied to software as the same module with Windows and Ubuntu shows different firmware versions...Odd why would hardware firmware version change with OS disk change? Must be tied to different Gobi.API/blob in the different OS's- i really dunno...

I loaded a fresh Windows and Sierra drives and Watcher. Was still seeing my LTE drop to 3G and i noticed it seemed tied to usage. When I ran speed test the speed would jump to LTE.
I found this thread which explains why.

Telstra cells drop you back to basic "UMTS" when there's no data in transmission between yourself and the tower. Once the data flows again, you'll typically go to "HSPA+" (single carrier) or "DC-HSPA+" (dual carrier) on Telstra, depending on the capabilities of the tower. That's got nothing to do with diversity.

http://forums.whirlpool.net.au/archive/2178921

This is exactly what I am seeing with ATT. Drops back to lower connection speed when not in use.

Juancho1972

Phishfry, I'm from Argentina and need your help to convert a Dell DW5808 to the native Sierra MC7355.

Thanks

Juancho1972

Finally I can convert the Dell DW5808 into a Generic Sierra Wireless MC7355 to work with Windows 7.

Thanks to Phishfry and other web sites with a perl script to change the UDUSBCOMP to work in Ubuntu.

FranciscoFranco

Here is the skinny. The Dell DW5808 (MC7355) module was locked in USBCOMP #9 which is MBIM only -No AT command port at all. This is the new method by which Windows 8.1 and 10 work. They flash your modem to only use MBIM effectivly locking you out of the module.

So Juancho had to not only switch his module from Dell VID+PID to Sierra VID+PID but also regain the AT console to do it.
Luckily one of the developers of Linux modem-manager has a perl script which can bring back the AT console.

Here is how without installing Linux on host.
Setup Ubuntu USB Flash/Pen drive with persistent storage like Unetbootin offers. Run you disk and install the following

sudo apt-get install perl
sudo apt-get install libuuid-tiny-perl
sudo apt-get install libipc-shareable-perl
sudo apt-get install modem-manager-gui
sudo apt-get install putty

So first off you need to run networkmanager and modemmanager in debug mode:
https://www.freedesktop.org/wiki/Software/ModemManager/Debugging

Then run this script:
http://git.mork.no/?p=wwan.git;a=blob_plain;f=scripts/swi_setusbcomp.pl

Here are some supporting docs.
https://www.freedesktop.org/software/ModemManager/man/1.0.0/mmcli.8.html
https://www.chromium.org/chromium-os/how-tos-and-troubleshooting/debugging-3g/modem-debugging-with-mmcli

Even on other Sierra modems you get worked:
https://forum.sierrawireless.com/viewtopic.php?f=117&t=8863

Different Modem same effects-different cure:
http://zukota.com/how-to-enable-at-command-and-diagnostic-ports-and-gps-for-sierra-wireless-em7345/

This has exactly what is needed. Note EM7305 is the same firmware wise.
http://www.0xf8.org/2016/04/changing-dell-wireless-5809e-sierra-wireless-em7305-usb-composition-mbim-qmi-at-interface-nmea/

Note: ModemManager and Networkmanager in debug mode might not be needed. We were trying to send the modem AT commands with mmcli and that required debug mode. Unfortunately it did not work as AT command console was not exposed. Ultimately it required the perl script. Here is sample command.
sudo perl swi_setusbcomp.pl –usbcomp=14

Putty set to Serial and ttyUSB1. Putty requires sudo putty from terminal to work.

ModemManager GUI is useful in showing the mode the modem is in. For example MIBM was apparent without AT console.

From Juancho:
Maybe with Perl script and Putty is enough to do the conversión from Dell to Sierra Wireless.

So the above post about bricking your module is now the new normal from Microsoft. MBIM only mode. Luckily Bjorn Mork and his magic perl script saves the day.

FranciscoFranco

The reason I post this is the procedures are exactly the same for pfSense, FreeBSD and Windows. #14 USBCOMP works and as a bonus proves a dual configuration with DirectIP for FreeBSD/pfSense and QMI modes for Linux.

thanatos2k

So I was in the middle of changing a DW5808 into a generic module when the laptop lost power. I got the VID successfully updated to 1199 but the PID is still 81A8… This means the drivers won't pick up the card as a valid combo to get me into a serial terminal to change the PID. Is there some way in Linux I can trick it into working so I can issue the PID change or is the card toast? Card is in USBcomp=6 right now if that changes anything, I see 4 interfaces exposed in windows but can't get drivers to load for them.

thanatos2k

Ended up comparing the modules loaded with a 5808e in a USB carrier and manually installing the modules using modprobe. That exposed the /dev/ttyUSB0 through /dev/ttyUSB2, ttyUSB2 was the AT command interface so I was able to issue the AT!UDVID=9041 to fix. Runs like a champ now.


/sbin/modprobe cdc_wdm product=0x81a8 vendor=0x1199
/sbin/modprobe usbserial product=0x81a8 vendor=0x1199
/sbin/modprobe qmi_wwan product=0x81a8 vendor=0x1199
/sbin/modprobe qcserial product=0x81a8 vendor=0x1199

tubs89

How are you guys connecting the mPCIe card to a host? Does a USB adapter work or do I need to find a mainboard that supports mPCIe to do these steps?

thanatos2k

3g/4g modems almost all use the USB interface of Mini-PCI(e) so a USB adapter will work if you don't have a slot internally.

tubs89

@thanatos2k:

3g/4g modems almost all use the USB interface of Mini-PCI(e) so a USB adapter will work if you don't have a slot internally.

Thanks!

tubs89

Thanks for these instructions! I also followed this post to re-flash the firmware to Sprint: https://forum.pfsense.org/index.php?topic=120614.0.

I had to run !BOOTHOLD via the AT interface prior to flashing the firmware to get the firmware to flash without an error.

tubs89

I'm still having trouble - it appears the device is stuck in "Low Power Mode", similar to this post: https://sigquit.wordpress.com/2015/02/09/dell-branded-sierra-wireless-3g4g-modem-not-online/

The output of !PCINFO:
State: LowPowerMode
LPM force flags - W_DISABLE:0, User:0, Temp:0, Volt:0, BIOS:1, GOBIIM:0
W_DISABLE: 0
Poweroff mode: 0
LPM Persistent: 0

Indicates the BIOS system is somehow converting the module to low power. Googling reveals linux can do this with it's ACPI modules by accident/bug:
https://lists.freedesktop.org/archives/libqmi-devel/2015-January/001084.html

Anyone else struggling with this?

thanatos2k

I've found with the DW5805 and DW5805e that even if you change the VID/PID to non-Dell configurations, they still need the FCC auth command to be sent before they will go out of low power mode. I've not found a way to send that command under FreeBSD. I'm currently using ModemManager on OpenWRT in a secondary atom box to do the 4G connections.

tubs89

I think I found the key:
https://forum.sierrawireless.com/viewtopic.php?f=117&t=9771&sid=3c2d1cc46ccb965d2e37535a9ab142b1&start=15#p39184

The PCFCCAUTH can only be changed in a special mode with OPENLOCK command. Trying to disable it via the AT console without OPENLOCK results in an error.

The older Sierra chips could be unlocked from here - https://github.com/bkerler/SierraWirelessGen. But it sounds like the algorithm is updated.

I think there is a tool out there that let's you modify these settings outside the AT interface. May be easier.

meteora

@juancho1972 . can you upload the whole process? I am facing the same problem.

fips

@juancho1972 Yes, please show the process.

alexTM

So MC7355 and EM7355, which one is better to use with OpenWRT?

stephenw10

Hmm, seems like spam but...

That is the same card effectively. The MC version is mPCIe and the EM is m.2.

Steve

alexTM

@stephenw10 thanks

mrneutron

Is anyone able to get this to work on a DW5808 (MC-7355) in 2020?
I've been following the procedure outlined by user FranciscoFranco in forum post #17.
I've experienced that the procedure does not work in 2020.
I tried to perform the process using Ubuntu 19.10 on a 8GB bootable USB flash drive with 2GB of persistent memory space. I also tried using a Ubuntu 19.10 Virtual Machine inside VMware Workstation. Both setups yielded the same failures.
One of the first steps is to run networkmanager and modemmanager in debug mode:
see https://www.freedesktop.org/wiki/Software/ModemManager/Debugging
$> sudo /usr/sbin/ModemManager --debug
This process fails to complete. The screen scrolls with text for about 1 minute then stops and never completes - even after waiting/watching for an hour. If you perform a CTRL-C to stop the command, it stops with a message that it failed.
Since this is one the first steps in the process, the rest cannot be accomplished.

has anyone been able to get this to work on a DW5808 (MC-7355) in 2020?
if not, why this would work in 2016 and not in 2020?
do I need to use a 2016 version of Ubuntu for this to work?

Thanks for any updates on this procedure.

stephenw10

Do you have the error it actually shows?

Did you try modprobing the serial driver onto the device?
https://forum.netgate.com/post/677772

mrneutron

@stephenw10
No, I didn't try any modprobe commands.

from: sudo /usr/sbin/ModemManager --debug
This is the last part of the output...says it didn't connect to the AT port and stops and waits, forever.

ModemManager[5079]: <debug> [1595789997.248458] [plugin manager] task 0: min probing time elapsed
ModemManager[5079]: <debug> [1595789997.248676] [plugin Manager] task 0: still 1 running probes (0 active): ens33
ModemManager[5079]: <debug> [1595789997.248711] [plugin manager] task 0: no active tasks to probe
ModemManager[5079]: <debug> [1595789997.248745] [plugin manager] task 0,ens33: deferred task completed, no suggested plugin
ModemManager[5079]: <debug> [1595789997.248777] [plugin manager] task 0,ens33: finished in '2.502197' seconds
ModemManager[5079]: <debug> [1595789997.248943] [plugin manager] task 0,ens33: not supported by any plugin
ModemManager[5079]: <debug> [1595789997.248983] [device /sys/devices/pci0000:00/0000:00:11.0/0000:02:01.0] fully ignoring port 'net/ens33' from now on
ModemManager[5079]: <debug> [1595789997.249009] [plugin manager] task 0: no more ports to probe
ModemManager[5079]: <debug> [1595789997.249028] [plugin manager] task 0: finished in '2.502562' seconds
ModemManager[5079]: <info> [1595789997.249077] Couldn't check support for device '/sys/devices/pci0000:00/0000:00:11.0/0000:02:01.0': not supported by any plugin
ModemManager[5079]: <debug> [1595789997.249342] [plugin manager] task 1: min probing time elapsed
ModemManager[5079]: <debug> [1595789997.249390] [plugin Manager] task 1: still 2 running probes (1 active): cdc-wdm0, wws35u1c2i12
ModemManager[5079]: proxy configuration failed: closed
ModemManager[5079]: <debug> [1595790026.390983] (usbmisc/cdc-wdm0) error checking MBIM support: 'Transaction timed out'
ModemManager[5079]: <debug> [1595790026.391018] (usbmisc/cdc-wdm0) port is not MBIM-capable
ModemManager[5079]: [/dev/cdc-wdm0] channel destroyed
ModemManager[5079]: <debug> [1595790026.391300] [plugin manager] task 1,cdc-wdm0: found best plugin for port (Dell)
ModemManager[5079]: <debug> [1595790026.391498] [plugin manager] task 1,cdc-wdm0: finished in '31.644019' seconds
ModemManager[5079]: <debug> [1595790026.391536] [plugin manager] task 1,cdc-wdm0: found best plugin: Dell
ModemManager[5079]: <debug> [1595790026.391557] [plugin manager] task 1,wws35u1c2i12: deferred task completed, got suggested plugin (Dell)
ModemManager[5079]: <debug> [1595790026.391586] [plugin Manager] task 1: still 1 running probes (1 active): wws35u1c2i12
ModemManager[5079]: <debug> [1595790026.391663] [plugin manager] task 1,wws35u1c2i12: checking with plugin 'Dell'
ModemManager[5079]: <debug> [1595790026.391712] (Dell) [wws35u1c2i12] probing deferred until result suggested
ModemManager[5079]: <debug> [1595790026.391748] [plugin manager] task 1,wws35u1c2i12: completed, got suggested plugin (Dell)
ModemManager[5079]: <debug> [1595790026.391768] [plugin manager] task 1,wws35u1c2i12: finished in '31.644749' seconds
ModemManager[5079]: <debug> [1595790026.391792] [plugin manager] task 1,wws35u1c2i12: best plugin matches device reported one: Dell
ModemManager[5079]: <debug> [1595790026.391810] [plugin manager] task 1: no more ports to probe
ModemManager[5079]: <debug> [1595790026.391827] [plugin manager] task 1: finished in '31.644885' seconds
ModemManager[5079]: <info> [1595790026.391870] [device /sys/devices/pci0000:00/0000:00:11.0/0000:02:03.0/usb1/1-1] creating modem with plugin 'Dell' and '2' ports
ModemManager[5079]: <debug> [1595790026.391934] Dell-branded generic modem found...
ModemManager[5079]: <warn> [1595790026.399711] Could not grab port (usbmisc/cdc-wdm0): 'Cannot add port 'usbmisc/cdc-wdm0', unsupported'
ModemManager[5079]: <debug> [1595790026.400072] (wws35u1c2i12) type 'net' claimed by /sys/devices/pci0000:00/0000:00:11.0/0000:02:03.0/usb1/1-1
ModemManager[5079]: <debug> [1595790026.400336] Modem (Dell) '/sys/devices/pci0000:00/0000:00:11.0/0000:02:03.0/usb1/1-1' completely disposed
ModemManager[5079]: <warn> [1595790026.401515] Couldn't create modem for device '/sys/devices/pci0000:00/0000:00:11.0/0000:02:03.0/usb1/1-1': Failed to find primary AT port

Then I did a CTRL-C

^CModemManager[5079]: <info> [1595790115.488383] Caught signal, shutting down...
ModemManager[5079]: <debug> [1595790115.488753] Stopping connection in object manager server
ModemManager[5079]: <info> [1595790115.497328] ModemManager is shut down
ModemManager[5079]: <debug> [1595790115.497632] disposing MMSleepMonitor singleton (0x5614725aba90)
ModemManager[5079]: <debug> [1595790115.497964] [sleep-monitor] dropping systemd sleep inhibitor
eric@ubuntu-vm:~$ ^C
eric@ubuntu-vm:~$

stephenw10

I assume you stopped and disabled modem manager before that?

You need to access an AT port somehow. If that cannot be done in Ubuntu it's probably possible in Windows.

Steve

mrneutron

@stephenw10 Yes, from reading this thread, I can tell that getting AT command access is crucial to changing the PID and VID numbers.
Yes, I followed the documented process, and disabled and stopped the processes noted. But, since command 3 fails, I can't go on to command 4.
sudo systemctl disable NetworkManager ModemManager
sudo systemctl stop NetworkManager ModemManager
sudo /usr/sbin/ModemManager --debug
sudo /usr/sbin/NetworkManager --debug --log-level=DEBUG

I just tried the modprobe lines. 1 gave an error. I tried starting ModemManager in debug mode after entering these 4 lines...no difference.
eric@ubuntu-vm:~$ /sbin/modprobe cdc_wdm product=0x81a8 vendor=0x1199
eric@ubuntu-vm:~$ /sbin/modprobe usbserial product=0x81a8 vendor=0x1199
eric@ubuntu-vm:~$ /sbin/modprobe qmi_wwan product=0x81a8 vendor=0x1199
modprobe: ERROR: could not insert 'qmi_wwan': Operation not permitted
eric@ubuntu-vm:~$ /sbin/modprobe qcserial product=0x81a8 vendor=0x1199
eric@ubuntu-vm:~$

Given that the documented procedure doesn't work in Ubuntu 19.10, I have to guess:

the procedure was not documented completely/accurately back in 2016
the 2019 versions of software in the Ubuntu apt repository are not compatible with this procedure

mrneutron

Here's the output of the perl script.
eric@ubuntu-vm:~$ sudo perl swi_setusbcomp.pl -usbcomp=14
Running in MBIM mode (driver=cdc_mbim)
MBIM OPEN succeeded
QMI msg '0x0021' returned status = 1
MBIM QMI support verified
supports 26 QMI subsystems:
0x00 (1.5) 'QMI_CTL' - Control service
0x01 (1.36) 'QMI_WDS' - Wireless data service
0x02 (1.14) 'QMI_DMS' - Device management service
0x03 (1.25) 'QMI_NAS' - Network access service
0x04 (1.3) 'QMI_QOS' - Quality of service, err, service
0x05 (1.10) 'QMI_WMS' - Wireless messaging service
0x06 (1.0) 'QMI_PDS' - Position determination service
0x07 (1.2) 'QMI_AUTH' - Authentication service
0x08 (1.2) 'QMI_AT' - AT command processor service
0x09 (2.1) 'QMI_VOICE' - Voice service
0x0a (2.16) 'QMI_CAT2' - Card application toolkit service (new)
0x0b (1.25) 'QMI_UIM' - UIM service
0x0c (1.4) 'QMI_PBM' - Phonebook service
0x11 (1.0) 'QMI_SAR' - Specific absorption rate service
0x12 (1.0) 'QMI_IMSS' - IMS settings service
0x17 (1.0) 'QMI_TS' - Thermal sensors service
0x18 (1.0) 'QMI_TMD' - Thermal mitigation device service
0x1a (1.10) 'QMI_WDA' - Wireless data administrative service
0x1d (1.0) 'QMI_CSVT' - Circuit switched videotelephony service
0x1e (1.0) 'QMI_QCMAP' - Qualcomm mobile access point service
0x1f (1.0) 'QMI_IMSP' - IMS presence service
0x20 (1.0) 'QMI_IMSVT' - IMS videotelephony service
0x22 (1.0) 'QMI_COEX' - Coexistence service
0x29 (1.0) 'QMI_RFRPE' - RF radiated performance enhancement service
0xe1 (1.0) 'QMI_RMS' - Remote management service
0xf0 (1.0) 'unknown' -
QMI msg '0x0022' returned status = 1
Got QMI DMS client ID '44'
QMI msg '0x555b' returned status = 1
Current USB composition: 14
USB compositions:
0 - HIP DM NMEA AT MDM1 MDM2 MDM3 MS NOT SUPPORTED
1 - HIP DM NMEA AT MDM1 MS SUPPORTED
2 - HIP DM NMEA AT NIC1 MS NOT SUPPORTED
3 - HIP DM NMEA AT MDM1 NIC1 MS NOT SUPPORTED
4 - HIP DM NMEA AT NIC1 NIC2 NIC3 MS NOT SUPPORTED
5 - HIP DM NMEA AT ECM1 MS NOT SUPPORTED
6 - DM NMEA AT QMI SUPPORTED
7 - DM NMEA AT RMNET1 RMNET2 RMNET3 SUPPORTED
8 - DM NMEA AT MBIM SUPPORTED
9 - MBIM SUPPORTED
10 - NMEA MBIM SUPPORTED
11 - DM MBIM SUPPORTED
12 - DM NMEA MBIM SUPPORTED
13 - Config1: comp6 Config2: comp8 NOT SUPPORTED

14 - Config1: comp6 Config2: comp9 SUPPORTED
15 - Config1: comp6 Config2: comp10 NOT SUPPORTED
16 - Config1: comp6 Config2: comp11 NOT SUPPORTED
17 - Config1: comp6 Config2: comp12 NOT SUPPORTED
18 - Config1: comp7 Config2: comp8 NOT SUPPORTED
19 - Config1: comp7 Config2: comp9 SUPPORTED
20 - Config1: comp7 Config2: comp10 NOT SUPPORTED
21 - Config1: comp7 Config2: comp11 NOT SUPPORTED
22 - Config1: comp7 Config2: comp12 NOT SUPPORTED
Current setting is already '14'
QMI msg '0x0023' returned status = 1

stephenw10

Hmm, well if it can read that and is in composition 14 then you should be able to get an AT port.

You might have to change the current config index to 6 if it's 9.

What does lsusb -d 413C -v show.

Also it looks like you loaded the usbserial driver using Sierra vendor and product ID. You need to use the current values so it attaches. So I assume?:
modprobe usbserial product=0x818A vendor=0x413C

Steve

mrneutron

@stephenw10 that command line doesn't show us anything.

eric@ubuntu-vm:~$ lsusb -d 413C -v
Usage: lsusb [options]...
List USB devices
-v, --verbose
Increase verbosity (show descriptors)
-s [[bus]:][devnum]
Show only devices with specified device and/or
bus numbers (in decimal)
-d vendor:[product]
Show only devices with the specified vendor and
product ID numbers (in hexadecimal)
-D device
Selects which device lsusb will examine
-t, --tree
Dump the physical USB device hierarchy as a tree
-V, --version
Show version of program
-h, --help
Show usage and help
eric@ubuntu-vm:~$

From what I read in this thread, the problem is that the driver for the standard Sierra Wireless 7355 is linked to a differen PID and VID, so the driver won't connect with the Dell version of the 7355.
And, since the Dell version has no AT serial port, you can't sent it a command to change teh PID VID.

Also from the tread, MBIM is supported. I believe that is an alternate port to send AT commands to? Someone posted that you can configure PUTTY to send commands to the MBIM port, but I'm not sure what port or BAUD rate to set, because they didn't list the specifics.

Its not clear how they got this to work in 2016, and it doesn't seem to work in 2020....

Sierra MC73xx working

MC7355/MC7355 – VID=0x1199 User defined PID with AT!UDPID=xxxx 0x9041 0x9054 0x9056 0x68A2 0x68B1 0X68C0

MC7355/MC7355 – VID=0x1199
User defined PID with AT!UDPID=xxxx
0x9041
0x9054
0x9056
0x68A2
0x68B1
0X68C0