Can't get Sierra AC875 Aircard to work in 2.14 (SOLVED)

[tempestuous]

I just found a link specific to the Sierra 875 here -
http://tirania.org/blog/archive/2007/Feb-21-2.html

I had always suspected that the key would be to force the relevant kernel module to recognise your device ... and according to that post the relevant module is "usbserial". This is the command -

modprobe usbserial vendor=0x1199 product=0x6820
(but make sure you first unload the usbserial module)

So we were on the wrong track following Mikrodot's instructions, because the correct kernel module for his Sierra 860 is "serial_cs".
I don't know whether the "sierra" module is involved at all. I suspect it is for some other Sierra device, not yours.

... and don't forget to create the device nodes - /dev/ttyUSB0 and /dev/ttyUSB1

[gamfa]

I just found a link specific to the Sierra 875 here -
http://tirania.org/blog/archive/2007/Feb-21-2.html

I have been there and tried that. Actually before I started this thread. It didn't help I but have tried it again just incase. I have tried that by itself and also before loading the sierra module.

.. and don't forget to create the device nodes - /dev/ttyUSB0 and /dev/ttyUSB1

Oh yes, those have been created plus some in /dev/usb. Still no brass ring.

I also have run the lspci -v|grep subordinate on both VL and Pup and got the same data which is what I expected. If I understood what I read on that web page, every thing is OK here.

Bus: primary=00, secondary=01, subordinate=01, sec-latency=0
Bus: primary=00, secondary=02, subordinate=05, sec-latency=176

I did a little comparing between Puppy and VL this evening and came
across something odd that might be causing the problem.

On VL there are these usb buses

/proc/bus/usb/001
/proc/bus/usb/002
/proc/bus/usb/003
/proc/bus/usb/004
/proc/bus/usb/005
/proc/bus/usb/006

With the modem showing up on usb bus 005

On puppy I have the following buses...

/proc/bus/usb/001
/proc/bus/usb/002
/proc/bus/usb/003
/proc/bus/usb/004

With the modem not showing up at all.

Kinda odd to say the least????????

[gamfa]

Well tempestuous I finally discovered what was causing the problem. I needed to probe the ohci_hcd module. What I did was to load the udev package from Slack 11.0 to puppy and ran it. When I opened puppy's little usb app. there it was looking me in the face. All 6 usb buses and the Sierra AirCard. Every thing is working now. I would like a little more advice though. I am loading that module in rc.local which is working but I see that the other such modules, ehci_hcd, etc, load early in the boot. Is there a better point at which the ohci module should be loaded?

Thanks again for your help.

[tempestuous]

Well loading the ohci-hcd module from rc.local is fine, but I think it's bad that Puppy's startup script fails to load this module in the first place.
Maybe you should report this as a bug. Barry and Dougal have done a lot of work to refine Puppy's hardware detection.

You would probably need to report your output from "lspci" and "lspci -n" so that the pci id's of your USB interfaces can be identified.