James H. Cloos Jr. wrote:
I finally tried moving to test4 (from about halfway between test3 and
test4) last night.  I had to drop back, though, because I was unable
to get my acm modem to work.
...

Dmesg shows:

##############
drivers/usb/class/cdc-acm.c: need inactive config #2
drivers/usb/class/cdc-acm.c: need inactive config #2
###############

...

I will try reversing that for now; is there a more correct solution?

Not just yet. FWIW that means CDC-acm will rely on partially bogus usbcore state (driver model will be matching the wrong config's state) not causing any serious breakage ... not good long-term.

Can you forward /proc/bus/usb/devices output, or
better yet 'lsusb -v' output, for this device?

I want to see what its configuration descriptors look
like.  Seems like usbcore or cdc-acm is going to have
to get more intelligence about multi-config devices.

- Dave




------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel

Reply via email to