Red Hat Bugzilla – Bug 471752
freerunner usb device unable to be enumerated/connected
Last modified: 2008-11-21 21:29:32 EST
Description of problem:
Just updated to 22.214.171.124-37.fc9.x86_64 and when I plug in my freerunner usb phone, i can't ssh into it or ping it like I could in 2.6.26 and I get a lot of the following dmesg:
hub 2-0:1.0: unable to enumerate USB device on port 1
I will attach as much info as I can. If I'm missing anything let me know, I can get it.
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1. plug in a freerunner into usb port
2. kernel "unable to enumerate USB device on port 1"
phone isn't contactable with ssh or ping.
phone is contactable with ssh or ping.
In the 2.6.26 dmesg log I'll attach, you can see a Call Trace Warning. That seems to have been fixed in 2.6.27, but it usually didn't keep the phone from being contactable. If it did, I usually just replugged the phone in again and it would come up.
Created attachment 323696 [details]
dmesg of kernel-126.96.36.199-37.fc9.x86_64
Created attachment 323697 [details]
dmesg of kernel-188.8.131.52-79.fc9.x86_64
Created attachment 323698 [details]
Created attachment 323699 [details]
This is strange. The "unable to enumerate" state should not be reoccuring.
Would you capture the usbmon trace for this? There's a sort of a howto in
You're not gonna believe this, but I can't make it happen now no matter what I try. :/
I think there is still something funky here, but feel free to close this bug... I can reopen with a usbmon trace, now that I know how to do that. thanks!
The poor signal integrity could do that, actually. I saw things as weird
as this. It may come and go with the level of battery charge in the Neo,
I'll close for now.
BTW, if you have any pull with Open Moko, it would be awesome to rework
the gadget side so it provides both ACM and ETH descriptors at the start
and so does not need the reconnect. Too many things can go wrong if the
gadget does such tricks.