Bug 625758

I'm seeing this too, but not on every machine. What kind of hardware is it happeing on for you? Please attach (as attachments, not inline text) output of the 'lspci -vnn' command and contents of /proc/cpu.

Also, is your keyboard USB or PS/2?

Created attachment 440093 [details]
lspci -vnn

Created attachment 440094 [details]
/proc/cpuinfo

Created attachment 440095 [details]
lsusb -v

The keyboard is USB. It's the Lite-On Tech IBM USB Travel Keyboard with UltraNav. Here's what the kernel prints about it:

usb 2-1.5.3: new low speed USB device using ehci_hcd and address 4
usb 2-1.5.3: New USB device found, idVendor=04b3, idProduct=3019
usb 2-1.5.3: New USB device strings: Mfr=1, Product=2, SerialNumber=0
usb 2-1.5.3: Product: IBM USB Travel Keyboard with UltraNav
usb 2-1.5.3: Manufacturer: Lite-On Tech
input: Lite-On Tech IBM USB Travel Keyboard with UltraNav as /devices/pci0000:00/0000:00:1d.0/usb2/2-1/2-1.5/2-1.5.3/2-1.5.3:1.0/input/input9
generic-usb 0003:04B3:3019.0001: input,hidraw0: USB HID v1.00 Keyboard [Lite-On Tech IBM USB Travel Keyboard with UltraNav] on usb-0000:00:1d.0-1.5.3/input0
input: Lite-On Tech IBM USB Travel Keyboard with UltraNav as /devices/pci0000:00/0000:00:1d.0/usb2/2-1/2-1.5/2-1.5.3/2-1.5.3:1.1/input/input10
generic-usb 0003:04B3:3019.0002: input,hidraw1: USB HID v1.00 Device [Lite-On Tech IBM USB Travel Keyboard with UltraNav] on usb-0000:00:1d.0-1.5.3/input1

*** Bug 626219 has been marked as a duplicate of this bug. ***

Does booting with the kernel option "usbcore.autosuspend=-1" make any difference?

(In reply to comment #6)
> Does booting with the kernel option "usbcore.autosuspend=-1" make any
> difference?

That makes it worse: no input from the USB keyboard is registered with that option.

Try running this command as root instead of using the usbcore.autosuspend option:

  for node in /sys/bus/usb/devices/*/power/control; do echo "on" >$node ; done

(In reply to comment #8)
> Try running this command as root instead of using the usbcore.autosuspend
> option:
> 
>   for node in /sys/bus/usb/devices/*/power/control; do echo "on" >$node ; done

This seems to fix it, yes.

With:

# uname -a
Linux zeus.lan 2.6.34.5-44.fc13.x86_64 #1 SMP Mon Aug 23 00:25:05 UTC 2010 x86_64 x86_64 x86_64 GNU/Linux

The following:

# echo on > '/sys/class/input/input4/device/power/control'

fixes the problem.

Which is the end effect of:

for i in /sys/class/input/input*/device; do
  C=`cat $i/bInterfaceClass`
  SC=`cat $i/bInterfaceSubClass`
  P=`cat $i/bInterfaceProtocol`
  # 03 01 01 == Human Interface Device, Boot Interface SubClass, Keyboard
  if [[ $C$SC$P == 030101 ]]; then
    echo "${i}"
    echo on > "${i}/power/control"
  fi
done 2>/dev/null

Should be fixed in 2.6.34.5-45

Posted upstream:

http://marc.info/?l=linux-usb&m=128272657410313&w=2

I confirm this is a problem in f13 updates-testing, and that it is fixed by kernel-PAE-2.6.34.6-47.fc13.i686 (and probably earlier).

Closing this now since the bug never hit a released kernel.