On Wed, Oct 12, 2011 at 1:40 AM, Alexandr Shadchin <alexandr.shadc...@gmail.com> wrote: > On Wed, Oct 12, 2011 at 2:29 PM, Alexandr Shadchin > <alexandr.shadc...@gmail.com> wrote: >> On Wed, Oct 12, 2011 at 1:59 PM, patrick keshishian <pkesh...@gmail.com> wrote: >>> On Tue, Oct 11, 2011 at 2:09 PM, Philip Guenther <guent...@gmail.com> wrote: >>>> On Tue, Oct 11, 2011 at 12:34 PM, S V <ner...@gmail.com> wrote: >>>>> I have one pretty specific problem, i have two USB keyboards >>>>> attached to the samsung nc10 laptop. And then i want to >>>>> attach different layouts to each i >>>>> stumble into one problem. >>>>> >>>>> $ xinput -list >>>> ... >>>>> So i can't separate each keyboard from others by device ids... >>>>> Can anybody help me to find where source of this behavior >>>>> lie and how i can fix it? >>> >>> this is interesting to me, because I wondered about this a few days ago. >>> >>>> The multiplexing of the two keyboards to one input is done here, I >>>> believe, in the wscons layer. If so, it should be possible to use >>>> wsconsctl to set the mappings of the two independently. For example, >>>> on my laptop, when docked with a USB keyboard plugged in, wsconsctl >>>> shows entries for "keyboard.map" (the built-in keyboard) and >>>> "keyboard1.map" (the USB keyboard). >>> >>> I have a netbook where I have: >>> >>> $ grep keyboard /etc/wsconsctl.conf | grep -v ^# >>> keyboard.encoding=us.swapctrlcaps >>> >>> Then I want to connect a Sun keyboard (with the Control key in its >>> proper location) via USB connection. >>> >>> $ sudo wsconsctl | grep keyboard | grep -e encoding -e type >>> wsconsctl: Use explicit arg to view keyboard.map. >>> wsconsctl: Use explicit arg to view keyboard1.map. >>> keyboard.type=pc-xt >>> keyboard.encoding=us.swapctrlcaps >>> keyboard1.type=usb >>> keyboard1.encoding=us >>> >>> In the virtual terminal (this is -current-ish amd64) each keyboard >>> acts independently with their respective encoding. On the netbook >>> keyboard, CapsLock acts as the Control key, while on the Sun keyboard, >>> Control acts as Control (as it should). Enabling caps-lock on the >>> netbook keyboard, does not affect the Sun keyboard and vice versa (as >>> expected). >>> >>> However, in X11, both keyboards act as if swapctrlcaps are enabled. >>> Pressing Control on the Sun keyboard turns on caps-lock on both >>> keyboards. >>> >>> I attempted to play with xorg.conf to see if I can define a separate >>> InputDevice section for the Sun keyboard, but gave up after a few >>> permutations. The X documentation (man-pages) pretty much suck. >>> >>> I wasn't going to post on this topic, but since it has been brought >>> up, might has well ask: Is what I want to do possible in Xorg? >>> >>> Cheers, >>> --patrick >>> >>> >> >> X use /dev/wskbd (it is mux) on default. it is one keyboard for X. >> >> for separate settings need to define section "InputDevice" for every >> keyboards (/dev/wskbdX) >> >> as example xorg.conf:Section "ServerLayout" ... InputDevice >> "Kb1" "CoreKeyboard" InputDevice "Kb2" "SendCoreEvents"EndSection >> Section "ServerFlags" # This off auto configure, therefore need >> configure mouse manualy Option "AutoAddDevices" "False"EndSection >> Section "InputDevice" Identifer "Kb1" Driver "kbd" Option >> "Device" "/dev/wskbd0" Option "XkbLayout" "..." ...EndSection >> Section "InputDevice" Identifer "Kb2" Driver "kbd" Option >> "Device" "/dev/wskbd1" Option "XkbLayout" "..." ...EndSection >> -- >> Alexandr Shadchin >> > > Errr, sorry. bad web gmail. > > See http://koba.devio.us/distfiles/xorg.conf
umm.... following that example, got the me to a state where the netbook keyboard was generate "gibberish". I couldn't log in. In console mode it worked just fine. Reverting the changes made to xorg.conf and restarting X didn't fix the problem, which is the weird part. I restarted X (actually xdm) multiple times, but something must have gotten cached somewhere, because the problem persisted. Note that switching to console mode the keyboard functioned fine. I had to reboot to get the netbook's keyboard working in X again. --patrick