Re: keyboard

Re: keyboard

[Michael Schmitz]

> I have a microsoft (bleah) "internet keyboard" attached to my G3.
> None of the new 19 keys works. When I press them it writes:
>
> Aug 15 15:56:35 Jay kernel: keyboard.c: can't emulate rawmode for keycode 165
>
> I see that there are some unused codes in the translation table in
> usb/keybdev.c. May I use them to map the new keys ?

Big win if all of these keys produce the same keycode :-)

You can solve your problem much easier with loadkeys and a tailored
console keymap. Why pollute the kernel with that stuff (except for the
learning experience)?

	Michael


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

keyboard

[Giuliano Pochini]

I have a microsoft (bleah) "internet keyboard" attached to my G3.
None of the new 19 keys works. When I press them it writes:

Aug 15 15:56:35 Jay kernel: keyboard.c: can't emulate rawmode for keycode 165

I see that there are some unused codes in the translation table in
usb/keybdev.c. May I use them to map the new keys ?

Bye.

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: keyboard

[Giuliano Pochini]

>> Aug 15 15:56:35 Jay kernel: keyboard.c: can't emulate rawmode for keycode
>> 165
>>
>> I see that there are some unused codes in the translation table in
>> usb/keybdev.c. May I use them to map the new keys ?
>
> Big win if all of these keys produce the same keycode :-)
>
> You can solve your problem much easier with loadkeys and a tailored
> console keymap.

No, I can't. mac_keyb.c uses a byte to store the pressed key. 1 bit is used
for pressed/released flag, so key codes range is 1-127. That special keys
have a code > 127 so currently they're not mapped at all.
I must change the kernel to remap that codes to something < 128.

Or perhaps I'll fix it by allowing int16_t keycodes just like on peecees.

Bye.
    Giuliano Pochini ->)|(<- Shiny Corporation {AS6665} ->)|(<-


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: keyboard

[Giuliano Pochini]

>> I see that there are some unused codes in the translation table in
>> usb/keybdev.c. May I use them to map the new keys ?
>
> Big win if all of these keys produce the same keycode :-)

Ehm, no. keypress code is different, but keyrelease code if 180
for all of them. What a mess.


Bye.
    Giuliano Pochini ->)|(<- Shiny Corporation {AS6665} ->)|(<-


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: keyboard

[Michael Schmitz]

> > You can solve your problem much easier with loadkeys and a tailored
> > console keymap.
>
> No, I can't. mac_keyb.c uses a byte to store the pressed key. 1 bit is used
> for pressed/released flag, so key codes range is 1-127. That special keys
> have a code > 127 so currently they're not mapped at all.

So it's a more generic problem ...

> I must change the kernel to remap that codes to something < 128.
>
> Or perhaps I'll fix it by allowing int16_t keycodes just like on peecees.

The latter, if at all possible (because that allows for using the console
keymap tools again if I'm not mistaken).

	Michael


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: keyboard

[Giuliano Pochini]

>> Or perhaps I'll fix it by allowing int16_t keycodes just like on peecees.
>
> The latter, if at all possible (because that allows for using the console
> keymap tools again if I'm not mistaken).

?  I don't know anything about that. Please tell me more.

Bye.
    Giuliano Pochini ->)|(<- Shiny Corporation {AS6665} ->)|(<-


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: keyboard

[Michael Schmitz]

> >> Or perhaps I'll fix it by allowing int16_t keycodes just like on peecees.
> >
> > The latter, if at all possible (because that allows for using the console
> > keymap tools again if I'm not mistaken).
>
> ?  I don't know anything about that. Please tell me more.

Come again? You said you could not use the user space keymap tools
(loadkeys) because only keycodes < 127 are currently used. You offered to
fix the kernel side problem by allowing for 16 bit keycodes, and I said
fine, do that if possible, it might solve the loadkeys problem in the
process. So where's the misunderstanding?

	Michael


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/