[PATCH] HAL Keymap Quirks: hal-info keyboard support for BenQ R22e

Lennart Poettering mzuny at 0pointer.de
Mon Jul 30 08:09:58 PDT 2007


On Mon, 30.07.07 17:00, Danny Kukawka (danny.kukawka at web.de) wrote:

> 
> On Montag, 30. Juli 2007, Lennart Poettering wrote:
> > On Mon, 30.07.07 16:28, Danny Kukawka (danny.kukawka at web.de) wrote:
> > > On Montag, 30. Juli 2007, Paul Menzel wrote:
> > > > Am Montag, den 30.07.2007, 12:43 +0100 schrieb Richard Hughes:
> > > > > 6d for both brightness up and down?
> > > >
> > > > Yes. I checked it several times.
> > >
> > > But then you never know if it go up or down. You can't (?) map the key
> > > two times.
> >
> > I guess this one of the cases where the brightness level may be
> > queried from the hadrware via some special driver and the keypress
> > event is basically just a notification for the software that the
> > brightness has changed and that the software should read the new
> > brightness from the hardware and possibly show a nice OSD gauge when
> > it changed from what was read previously.
> 
> But wouldn't it be better in this case to not map the keys and let the 
> software/driver/whatever get the unknown keyevent, check which one it is, 
> send an new key event (via the input layer) and do the related mapping (to 
> brightness up/down) before? The current setting make IMO no sense.

Indeed, the keymap that has been proposed makes no sense.

I think we should add something like a generic KEY_BRIGHTNESSCHANGED
key and than have g-p-m handle that stuff in a generic fashion.

I see no reason why this key needs to be handled in-kernel. If you did
this in-kernel the right way would be to expose this as some kind of
notification event from the backlight class driver. However the 
backlight class in the kernel doesn't support change notification
right now, hence doing this in-kernel would require a bit of
non-trivial code.

Just my 2¢.

Lennart

-- 
Lennart Poettering                        Red Hat, Inc.
lennart [at] poettering [dot] net         ICQ# 11060553
http://0pointer.net/lennart/           GnuPG 0x1A015CC4


More information about the hal mailing list