[i2c] [PATCH 1/1] Input/Joystick Driver: add support AD7142 joystickdriver (v2)
rgetz at blackfin.uclinux.org
Sun Apr 27 03:09:09 CEST 2008
On Fri 25 Apr 2008 11:39, Dmitry Torokhov pondered:
> On Fri, Apr 25, 2008 at 06:09:52PM +0800, Bryan Wu wrote:
> > On Sat, Apr 19, 2008 at 1:15 AM, Dmitry Torokhov
> > <dmitry.torokhov at gmail.com> wrote:
> > > Hi Bryan,
> > >
> > > Why don't you use REL_X/REL_Y to report directional movement? Using
> > > buttons for that is pretty unusual.
> > >
> > In our development board, it includes UP/DOWN/LEFT/RIGHT 4 buttons. So
> > this driver matches this.
> > Please found some information about the hardware as below:
> What is the expected behaviur when a person touches one of these
It depends on the design of the button on the target hardware. They can be
pretty much any number of different shapes and sizes. Buttons, wheels,
scroll-bar, joypad, and touchpad shapes can be laid out as traces on the
sensor PCB. While we tested/wrote something for what we had hardware to
verify, it is not restricted to this.
As Bryan indicated - have a look at:
for a variety of options.
> Is it generatig a left/rigth mouse button clocks or moving a
> pointer in some direction? Judging by the category of the devices you
> are placing it in (joystick) it seems that latter is what you want,
> therefore it should generate REL_X and REL_Y events. Otherwise it
> should probably be called a button driver and live either in
> drivers/input/keyboard or driver/input/misc. Do you agree?
If someone makes a joypad, it is a joypad. If someone makes a button, it is a
button. The specific ad7142_work function will need to be modified to match
the hardware design.
So, I am not sure where it should live - maybe driver/input/misc since it can
be so many different things. The issue (in my mind) is how it gets exposed to
userspace. If someone makes a joypad - they want it to show up as a joystick,
so they don't have to modify any standard userspace applications to use it.
More information about the i2c