Re: [PATCH v6 3/4] HID: playstation: add DualSense player LEDs support.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Tue, Feb 16, 2021 at 3:37 AM Marek Behun <marek.behun@xxxxxx> wrote:
>
> On Mon, 15 Feb 2021 17:11:14 -0800
> Roderick Colenbrander <roderick@xxxxxxxxxx> wrote:
>
> > On Mon, Feb 15, 2021 at 4:33 PM Marek Behun <marek.behun@xxxxxx> wrote:
> > >
> > > On Mon, 15 Feb 2021 15:00:30 -0800
> > > Roderick Colenbrander <roderick@xxxxxxxxxx> wrote:
> > >
> > > > What is the desired naming for these player LEDs? There is not an
> > > > officially designed function based on DT bindings. So far they used
> > > > "playstation::mac::ledX". When changing the naming scheme towards
> > > > "hid" and removing MAC, they would be: "hid%d::led1" etcetera.
> > >
> > > Hi,
> > >
> > > there is one more thing I forgot to mention in the LED name schema:
> > >   devicename:color:function-functionEnumerator
> > >
> > > So LED core can for example compose a names in the format:
> > >   switch0:green:lan-1
> > >   switch0:green:lan-2
> > >   switch0:green:lan-3
> > >   switch0:green:lan-4
> > >
> > > In your case I think the most appropriate name would be something like
> > >   hid0:color:indicator-1
> > >   hid0:color:indicator-2
> > >   ...
> >
> > I am trying to think if indicator is clear enough. Currently devices
> > use a mixture of names, which is obviously bad (wiimote uses p1-p4 at
> > the end, sony uses sony1-4 for DualShock 3, hid-nintendo uses
> > player1-4). I would at least like new drivers to standardize. In
> > particular in Android frameworks we have a need to map these LEDs back
> > to the Java InputDevice. Finding the LEDs has been quite painful so
> > far.
>
> Thinking about it more, function "player" should theoretically be
> reasonable. Maybe we should try sending a patch for review, adding this
> funciton to include/dt-bindings/leds/common.h, and see what others
> think of it...

FWIW, I am not sure 'player' would be a good fit here. I personally
preferred "indicator".
My reasons are because those LEDs are basically a matrix of LEDs, and
are supposed to be read as a whole. For player 1, you would light up
the 3rd LED only. And for player 2, you would light up LEDS #2 and #4.

So I would say that in this particular case, "player" would lead to
confusion because users would want to set player 1 on the controller
and would have to talk to the "player-3" LED.

If we keep the more generic "indicator", the one-to-one mapping is
removed, and it's clearer that userspace needs an adaptor to convert
"players" into "indicator".

For the older controllers with a dedicated "player" LED, like the PS3
and the Wiimote, "player" would make sense, yes.

Cheers,
Benjamin

>
> > If this is what is decided, I guess we should update the Linux gamepad
> > document at some point as well.
> >
> > > Are these LEDs of different colors which are impossible to determine?
> > > The string "hid%d::led1" you mention above does not indicate color.
> >
> > The DualSense LEDs are all white (at least so far?). On controllers
> > from other brands I have seen them be red or green. So could indeed
> > use: "hid%d:white".
>
> Yes, a constant for white color is defined in headers.
>
> > > Marek
> >
> > Thanks,
> > Roderick
>




[Index of Archives]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Security]     [Bugtraq]     [Linux OMAP]     [Linux MIPS]     [ECOS]     [Asterisk Internet PBX]     [Linux API]

  Powered by Linux