Re: [PATCH v8 4/7] input: keyboard: Add driver for ASUS Transformer dock multimedia keys

[Svyatoslav Ryhel]

ср, 17 черв. 2026 р. о 00:23 Dmitry Torokhov <dmitry.torokhov@xxxxxxxxx> пише:
>
> On Tue, Jun 16, 2026 at 09:25:25AM +0300, Svyatoslav Ryhel wrote:
> > вт, 16 черв. 2026 р. о 07:26 Dmitry Torokhov <dmitry.torokhov@xxxxxxxxx> пише:
> > >
> > > Hi Svyatoslav,
> > >
> > > On Thu, May 28, 2026 at 08:32:00AM +0300, Svyatoslav Ryhel wrote:
> > > > From: Michał Mirosław <mirq-linux@xxxxxxxxxxxx>
> > > >
> > > > Add support for multimedia top button row of ASUS Transformer's Mobile
> > > > Dock keyboard. Driver is made that function keys (F1-F12) are used by
> > > > default which suits average Linux use better and with pressing
> > > > ScreenLock + AltGr function keys layout is switched to multimedia keys.
> > > > Since this only modifies codes sent by asus-ec-keys it doesn't affect
> > > > normal keyboards at all.
> > >
> > > I think using input handler to intercept ScreenLock + AltGr is quite
> > > awkward. I think this also passes the original key events (unless you
> > > make it a filter not a regular handler).
> > >
> > > I do not see benefit for reacting to AltGr+ScreenLock on other keyboards
> > > to activate the special mode on this one. So given the fact that you
> > > already mange the data stream when you split it into "serio" ports,
> > > maybe just intercept this key combo right there and create the input
> > > device and signal input events right there?
> > >
> >
> > Though it seems awkward at a first glance, media keys are integrated
> > with a standard keyboard in a detachable dock. It is highly unlikely
> > that media keys will be used with a different keyboard then the one
> > that is integrated with dock. Additionally, the ScreenLock key has a
> > code specific to this driver and is not in general use, so even if any
> > standard keyboard has AltGr but none has ScreenLock specific to this
> > driver except the dock itself. Handler is also set as observer so it
> > should not interfere with work of other input devices.
>
> I am not concerned about it interfering with other drivers, I am
> concerned about it unnecessarily connecting to unrelated devices
> (anything that declares EV_KEY).
>
> Again, I think having input handler is not appropriate here. I would
> fold this patch into the patch that introduces the 2 serio ports,
> enhanced the data stream analysis to detect your key combo, and then
> report through this new input device. You do not need to have the round
> trip through atkbd and the new input handler for this.
>

I will try with filtering first if you don't mind. If that will not
work I'll consider folding. Do you have any good examples of handler
with filter to use as an inspiration? Thank you.

> Thanks.
>
> --
> Dmitry