[RFC] Automatic modifier update of slave devices

[Peter Hutterer]

On Sun, Feb 23, 2014 at 09:53:13PM -0800, Keith Packard wrote:
> Peter Hutterer <peter.hutterer at who-t.net> writes:
> 
> > Right now, if any of the LEDs are on (e.g. NumLock) and a keyboard is
> > plugged in, that new keyboard won't have the LED light up. Likewise, hitting
> > NumLock on one keyboard won't light it up on the second keyboard.
> > For many, this was largely a nonissue because GNOME would overwrite the
> > numlock LED, but a recent change has brought this to light.
> >
> > This it is a long-standing issue (at least server 1.7, possibly back to 1.4
> > or even longer).
> >
> > This is one attempt of solving it by forcing a full modifier state update
> > down the slave device, once when it is attached, once when it is enabled. So
> > attaching a floating slave to a master keyboard with numlock on will light
> > up the LED too.
> 
> That sure sounds like what we want -- multiple keyboards driving the
> same master should all have the same modifier state. After all, an app
> listening to the master will see only one modifier state, and if the
> slaves have different states, what will the master report?

yeah, it does seem like the most sensible behaviour but right now the
keyboards are completely independent and the event routing is based on that.
so if you hit ctrl on one keyboard, alt on the other, the master will have
ctrl+alt down, while each keyboard only has either modifier set.
somewhat similar to the button state.

Cheers,
   Peter