Hi,
Before I get to work on inventing a driver for the E3 mailboard that could be
accepted upstream, I'd like to learn more about the keyboard interface.
In 2006, Jasmine Strong wrote:
> > Huh --- I did look at the defconfig file, but it never occurred to me to
> > look at that bit (after all, the mailboard is most decidedly *not* a PC AT
> > keyboard)...
>
> Actually, it is- it's a PS/2 device. It just happens to be connected to
> some IO lines, rather than a FIFO, like it ought to be.
Jasmine (I hope you still read this ML), or anyone else who knows anything
more about that, could you please provide me with as much information on that
matter as possible? Can it be proved in any way that the keyboard is a PS/2
device?
For me, the fact that Matt's serio driver interacts correctly with a standard
atkbd keyboard driver means that reusing atkbd is not as wrong solution as
Dmitry suggests saying a new keyboard driver should be created. Maybe just
removing the scancode translation functionality from the Matt's serio driver
and somehow pushing a custom keymap for E3 into the atkbd instead could be an
acceptable solution here?
But before I take this way, I think I have to be able to prove that this
keyboard is not really different from a standard AT keyboard, and writing a
separate keyboard driver for it, just because the E3 has a non-standard
keyboard port, may not be a good idea.
Thanks,
Janusz