Janusz,
There are several emails on the subject in the archives. I think these 2, from Cliff Lawson (who works for/did work for Amstrad) make it pretty clear that it is a PS/2 device
http://www.earth.li/pipermail/e3-hacking/2006-April/000445.html http://www.earth.li/pipermail/e3-hacking/2006-April/000449.html
Matt
2009/12/11 Janusz Krzysztofik jkrzyszt@tis.icnet.pl:
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
e3-hacking mailing list e3-hacking@earth.li http://www.earth.li/cgi-bin/mailman/listinfo/e3-hacking