> > 2) It touches things it should not touch. It replaces big parts of
> > drivers/char. I know that keyboard etc. are somehow related to
> > graphics, but you really do not want to _replace_ low level
> > drivers. This way, ggi being config option would mean having drivers
> > twice.
>
> Yes, well, I am not _the_ kernel hacker of GGI, but I have had it
> explained to me that it was necessary to be able to send keycodes etc to
> the right head/window/app/whatever. I will forward to those responsible.
I'm not sure about the current state of GGI drivers, but the last time
I've looked at the GGI driver replacements, I suspected GGI from having
exactly the same sharing problems between keyboard and PS/2 mouse I've
fixed a long time ago in the mainstream kernel. I think GGI should re-use
existing kernel code as much as possible (e.g., the keyboard driver is now
cleanly separated between low-level [arch-dependent] and high-level
[generic -- the keymaps etc.] parts and the low-level part seems to be
ready for use even in EvStack environment).
Have a nice fortnight
-- Martin `MJ' Mares <mj@ucw.cz> http://atrey.karlin.mff.cuni.cz/~mj/ Faculty of Math and Physics, Charles University, Prague, Czech Rep., Earth "A professor is one who talks in someone else's sleep."- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu