Re: [MiNT] [Fwd: Re: [Emutos-devel] FYI [Fwd: Re: Keyboard input failswith 1.16.1]]

["Konrad Kokoszkiewicz" <draco@obta.uw.edu.pl>]

>> So, unless I overlook something, it is to be concluded, that EmuTOS,
>> eventhough it claims to be TOS version 2.06, lacks the TOS 2.06 BIOS
>> functionality in this area, i.e. the kbdvec vector, which is used by
>> FreeMiNT to intercept keyboard scancodes.
>
> Emutos implements the kbdvecs structure as documented in the
> compendium. (see bios/aciavecs.[Sh]). It even uses it internally.

I guess so. But the Compendium does not list all keyboard vectors,
unfortunately. Strictly speaking, it lists vectors used by the TOS 1.0x.
Unfortunately, TOS 2.0 introduces one more vector, located at Kbdvbase()-4.
This vector is called with the keyboard scancode in d0. You may refer to the
freemint/sys/arch/acia.S to look at the actual code.

As I said, EmuTOS, eventhough it claims to be TOS 2.06, does not provide TOS
2.06 functionality. Funnily, FreeMiNT currently contains a workaround
installed for TOS 1.0x. But since EmuTOS claims to be TOS 2.06, the ACIA
interrupt routine does not get replaced, of course.

And, btw., you seem to trust Atari Compendium too much. It is full of
errors.

--
CVV
Konrad M.Kokoszkiewicz, http://draco.atari.org

** Ea natura multitudinis est,
** aut seruit humiliter, aut superbe dominatur.
** Taka to już natura pospólstwa, albo służalczo się płaszczy,
** albo bezczelnie się panoszy. (Liwiusz XXIV, 25).