At 08:00 PM 7/20/99 +0200, you wrote:
>> That only works if the opcodes are unused (no instruction is mapped to
>> them). But if a different Z380 instruction is mapped to the R800 opcode,
>> it is not trapped but the (wrong) Z380 instruction is executed instead.
>> I think the latter is the case: R800 opcodes speficy different
>> instructions on Z380.
>
>Are you talking about the doubled instructions like the 'second' LD HL,(nn)
>etc.??? I think thase aren't used anyway, because their counterpart is
>smaller and faster.
No, I'm talking about MULUB and other R800-specific instructions. If the
opcode for MULUB would be unused on Z380, it could be trapped (kind-of
software interrupt) and a handler could be written that emulated MULUB. But
if the opcode is used for a new Z380 instruction, the trapping mechanism
doesn't apply.
>> Yes, the web page will serve as a reference and a log, but it won't be the
>> place where the discussion is held.
>
>Agree.
But I'm still worried. The two of us seem to agree on most points, but
we're not hardware developers. Could they please give their opinion? I
would even prefer a negative opinion over silence.
Bye,
Maarten
****
MSX Mailinglist. To unsubscribe, send an email to [EMAIL PROTECTED] and put
in the body (not subject) "unsubscribe msx [EMAIL PROTECTED]" (without the
quotes :-) Problems? contact [EMAIL PROTECTED] (www.stack.nl/~wiebe/mailinglist/)
****
