🚀 go-pugleaf

This morning I wrote a quick and simple program that, in it's current
state generates a byte-4-byte equivalent of the opcode table in TP6, and
although I don't have it installed, BP7. (I'm pretty sure that it uses
the same table - CAN ANYONE CONFIRM THIS?)

Obviously I need to translate the program into Pascal (I wrote it in
PL/I on the MF), but it will allow me, with certain restrictions, to
modify the builtin assembler hash table. Once it's ready, I will make it
available by email - sorry, don't have a web site. (Franz G, feel free
to add it to your site!)

As I mentioned in an earlier email about this, I will mercilessly chop
the five obsolete FPU instructions (F(N)DISI/F(N)ENI/FSETPM) but what
else would be useful to throw out?

ENTER/LEAVE, FNOP? Some of the synonyms for conditionals & repeat
prefixes, (REP=REPE=REPZ, REPNE=REPNZ, LOOPE=LOOPZ, LOOPNE=LOOPNZ) and
some Jcc's have three variants (anyone ever uses JNAE instead of JB?)

And what would be useful to add?

MOVSD/STOSD, CDQ/CWDE, SEGFS/GS, an added $66 to PUSHA/POPA, RDTSC
without sacrificing stand-alone AAM. Of course, with the program you
would be able to tailor the table yourself, but I would like to provide
at least one tried and tested example, so give me your suggestions.

BTW, I'm still in the process of figuring out the contents of the two
flag bytes for every instruction in the table, so MMX & SIMD are most
certainly OOTQ! (Partial decode: the last 2 bits give the length of the
following instruction - 1, if the bit before those is set it indicates a
286 instruction, rest: as clear as mud, any help in decoding them would
be appreciated)

Robert
--
Robert AH Prins
prinsra@williscorroon.com


--== Sent via Deja.com http://www.deja.com/ ==--
---Share what you know. Learn what you don't.---