nyef wrote:
[...] the Ragsdale 6502 assembler, which looks as though it corresponds with the assembler syntax for the fig-FORTH source that I found
Back in the day, FIG-Forth and the Ragsdale assembler were central to an elaborate effort that allowed me to create a new Forth, using FIG-Forth as the "parent." I didn't call it meta-compiling but maybe that definition applies. It was a good thing the Ragsdale assembler was workable because it was all I had available. I guess I had acquired a DOS box by then but certainly I had no 65xx cross-assembler.
My main goal was to reassemble the FIG-Forth source supplied in the FIG Installation Manual, but incorporating optimizations based on my
KK computer's new instructions (such as 9-cycle ITC
NEXT )! FIG lives down at $0200, so -- skipping over a lot of ugly detail -- I basically moved the dictionary pointer DP up to $8000 and assembled and compiled the whole deal as a continuation of the original Forth. Then some ugly pokes severed the new name list from the old, and the new Forth at $8000 became independent and self-sufficient.
Did I mention ugly?! The only good thing you could say was that it worked. The new Forth was entirely satisfactory! But the hacks in its heritage were hideous, partly because there weren't any hooks in FIG to support what I was doing.
That became a subsequent goal -- to revise my new Forth so as to include some hooks. Doing so would let me repeat the process but in a more sanitary fashion. And I'd end up with a Forth that lived back down at $0200 again!
But it never happened -- my attention shifted elsewhere. I'd already implemented the KK optimizations, and that was by far my strongest motivation.
-- Jeff
_________________
In 1988 my 65C02 got six new registers and 44 new full-speed instructions!
https://laughtonelectronics.com/Arcana/ ... mmary.html