|info on portable assemblers firstname.lastname@example.org (1991-11-13)|
|Re: info on portable assemblers email@example.com (1991-11-14)|
|Re: info on portable assemblers firstname.lastname@example.org.OZ.AU (1991-11-18)|
|From:||email@example.com (Tom Wicklund)|
|Date:||Thu, 14 Nov 91 10:24:06 MST|
In comp.compilers firstname.lastname@example.org writes:
>I am interested in writing an assembler that will compile code for different
>architectures. If anyone has information on previous research regarding
>this, please let me know.
>By "compile code", I do not mean convert ASM code for an 80386 to 68030.
>Rather, I want to be able to compile 80386 ASM code and 68030 code.
>A primary consideration is speed.
Check with the Computer Based Education Research Lab (CERL) at the
University of Illinois. One of the people there (Don Lee if my memory is
correct) took the COMPASS assembler for CDC cybers and modified it to
handle a wide variety of machines.
I talked to him back around 1980 and he told me that he needed and 8080
assembler. He looked at COMPASS and discovered it was about 350 pages of
table manipulators and 17 pages of machine definition for the CPU and
peripheral processors. His assembler, called CROSS, was a fairly small
set of changes to compass which allow definition of the target assembler.
His goal was to handle microcode for a bit slice Cyber (300+ bit words) he
was designing (I've heard that he did complete both the machine and the
assembler, though it was a 3rd hand report).
When I left CDC in 1984 they had at least 8080, Z80, 8086, Z80 source
assembled to 8086 machine language, 6502, 68000, and 6800 (which I wrote).
I've got the CROSS definitions for most of these machines if anybody's
interested in what they look like (the definition language is not very
I don't know if anything was ever published about this work.
I'm also aware of a generalized cross assembler put out by the company
AnyWare Engineering in Boulder, CO. It's a commercial product running on
PCs but if your interest is more in having the assembler than the process
of writing it I know the author. He's a consultant and might be willing
to arrange something for other machines.
Return to the
Search the comp.compilers archives again.