[comp.sys.amiga] MIT Scheme 7.0 for the Amiga

smr@beach.cis.ufl.edu (Samual Rushing) (12/11/90)

	After about two weeks of off-on attempts at porting AKCL to
the Amiga (under AmigaDOS, not Unix) , frustration got the best of me,
and so I dragged out a copy of 7.0 MIT Scheme and dug into it a
little.  I had it up and running in under 24 hrs!  (about 6 hours of
work, mostly mangling the makefile).  What a victory for portability!
I'd just like to say that that was the easiest port I've done in a
long time.  Good Job!

	Now, of course, I didn't do much to make the 'unknown'
configuration understand the Amiga, so things are a little brain-dead,
i.e, it doesn't understand the file-system - no interrupt
characters...  I can load the runtime and save a world, but can't
load the SF system.  (I think it has to do with 'pathname-unparse'.
That configuration requires about 3 Megs of contiguous (non-chip)
memory in the default (not the disk-paging) mode.

	My question is this:  how long before the Next Release?  I'm
excited enough about this to do a complete port, graphics and all, but
I don't want to waste my time if it'll be obsolete in a month or two.

For the curious: I brought it up using GCC (the BuRP port, using
modified & extended PDC libraries, all available from
titan.ksc.nasa.gov).  As far as I can tell, no problems were
introduced by any of the assembly.  The bignum code seems to work,
too.  Performance is 'pretty quick' on a 28Mhz GVP 68030, I haven't
tried it yet on a stock Ami.

If anyone wants to play with it, I'll upload the executable and a
world load (band) of the runtime to titan.


-Sam Rushing
rushing@titan.ksc.nasa.gov
smr@beach.cis.ufl.edu
--
(not (not (not (not (not ... )))))
-------------------------------------------------------------------------------
smr@beach.cis.ufl.edu				    rushing@titan.ksc.nasa.gov