home *** CD-ROM | disk | FTP | other *** search
- Dear tester,
-
- This archive contains the Amiga version of CLISP, as of beginning of April
- 1993. Please see files clisp.man and README too.
-
- There are now two distinct versions of CLISP. These will be referred as
- *-high and *-low. Depending on your machine, you'll be able to run one or
- the other, or both, or - hopefully not - none.
-
- o The *-low Version:
-
- + has been compiled with a 68000 in mind. It can only deal with a 16MB
- address space ranging from 0 to 0x00FFFFFF. It clears the upper 8 bits of a
- 32 bit address before putting it on the bus and won't forget to check the
- memory it gets from the system.
-
- + uses the MEMF_24BITDMA flag to AllocMem() to ensure that it does not get
- memory outside the 16MB address range. DMA-able memory is a bit too hard a
- requirement, but there are no other flags.
-
- + can be made to run even on a A3000 or A4000 by patching the executable to
- load the code and data hunks into CHIP (24BITDMA would suffice) memory. The
- supplied ratom ARexx script can do this for you, but you really don't want
- to run CLISP out of CHIP memory now that there is the *-high version.
-
- o The *-high version:
-
- + has been compiled for 68020 and higher, won't run on 68000 CPUs.
-
- + is able to deal with addresses from 0 to 0x07FFFFFF (0x0FFFFFFF soon),
- thus working on A3000 and A4000 machines with basic fast memory. I have
- been told that GigaMem uses addresses between 0x20000000 and 0x60000000.
- CLISP cannot run with that.
-
- o If you start (REXX-LOOP), the only exit is the exit-loop.cl ARexx script.
-
-
- Changes:
-
- o Fixed all known problems with division and float conversion of all
- types (short, single, double and long floats).
-
- o Fixed bignum multiplication.
-
- o CLISP now always allocates its own stack and should not cause MungWall or
- Enforcer hits. I don't know whether this stack allocation feature may cause
- compatibility issues with future Amigas.
-
- o (SETQ *PRINT-PRETTY* T) is set in lispinit.mem for faster output,
- otherwise most console or handle-stream output is done one character at a
- time, causing a lot of DOS packets traffic and making it really slow.
-
- o Slightly fixed Workbench Startup so that you can run it from
- workbench, but it will still only read project files from the current
- directory. I still need to add NameFromLock() for 1.3.
-
- o The lisp.run executable now contains symbolic information. I did
- ; this because I'm using gcc-2.3.3 -O2 -fomit-frame-pointer which puts
- ; several bugs in the compiler output, two of them I corrected by
- ; hand-patching the assembly files. You may find more :-( and report in
- ; what function these happened.
- The present version has been compiled with gcc -O1 and still needed
- one patch. Symbolic information has been left in.
-
- Enjoy and send me all sorts of comments, bug reports and suggestions.
-
- Joerg.
- hoehle@inf-wiss.ivp.uni-konstanz.de
-
-