Greetings!
May I please suggest one more item prerelease? The function
proclamation in maxima for gcl is completely disabled at present,
leading to highly suboptimal compiles. When ansi-cl is on, the
declaim file is used, which is commented out entirely. I suspect,
though have not yet verified, that even then the proclaim file is
being loaded, it is being loaded in a manner which does not preserve
its effects for the compilation of subsequent (and previous) files.
Beyond this, the file needs updating, as maxima has put on a whole
host of new slatec stuff (for example) which is not included in this
static proclaims file.
The best way to do this to transparently keep up with maxima
development is to (for gcl)
1) (compiler::emit-fn t)
2) compile once
3) (compiler::make-all-proclaims ....)
4) (load "sys-proclaim.lisp")
5) compile again
I can show you what the C source looks like in both cases if you are
interested. One can also tell that functions have been proclaimed (in
gcl) by looking at their symbol-plist, which is not being set at all in
the current build process.
I don't know how to effect the above within defsystem, but I'd be
happy to commit a patch either within defsystem or make (as people may
wish) to achieve the above if there is consensus to do so.
Take care,
P.S. In more detail, without editing any current maxima files, one can
do
1) echo '(compiler::emit-fn t)(si::save-system "/tmp/gcl")' | gcl
2) export PATH=/tmp:$PATH
3) make
4) echo '(load "src'maxima-package.lisp")(load
"src'numeric/f2cl-package.lisp")(load "src'slatec.lisp")(load
"src'command-line.lisp")(compiler::make-all-proclaims
"*/*/*.fn" "*/*/*/*/*.fn")(load
"sys-proclaim.lisp")(si::save-system "/tmp/gcl1")' | gcl
5) mv /tmp/gcl1 /tmp/gcl
6) make clean && make
James Amundson <amundson@users.sourceforge.net> writes:
> Everyone,
>
> Here is what I would like to see done for the second beta release of
> 5.9.1:
>
> 1) Updates to windows-related items. (Done!)
>
> 2) Fix for CMUCL bug reported by Vadim V. Zhytnikov.
> -This can't be too difficult. I should probably be the one to do
> it.
>
> 3) Resolve the problem seen with invert() under GCL 2.6.2.
>
> 4) Add documentation for new plotting routines.
> -I was somehow convinced that I had already done this, but
> obviously I have not. Most of the necessary information is on the demo
> web page I created.
>
> 5) Apply the compilation-warning-reducing patches from Andreas Eder.
> Andreas has sent me a set of changes that dramatically reduce the number
> of compiler warnings in a maxima compile. I think this is very important
> work. We should be very grateful for it. Since it is a large patch,
> however, I would like to see at least one pair of eyeballs go over it
> before it goes in. I would also like to see testing on gcl, cmucl and
> clisp. If you would like to volunteer to look at the changes, please
> contact me and I will send you the changes. (It's a ~400k tarball.)
>
> I had hoped to get the beta 2 release out by now, but I've been ill
> since I returned from my last trip. Outside help would definitely speed
> things along.
>
> --Jim
>
>
> _______________________________________________
> Maxima mailing list
> Maxima@www.math.utexas.edu
> http://www.math.utexas.edu/mailman/listinfo/maxima
>
>
>
--
Camm Maguire camm@enhanced.com
==========================================================================
"The earth is but one country, and mankind its citizens." -- Baha'u'llah