Hi,
Perhaps I can shed some light here. I wrote the socket.d patch
for clisp.
The socket patch fixes a bug related to uninitialized sockaddr
structures
on old style BSDs and OS X. You can _usually_ see the error
by running 'make check' ; the socket tests will fail. They don't
always fail because, as I mentioned, the bug relates to uninitialized
memory. If by chance the pad space in the sockaddr struct gets
filled with zeros, everything seems to work fine.
I also maintain the maxima port under the darwinports infrastructure.
Usually maxima has depended on clisp, but for a time it built
using openmcl by default. If anyone is interested in what patches
were used, they can get the darwinports cvs tree (see darwinports.org)
and look backward through the maxima tree to see how it was done.
I don't have copies of the patches on my machine anymore.
Note that I switched back to clisp because while maxima worked with
openmcl, the maxima + openmcl combination did not work with TeXmacs.
TeXmacs seems to be the favored front end for maxima on OS X.
(I also was able to build maxima under SBCL and had similar
problems with maxima + SBCL + TeXmacs.) The problems with
TeXmacs seem to reflect the slight differences in socket handling
in the different dialects, since TeXmacs talks to maxima through a
socket.
Best Wishes,
Greg
On Feb 18, 2006, at 11:17 PM, Robert Dodier wrote:
> Hello Yasuaki,
>
>> In case of OpenMCL, you need to modify 5.9.2 source
>> code slightly to make maxima run on it. This runs very fast,
>> though. The necessary modifications are posted while ago here.
>
> Now that you have write access to Maxima CVS,
> please commit the modifications to CVS, if you can.
>
> For what it's worth, I searched the mailing list archive
> but I was unable to find patches for OpenMCL.
>
> I am glad to hear there is interest in the OpenMCL port of Maxima.
> Thank you very much for working on it.
>
> All the best,
> Robert Dodier
>
> _______________________________________________
> Maxima mailing list
> Maxima at math.utexas.edu
> http://www.math.utexas.edu/mailman/listinfo/maxima