Re: [Gcl-devel] fasloading on Mac OS X

[Aurelien Chanudet]

On Wednesday, Aug 27, 2003, at 23:39 Europe/Paris, Camm Maguire wrote:

> Greetings!  Great!  Did you use sfaslelf.c as an example?  I take it
> from your earlier email that bfd is out for this task.  Any idea if
> this is a permanent situation?  It will be easier to maintain in the
> long term to the extent we can centralize on one relocation code.  In
> any case, at least in the near term, I think any files you'd like to
> commit along the lines of sfaslmacho.c and rsym-macho.c would be
> fine.  You da man on macho, in other workds :-).  But please post a
> proposed patch to the list before committing for the first couple of
> times around, if you don't mind.  Just the output of cvs diff -u would
> be fine.   Have you tested this with the normal (i.e. no alt-link)
> maxima build?
>
> Does the acl2 server problem persist?  I can get you the source if
> so.
>
> Take care,

As far as fasloading is concerned, what I've done so far is just use the
available APIs for dynamically loading code at runtime. This is just 
meant
as a temporary solution because what is needed for GCL goes past what
those APIs offer in terms of flexibility. I still have some work to do 
before
being able to build maxima without the alternate linking mechanism but
I promise I'll have a look at this as soon as possible. I guess it will 
be
incredibly difficult to maintain a separate relocation code in the long 
run,
therefore I'm planning to move on to bfd, which involves some more work.

Btw, the ftp server for acl2 at the University of Texas is still 
unreachable.

Aurelien