[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gcl-devel] condition-proposal UPDATE

From: Camm Maguire
Subject: Re: [Gcl-devel] condition-proposal UPDATE
Date: 09 Nov 2002 13:14:43 -0500

Greetings, all, and thanks for your patience with my recently busy

Peter Wood <address@hidden> writes:

> On Fri, Nov 08, 2002 at 06:11:14PM +0100, I wrote:
> > Hi
> > 
> > Much to my surprise, ACL2 _does_ compile with a GCL incorporating the
> > changes I am trying for an ansi condition system.  Maxima,
> > unfortunately, does not.  The previous time I tried compiling ACL2 I
> > was using the saved_gcl image from a build with --enable-ansi, and the
> > ensuing package confusion broke the ACL2 build.
> > 
> WOOHOO.  Maxima _does_ compile.  I suspected it might, but was feeling
> a little bitter about having to grovel around in its build.  Anyway I
> bit the bullet and this is what I got:

Wonderful!  Actually, I feel a bit guilty that you had to discover
this yourself, as I knew from reading your previous emails that what
you were doing would not necessarily destroy these builds, but I
didn't have the time to reply.

> I was using GCL configured with --enable-ansi, and my changes to the
> condition system. My development GCL never gets installed, so I was
> running it in place.  This might have confused Maxima's build.

You can build with a compiled but uninstalled gcl source tree via
export PATH=<gcl source>/bin:$PATH.  The C_INCLUDE_PATH is not set

> I had to manually copy the cmpinclude.h file into Maxima's
> binary-gcl/, numerical/, and slatec/ directories.  I also made some

See above.  If you use the shell script wrapper in your source, the
include path should be set for you, and this should be unnecessary.

> small changes to Maxima's defsystem.lisp, otherwise it thinks GCL
> really is ANSI compliant, and wants to have WITH-COMPILATION-UNIT
> defined. 

This is a known item, on which several emails have been sent with a
patch to the maxima list, but I have not succeeded in either getting
cvs access, nor getting the attention of someone who does.

> With these interventions, it compiled all the way through.  I had to
> install it to run it (grrr) though, because it wont run in place.


> So, the changes I have made:
> 1) Work with ACL2 if GCL is configured with --disable-ansi

This currently must be the case, as acl2 explicitly blocks our ansi
build and its common-lisp package at present.

> 2) Work with Maxima (with very minor changes to defsystem.lisp)

And make test?  (Probably works I'd bet)

> 3) Work with the traditional image, although with new (ansi) condition
> types.


> The only requirement left is that it not break any builds for the 11
> Debian architectures.  I am sure this can be achieved.

Me too.

> So whaddya think, Camm?  Like to see the patches?

Yes I would!  Thanks again for all your work!

I have a few other thoughts about maxima/serror, and the build method
for the ansi image, but I'll post that separately.

Take care,

> Regards,
> Peter
> _______________________________________________
> Gcl-devel mailing list
> address@hidden
> http://mail.gnu.org/mailman/listinfo/gcl-devel

Camm Maguire                                            address@hidden
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah

reply via email to

[Prev in Thread] Current Thread [Next in Thread]