From jac@divsun.unige.ch  Sun Feb 11 04:21:59 1990
Received: by iris.ucdavis.edu (5.57/UCD.EECS.2.0)
        id AA07277; Sun, 11 Feb 90 04:21:59 PST
Received: from kwai.inria.fr by clover.ucdavis.edu (5.59/UCD.EECS.1.11)
        id AA05188; Sun, 11 Feb 90 04:26:13 PST
Received: by inria.fr with X.400; 11 Feb 90 13:23:17+0100
Received: by ch; 11 Feb 90 13:23:47+0100
Date: 11 Feb 90 13:23:47+0100
From: John Carroll <jac@divsun.unige.ch>
To: <info-hol@clover.ucdavis.edu>
Subject: Re: lisps
Message-Id: <109*jac@divsun.unige.ch>

In reply to Phil Windley <windley@cheetah.ucdavis.edu>
>I'm trying to build HOL using Allegro CL.  The cl-change-notes file says
>
>7. Implementations tested:
>
>      Procyon Common Lisp 2.1.3
>      Sun Common Lisp (Lucid) 2.1.1
>      Kyoto Common Lisp June 3, 1987
>      Austin Kyoto Common Lisp
>      Allegro Common Lisp 3.0.3 (but bug in interpreter means doesn't work)
>                                 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
>I'm getting a case sensative bug when I try and load the f-cl.fasl object
>file.  Is this the bug in the interpreter that makes it so that it doesn't
>work?  Does anyone have HOL working using Allegro?

The case sensitive error sounds as though your version of Allegro CL
was originally built with the case sensitive mode set (which is not standard
Common Lisp). Check this, and if necessary do (set-case-mode
:case-insensitive-upper) and dump a new core image. The Lisp part of HOL
should build properly in this new image.

However, Allegro 3.0.3 will not load all of the HOL ML code, and will not
build some of the theories - communication with Franz over this leads
me to suspect it is due to a garbage collector (not an interpreter) bug.
The bug is apparently fixed in version 3.1.beta.22, the currently
distributed version. I recommend you upgrade to this and you should
have no problems.

John Carroll


