gcl-devel
[Top][All Lists]
Advanced

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

Re: [Gcl-devel] GCL + xemacs + ilisp-5.12.0


From: Gregory Wright
Subject: Re: [Gcl-devel] GCL + xemacs + ilisp-5.12.0
Date: 10 Jun 2002 18:13:47 -0400

Hi Camm,

I'll start delving into the details of the gcl + ilisp failure. I looked
into the ilisp.el files and it seems to be correctly identifying gcl.
Then it goes flooey. I can't enter anything (everything in xemacs is
locked up) but the CPU usages goes way up. That emacs segfaults is
probably telling us something.

ACL2 is indeed used for more than research on automatic theorem proving.
It is also used for verification of digital logic designs (one of AMD's
floating point units was verified in ACL2). We use it to verify network
protocols.

Best Wishes,
Greg


On Mon, 2002-06-10 at 18:01, Camm Maguire wrote:
> Greetings!
> 
> Gregory Wright <address@hidden> writes:
> 
> > Hi,
> > 
> > gcl (either 2.5.0 or 2.4.3) fails when used under xemacs (or emacs for
> > that matter) when used with the ilisp-5.12.0 package. SBCL, cmucl, clisp
> > are all OK. I had hoped that the bug was in the loop macro (as mentioned
> > by Vadim Z.) but switching to loop2 doesn't help.
> > 
> 
> Well, I've confirmed strange behavior with ilisp 5.11 in emacs and
> xemacs under Debian.  But I really don't know how to use ilisp.  What
> I see is no response on entering (+ 2 3).  No cpu time used at all.
> I'd guess a misdefinition of the gcl dialect somewhere in the
> ilisp.el files, but I really don't have the expertise to quickly
> confirm this.  Have you asked at the ilisp devel site?
> 
> > The principal symptom of the failure under xemacs is a hang with cpu
> > usage going to 90%. Emacs segfaults when ilisp is starting gcl as an
> > inferior lisp.
> > 
> > Methinks this is because ilisp is stressing the ANSI-ness of gcl. How
> > should I begin to track down? Any pointers would be apprecaited.
> > 
> 
> I don't know what you are doing (i.e. entering/evaluating) to get this
> cpu usage, but if your results are exhibited like mine on immediate
> startup of gcl, I doubt it has much to do with ansi compliance.  It
> could be the defpackage, which I know ilisp uses, and which we've done
> some work on recently in adding it.  But you'd at least see different
> behavior with 2.4.3, where defpackage isn't installed.
> 
> > 
> > To briefly revisit an older topic---what projects depend on gcl---I want
> > to second the note that pointed out that ACL2 depends on gcl. It's my
> > main interest in lisp, and it only builds out of the box with gcl. CMUcl
> > and clisp are supposed to work, but don't on ACL2.6 with recent versions
> > of CMUcl or clisp (RH6.2 linux).
> > 
> 
> Thanks for this tip!  Now we have two open source reasons for gcl to
> exist! :-).  I've looked at the acl2 web page briefly, and it looks
> quite complex.  I haven't yet gone through the tour on the website.
> Anyone find this useful in day to day work, as apart from theoretical
> studies in computational logic?
> 
> Take care,
> 
> 
> > Best Wishes,
> > Greg
> > 
> > 
> > -- 
> > 
> > Gregory Wright
> > Chief Technical Officer
> > PacketStorm Communications, Inc.
> > 20 Meridian Road
> > Eatontown, New Jersey 07724
> > 
> > 1 732 544-2434 ext. 206
> > 1 732 544-2437 [fax]
> > address@hidden
> > 
> > 
> > 
> > _______________________________________________
> > 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
-- 

Gregory Wright
Chief Technical Officer
PacketStorm Communications, Inc.
20 Meridian Road
Eatontown, New Jersey 07724

1 732 544-2434 ext. 206
1 732 544-2437 [fax]
address@hidden





reply via email to

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