[Ecls-list] CL-Containers error on ECL

[Pascal Costanza]

Sorry for the long delay. The :optimize-slot-access thing is weird  
and got me thinking: What version of ECL are you currently trying to  
use?

I am developing Closer to MOP support only for the current CVS  
version of ECL...


Pascal

On 6 Jun 2006, at 04:12, Gregory Martin Pfeil wrote:

> On 31 May 2006, at 13:25, Pascal Costanza wrote:
>
>> On 25 May 2006, at 15:42, Gary King wrote:
>>
>>>> ;;; OPTIMIZE levels: Safety=2, Space=0, Speed=3
>>>> ;;; Finished compiling /Users/gwking/.asdf-install-dir/site/lw-
>>>> compat/lw-compat.lisp.
>>>> ;;; Loading "/Users/gwking/.asdf-install-dir/site/lw-compat/
>>>> ecl-0.9h-darwin-unknown/lw-compat.fas"
>>>> ;;; Compiling /Users/gwking/.asdf-install-dir/site/closer-mop/ecl/
>>>> closer-mop-packages.lisp.Cannot find symbol "COMPUTE-
>>>> DISCRIMINATING-FUNCTION" in package #<"CLOS" package>
>>>> Broken at EVAL.
>>>> CLOSER-MOP>>
>>
>> I am not sure whether that's what you are looking for here, but it is
>> indeed the case that compute-discriminating-function doesn't exist in
>> ecl. Generic functions in ecl are not implemented according to the
>> CLOS MOP (as in MCL and OpenMCL).
>
> So, while Closer-MOP seems to be working for me in ECL on x86 (both  
> OS X and Linux), I've managed to finally reach parity with Gary  
> King in ECL on PPC. Attached are changes I've made to Closer-MOP to  
> make errors go away. I'm not claiming that they're fixes at all,  
> but figured that getting errors out of the way might show us how  
> far there is to go to get a solid MOP on ECL.
>
> Oh, Pascal, there is one thing I consider a fix in that patch: ECL  
> on PPC seems to not support :OPTIMIZE-SLOT-ACCESS, so I guarded  
> that option with #-powerpc.
>
> With this patch applied, I still get an error, but it happens  
> loading the FASL as opposed to during compilation, so my normal  
> "look at what was just emitted" strategy doesn't really work as well:
>
> 	;;; Loading "/Users/greg/Documents/Lisp/closer-mop/ecl/ecl-0.9h- 
> darwin-powerpc/closer-mop.fas"
> 	;;; Warning: Redefining class STANDARD-GENERIC-FUNCTION
> 	;;; Warning: Redefining class STANDARD-ACCESSOR-METHOD
> 	;;; Warning: Redefining class STANDARD-READER-METHOD
> 	;;; Warning: Redefining class STANDARD-WRITER-METHOD
> 	;;; Warning: Redefining class STANDARD-READER-METHOD
> 	;;; Warning: Redefining class STANDARD-WRITER-METHOD
> 	;;; Warning: Redefining class EQL-SPECIALIZER*
> 	
> 	Unable to display error condition
> 	   [Condition of type SIMPLE-ERROR]
>
> 	Backtrace:
> 	  0: LAMBDA
> 	  1: CALL-WITH-DEBUGGING-ENVIRONMENT
> 	  2: LAMBDA
> 	  3: LAMBDA
> 	  4: CALL-WITH-BINDINGS
> 	  5: DEBUG-IN-EMACS
>
> Using Closer-MOP in ECL on OS X/PPC isn't very important to me, but  
> I would like to find a fix to that CL-Containers error that started  
> this thread. If that means getting Closer-MOP to work on Gary's  
> platform, that's what I'll try to do. My intuition is that the bug  
> is in ECL, since it seems to work on a number of other CLs and the  
> error doesn't happen on the first occurence of the form.
> <closer-mop-darwin-ppc.patch>
>
> _______________________________________________
> Ecls-list mailing list
> Ecls-list at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/ecls-list

-- 
Pascal Costanza, mailto:pc at p-cos.net, http://p-cos.net
Vrije Universiteit Brussel, Programming Technology Lab
Pleinlaan 2, B-1050 Brussel, Belgium