Small finding about NewObjectMemory>>clone:

 
Hi,

I have just spotted that the method NewObjectMemory>>clone as found in
VMMaker-oscog-IgorStasenko.116.mcz (with hyphen):

clone: oop
        "Return a shallow copy of the given object. May cause GC"
        "Assume: Oop is a real object, not a small integer.
         Override to assert it's not a married context"
        self assert: ((coInterpreter isContext: oop) not
                                or: [(coInterpreter isMarriedOrWidowedContext: oop) not]).
        ^super clone: oop

causes compilation error in the interpreter.

It seems that it has been fixed in VMMaker.oscog-eem.116.mcz (with dot)

clone: obj
        "Return a shallow copy of the given object. May cause GC"
        "Assume: Oop is a real object, not a small integer.
         Override to assert it's not a married context"
        self assert: ((coInterpreter isContext: obj) not
                                or: [(coInterpreter isMarriedOrWidowedContext: obj) not]).
        ^super clone: obj

basically s/oop/obj/g - I had to make similar change in my local copy
of VMMaker.

So, Igor: please merge this into the CMMaker-oscog branch.

Thanks.

PS Is this some kind of inliner bug? When super method is inlined,
argument identifier remains the same instead of being substituted.

--
Dimitry Golubovsky

Anywhere on the Web

On Aug 15, 2011, at 6:58 PM, Dimitry Golubovsky <[hidden email]> wrote:

yes, bug or limitation.  The super send inliner is a very simple-minded quick hack that can't rename argument names when inlining the superclass method, and so argument names in the super- and sub-class methods must match.  In 115 I didn't notice the bug since it only affected code in an assert.  In 116 I made the argument names match and added an error check to the super inliner.

Eliot (phone)
> --
> Dimitry Golubovsky
>
> Anywhere on the Web

On 16 August 2011 03:58, Dimitry Golubovsky <[hidden email]> wrote: Will do.



--
Best regards,
Igor Stasenko AKA sig.