Re: [Scheme-reports] Procedural equivalence: the last debate
John Cowan 05 Jun 2013 01:42 UTC
Alex Shinn scripsit:
> Is the idea basically to transform:
>
> (let ((f (lambda ...)))
> (... f ...)
> ...
> (... f ...))
>
> into:
>
> (let ((f (lambda ...))
> (location (generate-unique-value)))
> (... (box f location) ...)
> ...
> (... (box f location) ...))
Just so.
> Using addresses wouldn't work, generate-unique-value
> would have to increment a global counter (which need
> not be unique to this location). Though if the counter
> ever overflows into a bignum all optimization is lost.
Just use the address of any one copy of the procedure.
--
Yes, chili in the eye is bad, but so is your John Cowan
ear. However, I would suggest you wash your cowan@ccil.org
hands thoroughly before going to the toilet. http://www.ccil.org/~cowan
--gadicath
_______________________________________________
Scheme-reports mailing list
Scheme-reports@scheme-reports.org
http://lists.scheme-reports.org/cgi-bin/mailman/listinfo/scheme-reports