Re: [Gcl-devel] Re: possible dotimes enhancement

[Camm Maguire]

Greetings!

"Paul F. Dietz" <address@hidden> writes:

> > Separately, at some point we have to figure out why the optimizer
> > cannot currently optimize
> > (declare (fixnum i)) (aref a (+ i 1))
> > but only
> > (let ((m 0)) (declare (fixnum m i)) (setq m (+ i 1)) (aref a m))
> 
> AREF requires each of its indices be a fixnum, so it's legal to
> treat
> 
>    (aref a i j ... k)
> 
> as
> 
>    (aref (the fixnum i) (the fixnum j) ... (the fixnum k))
> 

Paul, does this mean I can do something like

aref -> si::arefold

(defmacro aref (a &rest sub &aux i new-sub) 
  (dolist (i sub)
    (push `(the fixnum ,i) new-sub))
  `(arefold ,a ,@new-sub))

???

Take care,


>       Paul
> 
> 
> 
> 
> _______________________________________________
> 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