computation is now implicit and handled by rtlgen and dbgred.
Looking at the structure of some large programs (e.g. symb in
nscmutils) revealed that up to 32% of the text of a program was
%vector-index expressions. These get copied in every pass after
closconv/2. The `elegance' is not worth the performance hit. About
30% of the remaining text is %stack-closure-ref expressions. The
overhead could be reduced by introducing another KMP special form
(which would also mean that the DBG info would not need to compress
the expressions, saving time).
#| -*-Scheme-*-
-$Id: cleanup.scm,v 1.22 1995/07/24 17:38:03 adams Exp $
+$Id: cleanup.scm,v 1.23 1995/08/06 22:26:35 adams Exp $
Copyright (c) 1994-1995 Massachusetts Institute of Technology
(else
(internal-error "Unrecognized easy form" form))))
-(define cleanup/trivial/ops
- (list %vector-index))
+(define cleanup/trivial/ops '())
+; (list %vector-index)
(define cleanup/easy/ops
(append cleanup/trivial/ops
#| -*-Scheme-*-
-$Id: simplify.scm,v 1.15 1995/07/06 21:27:14 adams Exp $
+$Id: simplify.scm,v 1.16 1995/08/06 22:26:53 adams Exp $
Copyright (c) 1994-1995 Massachusetts Institute of Technology
(define (cost operator value)
(hash-table/put! *simplify/operator-open-coding-costs* operator value))
(cost not 0)
- (cost %vector-index -2)
+ ;;(cost %vector-index -2)
(cost %heap-closure-ref -2)
(cost %stack-closure-ref -2))