Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!mcvax!enea!ttds!draken!sics!lhe
From: lhe@sics.se (Lars-Henrik Eriksson)
Newsgroups: comp.lang.prolog
Subject: Re: optimizing append/nrev [was Re: Standard of Prolog code]
Message-ID: <1554@sics.se>
Date: Wed, 21-Oct-87 06:04:57 EDT
Article-I.D.: sics.1554
Posted: Wed Oct 21 06:04:57 1987
Date-Received: Sat, 24-Oct-87 12:53:13 EDT
References: <2265@mulga.oz> <5005@utah-cs.UUCP> <2217@megaron.arizona.edu> <1528@sics.se> <2431@megaron.arizona.edu>
Reply-To: lhe@sics.se (Lars-Henrik Eriksson)
Organization: Swedish Institute of Computer Science, Kista
Lines: 10

In article <2431@megaron.arizona.edu> debray@arizona.edu (Saumya Debray) writes:
>Hmmm ... I wonder if changing the name of the predicate from "append"
>to "foo" results in a significant slowdown! :-)

It does not. People who have looked at the code tell me there is a
special WAM instruction for clauses of the form

	foo([X|A],B,[X|C]) :- foo(A,B,C)

where "foo" can be any function symbol.