Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!seismo!sundc!pitstop!sun!amdcad!ames!ucbcad!ucbvax!sdcsvax!ucsdhub!jack!man!nu3b2!nusdhub!rwhite
From: rwhite@nusdhub.UUCP (Robert C. White Jr.)
Newsgroups: comp.lang.c
Subject: Re: another auto-increment problem
Message-ID: <110@nusdhub.UUCP>
Date: Mon, 28-Sep-87 15:01:24 EDT
Article-I.D.: nusdhub.110
Posted: Mon Sep 28 15:01:24 1987
Date-Received: Wed, 30-Sep-87 04:35:14 EDT
References: <3680001@wdl1.UUCP> <6484@sgi.SGI.COM>
Organization: National University, San Diego
Lines: 24
Summary: Evaluation Order... A good reason.

In article <6484@sgi.SGI.COM>, archer@elysium.SGI.COM (Archer Sully) writes:
> 
> As for reasons why arguments to functions are evaluated right to left,
> here's a pretty good one.  It may be more efficient to evaluate each 
> argument just before it is pushed, rather than evaluating
> the argument list and then pushing.  Remember that most C-compilers
> push arguments right to left rather than left to right, for various
> and sundry reasons.  If you find one that pushes arguments right
> to left, it probably evaluates that way, too.


	There is a very good reason that the evaluation of arguments
to a function are evaluated in a right to left fassion. [on some
machines anyway, I don't know if it is a "universal" truth or not.]

	In any function which takes a variable number of arguments
of possibly variant size, it is best to have the argument which
defines the size and type of future arguments on the TOP of the 
stack.  If it were pushed first, it would be under N words of stack
where N is an undefined number between 0 and MAXSTACK-4(or more or less).
This is not as uncommon as it seems all of the [sf]printf and [sf]scanf
functions use the format string for just such a variant descriptor.

Robert ("So who asked you anyway?") White.