Path: utzoo!mnetor!uunet!lll-winken!lll-lcc!ames!umd5!brl-adm!brl-smoke!gwyn
From: gwyn@brl-smoke.ARPA (Doug Gwyn )
Newsgroups: comp.lang.c
Subject: Re: Optimization (was: Re^3: gotos)
Message-ID: <7819@brl-smoke.ARPA>
Date: 3 May 88 13:53:44 GMT
References: <2586@geac.UUCP> <3950003@hplvly.HP.COM> <1988Apr24.004842.3251@utzoo.uucp> <2606@ttrdc.UUCP> <487@sas.UUCP>
Reply-To: gwyn@brl.arpa (Doug Gwyn (VLD/VMB) <gwyn>)
Organization: Ballistic Research Lab (BRL), APG, MD.
Lines: 9

In article <487@sas.UUCP> bts@sas.UUCP (Brian T. Schellenberger) writes:
>In any case, any quality implementation allows you turn off all optimization.

You've taken a somewhat simplistic view of optimization.  There is no well-
defined meaning for "turning off all optimization", since the C language
does not map one-to-one unambiguously onto machine language.  All C
compilers with which I am familiar, and very likely all serious compilers,
automatically perform certain code generation "optimizations" (these should
be called "improvements") that save both time and space over naive code
generation.  One virtually never has the option of disabling these.