Path: utzoo!mnetor!uunet!husc6!mailrus!ames!pasteur!ucbvax!hplabs!sdcrdcf!csun!sef
From: sef@csun.UUCP (Sean Fagan)
Newsgroups: comp.lang.c
Subject: Re: noalias comments to X3J11
Message-ID: <1160@csun.UUCP>
Date: 28 Mar 88 14:55:20 GMT
References: <12578@brl-adm.ARPA> <1988Mar25.172355.348@utzoo.uucp> <588@imagine.PAWL.RPI.EDU>
Reply-To: sef@csun.UUCP (Sean Fagan)
Organization: CSU, Northridge Sun Computer
Lines: 33

In article <588@imagine.PAWL.RPI.EDU> beowulf!lunge!jesup@steinmetz.UUCP writes:
>In article <1988Mar25.172355.348@utzoo.uucp> henry@utzoo.uucp (Henry Spencer) writes:
>>>>`Volatile,' in particular, is a frill for esoteric applications...
>>
>>Interrupt routines are almost by definition esoteric, not to mention highly
>>machine-specific.
>	'Volatile' is extremely important for dealing with real-world
>hardware.

Another use I haven't seen mentioned is for shared memory.  Since this can
be accessed (and modified) by more than one process, having a heavily
optimizing compiler move one location (or more) into a register would be a
bad idea.  Doing something like

	volatile char shm_seg[1024];
	shmat(...,shm_seg,...);

seems like it would be nice to have (I may have screwed up the syntax of the
shared memory stuff, pardon me).

>>"Noalias must go.  This is           |  Henry Spencer @ U of Toronto Zoology
>>non-negotiable."  --DMR              | {allegra,ihnp4,decvax,utai}!utzoo!henry
> ^^^^^^^^^^^^^^^^
>Now that we can agree on!
As has been stated before, when this man (dmr@alice) speaks, ANSI should
listen!

>     //	Randell Jesup			      Lunge Software Development
-- 

Sean Fagan                   uucp:   {ihnp4,hplabs,psivax}!csun!sef
CSUN Computer Center         BITNET: 1GTLSEF@CALSTATE
Northridge, CA 91330         (818) 885-2790