home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
ftp.cs.arizona.edu
/
ftp.cs.arizona.edu.tar
/
ftp.cs.arizona.edu
/
icon
/
newsgrp
/
group98c.txt
/
000026_icon-group-sender _Mon Sep 14 12:36:13 1998.msg
< prev
next >
Wrap
Internet Message Format
|
2000-09-20
|
2KB
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id MAA20475
for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 14 Sep 1998 12:36:10 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
id AA01374; Mon, 14 Sep 1998 12:35:42 -0700
Message-Id: <35FD3E12.56A4@gte.net>
Date: Mon, 14 Sep 1998 11:02:26 -0500
From: MJE <evans@gte.net>
Reply-To: evans@gte.net
Organization: None
X-Mailer: Mozilla 3.01 (Win95; I)
Mime-Version: 1.0
To: icon-group@optima.CS.Arizona.EDU
Subject: Re: Context Switching
References: <199809112225.RAA07906@segfault.cs.utsa.edu>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Content-Transfer-Encoding: 7bit
Status: RO
Does it really require threads to do context switching? I can see the
possibility, but it's not as though the context switching that happens
in Icon requires a "whole separate executable," is it?
I have written subroutines in C that use static local variables to
remember their "current state," so that every time you call them, they
remember what was going on at the last call, and just pick up from
there. They work like charms.
It seems to me something similar could happen in Icon without the
complication of threads.
Maybe I'm missing something fundamental about the purpose of context
switching in Icon? I haven't been using the language all that long.
Mark
Clinton Jeffery wrote:
>
> Cheyenne Wills and Todd Proebsting recently pointed out (or confirmed)
> that co-expressions can relatively easily be built on top of threads.
>
> Has anyone implemented and compared the performance of co-expressions using
> a modern threads package to Icon's assembler context switch on the same
> platform? It would be interesting to know whether we might just switch over
> to threads wholesale, or whether they still have a significant performance
> cost because the synchronization and concurrency of threads are "overkill"
> for co-expressions.
>
> Clint Jeffery, jeffery@cs.utsa.edu
> Division of Computer Science, The University of Texas at San Antonio
> Research http://www.cs.utsa.edu/research/plss.html