Path: utzoo!attcan!uunet!super!rminnich
From: rminnich@super.ORG (Ronald G Minnich)
Newsgroups: comp.sys.amiga.tech
Subject: Re: Could somebody tell me ...
Message-ID: <2209@super.ORG>
Date: 16 Dec 88 17:02:59 GMT
References: <8812151937.AA17954@postgres.Berkeley.EDU>
Sender: uucp@super.ORG
Reply-To: rminnich@duper.UUCP (Ronald G Minnich)
Organization: Supercomputing Research Center, Lanham, MD
Lines: 19

In article <8812151937.AA17954@postgres.Berkeley.EDU> dillon@POSTGRES.BERKELEY.EDU (Matt Dillon) writes:
> Some interesting stuff on lwp functions he is proposing.
Two thoughts: 
1) Have you looked at the way either Sequent's microtasking library or the 
   Spoc multiprocessor did it? 
Seems to me you want to create a set of context's for a set of lwps, then
fire up the lwp's using those pre-created segments. Bot Spoc and the Sequent
system do this well.
>	But the real reason is to get out of the main-loop driven standard
>of programming (mac users are fond of quoting the fact that multitasking
>does not avoid the need for a main-event-wait-loop).  Here, one would
>be able to segment a program into a lot of little programs each waiting for
>a different kind of event (or handling a different part of the data stream).
2) But doesn't there have to be a loop somewhere? I can't figure out how to make
it go away. There has to be something waiting for events and then dispatching
them. Or each of your lwp's has to poll, doesn't it? Even select() in the 
Unix kernel does polling in some form. I just can't see how to avoid it.
   What am i missing?
ron