From owner-freebsd-hackers  Thu Apr 15  6: 1: 2 1999
Delivered-To: freebsd-hackers@freebsd.org
Received: from alumni.cs.colorado.edu (alumni.cs.Colorado.EDU [128.138.192.80])
	by hub.freebsd.org (Postfix) with ESMTP id 555F114CCF
	for <freebsd-hackers@freebsd.org>; Thu, 15 Apr 1999 06:00:57 -0700 (PDT)
	(envelope-from atk@alumni.cs.colorado.edu)
Received: (from atk@localhost)
	by alumni.cs.colorado.edu (8.9.2/8.9.2) id GAA24333
	for freebsd-hackers@freebsd.org; Thu, 15 Apr 1999 06:58:33 -0600 (MDT)
Date: Thu, 15 Apr 1999 06:58:33 -0600 (MDT)
From: Alan T Krantz <atk@alumni.cs.colorado.edu>
Message-Id: <199904151258.GAA24333@alumni.cs.colorado.edu>
To: freebsd-hackers@freebsd.org
Subject: kernel threads
Sender: owner-freebsd-hackers@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.ORG


Hello,

I'm curious who is working on kernel threads and what the user and implementation
models will be. Will the threads work more like threads under solaris, linux or sgi
(I think linux and sgi hack the process table and threads are sort of like a 
special process where in solaris the threads seem to be a bit cleaner
I suspect they are similar to Mach threads). 

Also, someone posted that there would be N kernel threads (where N was the number of
processor). ifthis is the case does this mean the programmer will be limited to
creating N threads and the M other threads (M = T - N) will really be statically d
defined user threads or did the person really mean that only N threads would be 
active at a time?

I guess there are a couple of questions here - one is the user model
(which I suspect is suppose to be the same on all systems) and the other
is the implementation.

I'm not on this mailing list so please cc replies to
atk@cs.colorado.edu


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message