From owner-freebsd-hackers  Sun Aug 17 18:48:21 1997
Return-Path: <owner-freebsd-hackers>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.5/8.8.5) id SAA10958
          for hackers-outgoing; Sun, 17 Aug 1997 18:48:21 -0700 (PDT)
Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127])
          by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id SAA10952
          for <freebsd-hackers@FreeBSD.ORG>; Sun, 17 Aug 1997 18:48:17 -0700 (PDT)
Received: (from root@localhost)
	by dyson.iquest.net (8.8.6/8.8.5) id UAA00359;
	Sun, 17 Aug 1997 20:48:12 -0500 (EST)
From: "John S. Dyson" <toor@dyson.iquest.net>
Message-Id: <199708180148.UAA00359@dyson.iquest.net>
Subject: Re: The low priority items
In-Reply-To: <33F78D7E.2610@theshop.net> from Charles Ebert at "Aug 17, 97 06:47:10 pm"
To: kd5ob@theshop.net
Date: Sun, 17 Aug 1997 20:48:12 -0500 (EST)
Cc: freebsd-hackers@FreeBSD.ORG
Reply-To: dyson@FreeBSD.ORG
X-Mailer: ELM [version 2.4ME+ PL31 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-freebsd-hackers@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

> I'm a little slow.  Did I read correctly that Threaded processes are
> not supported?  And the next line down from that mentioned something
> of the same, however, can I assume it is to deal with multiprocessor
> CPU boards as well?
>
Threaded processes are not "supported" yet, but code exists in the
-current FreeBSD kernel.  I am one of the people creating that support.
We do have a pthread(s) library, but that doesn't take advantage of
SMP (yet.)

> channel scsi cards and all that.  
> 
> I was always told of UNIX's unique ability to run multiple processor
> systems.  I never imagined that this system would have a problem with
> threading.  I was hoping that the multiple processor issue would
> have a higher priority than it does.  
>
There aren't any problems.  Different OSes have different advantages
and disadvantages.  For example, you can create an entirely new process on
FreeBSD in a few hundred microseconds.  A kernel-based thread can be created
in several ten's of microseconds.  Creating new processes on NT (or
many commercial versions of U**X) is/are much much slower.

The multiple processor support is a pretty much fully funded effort, and
user-land code (like compiles) already sees a significant performance
improvement on SMP machines, when running the -current kernel.  When
we "support" an SMP release, it will be very very good.  I am also working
on the SMP team.

John