From owner-freebsd-net  Sat Jan  8 10: 6:21 2000
Delivered-To: freebsd-net@freebsd.org
Received: from cs.rice.edu (cs.rice.edu [128.42.1.30])
	by hub.freebsd.org (Postfix) with ESMTP id F1A2C1576F
	for <freebsd-net@freebsd.org>; Sat,  8 Jan 2000 10:06:17 -0800 (PST)
	(envelope-from aron@cs.rice.edu)
Received: (from aron@localhost)
	by cs.rice.edu (8.9.0/8.9.0) id MAA13508;
	Sat, 8 Jan 2000 12:06:00 -0600 (CST)
From: Mohit Aron <aron@cs.rice.edu>
Message-Id: <200001081806.MAA13508@cs.rice.edu>
Subject: Re: performance of FreeBSD-current as SMP
To: alex@big.endian.de (Alexander Langer)
Date: Sat, 8 Jan 2000 12:06:00 -0600 (CST)
Cc: weyrich@goodnet.com, freebsd-net@freebsd.org
In-Reply-To: <20000108183130.A13891@cichlids.cichlids.com> from "Alexander Langer" at Jan 8, 2000 06:31:30 pm
X-Mailer: ELM [version 2.4 PL25]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-freebsd-net@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.org


> 
> Yes, I know. I just wanted to mention it.
> Well, it's known that SMP produces this overhead.
> 
> The same is for NT and Linux, if you enable SMP and use only 1
> prozessor.
> 

Neither are very good OS's as far as SMP support is concerned. FreeBSD beats
them in networking and there's no reason why the case should be any different
for SMP support.


> It's the protocol-overhead or other overhead or such (don't know the
> internals)
> 

Its probably the overhead of lock acquirement/release. Which means the
implementation of locking in FreeBSD needs some improvement.




- Mohit


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