From owner-freebsd-hackers  Mon Mar 30 11:24:32 1998
Return-Path: <owner-freebsd-hackers@FreeBSD.ORG>
Received: (from majordom@localhost)
          by hub.freebsd.org (8.8.8/8.8.8) id LAA26991
          for freebsd-hackers-outgoing; Mon, 30 Mar 1998 11:24:32 -0800 (PST)
          (envelope-from owner-freebsd-hackers@FreeBSD.ORG)
Received: from terra.Sarnoff.COM (terra.sarnoff.com [130.33.11.203])
          by hub.freebsd.org (8.8.8/8.8.8) with SMTP id LAA26970
          for <freebsd-hackers@FreeBSD.ORG>; Mon, 30 Mar 1998 11:24:20 -0800 (PST)
          (envelope-from rminnich@Sarnoff.COM)
Received: (from rminnich@localhost) by terra.Sarnoff.COM (8.6.12/8.6.12) id OAA05144; Mon, 30 Mar 1998 14:22:51 -0500
Date: Mon, 30 Mar 1998 14:22:51 -0500 (EST)
From: "Ron G. Minnich" <rminnich@Sarnoff.COM>
X-Sender: rminnich@terra
To: Chris Csanady <ccsanady@friley585.res.iastate.edu>
cc: freebsd-hackers@FreeBSD.ORG
Subject: Re: Virtual Interface Architecture 
In-Reply-To: <199803301840.MAA03732@friley585.res.iastate.edu>
Message-ID: <Pine.SUN.3.91.980330141830.4075F-100000@terra>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-freebsd-hackers@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.ORG

http://www.extremelinux.org/activities/wkshop01/presentations/via/, 
also contact is
pbozeman@lbl.gov

pat's code does have some unet bits and pieces i understand. But it works 
and works well. 

word is that gigabit ethernet frames grow to 9300 bytes soon. Of course 
that will have interesting impacts on systems, with pci transactions of 
about 2048 cycles. Could be fun. 

> That should be good. :)  I was under the impression that you could just
> write a custom MCP to provide for VIA though, and get the same results.

I assume MCP here is the myrinet code. This has been tried, but the 
numbers I saw were not that great. 

Unet is not dead. I think unet should be tried by "someone" on freebsd. 
We actually put an entire separate switched ethernet onto our latest
cluster here for things like unet, hoping ... 

btw, a question for hackers: how hard would it be to get 10 microseconds 
latency on the tcp stack as it stands? Yes, i know: "impossible". how 
impossible? 

ron

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