From owner-freebsd-hackers  Thu Jun 19 16:18:23 1997
Return-Path: <owner-hackers>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.5/8.8.5) id QAA26140
          for hackers-outgoing; Thu, 19 Jun 1997 16:18:23 -0700 (PDT)
Received: from implode.root.com (implode.root.com [198.145.90.17])
          by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA26135
          for <hackers@FreeBSD.ORG>; Thu, 19 Jun 1997 16:18:15 -0700 (PDT)
Received: from localhost (localhost [127.0.0.1])
	by implode.root.com (8.8.5/8.8.5) with SMTP id QAA07656;
	Thu, 19 Jun 1997 16:20:15 -0700 (PDT)
Message-Id: <199706192320.QAA07656@implode.root.com>
X-Authentication-Warning: implode.root.com: localhost [127.0.0.1] didn't use HELO protocol
To: Chris Csanady <ccsanady@friley01.res.iastate.edu>
cc: sthaug@nethelp.no, hackers@FreeBSD.ORG
Subject: Re: Network concurrency problems!? 
In-reply-to: Your message of "Thu, 19 Jun 1997 18:08:44 CDT."
             <199706192308.SAA13293@friley01.res.iastate.edu> 
From: David Greenman <dg@root.com>
Reply-To: dg@root.com
Date: Thu, 19 Jun 1997 16:20:15 -0700
Sender: owner-hackers@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

>BTW, does the intel card use DBDMA?  From the driver, it looks like
>something similar.  How about the dec?

   What is DBDMA?

>Also, this is sortof an aside, but I am writing a new driver, and was
>wondering about using mem mapped vs io mapped registers.  How much of
>a difference in performance is there?  Or should I just do a lot of outl()'s?
>(like in linux.. but its really ugly imho..)
>
>The memory/PCI busses are already a significant bottleneck, so I dont think
>I can afford to sacrifice too much..

   Programmed I/O instruction on the P6 (and Pentium II) are serialized in a
way that makes them extremely expensive; memory mapped PCI operations are
much faster because of this.

-DG

David Greenman
Core-team/Principal Architect, The FreeBSD Project