From owner-freebsd-hackers  Tue Dec 16 20:21:46 1997
Return-Path: <owner-freebsd-hackers>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.7/8.8.7) id UAA05372
          for hackers-outgoing; Tue, 16 Dec 1997 20:21:46 -0800 (PST)
          (envelope-from owner-freebsd-hackers)
Received: from word.smith.net.au (vh1.gsoft.com.au [203.38.152.122])
          by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id UAA05358
          for <hackers@FreeBSD.ORG>; Tue, 16 Dec 1997 20:21:35 -0800 (PST)
          (envelope-from mike@word.smith.net.au)
Received: from word (localhost [127.0.0.1])
	by word.smith.net.au (8.8.8/8.8.5) with ESMTP id OAA01477;
	Wed, 17 Dec 1997 14:46:08 +1030 (CST)
Message-Id: <199712170416.OAA01477@word.smith.net.au>
X-Mailer: exmh version 2.0zeta 7/24/97
To: jak@cetlink.net (John Kelly)
cc: Mike Smith <mike@smith.net.au>, FreeBSD Hackers <hackers@FreeBSD.ORG>
Subject: Re: 3com 3c509 card 
In-reply-to: Your message of "Wed, 17 Dec 1997 05:09:35 GMT."
             <349a59c3.99678713@mail.cetlink.net> 
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Date: Wed, 17 Dec 1997 14:46:08 +1030
From: Mike Smith <mike@smith.net.au>
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by hub.freebsd.org id UAA05359
Sender: owner-freebsd-hackers@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

> On Wed, 17 Dec 1997 14:04:26 +1030, Mike Smith <mike@smith.net.au>
> wrote:
> 
> >"AT Bus Design" (Edward Solari, Annabooks)
> 
> BTW, I checked on getting that book and it seems to be no longer
> available.

Hmm.  I'd better steal this copy then.  If you're doing AT bus design 
work it's invaluable, if a bit terse.  8)

> >Now, if you take into account just a little CPU overhead, interrupt 
> >cycle overhead, metadata overhead, you will quickly see that you're 
> >not going to route 4MB/sec aggregate traffic through anything on the ISA
> >bus, period.
> 
> I seem to have one 486 board running at 12 MHz ISA and clock doubled
> 50/100MHz DX4-100 CPU will no apparent problems.  This has reduced my
> (measured) 8 bit I/O cycle time to 0.65 usec.  I don't have a suitable
> method for measuring the 16-bit cycle time, but I'm hoping for nearly
> 4MB through this supercharged bus in 16-bit mode. 

You can expect that it will be half, based on the 6 clock/3 clock 
ratio, so about 325ns.

> I'm only getting
> about 700K per card, so I don't think I need a full 4MB anyway.  I
> might even be able to use five interfaces -- they won't be saturated
> all the time.

Yup.  The other significant point is that you are routing, not 
bridging, so you are more concerned with through traffic than 
all-networks-aggregate.
 
> >Personally, I'd recommend a 4-port Zynx card.
> 
> I have a small truckload of SMC Ultras I need to use.

That does constrain you somewhat.  8)

mike