From owner-freebsd-hackers  Wed Sep 18 20:52:34 1996
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id UAA11201
          for hackers-outgoing; Wed, 18 Sep 1996 20:52:34 -0700 (PDT)
Received: from mail.crl.com (mail.crl.com [165.113.1.22])
          by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id UAA11168
          for <freebsd-hackers@FreeBSD.ORG>; Wed, 18 Sep 1996 20:52:29 -0700 (PDT)
Received: from root.com (implode.root.com) by mail.crl.com with SMTP id AA06414
  (5.65c/IDA-1.5 for <freebsd-hackers@FreeBSD.ORG>); Wed, 18 Sep 1996 20:53:04 -0700
Received: from localhost (localhost [127.0.0.1]) by root.com (8.7.5/8.6.5) with SMTP id UAA01225; Wed, 18 Sep 1996 20:50:52 -0700 (PDT)
Message-Id: <199609190350.UAA01225@root.com>
X-Authentication-Warning: implode.root.com: Host localhost [127.0.0.1] didn't use HELO protocol
To: rohit@cs.UMD.EDU
Cc: freebsd-hackers@FreeBSD.ORG
Subject: Re: Giant Sized Ethernet Packets 
In-Reply-To: Your message of "Wed, 18 Sep 1996 22:33:53 EDT."
             <199609190233.WAA18658@darling.cs.UMD.EDU> 
From: David Greenman <dg@root.com>
Reply-To: dg@root.com
Date: Wed, 18 Sep 1996 20:50:52 -0700
Sender: owner-hackers@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

>I was wondering if it is possible to send giant sized ethernet packets (>1500
>bytes say 1550) using the current 'de' driver for the SMC 10/100 DEC 21140 
>cards.

   No.

>If not, would somebody know if this is possible with the above h/w
>at all? 

   The answer to this is complicated, but it basically ends up being "no".
I just spent about 15 minutes looking over the DC21140 hardware reference
manual. It appears that the chip can except larger frames, but it signals
an error condition when this occurs, so I don't think you could do this
as a normal mode of operation. It also appears that it is possible to
generate larger than 1500 byte packets, but the frames wouldn't be ethernet
(the type/length field would not be IEEE 802.3) and you'd have to invent your
own encapsulation. ...that's how I read it, anyway. Perhaps Matt Thomas will
correct me on this.

-DG

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