From owner-freebsd-mobile Sat Jan 25 15: 8: 0 2003 Delivered-To: freebsd-mobile@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 8E65337B401 for ; Sat, 25 Jan 2003 15:07:58 -0800 (PST) Received: from postal1.es.net (postal1.es.net [198.128.3.205]) by mx1.FreeBSD.org (Postfix) with ESMTP id E292343EB2 for ; Sat, 25 Jan 2003 15:07:57 -0800 (PST) (envelope-from oberman@es.net) Received: from ptavv.es.net ([198.128.4.29]) by postal1.es.net (Postal Node 1) with ESMTP id MUA74016; Sat, 25 Jan 2003 15:07:57 -0800 Received: from ptavv (localhost [127.0.0.1]) by ptavv.es.net (Tachyon Server) with ESMTP id BC4195D06; Sat, 25 Jan 2003 15:07:56 -0800 (PST) To: Joerg Wunsch Cc: "M. Warner Losh" , mobile@freebsd.org Subject: Re: How to make an Intel EtherExpress known to xe(4)? In-reply-to: Your message of "Sat, 25 Jan 2003 23:55:20 +0100." <20030125235520.A97813@uriah.heep.sax.de> Date: Sat, 25 Jan 2003 15:07:56 -0800 From: "Kevin Oberman" Message-Id: <20030125230756.BC4195D06@ptavv.es.net> Sender: owner-freebsd-mobile@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org > Date: Sat, 25 Jan 2003 23:55:20 +0100 > From: Joerg Wunsch > Sender: owner-freebsd-mobile@FreeBSD.ORG > > As M. Warner Losh wrote: > > > : In addition to make it known to the files in /sys/dev/xe/, > > : what else needs to be done to handle this card? > > > > Index: if_xe_pccard.c > ... > > Should be all that's needed, unless it has a modem built-in. > > Yep, thanks, it is all to make it known: > > xe0: at port 0x100-0x10f irq 11 function 0 config 1 on pccard0 > xe0: Intel CE3, bonding version 0x45, 100Mbps capable > xe0: DingoID = 0x444b, RevisionID = 0x1, VendorID = 0 > xe0: Ethernet address 00:a0:c9:bc:b5:ef > > However, when trying to use it i get: > > xe0: discard oversize frame (ether type 800 flags 3 len 1518 > max 1514) > > This didn't happen with that card under 4.x. I might look > into the differences... The connection is not usable that > way, it seems all packets are thrown away as being oversized. > > Maybe as a workaround just assume it's a full-sized packet, > and only discard the `oversized' bytes? If this was in > error, the higher protocol layers will throw the packet > away anyway. This is typical of receiving 802.1q tagged frames, but I have been getting this error on my current system with a Xircom RE-100 card and I am reasonably sure that i am not getting any tagged frames where it's located. I suspect a bug in the driver, but have not had time to look at it to this point. (Like, maybe, it is counting the FCS bytes as part of the frame, which would make a full sized frame 1500 bytes of data plus 6 bytes source, 6 bytes destination, 2 bytes of protocol information, and 4 bytes of FCS for a total of (1500+6+6+2+4=1518). This is only conjecture until someone looks at the driver. R. Kevin Oberman, Network Engineer Energy Sciences Network (ESnet) Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab) E-mail: oberman@es.net Phone: +1 510 486-8634 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-mobile" in the body of the message