From owner-freebsd-hackers Thu Apr 17 10:13:06 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA21798 for hackers-outgoing; Thu, 17 Apr 1997 10:13:06 -0700 (PDT) Received: from elwood.pionet.net (ELWOOD.pionet.net [199.120.116.1]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id KAA21790 for ; Thu, 17 Apr 1997 10:13:00 -0700 (PDT) Received: from pm4_72.pionet.net by elwood.pionet.net (5.65v3.2/1.1.10.5/25Oct96-0835PM) id AA03715; Thu, 17 Apr 1997 12:15:35 -0500 Message-Id: <33564A75.476@pionet.net> Date: Thu, 17 Apr 1997 12:06:13 -0400 From: Tyson Boellstorff X-Mailer: Mozilla 2.02E (OS/2; I) Mime-Version: 1.0 To: hackers@freebsd.org Subject: Re: availability of 100Mb ethernet drivers Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Doug Rabson wrote: > I think the vx driver also supports 100 Mbps operation, although I > haven't tried it myself (I have a 3com 3C900 10/100Mbps controller, > but on a 10 Mbps LAN). I tried it. It gives appalling performance (50k/sec) on a 3c905. I can test any fixes in this area since I still have the 3com card sitting on my desk but I don't particularly want to hack on this driver. Before you go to all that work: 1) What is your BIOS version? It makes a BIG difference with this card. 2) Any 10 MB equipment on your LAN? Try it on a straight 100 MB segment, and see if it's a switching problem. 3) All else fails, if you have a different motherboard, try it. I've seen results that way, but no pattern to it. 4) Try another of the same card. Some switch better. 5) Try downloading & running the latest card configuration utility. I've heard that it makes a difference. 6) Here are stats on it on an Intel Thor motherboard, 1.00.03.CN0T BIOS (take with a grain of salt: this was a Win 95 client talking to Solaris 2.5.1 host, ran at midnight, so no traffic to speak of): Thor 03 BIOS, get, approx 24 sec, 70 mb file, 100 mb segment. ZDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD? 3 3Com SuperStack II Port Statistics 3 CDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD4 3 Unit ID: 2 Port ID: 4 3 3 Media Type: Twisted Pair (100BASE-TX) 3 3 3 3 Good Frames: 13465 FCS Errors: 0 3 3 Good Octets: 861804 Alignment Errors: 0 3 3 Unicast Frames: 13464 Short Events: 0 3 3 Multicast Frames: 0 Too Long Frames: 0 3 3 Broadcast Frames: 1 Very Long Events: 0 3 3 Data Rate Mismatches: 0 3 3 Total Collisions: 322 Late Events: 0 3 3 Runt Frames: 0 Total Errors: 0 3 3 AutoPartitions: 0 3 3 3 3 3 3 Source Address Changes: 0 3 3 Last Source Address: 00609793823C 3 3 3 3 CLEAR COUNTERS CANCEL 3 CDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD4 3 3 @DDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDY Same card, same system, 10 mb segment, 33 seconds my card, 03 bios, get, 33 sec ZDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD? 3 3Com Switch Port Traffic Statistics 3 CDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD4 3 3 3 Port ID: 15 (10BaseT) 3 3 3 3 Frames Received: 15340 Octets Received: 981760 3 3 Frames Transmitted: 31297 Octets Transmitted: 35812608 3 3 3 3 Multicasts Received: 0 Collisions: 22183 3 3 Broadcasts Received: 0 Fragments: 0 3 3 3 3 Frames Forwarded: 15340 Errors: 0 3 3 Frames Filtered: 0 IFM Count: 0 3 3 3 3 Frame Size Analysis. 3 3 64 Octets: 33 % 256 to 511 Octets: 5 % 3 3 65 to 127 Octets: 0 % 512 to 1023 Octets: 26 % 3 3 128 to 255 Octets 0 % 1024 to 1518 Octets: 36 % 3 3 3 3 CLEAR SCREEN COUNTERS CANCEL 3 CDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD4 3 3 @DDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDY