From mboxrd@z Thu Jan  1 00:00:00 1970
From: John DeGood <nu3e@arrl.net>
Subject: Re: [9fans] Has anyone got a 2114x ethernet card working under 4th
 Edition?
To: 9fans@cse.psu.edu
Message-id: <010901c226e3$b3f1dfc0$1b02a8c0@arnysm01.nj.comcast.net>
MIME-version: 1.0
Content-type: text/plain; charset=iso-8859-1
Content-transfer-encoding: 7BIT
References: <a16f7adb32f9aedc9c85e2a309ccbb37@plan9.bell-labs.com>
 <005101c1fa80$94ab59a0$0100a8c0@landra.org>
Date: Mon,  8 Jul 2002 20:58:12 -0400
Topicbox-Message-UUID: c5f58d70-eaca-11e9-9e20-41e7f4b1d025

> From: "Russ Cox" <rsc@plan9.bell-labs.com>
>
> What does
>
> pci
>
> print?  The line I care about is the one
> where the second field begins 02., as in
> the three marked below.

I know this is an old thread, but I just ran into this problem today with an
SMC 9332DST card and thought my experience might help others.  The broken
detection message was:

   #l0: 2114x: 0 Mbps port 0xfc80  irq 10  c0f401000000

Note that the speed and EA are both bogus.

Here's the interesting part:  I tried a newer variant of the card, an SMC
9332BDT, and it works perfectly with the 2114x driver.

The output from "pci" for *both* cards is identical:

   0.15.0:  02.00.00  1011/0009  10  0:0000fc81  128  1:ffbefc00  128

The Tulip chips on the 2 cards have the following markings:

   9332DST (doesn't work with 2114x driver):  21140 / DC1010BA / 21-40673-01

   9332BDT (works with 2114x driver):  21140-AE / DC1064B / 21-43864-03

I suspect transceiver or EEPROM differences may be the significant driver
issue in the case of these 2 cards rather than the variant of Tulip chip.
FWIW, I note that Don Becker's Tulip driver for Linux has an entry in
"eeprom_fixups[]" for the (broken in Plan 9) 9332DST, while I don't see
anything similar in ether2114x.c

John