[tulip] Getting two tulips to talk w/ x-over cable

[Thomas Gagne]

I've a PC with an older 10mbs card in it, and a laptop with an SMC PC Card in
it.

The laptop works fine at the office, don't worry.

When I try connecting them at home with a crossover cable (which also works at
the office when I plug my laptop into the straight-through port on the hub
normally reserved for chaining hubs) they seem unable to negotiate a link.

tulip-diag -ee for the PC says:

rocky:/home/tgagne cat outfile
tulip-diag.c:v1.19 10/2/99 Donald Becker (becker@cesdis.gsfc.nasa.gov)
Index #1: Found a Digital DC21041 Tulip adapter at 0x6500.
 Port selection is full-duplex.
 Transmit started, Receive started, full-duplex.
  The Rx process state is 'Waiting for packets'.
  The Tx process state is 'Idle'.
  The transmit unit is set to store-and-forward.
  The NWay status register is 000020cc.
EEPROM size is 6.
PCI Subsystem IDs, vendor 10b8, device 2009.
CardBus Information Structure at offset 00000000.
Ethernet MAC Station Address 00:E0:29:49:8D:B8.
EEPROM transceiver/media description for the Digital DC21041 Tulip chip.
Leaf node at offset 30, default media type 0800 (Autosense).
 2 transceiver description blocks:
  21041 media index 00 (10baseT).
  21041 media index 04 (10baseT-Full Duplex).
EEPROM contents:
  10b8 2009 0000 0000 0000 0000 0000 0000
  006c 0103 e000 4929 b88d 1e00 0000 0800
  0002 0004 0000 0000 0000 0000 0000 0000
  0000 0000 0000 0000 0000 0000 0000 0000
  0000 0000 0000 0000 0000 0000 0000 0000
  0000 0000 0000 0000 0000 0000 0000 0000
  0000 0000 0000 0000 0000 0000 0000 0000
  0000 0000 0000 0000 0000 0000 0000 588b
 ID block CRC 0x6c (vs. 0x6c).
  Full contents CRC 0x588b (read as 0x588b).
  Internal autonegotiation state is 'Ability detect'.

The Laptop says (and I admit I've been playing with -A and -F


tulip-diag.c:v1.19 10/2/99 Donald Becker (becker@cesdis.gsfc.nasa.gov)
Index #1: Found a Digital DS21143 Tulip adapter at 0x200.
 Port selection is 10mpbs-serial, full-duplex.
 Transmit started, Receive started, full-duplex.
  The Rx process state is 'Waiting for packets'.
  The Tx process state is 'Idle'.
  The transmit threshold is 72.
  The NWay status register is 000000c6.
EEPROM size is 6.
PCI Subsystem IDs, vendor 10b8, device 8034.
CardBus Information Structure at offset 0000003f.
Ethernet MAC Station Address 00:E0:29:5A:AE:AD.
EEPROM transceiver/media description for the Digital DS21143 Tulip chip.
Leaf node at offset 30, default media type 0800 (Autosense).
 1 transceiver description blocks:
  Media MII, block type 3, length 21.
   MII interface PHY 0 (media type 11).
   21143 MII initialization sequence is 2 words: 0886 0002.
   21143 MII reset sequence is 2 words: 0886 0002.
    Media capabilities are 7800, advertising 01e1.
    Full-duplex map 5000, Threshold map 1800.
    No MII interrupt.
EEPROM contents:
  10b8 8034 003f 0000 0000 0000 0000 0200
  985b 0104 e000 5a29 adae 1e00 0000 0800
  9501 0003 8602 0208 0200 0886 0002 7800
  01e0 5000 1800 0000 0000 0000 0000 0000
  0000 0000 0000 0000 0000 0000 0000 0000
  0000 0000 0000 0000 0000 0000 0000 8e6d
  0000 0000 0000 e000 5a29 adae 0140 0000
  0000 0000 0000 0000 0000 0000 0000 004e
 ID block CRC 0x5b (vs. 0x5b).
  Full contents CRC 0xeda5 (read as 0x004e).
 MII PHY found at address 1, status 0x7809.
 MII PHY #1 transceiver registers:
   1000 7809 0040 6212 0001 0000 0001 0000
   0000 0000 0000 0000 0000 0000 0000 0000
   5000 0001 0000 0000 0000 0000 0900 0000
   0038 3012 0f00 ff00 0028 4000 0020 000b.
  Internal autonegotiation state is 'Autonegotiation disabled'.

Anyway, I've tried everything to get them to talk.  After boot both up from
scratch, the PC says it doesn't get a link beat so it's switching back to
10Base2, though he starts out at 10BaseT.  mii-diag --watch on the laptop
doesn't seem to see anything from the other side.

Another thing I've tried is playing with /etc/conf.modules on the laptop.  In
there right now I have:

add options tulip debug=6 options=0
add options tulip_cb debu=6 options=0

But I'm not seeing anything appear in /var/log/messages that looks like
increased wordiness from the tulip driver.  A typical /var/log/messages
resembles:

Jun  1 00:50:43 rocky kernel: tulip_attach(bus 35, function 0)
Jun  1 00:50:43 rocky kernel: tulip.c:v0.91g-ppc 7/16/99
becker@cesdis.gsfc.nasa.gov (modified by danilo@cs.uni-magdeburg.de for XIRCOM
CBE, fixed by Doug Ledford)
Jun  1 00:50:43 rocky kernel: eth0: Digital DS21143 Tulip rev 65 at 0x200,
00:E0:29:5A:AE:AD, IRQ 11.
Jun  1 00:50:43 rocky kernel: eth0:  EEPROM default media type Autosense.
Jun  1 00:50:43 rocky kernel: eth0:  Index #0 - Media MII (#11) described by a
21142 MII PHY (3) block.
Jun  1 00:50:43 rocky kernel: eth0:  MII transceiver #1 config 3000 status
7809 advertising 01e1.
Jun  1 00:50:43 rocky cardmgr[435]: executing: './network start eth0'

Suggestions anyone, short of recommending a hub?

--
For Open Source Middleware Visit http://home.netcom.com/~tgagne