[tulip] Re: Digital Tulip Design

[Donald Becker]

On Mon, 8 May 2000, Homer Wilson Smith wrote:
> > >     I was not able to get v92g to run in linux 2.0.36.
> > You don't have 'pci-scan.o' loaded.
>     This worked beautifully, still no FD on 21143 Kingstons.
>     Using insmod tulip options=x  AFTER pci_scan is loaded,
> the only options that tulip diag reports as full duplex are
> 4,5,8,10,26 and 27.

Hmmm, did you try #14?
This should set MII- 100baseTx-FD

>     Perhaps it is a problem with Cisco 1900 Switches in Full Dup mode?

Ciscos switches frequently have broken autonegotiation.

> May 10 19:25:46 jane kernel: tulip.c:v0.92 4/17/2000  Written by Donald Becker <becker@scyld.com> 
> May 10 19:25:46 jane kernel:   http://www.scyld.com/network/tulip.html 
> May 10 19:25:46 jane kernel: eth0: Digital DS21143 Tulip rev 65 at 0x2820000, 00:40:F0:4C:E6:A9, IRQ 11. 
> May 10 19:25:46 jane kernel: eth0:  EEPROM default media type Autosense. 
> May 10 19:25:46 jane kernel: eth0:  Index #0 - Media MII 100baseTx (#13)
> described by a 21140 non-MII (0) block. 

Huh?  That's not right.  It should be just "MII transceiver".
I have what I believe is an identical board that has no problem with v0.92

> May 10 19:25:46 jane kernel: eth0:  MII transceiver #1 config 3100 status
> 782d advertising 03e0.


>   Media MII, block type 3, length 13.
>    MII interface PHY 0 (media type 11).

That looks normal.

>   2646 0001 0000 0000 0000 0000 0000 0300
>   10a6 0104 c000 4cf0 a9e6 1e00 0000 0800
>   8d01 0003 0000 7800 01e0 5000 1800 0000
>   0000 0000 0000 0000 0000 0000 0000 0000

The table on my Kingston-provided adatper, which reads correctly with v92,
is identical.
  2646 0001 0000 0000 0000 0000 0000 0000
  006b 0104 c000 3bf0 0200 1e00 0000 0800
  8d01 0003 0000 7800 01e0 5000 1800 0000
  0000 0000 0000 0000 0000 0000 0000 0000


Donald Becker				becker@scyld.com
Scyld Computing Corporation
410 Severn Ave. Suite 210
Annapolis MD 21403