Bug In v0.89 tulip.c

Donald Becker becker@cesdis1.gsfc.nasa.gov
Mon Oct 5 00:04:00 1998


On Sun, 4 Oct 1998, Jason Ledbetter wrote:

> From: Jason Ledbetter <jason@colltech.com>

[[ I replied to Jason privately. ]]

> I'm trying to get the integrated ethernet card in my Compaq Presario 5660
> working under 2.0.34. As far as I know, it is a 21143 based chipset .. at 
> least according to Win98. The proc/pci output below seems to contradict
> this.
>   Ethernet controller: DEC DC21142 (rev 65).

The 21142 and 21143 have the same PCI IDs.  (Grrrr..)

> tulip-diag.c:v1.06 9/18/98 Donald Becker (becker@cesdis.gsfc.nasa.gov)
> Index #1: Found a Digital DS21143 Tulip adapter at 0x2400.

I always report it as a 21143.  Few 21142 chips were made and this questions
was often asked.

> Digital DS21143 Tulip Tulip chip registers at 0x2400:
>   f9a08000 ffffffff ffffffff 00096028 00096228 f0660000 b2422202 fbfffbff
>   e0000000 fff483ff ffffffff 00000000 000000c6 ffff0000 fffbffff 8ff4c008
>  The Rx process state is 'Waiting for packets'.
>  The Tx process state is 'Idle'.

This is normal operation.

> Leaf node at offset 30, default media type 0800 (Autosense).
>  2 transceiver description blocks:
>   Media 10baseT,  block type 2.
>   Media 10baseT-Full Duplex,  block type 2.

You don't have 100mbps capabilitly with this design.
This is likely accurate.  (A very common mistake is using the Digital design
kit EEPROM verbatim.  If a designer changes the EEPROM they usually have a
clue.)

> ... now something interesting.. after 5-10 seconds, if I rerun this again,
> I get the following diff'ed output.
> 
> >  f9a08000 ffffffff ffffffff 00096028 00096228 f0066000 b3862002 fbfffbff
> >  e0000000 fff583ff ffffffff 00000000 000000c6 ffff0000 fffbff7f 8ff4c008
> 
> > Transmit started, Receive started, half-duplex.
> >  The transmit threshold is 128.
> >  Port selection is 100mbps-SYM/PCS 100BaseTx scrambler, half-duplex.

This is because I recently modified the driver to ignore the EEPROM table.
Why?  Because the EEPROM media table is so often wrong.

It should still detect the correct media type if you have a 10baseT cable
plugged in.


Donald Becker					  becker@cesdis.gsfc.nasa.gov
USRA-CESDIS, Center of Excellence in Space Data and Information Sciences.
Code 930.5, Goddard Space Flight Center,  Greenbelt, MD.  20771
301-286-0882	     http://cesdis.gsfc.nasa.gov/people/becker/whoiam.html