v0.91 driver released for testing

[David Godfrey]

* Donald Becker (becker@cesdis.gsfc.nasa.gov) [990417 23:05]:

> > > Load the module with "debug=3" to see the gory details of the media
> > Not the identical symptoms. Another card is not put into a usable
> > state by 0.91. Indeed, it is left in an unusable state and
> > will not work with 0.90p afer rebooting/resetting.
> > This card is in a dual-boot W95 machine - booting up W95 gets
> > the card working again.
> 
> Does the card work at 100mbps with W95 or the earlier driver?
> (I want to verify that this isn't a cable problem.)
> 

Yes. It works at 100mbs with W95 (though usually half-duplex) and
100mbs full-duplex with 0.90p.

> > eth0: 21143 link status interrupt 41e192cf, CSR5 f8668000, fffbffff.
> 
> Ooops, you lost 100baseTx link beat.  How quickly did this happen?
> 

Almost immediately. Much less than 1s until it settles.


> > eth0: 21143 negotiation status 41e192cf, 100baseTx-FD.
> 
> This time when you switched there just wasn't link beat. ???
> 

The status lights on the switch indicate 100mbs full-duplex but the
machine doesn't respond to ping. The console just spits out occasional

eth0: 21143 negotiation status 000022cf, 100baseTx-FD.
eth0: Using NWay-set 100baseTX-FD media, csr12 000022cf.


Rebooting without going into W95 just gives the following
eth0: 21143 link status interrupt 41e1d2ce, CSR5 f0668010, ffbffff.
eth0: Switching to 100bseTx-FD based on link partner advertisement
41e1.

before settling into the state above.

Regards
-- 
Dave Godfrey
dave@delta.demon.co.uk
Fax +44 181 286 1909
PGP key available