[tulip] LNE100TX... not solved

[Erik Steffl]

Donald Becker wrote:
> 
> On Thu, 14 Dec 2000, Erik Steffl wrote:
> 
> >   I have the same card, same problem (or similar). I have tried all the
> > drivers (current latest, current test, some patches from Dan Hollis) but
> > nothing helps so far. Is this card considered working? Under
> > development? Ignored?
> 
> The ADMtek Centaur 985 chip does work, and frequently tested.
> 
> The tulip driver is under active development -- I've spent about six hours
> on it this week, including tracking down a bug that only occured on fast
> (over 800Mhz) systems, using memory-mapped access to boards with slow
> EEPROMs.  The bug corrupted only the high bit of EEPROM word #32 (!), and
> looked like a transceiver selection problem.
> 
> > > tulip-diag.c:v2.04 9/26/2000 Donald Becker (becker@scyld.com)
> > > Index #1: Found a ADMtek AL985 Centaur-P adapter at 0xa800.
> > >  Port selection is 100mbps-SYM/PCS 100baseTx scrambler, half-duplex.
> 
> Note that 985 reports unusual values for the media settings, so the
> 'tulip-diag' report is usually wrong.  This is a 'tulip-diag' bug, and it
> will be fixed in an upcoming release.

  the tulip-diag output wouldn't be a problem, of course, the problem is
that I get a lot of carrier errors and the tx does not work (at least it
looks like tx because the transfer of large amounts of data FROM the
computer that has LNE100TX does not work (no error messages other then
application timeout), the transfer TO the same computer works OK).

  I have tried patch for tulip-diag from Dan Hollis that he said should
be able to correctly report the media but it still reports half-duplex,
not sure if that means anything...

  are you aware of this tx problem? can you (or other developers, I am
not sure who works on the driver except of you) reproduce the problem?
Is there anything I can do to help to fix the problem? I mean unless you
know about the problem and can reproduce it it's not going to get fixed,
right?

	erik