Kernel panic with Netgear Lite-On PNIC and v0.91

[Martin Weinberg]

Folks,

We have a cluster with mostly Kingston and Netgear 21140s 
and a few recently obtained Netgear Lite-On 82c168 boards.

We have experienced a number of freezes under very heavy
load which led me to switch our cluster running kernel 2.2.5 
from driver v0.89H to v0.91, partly based on discussion in
this group.  These are dual PIIs with ASUS P2B main boards.

The 21140s are working fine.  In fact, v0.91 has fixed the
freezing problem (so far)!

However, the Lite-On 82c168 machines kernel panic as soon 
as the network is initialized with v0.91 but v0.89H works 
ok, e.g.:

tulip.c:v0.89H 5/23/98 becker@cesdis.gsfc.nasa.gov
eth0: Lite-On 82c168 PNIC at 0xb000, 00 a0 cc 40 d6 38, IRQ 19.
eth0:  MII transceiver found at MDIO address 1, config 3000 status 7829.
eth0:  Advertising 01e1 on PHY 1, previously advertising 01e1.
eth0: Changing PNIC configuration to full-duplex, CSR6 812e0200.

Does anybody know what to do here? 

Thanks,

--Martin


===========================================================================

Martin Weinberg                      Phone: (413) 545-3821
Dept. of Physics and Astronomy       FAX:   (413) 545-2117/0648
530 Graduate Research Tower
University of Massachusetts
Amherst, MA  01003-4525