Netgear FA310TX panicking kernel with tulip driver in 2.0.34-36

[me@karma.reshall.berkeley.edu]

I'm running linux 2.0.36 with a netgear FA310TX connected to a 100baseTX hub
and a generic ne2000-compatible card connected to the outside (10baseT)
network. I've been using the tulip driver as a module to work the netgear card.

Every few days, I've been losing the console, and can only access the box
remotely. In /var/log/messages, there are blocks like this:

Jan 14 10:45:19 karma kernel: Unable to handle kernel paging request at virtual
address c883a098
Jan 14 10:45:19 karma kernel: current->tss.cr3 = 006ce000, pr3 = 006ce000
Jan 14 10:45:19 karma kernel: *pde = 00000000
Jan 14 10:45:19 karma kernel: Oops: 0000
Jan 14 10:45:19 karma kernel: CPU:    0
Jan 14 10:45:20 karma kernel: EIP:    0010:[tulip:tulip_probe+-26838004/22328]
Jan 14 10:45:20 karma kernel: EFLAGS: 00010282
Jan 14 10:45:20 karma kernel: eax: 04c4f9b0   ebx: 0441d04c   ecx: 04e94dac   ed
x: 00000038
Jan 14 10:45:20 karma kernel: esi: 04e94c1c   edi: 0441d068   ebp: 0441d04c   es
p: 05e4ceec
Jan 14 10:45:20 karma kernel: ds: 0018   es: 0018   fs: 002b   gs: 002b   ss: 00
18
Jan 14 10:45:20 karma kernel: Process rc5des (pid: 2521, process nr: 95, stackpa
ge=05e4c000)
Jan 14 10:45:20 karma kernel: Stack: 0441d04c 14a34250 00000014 001d9a18 0441d03
8 00000000 05e4cf80 00000001
Jan 14 10:45:20 karma kernel:        0013008b 04d9bd5c 00000002 0000faec 0441d03
8 01010014 001447db 0441d068
Jan 14 10:45:20 karma kernel:        0000faec 00000000 0101a8c0 00000014 0201a8c
0 00000000 001d9a18 00000000
Jan 14 10:45:20 karma kernel: Call Trace: [load_elf_interp+47/716] [ip_rcv+1195/
1580] [net_bh+244/276] [do_bottom_half+59/96] [handle_bottom_half+11/24]
Jan 14 10:45:20 karma kernel: Code: 18 9c 1d 00 00 00 00 00 00 00 00 00 f8 00 00
 00 03 00 00 00
Jan 14 10:45:20 karma kernel: Aiee, killing interrupt handler

The disturbing thing is, I've had this problem once in 2.0.34, and a few times
in 2.0.35. I've also replaced the card with another of the same model, with no
success.

I'm running redhat 5.1, and have gotten my kernel's by RPM. Is there anything
wrong with redhat's kernel rpm's?

I disabled the netgear card a couple days ago and haven't had any problems, but
I would like it to work eventually...

Thanks for your help. :-)