Once booting is complete, pulling/reinserting the card (hotplug) restores functionality.
/var/log/messages is full of errors like:
Nov 7 22:20:46 localhost kernel: NETDEV WATCHDOG: eth0: transmit timed out
Nov 7 22:20:46 localhost kernel: eth0: Transmit timed out, status fc69c1d7, CSR12 00000000, resetting...
Nov 7 22:20:54 localhost kernel: NETDEV WATCHDOG: eth0: transmit timed out
Nov 7 22:20:54 localhost kernel: eth0: Transmit timed out, status fc69c1d7, CSR12 00000000, resetting...
Nov 7 22:21:02 localhost kernel: NETDEV WATCHDOG: eth0: transmit timed out
Nov 7 22:21:02 localhost kernel: eth0: Transmit timed out, status fc69c1d7, CSR12 00000000, resetting...
Nov 7 22:21:10 localhost kernel: NETDEV WATCHDOG: eth0: transmit timed out
...these go away after hotplugging.
Here is some hotplug info:
Nov 7 23:03:50 localhost kernel: cs: cb_free(bus 6)
Nov 7 23:03:55 localhost kernel: cs: cb_alloc(bus 6): vendor 0x1317, device 0x1985
Nov 7 23:03:55 localhost kernel: PCI: Enabling device 06:00.0 (0000 -> 0003)
Nov 7 23:03:55 localhost kernel: eth0: ADMtek Comet rev 17 at 0xc88cf000, 00:10:7A:6B:19:21, IRQ 10.
Nov 7 23:03:55 localhost cardmgr[1462]: socket 1: CardBus hotplug device
Nov 7 23:03:55 localhost pci.agent[5501]: missing kernel or user mode driver tulip
Nov 7 23:03:56 localhost kernel: ip_tables: (C) 2000-2002 Netfilter core team
This is from dmesg; I wonder why the address changes between boot and hotplug:
# dmesg | grep ADMtek eth0: ADMtek Comet rev 17 at 0xc88e7000, 00:10:7A:6B:19:21, IRQ 10. eth0: ADMtek Comet rev 17 at 0xc88cf000, 00:10:7A:6B:19:21, IRQ 10.
Regards, John
