more NetGear mising ARPs

[Dave Platt]

} This morning (being Saturday) I just could not get this problem to perform
} - then I remembered the above and that there isn't much traffic in the
} office on a Saturday.  This evening I have now had it perform (i.e.
} incomplete ARP) 5/5 times by first setting up a 1400 byte ping to the
} Cisco from the other Linux host on the subnet.  Anyone hazard a guess as
} to the significance of this?  It also possibly means that on a switch you
} may be far less likely to see this problem (and with a 100Mb card there's
} a fair chance of being on a switch?).

The problem we've seen here with the Netgear card (the one with the
Netgear-labelled PNIC chip) definitely seems to be related to network
traffic levels and the hub/switch issue.  We were seeing a lot of card
lockups under heavy load (the "input FIFO gets confused, dumps a lot
of garbage into the ring buffers, and the driver gets even more
confused and stops listening") problem.  Since we rewired our network
room, and put all of these systems on switches (thus reducing network
contention, going full-duplex, and almost completely eliminating
collisions) we have seen _no_ lockups at all.

So, yes, something seems to be rotten in the state of Denmark on these
cards/chips when they have to deal with a shared network.