It would be nice if the driver would be shaken down before being
tossed into the base kernel. At McAfee, we've been playing around
with various versions of the 3c905 driver for WebShield, and have
discovered that the quickest way to kill a WebShield is to put a 3c905
into it, then tuck it onto a busy network. 45 seconds from enabling
the interface to a horrible death is the record; if this is happening
to a lot of the other testers of the driver, it seems like that would
be a good reason to delay the bug reports for a while longer.
____
david parsons \bi/ plus it's kinda slow in 100mb.
\/