To: None <>
From: der Mouse <mouse@Rodents.Montreal.QC.CA>
List: tech-net
Date: 08/10/2002 15:27:32
I have some questions about the lance Rev. C bug.

Is there any known way to provoke it?  In particular, I've added a
printf to my kernel which tells me when lance.c drops a packet because
it shouldn't have received it.  So far this hasn't triggered, but the
machine hasn't been using this kernel for long and it hasn't been
stressed.  I'm wondering when - if at all - I can conclude I don't have
a buggy chip.

Also, why does it preclude multicast support?  As far as I can see, all
it amounts to is that sometimes the chip will receive a packet which
from the host's point of view has been garbaged.  Won't the other
layers of the system deal with this correctly?  (I have a -3/60 that
I'd like to run IPv6 on, and I'm considering just turning off
LANCE_REVC_BUG in the config - but presumably whoever added the test
wasn't stupid, so there's probably something I'm missing.)

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B