sfc: Accumulate RX_NODESC_DROP count in rx_dropped, not rx_over_errors
authorBen Hutchings <bhutchings@solarflare.com>
Fri, 10 Sep 2010 06:41:06 +0000 (06:41 +0000)
committerDavid S. Miller <davem@davemloft.net>
Fri, 10 Sep 2010 19:27:29 +0000 (12:27 -0700)
rx_over_errors appears to be intended as a count of packets that
overflow a packet buffer in the NIC.  Given that we implement a
cut-through receive path, this should always be 0.

rx_dropped appears to be the correct counter for packets dropped due
to lack of host buffers.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

No differences found