Hi
The reason I want to disable RSS is that I doubt it will incur packet mis-order
in af_packet.c/PACKET_MMAP
i.e.
In multi-queue enabled kernel, tpacket_rcv will be run in parallel,right? So
that a later packet might get the spinlock earlier and it is copied to the ring
buffer prior to packet which timestamp is earlier?
if RSS is disabled the tpacket_rcv will be run in sequence, but what's the
performance? Can intel X520 nic reach 5gbps+ without RSS?
Thanks
jon
af_packet.c
static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev,
struct packet_type *pt, struct net_device *orig_dev)
{
...
spin_lock(&sk->sk_receive_queue.lock);
h.raw = packet_current_frame(po, &po->rx_ring, TP_STATUS_KERNEL);
if (!h.raw)
goto ring_is_full;
packet_increment_head(&po->rx_ring);
po->stats.tp_packets++;
if (copy_skb) {
status |= TP_STATUS_COPY;
__skb_queue_tail(&sk->sk_receive_queue, copy_skb);
}
if (!po->stats.tp_drops)
status &= ~TP_STATUS_LOSING;
spin_unlock(&sk->sk_receive_queue.lock);
skb_copy_bits(skb, 0, h.raw + macoff, snaplen);
}
------------------------------------------------------------------------------
Forrester recently released a report on the Return on Investment (ROI) of
Google Apps. They found a 300% ROI, 38%-56% cost savings, and break-even
within 7 months. Over 3 million businesses have gone Google with Google Apps:
an online email calendar, and document program that's accessible from your
browser. Read the Forrester report: http://p.sf.net/sfu/googleapps-sfnew
_______________________________________________
E1000-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/e1000-devel
To learn more about Intel® Ethernet, visit
http://communities.intel.com/community/wired
