* [dpdk-users] Packets beyond MTU size in i40evf
@ 2018-06-06 14:59 Varun Lakkur
0 siblings, 0 replies; only message in thread
From: Varun Lakkur @ 2018-06-06 14:59 UTC (permalink / raw)
To: users
Hey,
I am using rte_eth_stats_get() to get the stats of how many packets
are received/transmitted by my application.
I am running SRIOV on XL710 (using i40evf pmd). When I send packets
beyond the MTU size, my application does not receive them as expected.
But these are not reported in rx_dropped_packets/rx_errors. I see that
these are counted as rx_good_packets. How do I differentiate the
oversized packets and count them correctly? Shouldn't they be counted
under rx_dropped_packets?
I went through the source code for i40e_dev_stats_get and
i40evf_dev_stats_get. I see that i40e reads extra counters from
i40e_hw_port_stats. Is there anything similar for SRIOV interfaces?
Here is the xstats counters I got from dpdk-procinfo:
####################################################
NIC extended statistics for port 1
####################################################
rx_good_packets: 656
tx_good_packets: 556
rx_good_bytes: 225160
tx_good_bytes: 33360
rx_errors: 0
tx_errors: 0
rx_mbuf_allocation_errors: 0
rx_q0packets: 0
rx_q0bytes: 0
rx_q0errors: 0
tx_q0packets: 0
tx_q0bytes: 0
rx_bytes: 225160
rx_unicast_packets: 656
rx_multicast_packets: 0
rx_broadcast_packets: 0
rx_dropped_packets: 0
rx_unknown_protocol_packets: 0
tx_bytes: 33360
tx_unicast_packets: 556
tx_multicast_packets: 0
tx_broadcast_packets: 0
tx_dropped_packets: 0
tx_error_packets: 0
Regards,
Varun
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2018-06-06 14:59 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-06-06 14:59 [dpdk-users] Packets beyond MTU size in i40evf Varun Lakkur
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).