During TX in netdev-dpdk in OVS, there is a pkt drop counter but there are different reasons why that counter may be incremented: - Drops due to egress policing - Drops due to MTU exceeded - Drops reported from the DPDK PMD Tx function There is a proposed patch upstream [1] that adds the above as more fine grained custom stats for dpdk and vhost ports. This Bz is to help with upstreaming of that patch and possible backport to downstream. [1] https://mail.openvswitch.org/pipermail/ovs-dev/2019-September/362872.html
Verified the RFE based on the validation nodes in commnent #1. [root@localhost jhsiao]# cat testpmd.sh testpmd -l 0,1,2,3,4 -w 0000:00:03.0 -w 0000:00:09.0 --socket-mem 4096 -- -i --nb-cores=4 --rxq=2 --txq=2 --rxd=10 --txd=1024 [root@localhost jhsiao]# [root@netqe7 jhsiao]# ovs-vsctl get Interface vhost0 statistics:tx_dropped 222079 [root@netqe7 jhsiao]# ovs-vsctl get Interface vhost0 statistics:tx_dropped 222079 [root@netqe7 jhsiao]# ovs-vsctl get Interface vhost0 statistics:tx_dropped 426161 [root@netqe7 jhsiao]# ovs-vsctl get Interface vhost0 statistics:ovs_tx_failure_drops 2876063 [root@netqe7 jhsiao]# ovs-vsctl get Interface vhost0 statistics:ovs_tx_failure_drops 3237435 [root@netqe7 jhsiao]# ovs-vsctl get Interface vhost0 statistics:ovs_tx_failure_drops 3237435 [root@netqe7 jhsiao]# ovs-vsctl set Interface dpdk0 ingress_policing_rate=368000 [root@netqe7 jhsiao]# ovs-vsctl get Interface dpdk0 statistics:ovs_rx_qos_drops 635919821 [root@netqe7 jhsiao]# ovs-vsctl get Interface dpdk0 statistics:ovs_rx_qos_drops 635919821 [root@netqe7 jhsiao]# ovs-vsctl get Interface dpdk0 statistics:ovs_rx_qos_drops 635919821 [root@netqe7 jhsiao]# ovs-vsctl get Interface dpdk0 statistics:ovs_rx_qos_drop
[root@netqe7 jhsiao]# rpm -q openvswitch2.12 openvswitch2.12-2.12.0-21.el8fdp.x86_64 [root@netqe7 jhsiao]# uname -r 4.18.0-147.el8.x86_64 [root@netqe7 jhsiao]#
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2020:0745