DPDK: 18.11/19.11/20.11/21.02/21.05/21.08 NIC: Mellanox Technologies MT27800 Family [ConnectX-5] FW: firmware-version: 16.31.1014 (MT_0000000012) CPU: Intel(R) Xeon(R) Platinum 8170M CPU @ 2.10GHz KERNEL: 5.4.17-2102.200.13.uek PKTGEN: IxNetwork 9.00.1915.16 MLNX OFED: MLNX_OFED_LINUX-5.0-2.1.8.0 testpmd: testpmd -l 26-51 --socket-mem=4096,4096 -w d8:00.0,dv_flow_en=0,mprq_en=1,rxqs_min_mprq=1,rx_vec_en=1 -- -i --rxq=16 --txq=16 --nb-cor es=16 --forward-mode=icmpecho --numa --enable-rx-cksum -a --rxd=2048 --txd=2048 --burst=64 testpmd -l 26-51 --socket-mem=4096,4096 -w d8:00.0,dv_flow_en=1,mprq_en=1,rxqs_min_mprq=1,rx_vec_en=1 -- -i --rxq=16 --txq=16 --nb-cor es=16 --forward-mode=icmpecho --numa --enable-rx-cksum -a --rxd=2048 --txd=2048 --burst=64 flow: testpmd> flow create 0 ingress pattern eth / ipv4 dst is 1.1.1.1 / tcp / end actions queue index 15 / end Flow rule #0 created testpmd> flow create 0 ingress pattern eth / ipv4 dst is 1.1.1.1 / udp / end actions queue index 15 / end Flow rule #1 created testpmd> flow create 0 ingress pattern eth / ipv4 / udp dst is 53 / end actions count / rss / end Flow rule #2 created with these flows, no packet matched and testpmd only received 60.1Mpps: testpmd> show port stats all ######################## NIC statistics for port 0 ######################## RX-packets: 6512958506 RX-missed: 4476258 RX-bytes: 390777510360 RX-errors: 0 RX-nombuf: 0 TX-packets: 0 TX-errors: 0 TX-bytes: 0 Throughput (since last show) Rx-pps: 60163078 Rx-bps: 28878277584 Tx-pps: 0 Tx-bps: 0 ############################################################################ testpmd> with flush all flows, testpmd received 148.8Mpps testpmd> flow flush 0 testpmd> show port stats all ######################## NIC statistics for port 0 ######################## RX-packets: 10076834471 RX-missed: 4482703 RX-bytes: 604610068260 RX-errors: 0 RX-nombuf: 0 TX-packets: 0 TX-errors: 0 TX-bytes: 0 Throughput (since last show) Rx-pps: 148061620 Rx-bps: 71069577904 Tx-pps: 0 Tx-bps: 0 ############################################################################ testpmd>
Hi, Can you describe your traffic pattern? Please note all flows have the same priority (0). It is possible that all traffic matches flows #1 or #2 so you see the performance of a single queue. Another thing to try is to replace queue 15 with RSS to see the result.
Thanks a lot! all traffic use UDP protocol, and config 1024 source address(14.175.242.132 [RepeatableRandom: 0.0.0.0, 223.255.255.255, 1, 1024]), dest address is fixed(10.245.102.44)! I had confirmed all traffic not in a single rx queue.
After checking the configuration the issue is gone.
How did you test it? I tested it but the issue existed
I missunderstood and thought it was resolved for you. Can you remove flows #1 and #2? Let's leave only the RSS flow and see. Also, can you provide the xstats output? We can see the distribution of the packets among the queues.
Asaf, Assigning to you since you are following this up. Thanks
(In reply to Asaf Penso from comment #5) > I missunderstood and thought it was resolved for you. > Can you remove flows #1 and #2? Let's leave only the RSS flow and see. > Also, can you provide the xstats output? We can see the distribution of the > packets among the queues. If device add only one RSS flow, the device RX can reach to wire speed ! with more no matched flows, the device will reduce rx performance !
Let's start again, since i think we don't have a clear understanding. According to this traffic: >all traffic use UDP protocol, and config 1024 source address(14.175.242.132 >[RepeatableRandom: 0.0.0.0, 223.255.255.255, 1, 1024]), dest address is >fixed(10.245.102.44)! These flows would not hit, because: Flow rule #0 is matching on TCP Flow rule #1 is ipv4 dst 1.1.1.1 There is an impact with the packet processing that we first need to miss these 2 before hitting the other flow. So, if you only have Flow rule #2, what perf result do you see?
(In reply to Asaf Penso from comment #8) > Let's start again, since i think we don't have a clear understanding. > > According to this traffic: > >all traffic use UDP protocol, and config 1024 source address(14.175.242.132 > >[RepeatableRandom: 0.0.0.0, 223.255.255.255, 1, 1024]), dest address is > >fixed(10.245.102.44)! > > These flows would not hit, because: > Flow rule #0 is matching on TCP > Flow rule #1 is ipv4 dst 1.1.1.1 > There is an impact with the packet processing that we first need to miss > these 2 before hitting the other flow. > > So, if you only have Flow rule #2, what perf result do you see? I'll try test this example again and please wait my result !
Hi, I have tested again and the following is the test result! the IXIA traffic is 64 bytes UDP packet! the packet's ipv4 destnation address is 10.245.28.42, UDP dest port is 8080! so all packet don't match any flows! As the flow increases, the NIC RX performance (pps) will to reduce ! DPDK: 21.08 NIC: Mellanox Technologies MT27800 Family [ConnectX-5] FW: firmware-version: 16.32.1010 (MT_0000000012) Ethernet controller: Mellanox Technologies MT2892 Family [ConnectX-6 Dx] FW: firmware-version: 22.28.1002 (MT_0000000359) CPU: Intel(R) Xeon(R) Platinum 9242 CPU @ 2.30GHz x2 (48 core x 2) KERNEL: 5.4.17-2102.200.13.uek PKTGEN: IxNetwork 9.00.1915.16 MLNX OFED: MLNX_OFED_LINUX-5.0-2.1.8.0 DPDK testpmd: ./testpmd-v20.8-cascadelake -l 1,24-48 -m 4096 -n 4 -w '54:00.1,dv_flow_en=1,mprq_en=1,rxqs_min_mprq=1,rx_vec_en=1' \ --log-level="lib.eal":8 --log-level=pmd:8 --log-level="pmd.net.mlx5":3 \ -- -i --rxq=24 --txq=24 --nb-cores=24 --forward-mode icmpecho --numa \ --enable-rx-cksum --auto-start DPDK testpmd with no flow: testpmd> show port stats all ######################## NIC statistics for port 0 ######################## RX-packets: 32261848744 RX-missed: 3607570 RX-bytes: 1935710931360 RX-errors: 0 RX-nombuf: 0 TX-packets: 2 TX-errors: 0 TX-bytes: 120 Throughput (since last show) Rx-pps: 149175705 Rx-bps: 71604353904 Tx-pps: 0 Tx-bps: 0 ############################################################################ testpmd> flow list 0 testpmd> DPDK testpmd with flow: flow create 0 ingress pattern eth / ipv4 dst is 1.1.1.1 / tcp / end actions queue index 15 / end flow create 0 ingress pattern eth / ipv4 dst is 2.2.2.2 / tcp / end actions queue index 15 / end flow create 0 ingress pattern eth / ipv4 dst is 1.1.1.1 / udp / end actions queue index 15 / end flow create 0 ingress pattern eth / ipv4 dst is 2.2.2.2 / udp / end actions queue index 15 / end flow create 0 ingress pattern eth / ipv4 / udp dst is 53 / end actions count / rss / end flow create 0 ingress pattern eth / ipv4 / tcp dst is 5353 / end actions count / rss / end flow create 0 ingress pattern eth / ipv4 / tcp dst is 5060 / end actions count / rss / end flow create 0 ingress pattern eth / ipv4 / tcp dst is 8081 / end actions count / rss / end flow create 0 ingress pattern eth / ipv4 / udp dst is 5353 / end actions count / rss / end flow create 0 ingress pattern eth / ipv4 / udp dst is 5060 / end actions count / rss / end flow create 0 ingress pattern eth / ipv4 / udp dst is 8081 / end actions count / rss / end testpmd> show port stats all ######################## NIC statistics for port 0 ######################## RX-packets: 41411062408 RX-missed: 3607570 RX-bytes: 2484663748260 RX-errors: 0 RX-nombuf: 0 TX-packets: 2 TX-errors: 0 TX-bytes: 120 Throughput (since last show) Rx-pps: 96406314 Rx-bps: 46275071592 Tx-pps: 0 Tx-bps: 0 ############################################################################ testpmd> flow list 0 ID Group Prio Attr Rule 0 0 0 i-- ETH IPV4 TCP => QUEUE 1 0 0 i-- ETH IPV4 TCP => QUEUE 2 0 0 i-- ETH IPV4 UDP => QUEUE 3 0 0 i-- ETH IPV4 UDP => QUEUE 4 0 0 i-- ETH IPV4 UDP => COUNT RSS 5 0 0 i-- ETH IPV4 TCP => COUNT RSS 6 0 0 i-- ETH IPV4 TCP => COUNT RSS 7 0 0 i-- ETH IPV4 TCP => COUNT RSS 8 0 0 i-- ETH IPV4 UDP => COUNT RSS 9 0 0 i-- ETH IPV4 UDP => COUNT RSS 10 0 0 i-- ETH IPV4 UDP => COUNT RSS testpmd>
(In reply to Asaf Penso from comment #8) > Let's start again, since i think we don't have a clear understanding. > > According to this traffic: > >all traffic use UDP protocol, and config 1024 source address(14.175.242.132 > >[RepeatableRandom: 0.0.0.0, 223.255.255.255, 1, 1024]), dest address is > >fixed(10.245.102.44)! > > These flows would not hit, because: > Flow rule #0 is matching on TCP > Flow rule #1 is ipv4 dst 1.1.1.1 > There is an impact with the packet processing that we first need to miss > these 2 before hitting the other flow. > > So, if you only have Flow rule #2, what perf result do you see? Hi, I have tested again and the following is the test result! the IXIA traffic is 64 bytes UDP packet! the packet's ipv4 destnation address is 10.245.28.42, UDP dest port is 8080! so all packet don't match any flows! As the flow increases, the NIC RX performance (pps) will to reduce ! DPDK: 21.08 NIC: Mellanox Technologies MT27800 Family [ConnectX-5] FW: firmware-version: 16.32.1010 (MT_0000000012) Ethernet controller: Mellanox Technologies MT2892 Family [ConnectX-6 Dx] FW: firmware-version: 22.28.1002 (MT_0000000359) CPU: Intel(R) Xeon(R) Platinum 9242 CPU @ 2.30GHz x2 (48 core x 2) KERNEL: 5.4.17-2102.200.13.uek PKTGEN: IxNetwork 9.00.1915.16 MLNX OFED: MLNX_OFED_LINUX-5.0-2.1.8.0 DPDK testpmd: ./testpmd-v20.8-cascadelake -l 1,24-48 -m 4096 -n 4 -w '54:00.1,dv_flow_en=1,mprq_en=1,rxqs_min_mprq=1,rx_vec_en=1' \ --log-level="lib.eal":8 --log-level=pmd:8 --log-level="pmd.net.mlx5":3 \ -- -i --rxq=24 --txq=24 --nb-cores=24 --forward-mode icmpecho --numa \ --enable-rx-cksum --auto-start DPDK testpmd with no flow: testpmd> show port stats all ######################## NIC statistics for port 0 ######################## RX-packets: 32261848744 RX-missed: 3607570 RX-bytes: 1935710931360 RX-errors: 0 RX-nombuf: 0 TX-packets: 2 TX-errors: 0 TX-bytes: 120 Throughput (since last show) Rx-pps: 149175705 Rx-bps: 71604353904 Tx-pps: 0 Tx-bps: 0 ############################################################################ testpmd> flow list 0 testpmd> DPDK testpmd with flow: flow create 0 ingress pattern eth / ipv4 dst is 1.1.1.1 / tcp / end actions queue index 15 / end flow create 0 ingress pattern eth / ipv4 dst is 2.2.2.2 / tcp / end actions queue index 15 / end flow create 0 ingress pattern eth / ipv4 dst is 1.1.1.1 / udp / end actions queue index 15 / end flow create 0 ingress pattern eth / ipv4 dst is 2.2.2.2 / udp / end actions queue index 15 / end flow create 0 ingress pattern eth / ipv4 / udp dst is 53 / end actions count / rss / end flow create 0 ingress pattern eth / ipv4 / tcp dst is 5353 / end actions count / rss / end flow create 0 ingress pattern eth / ipv4 / tcp dst is 5060 / end actions count / rss / end flow create 0 ingress pattern eth / ipv4 / tcp dst is 8081 / end actions count / rss / end flow create 0 ingress pattern eth / ipv4 / udp dst is 5353 / end actions count / rss / end flow create 0 ingress pattern eth / ipv4 / udp dst is 5060 / end actions count / rss / end flow create 0 ingress pattern eth / ipv4 / udp dst is 8081 / end actions count / rss / end testpmd> show port stats all ######################## NIC statistics for port 0 ######################## RX-packets: 41411062408 RX-missed: 3607570 RX-bytes: 2484663748260 RX-errors: 0 RX-nombuf: 0 TX-packets: 2 TX-errors: 0 TX-bytes: 120 Throughput (since last show) Rx-pps: 96406314 Rx-bps: 46275071592 Tx-pps: 0 Tx-bps: 0 ############################################################################ testpmd> flow list 0 ID Group Prio Attr Rule 0 0 0 i-- ETH IPV4 TCP => QUEUE 1 0 0 i-- ETH IPV4 TCP => QUEUE 2 0 0 i-- ETH IPV4 UDP => QUEUE 3 0 0 i-- ETH IPV4 UDP => QUEUE 4 0 0 i-- ETH IPV4 UDP => COUNT RSS 5 0 0 i-- ETH IPV4 TCP => COUNT RSS 6 0 0 i-- ETH IPV4 TCP => COUNT RSS 7 0 0 i-- ETH IPV4 TCP => COUNT RSS 8 0 0 i-- ETH IPV4 UDP => COUNT RSS 9 0 0 i-- ETH IPV4 UDP => COUNT RSS 10 0 0 i-- ETH IPV4 UDP => COUNT RSS testpmd>
(In reply to Asaf Penso from comment #8) > Let's start again, since i think we don't have a clear understanding. > > According to this traffic: > >all traffic use UDP protocol, and config 1024 source address(14.175.242.132 > >[RepeatableRandom: 0.0.0.0, 223.255.255.255, 1, 1024]), dest address is > >fixed(10.245.102.44)! > > These flows would not hit, because: > Flow rule #0 is matching on TCP > Flow rule #1 is ipv4 dst 1.1.1.1 > There is an impact with the packet processing that we first need to miss > these 2 before hitting the other flow. > > So, if you only have Flow rule #2, what perf result do you see? with match flow rule, the CX5 RX performance also reduced testpmd> flow create 0 ingress pattern eth / ipv4 dst is 10.245.28.42 / end actions mark id 42 / count / rss / end Flow rule #0 created testpmd> flow list 0 ID Group Prio Attr Rule 0 0 0 i-- ETH IPV4 => MARK COUNT RSS testpmd> show port stats all ######################## NIC statistics for port 0 ######################## RX-packets: 7634030896 RX-missed: 36190 RX-bytes: 458041856406 RX-errors: 0 RX-nombuf: 0 TX-packets: 1 TX-errors: 0 TX-bytes: 66 Throughput (since last show) Rx-pps: 142120399 Rx-bps: 68217734104 Tx-pps: 0 Tx-bps: 0 ############################################################################
If you add more rules the HW needs to process the packet against these rules to determine what actions to do. Even if the packet doesn't match any rule, the HW still needs time to go over *all* the flows*. It is expected to see some degradation when adding more rules, even if they don't match. It's also expected to see some degradation when the packet is matched. In this case, the HW not only does the matching, but also the actions. When you offload flows, you want to offload the logic from the SW and accelerate in the HW. As I wrote above, this is expected. Do you still have some concern or an issue?
(In reply to Asaf Penso from comment #13) > If you add more rules the HW needs to process the packet against these rules > to determine what actions to do. > Even if the packet doesn't match any rule, the HW still needs time to go > over *all* the flows*. > It is expected to see some degradation when adding more rules, even if they > don't match. > It's also expected to see some degradation when the packet is matched. In > this case, the HW not only does the matching, but also the actions. > When you offload flows, you want to offload the logic from the SW and > accelerate in the HW. > > As I wrote above, this is expected. > Do you still have some concern or an issue? Thanks a lot ! I see ! but I test multiple flow with intel E810 100G NIC, and not found this performance prolem!
(In reply to Asaf Penso from comment #13) > If you add more rules the HW needs to process the packet against these rules > to determine what actions to do. > Even if the packet doesn't match any rule, the HW still needs time to go > over *all* the flows*. > It is expected to see some degradation when adding more rules, even if they > don't match. > It's also expected to see some degradation when the packet is matched. In > this case, the HW not only does the matching, but also the actions. > When you offload flows, you want to offload the logic from the SW and > accelerate in the HW. > > As I wrote above, this is expected. > Do you still have some concern or an issue? and I start new test with 70 bytes UDP packet, IXIA send packets with 138888888pps! but CX5/CX6 can only receive 112744290pps! testpmd> show port stats all ######################## NIC statistics for port 0 ######################## RX-packets: 1235553968 RX-missed: 4095439 RX-bytes: 81546561888 RX-errors: 0 RX-nombuf: 0 TX-packets: 0 TX-errors: 0 TX-bytes: 0 Throughput (since last show) Rx-pps: 112744290 Rx-bps: 59528973864 Tx-pps: 0 Tx-bps: 0 ############################################################################ testpmd> but with 64 bytes, CX5/CX6 can receive all packets with wire speed(148809523pps)!
Can you try adding a single rule to group 0 to match all traffic and jump to group 1? something like: flow create 0 ingress pattern eth / end actions jump group id 1/ end Then add all of your flows to group 1. What results do you see now?
(In reply to Asaf Penso from comment #16) > Can you try adding a single rule to group 0 to match all traffic and jump to > group 1? > something like: > flow create 0 ingress pattern eth / end actions jump group id 1/ end > > Then add all of your flows to group 1. > > What results do you see now? Ok, I will test it and report it in next day!
(In reply to Asaf Penso from comment #16) > Can you try adding a single rule to group 0 to match all traffic and jump to > group 1? > something like: > flow create 0 ingress pattern eth / end actions jump group id 1/ end > > Then add all of your flows to group 1. > > What results do you see now? no flow rule: testpmd> show port stats all ######################## NIC statistics for port 0 ######################## RX-packets: 13767079146 RX-missed: 20838 RX-bytes: 826024748892 RX-errors: 0 RX-nombuf: 0 TX-packets: 0 TX-errors: 0 TX-bytes: 0 Throughput (since last show) Rx-pps: 148808971 Rx-bps: 71428306336 Tx-pps: 0 Tx-bps: 0 ############################################################################ testpmd> flow list 0 testpmd> with sample flow rules: flow create 0 ingress group 0 pattern eth / end actions jump group 1 / end flow create 0 ingress group 1 pattern eth / ipv4 / end actions rss / end testpmd> show port stats all ######################## NIC statistics for port 0 ######################## RX-packets: 18662603754 RX-missed: 29476 RX-bytes: 1119756225384 RX-errors: 0 RX-nombuf: 0 TX-packets: 0 TX-errors: 0 TX-bytes: 0 Throughput (since last show) Rx-pps: 109444777 Rx-bps: 52533493136 Tx-pps: 0 Tx-bps: 0 ############################################################################ testpmd> flow list 0 ID Group Prio Attr Rule 0 0 0 i-- ETH => JUMP 1 1 0 i-- ETH IPV4 => RSS testpmd> with more flow rules: flow create 0 ingress group 0 pattern eth / end actions jump group 1 / end flow create 0 ingress group 1 pattern eth / ipv4 dst is 1.1.1.1 / tcp / end actions queue index 15 / end flow create 0 ingress group 1 pattern eth / ipv4 dst is 2.2.2.2 / tcp / end actions queue index 15 / end flow create 0 ingress group 1 pattern eth / ipv4 dst is 1.1.1.1 / udp / end actions queue index 15 / end flow create 0 ingress group 1 pattern eth / ipv4 dst is 2.2.2.2 / udp / end actions queue index 15 / end flow create 0 ingress group 1 pattern eth / ipv4 / udp dst is 53 / end actions count / rss / end flow create 0 ingress group 1 pattern eth / ipv4 / tcp dst is 5353 / end actions count / rss / end flow create 0 ingress group 1 pattern eth / ipv4 / tcp dst is 5060 / end actions count / rss / end flow create 0 ingress group 1 pattern eth / ipv4 / tcp dst is 8081 / end actions count / rss / end flow create 0 ingress group 1 pattern eth / ipv4 / udp dst is 5353 / end actions count / rss / end flow create 0 ingress group 1 pattern eth / ipv4 / udp dst is 5060 / end actions count / rss / end flow create 0 ingress group 1 pattern eth / ipv4 / udp dst is 8081 / end actions count / rss / end flow create 0 ingress group 1 pattern eth / ipv4 / end actions rss / end testpmd> show port stats all ######################## NIC statistics for port 0 ######################## RX-packets: 26920802980 RX-missed: 32024 RX-bytes: 1615248178968 RX-errors: 0 RX-nombuf: 0 TX-packets: 0 TX-errors: 0 TX-bytes: 0 Throughput (since last show) Rx-pps: 43985714 Rx-bps: 21113143032 Tx-pps: 0 Tx-bps: 0 ############################################################################ testpmd> flow list 0 ID Group Prio Attr Rule 0 0 0 i-- ETH => JUMP 1 1 0 i-- ETH IPV4 TCP => QUEUE 2 1 0 i-- ETH IPV4 TCP => QUEUE 3 1 0 i-- ETH IPV4 UDP => QUEUE 4 1 0 i-- ETH IPV4 UDP => QUEUE 5 1 0 i-- ETH IPV4 UDP => COUNT RSS 6 1 0 i-- ETH IPV4 TCP => COUNT RSS 7 1 0 i-- ETH IPV4 TCP => COUNT RSS 8 1 0 i-- ETH IPV4 TCP => COUNT RSS 9 1 0 i-- ETH IPV4 UDP => COUNT RSS 10 1 0 i-- ETH IPV4 UDP => COUNT RSS 11 1 0 i-- ETH IPV4 UDP => COUNT RSS 12 1 0 i-- ETH IPV4 => RSS testpmd>
(In reply to Asaf Penso from comment #16) > Can you try adding a single rule to group 0 to match all traffic and jump to > group 1? > something like: > flow create 0 ingress pattern eth / end actions jump group id 1/ end > > Then add all of your flows to group 1. > > What results do you see now? not create group: flow create 0 ingress pattern eth / ipv4 dst is 1.1.1.1 / tcp / end actions queue index 15 / end flow create 0 ingress pattern eth / ipv4 dst is 2.2.2.2 / tcp / end actions queue index 15 / end flow create 0 ingress pattern eth / ipv4 dst is 1.1.1.1 / udp / end actions queue index 15 / end flow create 0 ingress pattern eth / ipv4 dst is 2.2.2.2 / udp / end actions queue index 15 / end flow create 0 ingress pattern eth / ipv4 / udp dst is 53 / end actions count / rss / end flow create 0 ingress pattern eth / ipv4 / tcp dst is 5353 / end actions count / rss / end flow create 0 ingress pattern eth / ipv4 / tcp dst is 5060 / end actions count / rss / end flow create 0 ingress pattern eth / ipv4 / tcp dst is 8081 / end actions count / rss / end flow create 0 ingress pattern eth / ipv4 / udp dst is 5353 / end actions count / rss / end flow create 0 ingress pattern eth / ipv4 / udp dst is 5060 / end actions count / rss / end flow create 0 ingress pattern eth / ipv4 / udp dst is 8081 / end actions count / rss / end flow create 0 ingress pattern eth / ipv4 / end actions rss / end testpmd> show port stats all ######################## NIC statistics for port 0 ######################## RX-packets: 42109410797 RX-missed: 66319 RX-bytes: 2529872463822 RX-errors: 0 RX-nombuf: 0 TX-packets: 0 TX-errors: 0 TX-bytes: 0 Throughput (since last show) Rx-pps: 54996566 Rx-bps: 26398351792 Tx-pps: 0 Tx-bps: 0 ############################################################################ testpmd> flow list 0 ID Group Prio Attr Rule 0 0 0 i-- ETH IPV4 TCP => QUEUE 1 0 0 i-- ETH IPV4 TCP => QUEUE 2 0 0 i-- ETH IPV4 UDP => QUEUE 3 0 0 i-- ETH IPV4 UDP => QUEUE 4 0 0 i-- ETH IPV4 UDP => COUNT RSS 5 0 0 i-- ETH IPV4 TCP => COUNT RSS 6 0 0 i-- ETH IPV4 TCP => COUNT RSS 7 0 0 i-- ETH IPV4 TCP => COUNT RSS 8 0 0 i-- ETH IPV4 UDP => COUNT RSS 9 0 0 i-- ETH IPV4 UDP => COUNT RSS 10 0 0 i-- ETH IPV4 UDP => COUNT RSS 11 0 0 i-- ETH IPV4 => RSS