This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TDA4VM: EST est problem with native driver

Part Number: TDA4VM

When using Native Driver, the verification of the TAS function using CPSW9G alone is normal. However, if you use the CPSW9G Switch-Mode, whether it is Mac Bridge or Vlan Bridge, the packets cannot send from the designated queue。

  • Hi,

    Can you share the SDK version you are using and scripts/applications you run when you receive the error. 

    Regards,
    Tanmay

  • hi,

    My sdk version is sdk8.5 ,below is my operating environment, where T is a VM product board, VH and VM are the development board:

    On the VH board, I ran the following script,

    tas.txt
    Fullscreen
    1
    2
    3
    4
    5
    6
    7
    8
    #!/bin/sh
    ifconfig eth4 down
    ifconfig eth1 down
    ifconfig eth2 down
    ifconfig eth3 down
    ethtool -L eth2 tx 3
    ethtool -L eth3 tx 3
    sleep 10
    XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

    On the VM board, I ran the following script,

        ip link add link eth0 name eth0.402 type vlan id 402
        sleep 3

        ifconfig eth0.402 192.168.5.111

        while true; do ./plget -m rx-lat -f ipgap -n 128 -i eth0.402 -t udp -u 5003 ; done

    On the T board,  I ran the following script,

        ifconfig eth1 192.168.5.90

        ./plget -i eth1 -m pkt-gen -t udp -u 5003 -a 192.168.5.111 -n 128 -l 80

    The picture below is the receiving result,The results show that eth3 realizes the TAS function。

    Then I continue to run on the VH as follows script:

        tc qdisc add dev eth3 clsact

        tc filter add dev eth3 egress protocol ip prio 1 u32 match ip dport 5003 0xffff action skbedit priority 3

    When I sends a message on the T board again, the result is as follows

    According to my expectations, the larger time interval in the result should be 300us instead of 600us。

    Please tell me where there is a problem!

    thanks!

  • Dear Ruiqing,

         Please check the script you are using, could you help check not using Vlan and see if it helps.

    BR

    Sikai

  • Dear SiKai,

    could you help check not using Vlan and see if it helps.

    I have tried,I have tried it, the following is the script.

    Fullscreen
    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    20
    21
    ifconfig eth3 down
    ethtool -L eth2 tx 4
    ethtool -L eth3 tx 4
    ethtool --set-priv-flags eth2 p0-rx-ptype-rrobin off
    ethtool --set-priv-flags eth3 p0-rx-ptype-rrobin off
    sleep 10
    devlink dev param set platform/c000000.ethernet name switch_mode value true cmode runtime
    ip link add name br0 type bridge
    ip link set dev br0 type bridge ageing_time 1000
    ip link set dev eth2 up
    ip link set dev eth3 up
    sleep 5
    ip link set dev eth2 master br0
    ip link set dev eth3 master br0
    ip link set dev br0 up
    phc2sys -s clock_realtime -c eth2 -m -O 0 > /dev/null &
    phc2sys -s clock_realtime -c eth3 -m -O 0 > /dev/null &
    tc qdisc replace dev eth2 parent root handle 100 taprio \
    XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

    Other processes are the same as before.

    The following is the result of running, it still does not look effective.

  • Dear Rui qing,

        We think this is due to two eth interface and the traffic goes through a longer path. 

    BR

    Sikai

  • Dear SiKai,

    The board cards are directly connected through the network cable,I think the 600us time above is the interval of 400+200,The phenomenon looks more like flowing filtering are not effective.

    If the flow filtration takes effect, I think the time interval will become 300us.

  • Hi,

    The traffic might not be falling into correct queues. This is causing the all the traffic to be sent from queue0 which can explain this behaviour.

    To check this, can you share the output of :

    • ethtool -S eth2
    • ethtool -S eth3

    Then I continue to run on the VH as follows script:

        tc qdisc add dev eth3 clsact

        tc filter add dev eth3 egress protocol ip prio 1 u32 match ip dport 5003 0xffff action skbedit priority 3

    When I sends a message on the T board again, the result is as follows

    In the above case, are you sending packets from VM or VH?

    Regards,
    Tanmay

  • hi,

     sorry for the delay   

    The following is the information of ETH2 and ETH3

    67662.log.txt
    Fullscreen
    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    root@j784s4-evm:~/rqyang/tas_script/VH/mac_bridge_tas# ethtool -S eth2
    NIC statistics:
    p0_rx_good_frames: 156
    p0_rx_broadcast_frames: 25
    p0_rx_multicast_frames: 131
    p0_rx_crc_errors: 0
    p0_rx_oversized_frames: 0
    p0_rx_undersized_frames: 0
    p0_ale_drop: 1
    p0_ale_overrun_drop: 0
    p0_rx_octets: 24920
    p0_tx_good_frames: 4
    p0_tx_broadcast_frames: 1
    p0_tx_multicast_frames: 3
    p0_tx_octets: 624
    p0_tx_64B_frames: 4
    p0_tx_65_to_127B_frames: 88
    p0_tx_128_to_255B_frames: 42
    p0_tx_256_to_511B_frames: 26
    XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

    In the above case, are you sending packets from VM or VH?

    I used three boards,A VM board is used as sender,a VH board is used as TSN switch,a VM board is used as receiver.

  • hello,

    Is there any update?