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.

Linux/AM4376: EtherCAT Master frame loss

Part Number: AM4376

Tool/software: Linux

Hi SIr 

we used AM4376 for ethercat master development and put third-party ethercat master stack in it. 

We found frame lost is series. So we bought another 3rd-party EVM for comparison. below is the information.

Our MB 

Kernel: RT Linux version 4.9.69

Ethernet: CPSW MII with  RTL8211FI-CG

3rd-party MB

Kernel:Linux version 4.1.18

Ethernet: CPSW RGNUU with AR8035

Our MB has pass Ethernet compliance test 

Question:

1. What is the difference between these two kernel versions ?  any effort for ethercat master implementation ? 

         Ours: Kernel RT Linux version 4.9.69

3rd-party: Kernel:Linux version 4.1.18

2. The HW difference is one used MII and another used RGMII

    Is it the factor to cause frame lost ? Does TI have any tool to test ethercat master function?

Thanks for suggestion

BR
Yimin

RTL8211FI-CG

  • Please complete the following checklist and post the results here: processors.wiki.ti.com/.../5x_CPSW
  • Hi,

    TI does not a maintain a list of differences between kernel versions. TI does not support questions on Ethercat Master as these SW packages are typically supplied by a 3rd party.

    Which Ethernet ports are being used here? CPSW or PRU Ethernet ?

    For the Ethernet compliance test could you please describe how this works? Was this done with non-EtherCAT but IP traffic? In the checklist that was provided in the previous post please note the request to look at the HW statistics and report any errors detected by the HW. We need to check that Ethernet traffic in general works before trying to run EtherCAT.

    The type of PHY should not matter, but the link speed might. While both PHYs are rated for 1Gpbs are both connecting at 1Gbps?

    One last thing to look at is the user environment. If you are you using a TI SDK running SystemD there will be daemons that try to send on discovery packets eth1 for example if the CPSW ports are being used.

    Best Regards,
    Schuyler
  • Hi,

    Since we have not heard back we are assuming that you were able to move past the issue and will close the thread.

    Best Regards,
    Schuyler