• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 100 subscribers
  • Views 221 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

SK-AM62P-LP: Link-up failed without auto-negotiation by strap

Sung-IL
Mastermind 31440 points

Part Number: SK-AM62P-LP
Other Parts Discussed in Thread: AM62P, SYSCONFIG

Tool/software:

Dear Champs,

My customer is trying to connect Ethernet with 100Mb/s fixed, and modified HW strap of Ethernet PHY(DP83867)  in AM62P SK EVM.

But, with this HW strap change, they failed to connect EVM with PC(Windows and Ubuntu) by manually setting of duplex/speed. EVM was connected to PC using cross cable.

They also connected between PORT1 and 2 of EVM using cross cable and tried to test it, but still failed to connect each other.

They modified HW strap of AM62P SK EVM to set "Mode 4" to disable 'auto negotiation' in below, and they removed R374/332 registers and change R373/331 register value to 2.4K ohm by referring below e2e.

 RE: SK-AM62P-LP: strap to disable auto-negotiation of Ethernet(forced set to 100Mb/s)  

From DP83867 datasheet,

They modified sysconfig as below.

Could you please guide them how they can link-up ethernet of AM62P SK EVM without auto-negotiation?

Thanks and Best Regards,

SI.

  • 0
    TI__Expert 7590 points

    Hi,

    If the requirement is to disable auto-negotiation, then you can achieve it without using strapped mode. But if the customer needs strapping, then the following configuration looks correct. Can you please check the values of BMSR register in the PHY and share with us? I will loop in the PHY team based on the config observed on the status registers.

    Regards,
    Teja.

  • 0 in reply to Teja Rowthu
    TI__Mastermind 31440 points

    Hi Teja,

    As it is important to reduce Ethernet link-up time, HW strapping should be used.

    Customer found there was holding/hang occurred in below ENET_IOCTL functions in below.

    And also, they tried to read BMSR register, but they also found there was hold/hang occurred while trying to read BMSR register.

    It looks there is something wrong in Ethernet PHY. Is there anything to check why EthernetPHY is not worked?

    Thanks and Best Regards,

    SI.

  • 0 in reply to Sung-IL
    TI__Expert 7590 points

    Hi SI,

    Being stuck in this function points to the fact that there is no link up seen at the PHY. What I would suggest the customer to do is to check the BMSR value just before entering into this function. That way, we can also have the sections which were loaded onto the PHY can be seen, and the BMSR can also be read.

     Can you please send few more details on the application behavior like the call stack, and the prints of the application during the failure?

    Regards,
    Teja.

  • +1 in reply to Teja Rowthu
    TI__Mastermind 26791 points

    Hi SI, 

    I am from the PHY team. Were you able to read the register before the HW strap changes? If so, could you check if the PHY is alive by possibly probing RBIAS pin and RX_CLK pin to see if the RBIAS pin gives out 1V and RX_CLK shows any signal?

    Best,
    J