• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 3 answers
  • Subscribers 33 subscribers
  • Views 239 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.

DS320PR410: Swaping for lane and pin

ttd
Mastermind 9295 points

Part Number: DS320PR410

Tool/software:

Dear Technical Support Team,

I have two questions and confirm the functionality of the Re-driver “DS320PR410RNQR” and “DS320PR810NJXR”.

Q1
Is it acceptable to reverse the P and N connections for each PCIe signal on these Re-drivers?
For example, connecting TXn/RXn signals to TXp/RXp pins, and connecting TXp/RX signals to TXn/RXn pins.

Q2
Is it acceptable to swap the lanes for these Re-drivers?
For example, connecting TX/RX[1], TX/RX[0], TX/RX[3], TX/RX[2] signals to the TX/RX[0], TX/RX[1], TX/RX[2], TX/RX[3] pins

Best Regards,

ttd

  • 0
    TI__Genius 13020 points

    Hi,

    Polarity swapping can be done and is supported in PCIe.

    Lane reversal is supported but lanes cannot be randomly swapped.  (ie, lanes 0-3 can be inverted to lanes 3-0, but must be done in order)

    Regards,

    Undrea

  • 0 in reply to Undrea Fields
    Mastermind 9295 points

    Hi Undrea,

    Thank you for your reply.

    Lane reversal is supported but lanes cannot be randomly swapped.  (ie, lanes 0-3 can be inverted to lanes 3-0, but must be done in order)

    →As an aside, I understand it's not possible for general target devices, but why is it not possible with the Re-driver?

      For my reference, please let me know if there is a reason.

    Best Regards,

    ttd

  • 0 in reply to ttd
    Mastermind 9295 points

    Hi Undrea,

    Do you have any update so far?

    Best Regards,

    ttd

  • 0 in reply to ttd
    TI__Genius 13020 points

    Sorry, I missed following up on this.

    There's nothing preventing this to be done through the redriver, but the lanes should be corrected to the Endpoint.  The redriver has no knowledge of the lane order.

    Regards,

    Undrea

  • 0 in reply to Undrea Fields
    Mastermind 9295 points

    Hi Undrea,

    Thank you for your comment.

    If the lanes from the CPU to the redriver are randomly swapped, the output of the redriver should work if I rearrange the signals in the correct order on the PCB and then connect them to the endpoint.
    Is this understanding correct?


    Example:
    CPU (Lanes 0,1,2,3) → Redriver (Lanes 1,0,3,2 randomly) → Endpoint (lanes 0,1,2,3 correct order)

    Best Regards,

    ttd

  • 0 in reply to ttd
    TI__Genius 13020 points

    Hi ttd,

    Your understanding is correct.

    Regards,

    Undrea