• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 2 answers
  • Subscribers 33 subscribers
  • Views 360 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.

DP83867CS: It takes a few minutes to link up.

FUSANOBU KAMITANI
Prodigy 20 points
Part Number: DP83867CS

I'm having a problem where it takes a few minutes for the DP83867CS to link up.

The circuit configuration is as follows.

・ Multiple DP83867 are mounted on our circuit board.

・ FPGA and each DP83867 are connected using SGMII.

・ Reset to each DP83867 will be released from the FPGA at the same timing.

・ DP83867 and RJ-45 are mirror-connected.

・ I use a straight cable to connect to RJ-45 of DP83867 (1) and DP83867 (2) on the circuit board.

The strap settings are listed below.

・ Auto-negotiation enabled

・ Advance ability of 10/100/1000

・ Port mirroring strapped to enable

・ SGMII strapped to enable

The problem is that if I unreset with the cable connected, it takes a few minutes for the DP83867 to link up.

After the DP83867 links up, it links up as soon as I unplug and plug it in.

The registers when DP83867 is not linked up are as follows.

・ Auto-Negotiation Link Partner Ability Register (ANLPAR), Address 0x0005

    → It is 0xC1E1 and the advertisement seems to be completed.

・ PHY Status Register (PHYSTS), Address 0x0011

    → 0x1000 and 0x1300 are repeated alternately.

    → When DP83867 is linked up, it will be 0xAC02.

Please give me some advice to speed up the link up.

  • 0
    TI__Genius 17060 points

    Hello,

    I have acknowledged your post and I hope to get back to you by the end of business Friday (8/27). In the meantime, can you please provide a block diagram?

    Kind Regards,

    Joe

    All information in this correspondence and in any related correspondence is provided “as is” and “with all faults”, and is subject to TI’s Important Notice (www.ti.com/.../important-notice.shtml)

  • 0 in reply to Joe Vanacore
    Prodigy 20 points
    Thank you for your reply.
    We will send you a block diagram.
    Regards,
    Fusanobu Kamitani
    5315.Block Diagram.pdf
  • 0 in reply to FUSANOBU KAMITANI
    TI__Genius 17060 points

    Hello,

    Thank you for providing the block diagram. I have a couple of follow-up questions.

    1. Have you tried forcing the link speed on for PHY 1 and PHY 2?
      1. Disable Auto-Negotiation
      2. Set your desired speed
      1. May I ask what you mean by "unreset"?
      2. Are you saying that after a power-cycle, the link-up issue is resolved?

    Kind Regards,

    Joe

    All information in this correspondence and in any related correspondence is provided “as is” and “with all faults”, and is subject to TI’s Important Notice (www.ti.com/.../important-notice.shtml)

  • 0 in reply to Joe Vanacore
    Prodigy 20 points

    Thank you for your reply to me.

    > 1. Have you tried forcing the link speed on for PHY 1 and PHY 2?

    > a. Disable Auto-Negotiation

    > b. Set your desired speed

         → The results of the trial are shown below.

        * Writing from FPGA to PHY1 and PHY2 at the same time through MDIO.

    ★ 10M

    (1) Write 0x0100 to [Basic Mode Control Register, Address 0x0000] (10M, Auto-neg: disable)

         → Link up immediately.

    ★ 100M

    (1) Write 0x2100 to [Basic Mode Control Register, Address 0x0000] (100M, Auto-neg: disable)

         → Link up immediately.

    ★ 1000M

    (1) Write 0x0140 to [Basic Mode Control Register, Address 0x0000] (1000M, Auto-neg: disable)

         → Do not link up.

    > 2. > a. May I ask what you mean by "unreset"?

    > b. Are you saying that after a power-cycle, the link-up issue is resolved?

         → I'm sorry. It seems that the explanation was bad. Here is the sequence that was performed.

    (1) Connect the cable to RJ-45 of PHY1 and PH2.

    (2) Turn on the power to the circuit board.

    (3) The power-on reset is released.

    (4) It takes a few minutes to link up.

    (5) Remove the cable.

    (6) Connect the cable to RJ-45 of PHY1 and PHY3.

    (7) Link up immediately.

    * The reason for connecting PHY1 and PHY3 is that connecting PHY1 and PHY2 takes several minutes to link up.

    (8) Remove the cable.

    (9) When I connect the cable to RJ-45 of PHY1 and PHY2, it links up immediately.

    (10) Reset PHY1 and PHY2 without turning off the power.

    (11) Release the reset of PHY1 and PHY2.

    (12) It takes a few minutes to link up.

    Regards,

    Fusanobu Kamitani

  • 0 in reply to FUSANOBU KAMITANI
    TI__Genius 17060 points

    Hello Fusanobu,

    Thank you for providing that information. I would like to request your friendship on e2e forums. Please accept my request at your earliest convenience and I will continue to support you via private message. I will now close this thread.

    Kind Regards,

    Joe

    All information in this correspondence and in any related correspondence is provided “as is” and “with all faults”, and is subject to TI’s Important Notice (www.ti.com/.../important-notice.shtml)

  • 0 in reply to Joe Vanacore
    Prodigy 20 points

    Hello Joe,

    I'm in trouble!

    > " I would like to request your friendship on e2e forums. Please accept my request at your earliest convenience."

    Please tell me what to do.

    Kind Regards,

    Fusanobu

  • 0 in reply to FUSANOBU KAMITANI
    TI__Genius 17060 points

    Hi Fusanobu,

    I may be mistaken but I thought that you accepted my request. 

    On the top right corner of the web page you should see the person icon for your profile. Then click 'Profile'.

    then navigate to 'Profile'.

    Once you are on your profile page, click on the 'friends' drop-down menu. Then click 'Requests to Review' and accept my request.

    Once complete, at the top right corner of the home page you should see this messages icon and this is where we will communicate. 

    Please let me know if you need any additional help/assistance accepting my request.

    Kind Regards,

    Joe

    All information in this correspondence and in any related correspondence is provided “as is” and “with all faults”, and is subject to TI’s Important Notice (www.ti.com/.../important-notice.shtml)

  • 0 in reply to Joe Vanacore
    TI__Genius 17060 points

    Hello Fusanobu,

    We are continuing support via email. I will now close this thread.

    Kind Regards,

    Joe

    All information in this correspondence and in any related correspondence is provided “as is” and “with all faults”, and is subject to TI’s Important Notice (www.ti.com/.../important-notice.shtml)