• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Subscribers 33 subscribers
  • Views 799 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.

DP83822I: Ethernet not working, no physical data flow.

Jan Grulich
Prodigy 50 points
Part Number: DP83822I
Other Parts Discussed in Thread: SK-AM62,

Hello,

I'm running the Sitara AM623x custom board with two Ethernet PHYs.

I have there two DP83822IRHBT with the following bootstrap config: FX_EN=0, PHY_AD0=0, AN_1=1, PHY_AD1=0, EEE_EN=0, PHY_AD2=0, FLD_EN=0, PHY_AD3=0, AN_EN=1, PHY_AD4=0, AN_0=1, LED_SPEED=1, LED_CFG=0, RGMII_EN=1, AMDIX_EN=1, XI_50=0, RMII_EN=0.

The only difference for the second PHY is the PHY_AD0=1

Connected the same way as the gigabit on the SK-AM62 EVK, means RGMII Interface TX_D[0:3], TX_CLK, TX_XTRL, RX_D[0:3], RX_CLK, RX_CTRL with 25MHz external clock to XI.

I'm running tisdk-base-image-am62xx-evm.tar.xz

Everything in the system looks fine, it detects connected cable, even Auto-MDIX works, speed is correctly set to 100M, ACT LED is flashing, but no data flow from any side.

dmesg:

[ 5.757522] remoteproc remoteproc0: 5000000.m4fss is available
[ 5.765139] remoteproc remoteproc0: Direct firmware load for am62-mcu-m4f0_0-fw failed with error -2
[ 5.776290] remoteproc remoteproc0: powering up 5000000.m4fss
[ OK ] Started Network Service.
[ 5.784400] random: systemd: uninitialized urandom read (16 bytes read)
[ 5.786144] remoteproc remoteproc0: Direct firmware load for am62-mcu-m4f0_0-fw failed with error -2
[ 5.798062] am65-cpsw-nuss 8000000.ethernet: down msc_sl e0000000 tmo 0
[ 5.811171] random: systemd-journal: uninitialized urandom read (16 bytes read)
[ 5.811792] remoteproc remoteproc0: request_firmware failed: -2
Starting Wait for Network to be Configured...
[ 5.838321] am65-cpsw-nuss 8000000.ethernet: set new flow-id-base 19
[ 5.853845] am65-cpsw-nuss 8000000.ethernet eth1: PHY [8000f00.mdio:01] driver [Generic PHY] (irq=POLL)
[ 5.854404] am65-cpsw-nuss 8000000.ethernet eth1: configuring for phy/rgmii-rxid link mode
[ 5.873332] am65-cpsw-nuss 8000000.ethernet: down msc_sl e0000000 tmo 0
[ 5.901394] am65-cpsw-nuss 8000000.ethernet eth0: PHY [8000f00.mdio:00] driver [Generic PHY] (irq=POLL)
[ 5.901418] am65-cpsw-nuss 8000000.ethernet eth0: configuring for phy/rgmii-rxid link mode

Please be so kind and point me what may be wrong, or what to test.

  • 0
    Prodigy 50 points

    Here is more exact dmesg:

    root@am62xx-evm:~# dmesg | grep cpsw

    [ 1.190957] am65-cpsw-nuss 8000000.ethernet: initializing am65 cpsw nuss version 0x6BA01103, cpsw version 0x6BA81103 Ports: 3 quirks:00000002
    [ 1.203896] am65-cpsw-nuss 8000000.ethernet: Use random MAC address
    [ 1.210161] am65-cpsw-nuss 8000000.ethernet: initialized cpsw ale version 1.5
    [ 1.217285] am65-cpsw-nuss 8000000.ethernet: ALE Table size 512
    [ 1.223963] am65-cpsw-nuss 8000000.ethernet: CPTS ver 0x4e8a010c, freq:500000000, add_val:1 pps:0
    [ 5.692621] am65-cpsw-nuss 8000000.ethernet: down msc_sl e0000000 tmo 0
    [ 5.732534] am65-cpsw-nuss 8000000.ethernet: set new flow-id-base 19
    [ 5.748645] am65-cpsw-nuss 8000000.ethernet eth1: PHY [8000f00.mdio:01] driver [Generic PHY] (irq=POLL)
    [ 5.763064] am65-cpsw-nuss 8000000.ethernet eth1: configuring for phy/rgmii-rxid link mode
    [ 5.808414] am65-cpsw-nuss 8000000.ethernet: down msc_sl e0000000 tmo 0
    [ 5.826104] am65-cpsw-nuss 8000000.ethernet eth0: PHY [8000f00.mdio:00] driver [Generic PHY] (irq=POLL)
    [ 5.841823] am65-cpsw-nuss 8000000.ethernet eth0: configuring for phy/rgmii-rxid link mode

  • 0 in reply to Jan Grulich
    TI__Mastermind 19250 points

    Hello Jan,

    It is strange that ACT LED is flashing, but there is no data transfer.

    Is the problem same with both the PHYs?

    Can you please probe RX_CLK, RX_D* pins and TX_CLK and TX_D* pins of the PHY and capture the waveforms?

    --
    Regards,
    Gokul.

  • 0 in reply to Gokul Koraganji
    Prodigy 50 points

    Hello, it's quite difficult for me to capture data hence it's no easy to reach these lines. May be this helps a bit that the line is not stable.

    Both lines works really the same wrong way.

    443.286421] am65-cpsw-nuss 8000000.ethernet eth0: Link is Up - 100Mbps/Full - flow control rx/tx
    [ 449.432518] am65-cpsw-nuss 8000000.ethernet eth0: Link is Down
    [ 452.502428] am65-cpsw-nuss 8000000.ethernet eth0: Link is Up - 100Mbps/Full - flow control rx/tx
    [ 501.654649] am65-cpsw-nuss 8000000.ethernet eth0: Link is Down
    [ 505.750424] am65-cpsw-nuss 8000000.ethernet eth0: Link is Up - 100Mbps/Full - flow control rx/tx
    [ 553.878317] am65-cpsw-nuss 8000000.ethernet eth0: Link is Down
    [ 556.950253] am65-cpsw-nuss 8000000.ethernet eth0: Link is Up - 100Mbps/Full - flow control rx/tx
    [ 570.262522] am65-cpsw-nuss 8000000.ethernet eth0: Link is Down
    [ 572.310417] am65-cpsw-nuss 8000000.ethernet eth0: Link is Up - 100Mbps/Full - flow control rx/tx
    [ 690.070350] am65-cpsw-nuss 8000000.ethernet eth0: Link is Down
    [ 697.110246] IPv6: ADDRCONF(NETDEV_CHANGE): eth1: link becomes ready
    [ 697.116570] am65-cpsw-nuss 8000000.ethernet eth1: Link is Up - 100Mbps/Full - flow control rx/tx
    [ 698.134344] am65-cpsw-nuss 8000000.ethernet eth1: Link is Down
    [ 701.206452] am65-cpsw-nuss 8000000.ethernet eth1: Link is Up - 100Mbps/Full - flow control rx/tx
    [ 729.878318] am65-cpsw-nuss 8000000.ethernet eth1: Link is Down
    [ 737.046229] am65-cpsw-nuss 8000000.ethernet eth1: Link is Up - 100Mbps/Full - flow control rx/tx
    [ 740.118453] am65-cpsw-nuss 8000000.ethernet eth1: Link is Down
    [ 742.166429] am65-cpsw-nuss 8000000.ethernet eth1: Link is Up - 100Mbps/Full - flow control rx/tx
    [ 751.382332] am65-cpsw-nuss 8000000.ethernet eth1: Link is Down

    One more question, is the cpsw-nuss the right driver for the DP83822I?

  • 0 in reply to Jan Grulich
    Prodigy 50 points

    The link instability was identified in wrong RJ45 connector. Connector replaced. Now it's stable. I've probed the lines and there is only 25MHz on TX_CLK and RX_CLK. When the cable is not connected, the RX_CLK has 2.5MHz clock. All data lines are idle. Some at GND, Some at 3.3V.

  • 0 in reply to Jan Grulich
    Prodigy 50 points

    TX_D[0:3] are all at GND, RX_D[0:3} are at 3V3.

  • 0 in reply to Jan Grulich
    TI__Mastermind 19250 points

    Hello Jan,

    Jan Grulich said:
    The link instability was identified in wrong RJ45 connector. Connector replaced. Now it's stable. I've probed the lines and there is only 25MHz on TX_CLK and RX_CLK. When the cable is not connected, the RX_CLK has 2.5MHz clock.

    This behavior is as expected when cable is disconnected.

    Jan Grulich said:
    TX_D[0:3] are all at GND, RX_D[0:3} are at 3V3.

    When cable is disconnected, the RX_D[3:0] can't be high. RX_D[2:0] should be 0V and RX_D3 can either be 0 or 1 depending on the settings.

    Can you please probe the signals when transmitting packets from either sides of the PHY?

    Jan Grulich said:
    One more question, is the cpsw-nuss the right driver for the DP83822I?

    You'll have to check with AM62 team regarding this. Can you raise another E2E request for these details?

    --
    Regards,
    Gokul.

  • 0 in reply to Gokul Koraganji
    Prodigy 50 points

    The whole issue was in the wrong connection of the LED1, which was connected to VCC instead of GND.