• State
  • Locked Locked
  • Replies 25 replies
  • Subscribers 32 subscribers
  • Views 779 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

DP83826E: DP83826 NO NET

Ning Li
Intellectual 690 points
Part Number: DP83826E

Dears,

Customer use Anyais 3760e can cooperate with dp83826 to make a terminal, now there is no network, Pls kindly help to check the schematic. Thanks

PHY83826.pdf

  • 0
    TI__Mastermind 24881 points

    Hi,

    Can you give some more detail about the failure signature. Are you able to read/write registers? Do you have link? I am unable to read the screenshot which you have shared.

    For schematic, please fill out the attached checklist.

    Thanks,

    David2235.DP83826_Schematic_Design_Review_Checklist.xlsx

  • 0 in reply to David Creger
    Intellectual 690 points

    Dear Dacid,

    1. If R492 R453 is deleted, the negotiation succeeds, but udhcp cannot obtain the IP address

    2. Connect PIN22 to DC3V3D_MAC. The IP address cannot be obtained from udhcp

    3. Connect PIN1 to the GDN. udhcp cannot obtain the IP address

    Pls kindly find the schematic_design_review form.

    Thanks

    2235.DP83826_Schematic_Design_Review_Checklist (1).xlsx

  • 0 in reply to Ning Li
    TI__Mastermind 24881 points

    Hi,

    What do you mean "If R492 R453 is deleted"? Does that mean DNP? How, then, is the PHY getting any reference clock?

    My first guess here is that strapping is the issue. The PHY must be configured to RMII slave mode to operate off a 50Mhz oscillator connected to XI pin. If you are not doing this in software, it must be done through bootstraps. Please have a look at section 9.4.1.1 of the datasheet to determine how to strap in RMII slave mode.

    Thanks,

    David

  • 0 in reply to David Creger
    Intellectual 690 points

    Dear David,

    Sorry for my mistake,Remove the R492 in the PCBA and add the R453 in the PCBA.

    For 50MHZ,Customer feedback that the SOC have provide 50MHZ pulse,so the schematic have issue or not？

    Thanks

  • 0 in reply to Ning Li
    TI__Mastermind 24881 points

    Hi Ning,

    This is an issue. R492 should not be connected in RMII slave mode (R453 should be connected). Please see section 9.3.10 of the datasheet for the difference between RMII master and RMII slave mode. 

    Looking at your schematic, you are strapped into MII mode. This explains why communication is not working. Please strap the device to RMII slave mode (or set it through register write).

    Thanks,

    David

  • 0 in reply to David Creger
    Intellectual 690 points

    Dear David,

    There is no R492 in the customer test now,and there is R453,but is no function.

    Thanks

  • 0 in reply to Ning Li
    Intellectual 690 points

    Dear David,

    Set it through register write,customer want to know witch register.

    Thanks

  • 0 in reply to Ning Li
    TI__Mastermind 24881 points

    Hi Ning,

    MII/RMII and RMII master/slave can be selected in register 0x468.

    Thanks,

    David

  • 0 in reply to David Creger
    Intellectual 690 points

    Dear David,

    Pls help to check below information:

    Debug: Operate the 0X468 register, configure BIT4, BIT3, code flow as attachment.

    Effect: After configuring register 0X468 and reading back the register value, udhcpc still cannot assign IP:

    = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =

    xzj 0x0468 reg val:0x2c87

    xzj read back 0x0468 reg val:0x2c87

    = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =

    Run log as attached.

    BTW,The schematic have any problem or not？

    Thanks

    初始化流程.txt
    Fullscreen
    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    status = ak_gmac_read_phy_reg((u32 *)gmacdev->MacBase, gmacdev->PhyBase, PHY_ID_HI_REG, &data_id1);
    if(status) {
    printk("phy read phy id1 waiting Error,data=0x%x\n", data_id1);
    goto err_phy_reg;
    }
    status = ak_gmac_read_phy_reg((u32 *)gmacdev->MacBase, gmacdev->PhyBase, PHY_ID_LOW_REG, &data_id2);
    if(status) {
    printk("phy read phy id2 waiting Error,data=0x%x\n", data_id2);
    goto err_phy_reg;
    }
    else if(data_id1 == 0x2000 && ((data_id2 == 0xa140) || (data_id2 == 0xa130) || (data_id2 == 0xa110)) ) { /* dp83825 and dp83826 initialization*/
    printk("cdh:id 0xa140 or 0xa130 or 0xa110 set!\n");
    /* RMII Reference Clock Selec: RMII Reference Clock Selec */
    data_ctl = 0;
    ak_gmac_read_phy_reg((u32 *)gmacdev->MacBase, gmacdev->PhyBase, 0x17, &data_ctl);
    XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
    运行log.txt
    Fullscreen
    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    /tmp # insmod ak_eth.ko
    ak_mac Ethernet Driver
    Configed MAC RMII interface!
    ak37e_clk: reset_module:0xf0008020 @ 16
    of_get_named_gpiod_flags: parsed 'reset-gpios' property of node '/soc/ethernet1@20500000[0]' - status (0)
    ak39_ethernet 20500000.ethernet1 eth0: irq 31, mapped at c571c000
    ak39_ethernet 20500000.ethernet1: using random MAC address 42:f9:37:64:68:0f
    ak_mac_probe, line:5777, mac_id:1
    insmod mac device ok
    /tmp # udhcpc > /dev/null
    udhcpc: started, v1.30.1
    Reset MAC1 controller and PHY!
    phy read phy id1 waiting ok,data=0x2000
    phy read phy id2 waiting ok,data=0xa130
    cdh:id 0xa140 or 0xa130 or 0xa110 set!
    xzj 0x0468 reg val:0x2c87
    xzj read back 0x0468 reg val:0x2c87
    phy read phy id1 waiting ok,data=0x2000
    phy read phy id2 waiting ok,data=0xa130
    XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

  • 0 in reply to Ning Li
    TI__Mastermind 24881 points

    Hi Ning,

    If you are using a 50Mhz clock connected to XI pin, then you must use RMII slave mode. 0x0468 = 0x2c87 selects MII mode. Which MAC interface is the customer trying to use?

    Please send me a register dump from 0x0 - 0x1f, as well as 0x467, 0x468.

    Thanks,

    David

  • 0 in reply to David Creger
    Intellectual 690 points

    Dear David,

    Pls find attachfile for 0X0-0X1f 0X467,0X468.

    Thanks

    6507.code.txt
    Fullscreen
    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    while(loop_count-- > 0){
    status = ak_gmac_read_phy_reg((u32 *)gmacdev->MacBase, gmacdev->PhyBase, PHY_STATUS_REG, &data);
    if(status)
    continue;
    if((data & Mii_AutoNegCmplt) != 0){
    if ((data_id1 == 0x2000) && ((data_id2 == 0xa140) || (data_id2 == 0xa130) || ((data_id2 == 0xa110))))
    {
    status = ak_gmac_read_phy_reg((u32 *)gmacdev->MacBase, gmacdev->PhyBase, PHY_SPECIFIC_CTRL_REG, &data);
    if(status)
    continue;
    if((data & 0x10) != 0){
    printk(KERN_ERR"Autonegotiation Complete,loop_count:%d\n", loop_count);
    #if 1 //xiaozijie , anjubao
    for(i = 0; i <= 0x1f; i++)
    {
    XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
    eth_log.txt
    Fullscreen
    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    /tmp # insmod ak_eth.ko
    ak_mac Ethernet Driver
    Configed MAC RMII interface!
    ak37e_clk: reset_module:0xf0008020 @ 16
    of_get_named_gpiod_flags: parsed 'reset-gpios' property of node '/soc/ethernet1@20500000[0]' - status (0)
    ak39_ethernet 20500000.ethernet1 eth0: irq 31, mapped at c5638000
    ak39_ethernet 20500000.ethernet1: using random MAC address 0e:5f:81:8e:46:61
    ak_mac_probe, line:5808, mac_id:1
    insmod mac device ok
    /tmp # udhcpc
    udhcpc: started, v1.30.1
    Reset MAC1 controller and PHY!
    phy read phy id1 waiting ok,data=0x2000
    phy read phy id2 waiting ok,data=0xa130
    cdh:id 0xa140 or 0xa130 or 0xa110 set!
    xzj 0x0468 reg val:0x2c87
    xzj read back 0x0468 reg val:0x2c87
    phy read phy id1 waiting ok,data=0x2000
    phy read phy id2 waiting ok,data=0xa130
    XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

  • 0 in reply to Ning Li
    TI__Mastermind 24881 points

    Hi Ning,

    If you are using a 50Mhz clock connected to XI pin, then you must use RMII slave mode. 0x0468 = 0x2c87 selects MII mode. Which MAC interface is the customer trying to use?

    Thanks,

    David

  • 0 in reply to David Creger
    Intellectual 690 points

    Dear David,

    I have send your reply to customer,I will confirm the MAC.

    So register dump from 0x0 - 0x1f have any problem or not?

    Thanks

  • 0 in reply to Ning Li
    TI__Mastermind 24881 points

    Hi Ning,

    The register dump shows link is up, so the problem is likely on the MAC interface side. Let me know the customer's reply.

    Thanks,

    David

  • 0 in reply to David Creger
    Intellectual 690 points

    Hi David,

    For MAC,they dont reply to me,they use DP83825 before,and now they check the hardware first,if any reply I will let you know.

    Thanks

  • 0 in reply to Ning Li
    TI__Mastermind 24881 points

    Hi Ning,

    This is required information. Let me know when they get back to you. 

    Thanks,

    David

  • 0 in reply to David Creger
    Intellectual 690 points

    Hi David,

    MAC is AK3760D,customer have Optimized hardware and provide a new register data,pls help to check it.

    Anyka安凯微电子

    Thanks

    Serial-COM5_03_21_17_54.log
    Fullscreen
    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    [20230321_17:54:21][20230321_17:54:21]
    [20230321_17:54:25]A
    [20230321_17:54:25]SPL 2013.10.0-V3.1.33
    [20230321_17:54:25]Jumping to U-Boot
    [20230321_17:54:25]
    [20230321_17:54:25]
    [20230321_17:54:25]U-Boot 2013.10.0-V3.1.33 (Mar 21 2023 - 09:21:15)
    [20230321_17:54:25]
    [20230321_17:54:25]uboot display lcd logo!
    [20230321_17:54:25]DRAM: 64 MiB
    [20230321_17:54:25]128 MiB
    [20230321_17:54:25]DTB no mmc info
    [20230321_17:54:26]ak_fb: srgb device is not enable
    [20230321_17:54:26]In: serial
    [20230321_17:54:26]Out: serial
    [20230321_17:54:26]Err: serial
    [20230321_17:54:26]Net: eth-0
    [20230321_17:54:26]
    [20230321_17:54:26]Hit any key to stop autoboot: 1
    XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

  • 0 in reply to Ning Li
    TI__Mastermind 24881 points

    Hi Ning,

    What changes were made to optimize the hardware? I need to know which MAC interface the customer wants to use, not just the MAC name. Do they want MII, RMII slave mode, RMII master mode, etc.?

    Thanks,

    David

  • 0 in reply to David Creger
    Intellectual 690 points

    Dear David,

    1.They refer to 9.4.1 hardware configuration,Add pull up resister 2.49K in PIN 14,PIN20,PIN22,now they want to know whether the hardware have any problem or not? If the hardware have no problem,they will check the software. So pls kindly help to check the changed schematic

    2.For work mode,The hardware boot configuration is as follows:ENHANCED 、RMII 、slave mode.

    PIN1：1

    PIN28：0

    PIN29：0

    PIN30：1

    PIN14：1

    PIN13：0

    PIN22：1

    PIN15：0

    PIN16：0

    PIN18：0

    PIN20：1

    3.For Mac,now the Phy can connect but no net,and anyka want to konw  whether the specific requirements for the mac or not?

    AJB_DP83826.pdf

    Thanks

  • 0 in reply to Ning Li
    Intellectual 690 points

    Hi David,

    BTW,pls kindly help to check the  register data too after they optimized the hardware.

    Thanks

    3324.Serial-COM5_03_21_17_54.log
    Fullscreen
    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    [20230321_17:54:21][20230321_17:54:21]
    [20230321_17:54:25]A
    [20230321_17:54:25]SPL 2013.10.0-V3.1.33
    [20230321_17:54:25]Jumping to U-Boot
    [20230321_17:54:25]
    [20230321_17:54:25]
    [20230321_17:54:25]U-Boot 2013.10.0-V3.1.33 (Mar 21 2023 - 09:21:15)
    [20230321_17:54:25]
    [20230321_17:54:25]uboot display lcd logo!
    [20230321_17:54:25]DRAM: 64 MiB
    [20230321_17:54:25]128 MiB
    [20230321_17:54:25]DTB no mmc info
    [20230321_17:54:26]ak_fb: srgb device is not enable
    [20230321_17:54:26]In: serial
    [20230321_17:54:26]Out: serial
    [20230321_17:54:26]Err: serial
    [20230321_17:54:26]Net: eth-0
    [20230321_17:54:26]
    [20230321_17:54:26]Hit any key to stop autoboot: 1
    XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

  • 0 in reply to Ning Li
    TI__Mastermind 24881 points

    Hi Ning,

    From the register data, I can see that the link is up in 100Mbps full duplex mode, and the PHY is in RMII slave mode. From the PHYs perspective, there are no issues here. 

    What do you mean by "For Mac,now the Phy can connect but no net." Please send any applicable screenshots.

    Thanks,

    David

  • 0 in reply to David Creger
    Intellectual 690 points

    Hi David,

    I have talk about customer,they said that "For Mac,now the Phy can connect but no net." you can find out this appearance at attachfile.

    And now they will double check the software.

    Thanks

    2671.Serial-COM5_03_21_17_54.log
    Fullscreen
    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    [20230321_17:54:21][20230321_17:54:21]
    [20230321_17:54:25]A
    [20230321_17:54:25]SPL 2013.10.0-V3.1.33
    [20230321_17:54:25]Jumping to U-Boot
    [20230321_17:54:25]
    [20230321_17:54:25]
    [20230321_17:54:25]U-Boot 2013.10.0-V3.1.33 (Mar 21 2023 - 09:21:15)
    [20230321_17:54:25]
    [20230321_17:54:25]uboot display lcd logo!
    [20230321_17:54:25]DRAM: 64 MiB
    [20230321_17:54:25]128 MiB
    [20230321_17:54:25]DTB no mmc info
    [20230321_17:54:26]ak_fb: srgb device is not enable
    [20230321_17:54:26]In: serial
    [20230321_17:54:26]Out: serial
    [20230321_17:54:26]Err: serial
    [20230321_17:54:26]Net: eth-0
    [20230321_17:54:26]
    [20230321_17:54:26]Hit any key to stop autoboot: 1
    XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

  • 0 in reply to Ning Li
    TI__Mastermind 24881 points

    Hi Ning,

    In the text file, I see eth0 is up and:

    "RX packets:95 errors:0 dropped:39 overruns:0 frame:0
    TX packets:3 errors:0 dropped:0 overruns:0 carrier:0
    collisions:0 txqueuelen:1000"

    What line in the file shows "but no net"?

    Can you perform a ping command?

    Thanks,

    David

  • 0 in reply to David Creger
    Intellectual 690 points

    Dear David,

    They have solve the problem,so you can close the case.

    Thanks

  • 0 in reply to Ning Li
    TI__Mastermind 24881 points

    Hi Ning,

    Can you please share what the solution was? 

    Thanks,

    David