• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 1 answer
  • Subscribers 97 subscribers
  • Views 3000 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.

Linux/AM5728: Ethernet errors

yongjun wang
Intellectual 725 points
Part Number: AM5728


Tool/software: Linux

Hi all,our custom board reference on AM5728 IDK,we met net error when kernel boot up.our sdk version is ti-processor-sdk-linux-rt-
am57xx-evm-03.03.00.04,the dts file we reference am572x-idk.dts,we just modify the GPIOs about STATUSLEDS. the part of log about
Starting Network Service as follow:

Starting Network Service...
[ OK ] Started System Logging Service.
[ 10.019226] omap_wdt: OMAP Watchdog Timer Rev 0x01: initial timeout 60 sec
[ 10.028591] omap-aes 4b500000.aes: OMAP AES hw accel rev: 3.3
[ OK ] Started strongSwan IPsec IKEv1/IKEv2 daemon using ipsec.conf.
Starting Print notice about GPLv3 packages...
Starting Avahi mDNS/DNS-SD Stack...
Starting Telephony service...
[ 10.155382] net eth1: initializing cpsw version 1.15 (0)
[ 10.175009] net eth0: initialized cpsw ale version 1.4
[ 10.201361] FAT-fs (mmcblk0p1): Volume was not properly unmounted. Some data.
[ OK ] Started Network Service.
[ 10.258747] ahci 4a140000.sata: SSS flag set, parallel bus scan disabled
[ 10.275280] ahci 4a140000.sata: AHCI 0001.0300 32 slots 1 ports 3 Gbps 0x1 ie
[ 10.278192] net eth0: ALE Table size 1024
[ 10.281182] libphy: PHY 48485000.mdio:01 not found
[ 10.281189] net eth1: phy "48485000.mdio:01" not found on slave 1, err -19
[ 10.321346] IPv6: ADDRCONF(NETDEV_UP): eth1: link is not ready
[ 10.358286] net eth0: initializing cpsw version 1.15 (0)
[ 10.361732] libphy: PHY 48485000.mdio:00 not found
[ 10.361744] net eth0: phy "48485000.mdio:00" not found on slave 0, err -19
[ 10.367145] IPv6: ADDRCONF(NETDEV_UP): eth0: link is not ready
[ OK ] Started Permit User Sessions.[ 10.425362] ahci 4a140000.sata: flags:

we searched someone have met the same problem,the link is this and this,but there's no method to solve it finally. The schematic as follow:

ETH1.pdfETH0.pdf

can someone help us to solve this problem?

  • 0
    TI__Guru**** 393215 points
    Please check and make sure that your PHY's are set with different addresses. From the schematics they seem to be at the same MDIO address.
  • 0 in reply to Biser Gatchev-XID
    Intellectual 725 points

    Hi Biser,we find out one bug in the both schematic ETH1.pdf and ETH0.pdf, in ETH1.pdf 

    the left net name is VETH1_AVDDL_PMOS  but the right is VETH1_AVDDL__PMOS ,added one _  what a careless. the same problem in ETH0.pdf. now we modify by jumper on our board.we test all the voltage is right. as you say it seem to the same MDIO address. we have look over the datasheet of KSZ9031, the address is  set by PIN35 15 17 when power up,in our schematic, the PIN17 in both pdf are connected different,it is connected to DVDDH in ETH1.pdf by a resistor(on our board is 10k),and connected to DGND by a resistor(on our board is 1k),so the address should be different,we get the log as next picture,

    is this right.?the same id means wrong? what's the serious problem is after the console output end with up picture,whole the board power down,it seem to the net problem trig ti reset. does this trig by the same id? 

  • 0 in reply to yongjun wang
    TI__Guru**** 393215 points
    Please follow this checklist and post the results here: processors.wiki.ti.com/.../5x_CPSW
  • 0 in reply to Biser Gatchev-XID
    Intellectual 725 points

    Hi Biser,thanks very much,we can get like this:
    [ 1.692105] libphy: Fixed MDIO Bus: probed
    [ 1.748039] davinci_mdio 48485000.mdio: davinci mdio revision 1.6
    [ 1.754163] davinci_mdio 48485000.mdio: detected phy mask fffffffc
    [ 1.764286] libphy: 48485000.mdio: probed
    [ 1.768318] davinci_mdio 48485000.mdio: phy[0]: device 48485000.mdio:00, driY
    [ 1.778043] davinci_mdio 48485000.mdio: phy[1]: device 48485000.mdio:01, driY
    [ 1.788508] cpsw 48484000.ethernet: Detected MACID = c8:fd:19:bd:ed:60
    [ 1.795154] cpsw 48484000.ethernet: cpts: overflow check period 800
    [ 1.802145] cpsw 48484000.ethernet: cpsw: Detected MACID = c8:fd:19:bd:ed:61


    as previous discribed. before the sytem boot up,the board powered down by itself.

    it is so sorry,we forgot one thing that  the  GB_ETH1_INTn and GB_ETH0_INTn connected to the cpu gpio different with AM5728 IDK.We looked for the dra7.dtsi have some place to depict the mac: ethernet@48484000, but have no idea to modify the gpio to interrupt.

  • 0 in reply to yongjun wang
    TI__Mastermind 40690 points
    Hi,

    Looking at the AM572x-IDK schematic I see the that interrupts you are pointing out are connected to the AM5728. While the signals are hooked up I will need to check into if they are actually used.

    Best Regards,
    Schuyler
  • 0 in reply to Schuyler Patton
    Intellectual 725 points
    thanks Schuyler,we will expect your reply
  • 0 in reply to yongjun wang
    TI__Mastermind 40690 points
    Hi,

    The interrupts signals from the PHYs have not been tested. The key issue is that we do not know if the PHY driver in the Linux kernel source tree from the PHY manufacturer will support interrupts or not. Our group does not support PHY drivers. Only polling the PHYs over MDIO has been tested.

    The best next step I can recommend is mailing the maintainer of the KSZ9031 PHY about if the PHY driver they have in the kernel tree will work with interrupts and how to add the gpio line to the DTS of the PHY node.

    Best Regards,
    Schuyler
  • 0 in reply to Schuyler Patton
    Intellectual 725 points
    Hi Schuyler,we tested the gpio about the KSZ9031 to AM5728 is not used in the driver,now our net can work well , the previous net problem is the dts config wrong.thanks for your reply .