• State Resolved
  • Locked Locked
  • Replies 32 replies
  • Subscribers 105 subscribers
  • Views 1025 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.

SK-AM64B: SK-AM64B with prebuild wic image SDK8.6 cannot boot?

Rio Chan
Genius 14325 points
Part Number: SK-AM64B
Other Parts Discussed in Thread: LM61460

Champs:

With the SDK8.6 WIC prebuilt image, the SK-AM64B cannot boot.

Any solution here?

BR Rio

  • 0
    TI__Guru**** 155911 points

    Rio,

    Rio Chan said:
    the SK-AM64B cannot boot.

    Please provide details, do you see some log message on the UART console or nothing at all? Have you tried the previous SDK8.5 and it fails the same?

  • 0 in reply to Bin Liu
    TI__Genius 14325 points

    Hi Bin:

    #1. I have pre-build for HS-FS image, based on SDK8.4.

         The SK-AM64B EVM we got have 2 versions: M1/ M2.

         M1 will complete booted okay, this is for sure.

         M2 evm will give us the attached fail log.

     [ 1.238325] sdhci-am654 fa10000.mmc: card claims to support voltages below defined range
    [ 1.250392] mmc1: new SDIO card at address 0001
    [ 1.274506] mmc0: error -110 whilst initialising SD card
    [ 1.521325] mmc0: error -110 whilst initialising SD card
    [ 1.763062] mmc0: error -110 whilst initialising SD card
    [ 2.029971] mmc0: error -110 whilst initialising SD card

    2023_5_3_SDK8.4_Tested_On_M2_SKAM64B_SDCard_Fail_due_to_Voltage

    #2. If testing the SDK8.6 WIC prebuild image.

          M1/ M2 will give us no any logs show on the console.

    So, there are 2 issues:

          Issue 1: SD card voltage failed on the M2 EVM.

          Issue 2: SDK8.6 cannot booted on M1 / M2 EVM. (IE: No any logs shown on the console.

    Question:

          What's the diff from M1/M2 EVM?

          How to fix the SDK8.6?

    BR Rio

  • 0 in reply to Rio Chan
    TI__Guru**** 155911 points

    Rio,

    I am routing your query to our U-Boot expert for comments.

  • 0 in reply to Bin Liu
    TI__Genius 14325 points

    Hi Bin:

    We have some customers are using AM64, so, this shall be fixed asap.

    Please assign the right owner to deal with.

    Thanks.

    BR Rio

  • 0 in reply to Rio Chan
    TI__Guru* 82595 points

    Rio

    This is still being debugged. We will keep you posted.

    Regards

    Mukul 

  • 0 in reply to Mukul Bhatnagar
    TI__Guru 65135 points

    Hello Rio,
    I did sanity run on SK-AM64B M1/M2 boards with AM64x SDK 8.6 release package (WIC image).
    Linux boots on both M1/M2 boards, and I'm attaching two log files for your reference.
    Best,
    -Hong

    am64B-M1_8.6_sd_boot.log
    Fullscreen
    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    U-Boot SPL 2021.01-g2ee8efd654 (Feb 27 2023 - 13:50:10 +0000)
    EEPROM not available at 80, trying to read at 81
    Resetting on cold boot to workaround ErrataID:i2331
    resetting ...
    U-Boot SPL 2021.01-g2ee8efd654 (Feb 27 2023 - 13:50:10 +0000)
    EEPROM not available at 80, trying to read at 81
    SYSFW ABI: 3.1 (firmware rev 0x0008 '8.6.4--v08.06.04 (Chill Capybar')
    SPL initial stack usage: 13424 bytes
    Trying to boot from MMC2
    Authentication passed
    Authentication passed
    Authentication passed
    Authentication passed
    Starting ATF on ARM64 core...
    NOTICE: BL31: v2.8(release):v2.8-226-g2fcd408bb3-dirty
    NOTICE: BL31: Built : 13:45:56, Feb 27 2023
    I/TC:
    XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

    am64B-M2_8.6_sd_boot.log
    Fullscreen
    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    U-Boot SPL 2021.01-g2ee8efd654 (Feb 27 2023 - 13:50:10 +0000)
    EEPROM not available at 80, trying to read at 81
    Resetting on cold boot to workaround ErrataID:i2331
    resetting ...
    U-Boot SPL 2021.01-g2ee8efd654 (Feb 27 2023 - 13:50:10 +0000)
    EEPROM not available at 80, trying to read at 81
    SYSFW ABI: 3.1 (firmware rev 0x0008 '8.6.4--v08.06.04 (Chill Capybar')
    SPL initial stack usage: 13424 bytes
    Trying to boot from MMC2
    Authentication passed
    Authentication passed
    Authentication passed
    Authentication passed
    Starting ATF on ARM64 core...
    NOTICE: BL31: v2.8(release):v2.8-226-g2fcd408bb3-dirty
    NOTICE: BL31: Built : 13:45:56, Feb 27 2023
    I/TC:
    XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

  • 0 in reply to Hong Guan64
    TI__Genius 14325 points

    Hi Hong:

    Please upload your image to TI Drive. + DIP switch settings.

    So, we can test it.

    BR Rio

  • 0 in reply to Rio Chan
    TI__Expert 8400 points

    Hello Rio,

    Can you verify if you have a GP or HS-FS device on the board that will not boot?

    The instructions to figure out device type can be found here: software-dl.ti.com/.../Foundational_Components_Migration_Guide.html

    regards

    Judith

  • 0 in reply to Judith Mendez
    TI__Guru 65135 points

    Hello Rio,
    1/. I used the wic image from SDK8.6 release link
    dr-download.ti.com/.../tisdk-default-image-am64xx-evm.wic.xz
    2/. Refer to this link on BOOTMODE pin, where the link is for AM64x-SK, but BOOTMODE pin apply to SK-AM64B (M1/M2)
    dev.ti.com/.../am64x_skevm_quick_start_guide.html
    Best,
    -Hong

  • 0 in reply to Judith Mendez
    TI__Genius 14325 points

    Surely I'm using the HS-FS EVM, that's why I mentioned the M1/M2 in the beginning.

  • 0 in reply to Rio Chan
    Intellectual 602 points
    Rio Chan said:
    Surely I'm using the HS-FS EVM, that's why I mentioned the M1/M2 in the beginning.

    This has been an ongoing chronic issue for at least a couple of months that has not been resolved.

    Just did as Hong said, I was under the impression that the default image might have been updated.

    Both our boards are dead.

    Only activity is a red led on the bottom side of the board.

    The ti3boot.bin was also changed per the docs.

    It does not send a string out the UART port when it is UART boot mode.

    Think about this, if the device is secure why would it send out a string so a key can be made from that string....Hence, its not secure. So docs are not correct or some how the boards are secured and cannot be unlocked. Were these boards even tested, were they secured by whomever made the board.

    I have 2 boards so probability of both of those random events having the same outcome is extremely low. So the board is dead for some unknown reason.

    Not sure if the docs are even correct, someplace you will see the red color board and someone did a photoshop on the boot switch settings.

    So if they were locked during testing, they are scrap.

    I see that you are an employee, can you bump this up to the next level. We are not getting any place on this forum. I have had several very helpful TI staff members get on this but it appears they are not getting the information that is needed to resolve the issue. One even sent a replacement board and it is dead too.

  • 0 in reply to Fred Fisher
    TI__Expert 8400 points

    Hello Fred,

    All boards will print out a string while in UART boot mode. If it does not, then the boot mode switch setting is wrong or another issue like the board is dead.

  • 0 in reply to Fred Fisher
    TI__Mastermind 25760 points
    Fred Fisher said:

    Both our boards are dead.

    Only activity is a red led on the bottom side of the board.

    I'm running several AM64x SK's regularly including SK-AM64B HS-FS. I don't see a red LED on the back side on on any of them. Looking in the SK users guide I see:

    And skimming through the schematics looks like LD15 is what you have red. It does look like the user's guide does not show where this LED physically is, appologies for that.

    So I'd suggest to use another USB-C power supply. I don't have an underpowered USB-C to try out to see if the red LED goes on. I tried a "badly" connected USB-C power cable and I saw the red LED LD15 go on.

      Pekka

  • 0 in reply to Pekka Varis
    Intellectual 602 points
    Pekka Varis said:
    So I'd suggest to use another USB-C power supply.

    What are you talking about, it is on a lab grade bench power supply. Same one that runs all the other boards we are testing. Then the boards are actually dead, now that we know for fact that is a power fault indicated by the LED.

  • 0 in reply to Pekka Varis
    Intellectual 602 points

    Hello Pekka,

    Do you know when the next lot of boards will become available? Date code on these are from 9/22/2022. When you get in a new batch I will reorder, no point in me re-ordering from this lot.

    Regards,

    Fred

  • 0 in reply to Fred Fisher
    TI__Mastermind 25760 points

    If the red LD15 is on it means the board power supply is saying power is not good and the board will not boot. So either the power supply is not good or the board is broken. Red LD15 on means the board will not boot.

      Pekka

  • 0 in reply to Pekka Varis
    Intellectual 602 points

    Hello Pekka,

    I can say with 100% certainty the the power supply is not broken, fully metered........

    That leaves the board.

    I will order another board however don't want a one from this current production run. Seems that several others on this forum along with one of my online friends has a bad SK-AM64B.

    Who can I contact regarding the arrival date of the next run of boards?

    If someone at TI would test it before it ships would be okay too.

  • 0 in reply to Fred Fisher
    TI__Genius 14325 points

    Hi Fisher:

    May you help me to test this image from Hong on M2?

    https://dr-download.ti.com/software-development/software-development-kit-sdk/MD-yXgchBCk98/08.06.00.42/tisdk-default-image-am64xx-evm.wic.xz

    Just want to see if you have the same fail result as mine.

    Thanks.

    BR Rio

  • 0 in reply to Rio Chan
    Prodigy 30 points

    Hello Rio Chan,

    I have also tested the 08.06.00.42 image from Hong on the M2.I confirm that I have the exact same failure that you have:

    Fullscreen
    1
    2
    3
    4
    5
    6
    7
    [ 1.530730] Waiting for root device PARTUUID=47b6d3e0-02…
    [ 1.539742] sdhci-am654 fa10000.mmc: card claims to support voltages below defined range
    [ 1.559217] mmc0: new SDIO card at address 0001
    [ 1.584120] mmc1: error -110 whilst initialising SD card
    [ 1.835602] mmc1: error -110 whilst initialising SD card
    [ 2.081288] mmc1: error -110 whilst initialising SD card
    [ 2.333152] mmc1: error -110 whilst initialising SD card
    XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

    I hope that TI can resolve the issue soon.

  • 0 in reply to Frederick Eckert
    TI__Guru* 82595 points

    Hello Fredrick , Rio, Fred

    Can you share the serial # of your board (it is on the bottom side of the board)

    Additionally can you clarify what brand SD card you are using and if you have tried to use a different SD card to see if it makes a difference. 

    With Rio and Fredrick I believe the problem is different , compared to Fred who seems to have DOA boards with just the red LED on LD15 (Unlike Rio)

    Fredrick

    Please clarify the status of the LED that is being talked about in this thread. My assumption is that you are not seeing that.


    Regards

    Mukul 

  • 0 in reply to Mukul Bhatnagar
    Prodigy 30 points

    Hello Mukul,

    My LED status is good. My board does not have the red LED on. I agree that Fred has different problem from Rio and myself.

    Board Info:

    SL# 37221000259
    ASSY# PROC100A(002)

    I am using a known good SanDisk Ultra 32GB. I did not try any other SD cards. If you have a TI certified SD card, please let me know and I will purchase one.

    boot log:

    Fullscreen
    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    20
    21
    U-Boot SPL 2021.01-g2ee8efd654 (Feb 27 2023 - 13:50:10 +0000)
    EEPROM not available at 80, trying to read at 81
    Resetting on cold boot to workaround ErrataID:i2331
    resetting ...
    U-Boot SPL 2021.01-g2ee8efd654 (Feb 27 2023 - 13:50:10 +0000)
    EEPROM not available at 80, trying to read at 81
    SYSFW ABI: 3.1 (firmware rev 0x0008 '8.6.4--v08.06.04 (Chill Capybar')
    SPL initial stack usage: 13424 bytes
    Trying to boot from MMC2
    Authentication passed
    Authentication passed
    Authentication passed
    Authentication passed
    Starting ATF on ARM64 core...
    NOTICE: BL31: v2.8(release):v2.8-226-g2fcd408bb3-dirty
    NOTICE: BL31: Built : 13:45:56, Feb 27 2023
    I/TC:
    I/TC: OP-TEE version: 3.20.0 (gcc version 9.2.1 20191025 (GNU Toolchain for the A-profile Architecture 9.2-2019.12 (arm-9.10))) #1 Mon Feb 27 13:46:53 UTC 2023 aarch64
    I/TC: WARNING: This OP-TEE configuration might be insecure!
    XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

    Best regards,

  • 0 in reply to Rio Chan
    Intellectual 602 points

    Hello Rio,

    Just burned the image and tested, red light is on.

  • 0 in reply to Fred Fisher
    TI__Genius 14325 points

    Hi Fred:

    What you mean the "red light is on?

    Are you saying the booting is fail?

    BR Rio

  • 0 in reply to Mukul Bhatnagar
    Prodigy 30 points

    Hello Mukul and All,

    I was able to boot the board using a very old SanDisk SDHC 16GB SD card.

    boot log works:

    Fullscreen
    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    20
    21
    U-Boot SPL 2021.01-g2ee8efd654 (Feb 27 2023 - 13:50:10 +0000)
    EEPROM not available at 80, trying to read at 81
    Resetting on cold boot to workaround ErrataID:i2331
    resetting ...
    U-Boot SPL 2021.01-g2ee8efd654 (Feb 27 2023 - 13:50:10 +0000)
    EEPROM not available at 80, trying to read at 81
    SYSFW ABI: 3.1 (firmware rev 0x0008 '8.6.4--v08.06.04 (Chill Capybar')
    SPL initial stack usage: 13424 bytes
    Trying to boot from MMC2
    Authentication passed
    Authentication passed
    Authentication passed
    Authentication passed
    Starting ATF on ARM64 core...
    NOTICE: BL31: v2.8(release):v2.8-226-g2fcd408bb3-dirty
    NOTICE: BL31: Built : 13:45:56, Feb 27 2023
    I/TC:
    I/TC: OP-TEE version: 3.20.0 (gcc version 9.2.1 20191025 (GNU Toolchain for the A-profile Architecture 9.2-2019.12 (arm-9.10))) #1 Mon Feb 27 13:46:53 UTC 2023 aarch64
    I/TC: WARNING: This OP-TEE configuration might be insecure!
    XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

    Is there a recommended modern SD card?

    Best regards,

  • 0 in reply to Rio Chan
    TI__Mastermind 25760 points
    Rio Chan said:
    What you mean the "red light is on?

    There is clearly at least 2 issues here (power supply, SD card , maybe more), unfortunately confusing. We know from the user's guide and schematics if the red LD15 (bottom side of the SK) is on, there is no use trying further different SD cards.  

    So to isolate the issues, if you see the red LD15 on, double check with another >15W USB-C charger, and if the LED is still on. Assume the SK power supply side is broken, LM61460_EN won't go on and the board does not get power.

      Pekka

  • 0 in reply to Frederick Eckert
    TI__Guru* 82595 points

    Hello Fredrick

    Thanks for confirming that a different SD card does work for you. 

    We expect this to work with most known brand SD cards that follow the specifications, so we may still need to spend time assessing why we are seeing variability. 

    For customers in US region, you can see the following links , these are typically some of the SD cards that we are typically using

    https://www.amazon.com/Sandisk-Ultra-Micro-UHS-I-Adapter/dp/B073K14CVB/ref=sr_1_1?crid=3OVBVGRG0M0SE&keywords=micro+sd+card&qid=1684242513&refinements=p_89%3ASanDisk%2Cp_n_feature_two_browse-bin%3A6518303011&rnid=6518301011&s=pc&sprefix=micro+sd+card%2Celectronics%2C118&sr=1-1

    https://www.amazon.com/SanDisk-Ultra-SDSQUNS-016G-GN3MN-UHS-I-microSDHC/dp/B074B4P7KD/ref=sr_1_3?crid=3OVBVGRG0M0SE&keywords=micro+sd+card&qid=1684242513&refinements=p_89%3ASanDisk%2Cp_n_feature_two_browse-bin%3A6518303011&rnid=6518301011&s=pc&sprefix=micro+sd+card%2Celectronics%2C118&sr=1-3

  • 0 in reply to Pekka Varis
    Intellectual 602 points
    Pekka Varis said:
    double check with another >15W USB-C charger, and if the LED is still on.

    Hello Pekka,

    Yes, that has been done several times along with different cables. When I can visually see the lab power supply indicating around 10 ma. I know the circuit is not broken. We cut the end off the USB cables and connect the cut end directly to the power supply while we are in the testing and design phase.

  • 0 in reply to Fred Fisher
    TI__Genius 14325 points

    All,

    As I remember AM64 has no supporting for the HS400 SD card.

    Not sure if this is the cause.

    Waiting TI RnD to confirm if this mastters.

    BR Rio

  • 0 in reply to Rio Chan
    TI__Guru 65135 points

    Hello Rio,
    It is good to know you’re able to boot from the swapped SD card on SK-64B boards (M1/M2) from our email exchanges.
    The SD discrepancy issue as observed are tracked and being debugged internally.In meantime I’d recommend to continue using the working SD card for your work. We’ll keep you posted once the issue is root-caused with the fix.

    HS400 is not for SD, but for eMMC as defined in the dtsi file.
    git.ti.com/.../k3-am64-main.dtsi

    I’ll move the e2e to “resolved” if you don’t have any follow-up questions.

    Best,
    -Hong

  • 0 in reply to Hong Guan64
    TI__Genius 14325 points

    Thanks, I will close this.