• State Resolved
  • Locked Locked
  • Replies 14 replies
  • Answers 1 answer
  • Subscribers 97 subscribers
  • Views 3278 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.

AM4372: Boot mode question

Sh.Sasaki
Guru 13625 points

Part Number: AM4372
Other Parts Discussed in Thread: TS5A3157,

Hi all,

I'd like to know about AM437x's boot mode.

My customer has selected the following boot mode.

Then, the application program is stored in the FLASH connected to the QSPI.

When power is input to the board in this state, it starts normally with QSPI boot.

However, when USB0 is connected in this state, QSPI boot will not be done.

Is this correct as the behavior of the device?

When the device is in this mode, it will boot with QSPI --> USB1 --> MMC0 --> USB0,

I think that QSPI boot will be done even if my customer connected USB0.

Best regards,
Sasaki

  • 0
    TI__Guru**** 393215 points
    Hi,

    Your understanding is correct. QSPI boot should be done in all cases. Have you checked all SYSBOOT pins to ensure that they are at the required levels? Can you post the SYSBOOT settings on the board?
  • 0
    TI__Mastermind 27890 points

    Sasaki-san,

    Yes, QSPI boot should work regardless of whether or not there is a USB device present.

    I ran this issue by our ROM expert and he suggested that perhaps the QSPI isn't ready on the ROM's first pass through the boot order and is latching the USB device instead. In the absence of a USB device, the QPSI will successfully boot on the second pass. This does assume that there is a boot image on the USB device.

    Could you measure the QSPI boot time in the passing case?

  • 0 in reply to -DK-
    Guru 13625 points

    Hi DK-san,

    Thank you for your reply.

    -DK- said:

    Could you measure the QSPI boot time in the passing case?

    Boot time is about 2.5 seconds.
    In this case, is not QSPI ready?
    Best regards
    Sasaki
  • 0 in reply to Sh.Sasaki
    TI__Mastermind 27890 points
    Hello Sasaki-san,
    This seems about right for QSPI boot. We will need to look at this internally. This thread will be updated with results.
  • 0 in reply to -DK-
    Guru 13625 points
    Hi DK-san,

    Thank you for your support.
    What's the status now?

    Best regards,
    Sasaki
  • 0 in reply to Sh.Sasaki
    TI__Guru* 96461 points
    Sasaki,
    can you try a cold or warm reset with USB0 connected? Does the boot work with QSPI in this case?

    Also, when the processor fails to boot, what is the contents of 0x44E10040 (CONTROL_STATUS) register?

    Regards,
    James
  • 0 in reply to JJD
    Guru 13625 points

    Hi James-san,

    Thank you for your reply.

    JJD said:

    can you try a cold or warm reset with USB0 connected? Does the boot work with QSPI in this case?
     

    QSPI boot does not work even if the device is reset.

    JJD said:

    Also, when the processor fails to boot, what is the contents of 0x44E10040 (CONTROL_STATUS) register?

    JTAG connection is impossible because the following error is output when the device is USB0 connected.

    So I can not read this register.

    By the way, in this state, when I reconnect JTAG USB cable and do "Launch Selected Configuration", the screen changes to the error below.

    Best regards,

    Sasaki

  • 0 in reply to Sh.Sasaki
    TI__Guru* 96461 points
    Hi Sasaki, that is strange, it should skip over USB0 boot. What are you connecting to USB0?
    Also, since you have JTAG issues, try reading the register after a successful QSPI boot

    Thanks,
    James
  • 0 in reply to JJD
    Guru 13625 points

    Hi James-san,

    Thnak you for your support.

    JJD said:

    What are you connecting to USB0?
    Also, since you have JTAG issues, try reading the register after a successful QSPI boot

    USB0 is connected to USB port of PC.

    Also, the register when QSPI boot succeeded is as follows.

    Best regards,

    Sasaki

  • 0 in reply to Sh.Sasaki
    Guru 13625 points

    Hi James-san,

    I have additional information.

    The power supply sequence may be incorrect.

    Let me know about the followings.

    If the USB is connected before the main power supply is on, the following power sequence occurs:

    Electrical concerns

    1.   IC15 "ts5a3157" because the voltage is loaded to the pin 1 before the power is hi, do not you have any problems such as latch-up?

    2.   The voltage is loaded into the U23 pin before starting the am4372. Is this the cause of the problem? ）

    Best regards,

    Sasaki

  • 0 in reply to Sh.Sasaki
    TI__Guru* 96461 points
    Hi Sasaki,
    VBUS is a fail-safe pin, so applying voltage to this pin out of sequence is fine.

    If possible, see if you can get a trace of the QSPI bus immediately following a power-on. There should be activity before it ever get to the other boot sources. Is there a QSPI reset signal that is preventing initial boot? Might want to check that signal relative to the QSPI clock.

    Regards,
    James
  • 0 in reply to JJD
    Guru 13625 points

    Hi JJD-san,

    Thank you for your support.

    I will confirm with my customer.

    I'll report again.

    Best regards,

    Sasaki

  • 0 in reply to Sh.Sasaki
    TI__Guru* 96461 points
    Sasaki, any resolution from the customer here?

    Thanks,
    james
  • 0 in reply to JJD
    Guru 13625 points
    Hi James-san,

    sorry for the late reply.

    Because there is no contact from the customer, the problem may have closed.

    Therefore, I close this thread once.

    Thanks and regards,
    Sasaki