• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 99 subscribers
  • Views 330 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.

Is this a ROM_BOOT bug?

Lihua Zhu
Intellectual 900 points

Hi,

I set the boot sequence of am335x board to boot[4..0]=00100, which means first UART, then MMC, and then NAND flash. But I can NOT start the board if I run reboot in the shell. It displays CCCCCCC and halt. after about 4 minute, it will restart again, and this time, the board can boot up. If I cut off the power directly, the board will boot up normally.

I inserted a MMC card trying to boot from it, the same thing happened, so it halted after it tried the UART.

Is this a bug of Boot Rom? Is there a solution or workaround?

Acturally I don't want to boot from UART, I want to try MMC first, if failed, then boot from NAND. but I can only find this boot configure that try MMC first and then NAND.

  • 0
    Intellectual 900 points

    anyone knows about this?

    Thanks

  • 0
    Guru 31670 points

    Zhu,

    Please read the AM335x TRM chapter 26 carefully. 00100b boot sequence is UART0, XIP,  MMC0 NAND. Also there is another bootmode where MMC0 is the first boot device-> 10111b. Have you tried this boot mode?

  • 0 in reply to Renjith Thomas
    Intellectual 900 points

    Hi Renjith,

    Thank you for the reply. Yes, you are correct, my boot wequence is UART0, XIP, MMC0, NAND, We don't have a XIP device in out board, so I omitted it. What puzzled me is why the board can boot up by power on but can't boot up by reboot from shell.

    I read the chapter, 10111 can NOT boot from NAND, that don't match my requirement. What I need is: if there is NOT a MMC card inserted, then boot from NAND (normal operation), else boot from MMC card ( recovery operation).

    Any other ideals? Thank you very much.

    Lihua

  • 0 in reply to Lihua Zhu
    TI__Guru* 86480 points

    Lihua, after a reboot, can you read the CONTROL_STATUS register (0x44e10040) with JTAG to see what the sysboot bits are.  It seems that the boot sequence is not what you are expecting.  Also, after a reboot, can you probe the MMC_CLK signal to see if the processor is attempting to boot from the SD card.

    Is this a custom board?

    Regards,

    James

  • 0 in reply to JJD
    Intellectual 900 points

    Hi James,

    I don't have a usable JTAG tool against Cortex A8 by now. I can't observe any clock on MMC_CLK after I reboot.

    Yes, this is a custom board. And I also tested with a board provided by a TI agency, It shows the same problem.

    Do you have a TI's evm board? Could you please set the boot sequence to 0b00100 and test for me?

    Thank you very much for your help and suggestion.

    Best regards,

    Lihua

  • 0 in reply to Lihua Zhu
    TI__Guru* 86480 points

    Lihua, i am not able to reproduce this on the TI EVM.  I have set the boot mode to 00100b, and it successfully reboots each time i issue the 'reboot' command in linux kernel.

    Regards,

    James

  • 0 in reply to JJD
    Intellectual 900 points

    Hi James,

    Thank you for the valuable information. Maybe it's a board specific problem.

    Regards,

    Lihua