• State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 100 subscribers
  • Views 439 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.

IPNC v5.0.0 boot hang up

Justin_Liu
Intellectual 260 points

Hi all:

I have a issue when I booting the IPNC DM368 version 5.0.0  and it hangs up at Starting NAND Copy...

[Sat Sep 07 01:39:15.757 2013] DM36x initialization passed!
[Sat Sep 07 01:39:15.788 2013] TI UBL Base Version: 1.50 
[Sat Sep 07 01:39:15.788 2013] Boot Loader BootMode = NAND 
[Sat Sep 07 01:39:15.788 2013] Starting NAND Copy...

It happens irregular.

 

Any comments are appreciated.

Thanks.

 

Best Regards,

Justin

  • 0
    Guru 31670 points

    Justin,

    I have seen this before. I've not spent time debugging this as it was quite random. I believe it has something to do with the NAND timings. Can you try adding more logs inside the NAND driver of UBL and try to trace the exact place where it hangs. 

    Also you can try to make the NAND timings maximum.

  • 0 in reply to Renjith Thomas
    TI__Genius 11000 points

    Hello,

    Ths issue can be due to the fact that in UBL, you need to intialize PLL2 first before setting PLL1, this is important to avoid this issue.

    Can you please check that you are using UBL from IPNC RDK version 5.1? 

    please let us know your results after using this

    Regards,

    Raghu

  • 0 in reply to Raghu Kudva
    Intellectual 260 points

    Dear Raghu:

    Much appreciate for your reply.

     

    The UBL is from IPNC RDK version 5.1 Binaries and it happens random.

    Also, I check the PLL initial setting, the PLL2 is init before than PLL1 on IPNC version 5.1.0.

     

    ipnc5.1.0/Utils/flash_utils_dm36x/DM36x/Common/src/device.c

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

    WDT_FLAG_ON();

      // DDR PLL setup  

    if (status == E_PASS) status |= DEVICE_PLL2Init();

      // System PLL setup  

    if (status == E_PASS) status |= DEVICE_PLL1Init();

      // DDR2 module setup  

    if (status == E_PASS) status |= DEVICE_DDR2Init();

    #endif  

      // AEMIF Setup  

    if (status == E_PASS) status |= DEVICE_EMIFInit();

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

     

    Regards,

    Justin

  • 0 in reply to Justin_Liu
    TI__Genius 11000 points

    Hello,

    Can you try with these binaries and check if the issue still persists?

    Regards,

    Raghu

    0160.Desktop.zip

  • 0 in reply to Raghu Kudva
    Intellectual 260 points

    Hi Raghu:

     

    Yes, all of them will stop at Starting NAND Copy...

     

    ubl_297Mhz: it stops at the 3. 84. 219 times booting up.

    ubl_432Mhz: it stops at the 15. 31. 35 times booting up.

    ubl_486Mhz: it stops at the 116. 117. 211 times booting up.

     

    Regards,

    Justin

     

  • 0 in reply to Justin_Liu
    Intellectual 700 points

    We have observed the exact same issue and it is driving us crazy!

    Happens rarely, but in all versions of UBL and uBoot that we tried, including 5.0. We did not try 5.1 since the file dates in the UBL folder are the same and it looks like UBL has not changed between 5.0 and 5.1 

    A major customer of ours is complaining and if we can't fix this we are facing a recall of many deployed cameras. We have no idea how to debug this.

    In trying to reproduce this, we clicked the reset button on the debug board many times. Then the UBL occasionally hangs as well, but gets a little bit further and after "Starting NAND Copy...." there is the message "NAND Boot failed." and "Aborting..."

    We don't know if the repeated pressing of the reset button is showing a different failure from the case where a camera is powering up and hanging just after "Starting NAND Copy...." with no further message or sign of life.