• State Resolved
  • Locked Locked
  • Replies 10 replies
  • Subscribers 81 subscribers
  • Views 852 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.

F28M35H52C: Problem with Flash burn

GZeev
Intellectual 270 points
Part Number: F28M35H52C
Other Parts Discussed in Thread: CONTROLSUITE

Hi,

I had a strange phenomena.

The Concerto Flash burn process includes erase the Flash sector and then burn it.

We wrote a PC Burn program which performs the above through Master subsystem (M3 ARM) UART0 peripheral. Of course we wrote also supporting programs on the M3 and C28 subsytems.

The PC program interacts well also with your ROM Burn program, when the corresponding GPIOs are set.

On the Concerto we have two programs on each subsystem Flash memory: Application and Burn porogram. On power up the application is running. The Burn program is activating as result of PC request. Each program is located in seperate sector in the Flash memory.

Yesterday we wanted to burn a new version application software on M3. Immediately after the Flash was erased, we had an electricity break. After the electricity returned we failed to have contact with the Concerto even with your ROM burn program. We succeeded to connect to the Concerto only through the JTAG and we succedded to burn the Flash ONLY using a RAM version of our burn program. Your ROM Burn program failed to connect to our PC Software (of course with the relevant GPIOs set).

Later on I simulated the problem by:

1. Erasing the M3 relevant sectors of the application program, including sector N which holds the Reset address

2. Power off and on the board.

3. Try and fail to connect with your ROM Burn program using my PC program (of course with the relevant GPIOs set).

I also checked with JTAG that your ROM Burn process is running. It seems running because the UART0 registers were configured.

Any idea how to solve this problem?

Thanks,

Zeev

  • 0
    TI__Guru 55700 points
    Zeev,

    Can you confirm you have this problem only with the unit which had electrical failure when the flash was erased?

    You have to check how your custom ROM burn program. Without any knowledge about how it is written we can't provide any meaningful suggestion.

    Regards,
    Manoj
  • 0 in reply to Manoj Santha Mohan
    Intellectual 270 points

    Manoj,

    I checked only the "infected" board.

    I prefer to get permission from my boss before try to simulate this problem (of course without electricity break) on other boards.

    How can I check your ROM burn program?

    Anyway when I burn again the Flash using JTAG and CCS the problem dissapeared and I can use your ROM Burn program to boot from UART0 with relevant GPIOs set. Doesn't it mean that your ROM burn program is OK?

    I will provide the exact details of the Concerto device and all other required information later today.

    Regards,

    Zeev

  • 0 in reply to GZeev
    TI__Guru 55700 points
    GZeev,

    When you say "ROM burn program", do you mean BOOTROM code? If it is BOOTROM code, flash programming failure (or) electrical failure shouldn't change the BOOTROM contents.

    Regards,
    Manoj
  • 0 in reply to Manoj Santha Mohan
    Intellectual 270 points

    Manoj,

    I think you are absolutely correct. Therefore probably there is a bug in your BOOTROM code.

    Unfortunately I didn't get permission to simulate the problem on another board.

    I wanted to debug your BOOTROM code, but I didn't find the source code in your controlSUITE of this device (Why? I found BOOTROM code of all other devices we have). Can you provide me the source code and its compiler version?

    Here are my device details:

    Text on the device:

    980

    F28M35H52CIREPT

    YFB - 36C7SPW

    G4

    Although unrelevant to your BOOTROM code, here are details of our IDE:

    CCS ver 6.0.0.00190

    M3 compiler version: 5.1.5

    Regards,

    Zeev

  • 0 in reply to GZeev
    TI__Guru 55700 points
    GZeev,

    I never said BOOTROM is defective. I just wanted clarification from you that "ROM burn program" means BOOTROM and not anything else.

    Have you compared the BOOTROM contents on defective device with other known good device? If so, BOOTROM contents match, then I wouldn't suspect BOOTROM.

    Regards,
    Manoj
  • 0 in reply to Manoj Santha Mohan
    Intellectual 270 points

    Manoj,

    I have compared the BOOTROM contents on defective device with other known good device. The contents are identical (Addresses 0-0xFFFF).

    Can you provide me the source code of this BOOTROM software?

    Regards,

    Zeev

  • 0 in reply to GZeev
    TI__Guru 55700 points

    Zeev,

    If the bootrom contents match, I wouldn't suspect BOOTROM problem. Most likely it should be some setup problem from your side. I don't know what you will debug in BOOTROm code.

    I will check with software folks in my team about source code and get back with you.

    Regards,
    Manoj

  • 0 in reply to Manoj Santha Mohan
    TI__Guru 55700 points
    Zeev,

    Looks like we can't provide BOOTROM source code without NDA in place.

    Regards,
    Manoj
  • 0 in reply to Manoj Santha Mohan
    Intellectual 270 points

    Manoj,

    We decided not to deal with this problem. This means that we will send this board to a costumer.

    Let's hope we will not face such a problem on other boards.

    Regards,

    Zeev

  • 0 in reply to Manoj Santha Mohan
    Intellectual 270 points
    Be more aware to your costumers requests