• State
  • Locked Locked
  • Replies 2 replies
  • Subscribers 82 subscribers
  • Views 1001 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.

f2812 with CCS4.1.3.00038 running SPRAB69 EEPROM Emulation example now behaves as if flash locked

N. Winoc
Prodigy 55 points


The last code I was able to successfully program onto this f2812 board & run was the EEPROM Emulation example code from SPRAB69. 

Somehow it appears as if all the flash sectors are now locked, even though I've never tried to use the CSM password features.  Attempting to unlock or erase the flash shows errors in CCS such as "C28xx: Flash Programmer: Error erasing flash memory. Device is locked or not connected. Operation cancelled" or "C28xx: Flash Programmer: Error unlocking flash memory. Device is still locked"

I am still able to successfully use the Example_281xCpuTimer, bootloaded from H0, so the card in general seems okay.

Is the card permanently locked or is there some hope of troubleshooting it?

Thank you,

N.

  • 0
    Prodigy 55 points

    Update to the above:

    Same thing happened again to a different f2812, again trying to learn from SPRAB69 (http://focus.ti.com/general/docs/lit/getliterature.tsp?baseLiteratureNumber=sprab69).  I've tried to narrow down possible causes but still don't see why the flash is getting locked.

    This time it was never unplugged/plugged from power during development, so it shouldn't have lost power during an erase cycle,

    jumper JP7 was set to position 2-3 the whole time (for it to boot from H0, not from flash, as we'd tried at some point with the first now-locked board)

    Bank_pointer was pointing to the start of flash sector J, and Page_Pointer was pointing to a location in flash sector J, so I seem to be loading the correct memory map,

    I've confirmed "Unlock_CSM();" is in the onReset() part of the f2812 .gel file used with the target configuration,

    and the linker given in the example appears to have already allocated memory for the CSM.  I haven't found any part of the code trying to write new passwords to the CSM registers or trying to lock it.

     

    From reading and asking around so far, our best guess is that if the program itself has somehow been written over where the CSM should be, we can:

    take the .out file from compiling/linking the program,

    open it in a hex editor,

    guess what the CSM passwords have become by going to whatever address of the program has the values that overwrote the CSM

    copypaste the values into the CSM unlocker in CCS.

    Does anyone have any ideas on whether that stunt would work, which address location we should scroll to to try, or any other ideas of how to unlock these boards?

     

    Thank you,

    N.

  • 0 in reply to N. Winoc
    TI__Intellectual 2575 points

    I got same issue without useful workaround.

    Does anybody know this?

    Thanks,

    Wei