• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 79 subscribers
  • Views 449 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.

MSP430 information memory

venkatesh b
Prodigy 85 points

Hi,

Am using information memory to store Hardware Part number.

Issue:

In test, System restarted for 1000 times.At Each restart system will read Hardware part number of module and if it is matching ,system continues to work else logs Error message.Out of 1000 times 970 times part number came correctly,last 30 times continually Part number came as invalid which is '------' but software gives the part number as '-------' if CRC is not matched or if it doesn't contain any Part number.After issue is observed  using MSP-FET430UIF memory is read and observed information memory has valid Hardware part number.System is running with AC.

I have no Idea what is causing this behavior.Issue also not occurred later times.

Please let me know the possible causes for this behavior.

Regards,

Venkatesh

  • 0
    Expert 1400 points

    Hi Venkatesh,

    There are plenty of reasons why the data read from flash may be corrupt (not sure that they necessarily apply to the MSP430), hence why there are CRCs. I admit it sounds suspicious, but before we go pointing fingers at the hardware, can we take a look at your code to see if there are any latent bugs that could potentially cause this behaviour?

    You may also want to take a look at the following post: e2e.ti.com/.../92849

  • 0 in reply to Chris Karaplis
    Prodigy 85 points
    Hi Chris,

    Brief Intro to Product:
    The whole system is a unity of Multiple boards along with Display unit Board.When ever system turns on Display Unit reads the part numbers from the all boards and does compatibility check with the part numbers it has.If any part number is not matched it logs error.The board which gave version as '-------' is a Power management boards which sources the supply to all modules in the system including Display Unit.

    Brief Intro to Test:
    We perform Reboot test, During this test Power management board will be always ON , Board output which is supply to all other boards in system will be turned ON and OFF with interval of 2 minutes.


    We had software reviews ,we are pretty confident on software.This software is running with different hardware versions without any issue from 2 years.This issue is occurred in only one board.We have observed RAM self Test failure in same board Twice in other Reboot test.Both issues are not reproducing in later time.

    Brief of RAM failure Issue:
    We have seen RAM test failure also (2 times out of >20,000 reboot cycles) in the same power subsystem board during reboot testing. Below is the description of the RAM test that software is performing
    1) Walking ones and walking zero test
    2) writing 0xFFFF and 0x0000 and reading back each

    Before performing the above test, the software disables the global interrupts and store the RAM content into CPU register. Once after the above test, the software restore the data from CPU register and then enable the global interrupts. The software performing the above test only two bytes of RAM at each seconds. So it may take few minutes to finished the complete RAM test.

    please let me know your opinion on these two issues.


    Regards,
    Venkatesh
  • 0 in reply to venkatesh b
    Expert 1400 points

    Hi Venkatesh,

    Sorry for the delayed response. It seems suspicious to me that you are getting RAM failures on that one board. If you are confident in your software and specifically your test its not impossible that the part itself is defective. An intermittent problem like you are are seeing could be taken as a board failure, and the board should probably be replaced. Depending on how your code is written, it is entirely possible that your problem above is related to this RAM issue. If you are storing the value from flash on the stack and a bit gets flipped, it makes sense that you would get an invalid result. Does the flash issue only occur on the same board that has the RAM issue?

**Attention** This is a public forum