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.

MSP430F5437A: Some of the RAM locations filled with 3fff

Part Number: MSP430F5437A

Hi,

I am using MSP430F5437A for one of the project. I observed that some of RAM memory locations filled with "3fff". Example RAM address 0x00000184 filled with "3fff". Address 0x00000182 used for SYSBSLC(SFRs) and 0x00000186 used for SYSJMBC.

If 0x00000184 is unused it should filled with ffff(16 bit data) and it should not filled with 3fff. Is my understanding correct? if so could you please let me know the reason for unused RAM locations filled with "3fff"

Thanks,

Gourav

  • Hi Gourav,

    This is should related with CPU instruction repertoire, so in the undefined RAM area, it will be initialized to 0x3fff.

    Best Regards

    Johnson

  • Thanks Johnson for quick reply. Actually we have one global variable stored at RAM address 0x00001CCC. we are using same variable for our calculations in the project. Some times it is getting filled with "3fff" leading wrong calculations the project.

    We will look into the code for possible corruption and used RAM address locations filled with "3FFF"

    Thanks,

    Gourav  

  • Hi Gourav,

    Could you provide some deatils about your issue?  "Some times it is getting filled with "3fff" leading wrong calculations the project."

     

    Best Regards

    Johnson

  • Hi Johnson,

    In my project collecting, ADC counts as samples from ADC12MEM0, ADC12MEM1 and ADC12MEM2 for every 521 micro sec. and selecting one as best sample among those based on threshold conditions we kept. For that best sample calculating square current(in Amps) and following same thing for 32 samples(treating one cycle) and the end of cycle calculating mean square of all 32 samples and stored at 1CCC RAM location using one global variable.

    We observed that global variable holding 3FFF some times actually should not possible. We simulated in the firmware to verify the possibility of mean square of all 32 samples results 3FFF. If all ADC samples(32 samples) filled with 4F3F then there is the scope for 3FFF but we are using 12 bit ADC and maximum possible ADC12MEM registers should be FFF(4095).

    Thanks,

    Gourav   

  • Hi Gourav,

    I'm sorry I can't clearly understand the issue you encountered.
    It seems that you use ADC module to captureonly some data and then calculations, so I think any variables in this process can be observed in Debug mode.

    Do you observe some parameter is error?

    Best Regards

    Johnson

**Attention** This is a public forum