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.

Uncorrectable flash ECC error from a RTI irq not involving flash on a RM48HDK

Other Parts Discussed in Thread: HALCOGEN, NOWECC

Hi,

I'm using the RM48HDK dev. kit with HALcoGEN generated code.

When the RTI generates an irq (compare0), ESM.Stat3 becomes 0x00000080. Often, ESM.Stat1 becomes 0x00000040, but not always. The nError pin is driven low and the nError LED is turned on. It clearly occurs when the interrupt is fired, since the interrupt period is 10 s, after which the NHET turns on a LED. But even with an empty ISR, the nError pins becomes low and the nError LED turns on, so it cannot be the ISR.

According to spns177, this means flash ECC errors, of which the 0x80 in group 3 is uncorrectable. However, the ISR seems to not involve flash. Does this mean my board is toast?

Thanks in advance!

Regards,

Tim

 

 

 

  • Tim,

    Your device initialization is done via HalCoGen, so I will assume all register init is done correctly.

    If you have an ECC error, that means you enable ECC checking. Is that correct?

    Are you using NowEcc to generate the ECC data corresponding to the code you flash in the device?

    In CCS, the compiler/linker generate an OUT file that does not contain ECC. There is an extra step to be done using NowECC.

    Please can you clarify?
    Also, if you can share your project, I can make a try on my environment and see what is going on.

  • Correct, ECC checking was enabled on flash memory. I did not generate the separate ECC file, because I assumed that either CSS would generate it automatically, or that the HALcoGEN code would recognize the absence of the ECC code and prevent complains from the ESM. Disabling flash ECC solved the problem. I'll enable it again later on in the project. Thanks for filling in my blind spot!