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.

Adjacent corruption risk during flash write

Hello,

I'm working on a design with the LM4F where backup of non-volatile data to flash could be as high as 10k cycles. 

What is the risk of corrupting code memory if it is placed adjacent to this non-volatile backup region?  Should I place an unused buffer area between the code and backup regions?  If so, is a 1kb page sufficient, or do I need to worry about even/odd pages being adjacent and instead use a 2kb buffer?

This is a high volume design, so even very rare occurrences of code memory corruption would be disastrous.  I would prefer to stay on the safe side and keep a buffer; however, not having to burn that code space is also desirable.

Note that I am already aware of the errata on rev. A1 where erasing a 1kb page erases the adjacent page as well.  I am not concerned about this - only about the current and future revisions that would actually be used in production.

Thanks in advance.

  • Hi Ben,

    I'm sorry for the delay in responding to you.  I checked with product engineering and got the following response:

    This should not be a concern for production (B0) material. Part of the process qual included cycling to 100k+ cycles with check of disturb on non-cycled adjacent sectors, so from a technology standpoint we have proven the capability.

    So you don't need to worry about this issue.

    Regards,

    Sue