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.

CCS/RM48L952: Flash ECC Errors when compiling with TI 18.4.1 LTS compiler, not present with 16.9.x compiler versions

Part Number: RM48L952


Tool/software: Code Composer Studio

I raised this issue previously on another thread

Upon installing the latest CCS v8.x to a new laptop and rebuilding my application from CCS v7.x I started to get Flash ECC errors whenever I wrote to a RAM variable through the debug server interface. This only occurs when I build with the latest compiler that ships with v8.x; install and specify the earlier CCS v7.x packaged compilers and my application runs OK. We are at production release so I will continue to just use the earlier compiler. But in parallel I need to identify the root cause.

Previous version was TI 16.9.0 LTS, I installed 16.9.10 and it works well. Version 18.1.4 throws the Flash ECC errors.

Thanks in advance for paths to identifying root cause. Any known differences in the compiler would be of interest. 

  • Jamie,

    Jamie Wardlaw said:
    Upon installing the latest CCS v8.x to a new laptop and rebuilding my application from CCS v7.x I started to get Flash ECC errors whenever I wrote to a RAM variable through the debug server interface.

    Could you provide a screenshot of the error displayed by CCS?

    Jamie Wardlaw said:
    Any known differences in the compiler would be of interest. 

    New features and bug fixes included in a particular compiler release are documented in the README.txt and Closed_defects.html files included in the compiler directory. You may review those to get a sense for what has changed. However, if there is a new bug introduced here we would need a reproducible test case to investigate the issue further. Are you able to share a simple, cut-down version of your project that reproduces the issue?

  • Jamie,

    I just wanted to follow up to check if you are able to send us a small reproducible test case that demonstrates this issue? If not, we will close out the thread for now, but you are always welcome to come back to it at a later time.