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 processor miss-identification when going into debugger

Other Parts Discussed in Thread: MSP430FR5738, MSP430FR5739, MSP430F5529

I have a system working using CCS v5.50, with the MSP430FR5738 (28 pin chip), and the MSP-FET430UIF (fw v3.04.00.020).  In this system, we are only using pins 1, 4, 7, & 9 of the JTAG connector.

A recent update to the EVM switched to the MSP430FR5739 (40 pin chip) and I updated the sources and configuration appropriately.  The new code compiles & links OK, but when I try writing the code to the EVM in DEBUG mode (F11),  Code Composer complains with a message box that says:

TI MPS430 USB1_0/MSP430

Error connecting to the target:
The target setup (MSP430F5529) does not match the actual target type (MSP430FR5739)

 

I can't seem to find  "5529" in any of the configuration files or my sources, so I don't know why it is doing this.

If I ignore this and let it write the code, it doesn't error out and the code written seems to be at least partly working (I may have some bugs in the 5738 ->5739 transition), so I think the JTAG connections are OK.  I also verified that I can write the firmware using the   FET-Pro430 (Lite) program (from Elpotronic) that also uses the MSP-FET430UIF pod (but I had to downgrade the firmware in the unit to get it to work).

Although the code update seems to work, by just ignoring the error message, the CC debugger is otherwise not working right.  If I try to single step the code, it doesn't seem to be working right.

Any suggestions on how to "fix" this (i.e to correctly identify the processor and get the debugger to work right)?

 

 

 

  • Hi Bruce,

    Check your cmd file settings. There's a possibility that by default CCS might've linked MSP430FR5739's cmd file. So check in your project properties and project tree.

    Regards,

    Gautam

  • Hi Bruce,

    make sure you changed any project settings, target configuration file, and header files.   Also try reindexing.  Right click on the project to do that.

    Please keep us informed.  

    Best Regards,

    Lisa

  • Gautam:

     

    The linker CMD file looks OK and has no number like 5529 in it.  Also, the compiler and linker command line data (as shown in some of the properties windows) show it using the FR5739.  Furthermore, if I connect the older EVM, with the FR5738, it does recognize this as a FR5738 and complains that it does not match the  FR5739 that the code was built with. 

    So it would "seem" that there is some failure in the process of recognizing the FR5739.  This is both a new chip (with private labeling on it)  and new EVM using it, so it is possible that some ID word written into the chip is either wrong or the Code composer table that it is used to look up this ID word is wrong.  It seems less likely that the wrong DIE was put in the package, as the F5529 is listed as an 80 pin part.

    It would be useful if there was some "tool" that can readout and display the ID word that is found, so that we can investigate this further.

     

    By the way, I tried this with compiler tools v4.1.2, v4.1.9, and v4.2.1  with the same result.

     

  • Lisa:

    See the message I left as a reply to Gautam.  I have changed the chip version back and forth several times from the Properties menu, always with the same result.   Also, you can see it uses the selection in the displays of the linker and compiler command lines that it shows.  I think the problem is really in the "recognition" of the chip.

     NEWS FLASH:  I had zipped up my project and sent it to a colleague that also has access to one of these EVMs (also with Code composer v5.50).  She just reported that it works OK for her!!   So I copied the project to somewhere else, deleted it from the workspace and re-imported it from my save location (sort of repeating what my colleague would do) and it now works for me also!!   Oh the mysteries of Code Composer…..

  • Hi Bruce,

    ok, great to hear and thanks for the update.  I wonder if the project got corrupted.

    All the best,

    Lisa

  • Lisa:

    I cross compared the directories (and all the files in them) of the "new, re-imported" version and the original one and they are identical (except for new LOG entries).  So the information that got corrupted was outside the project directory!! 

  • The curious case of "Project Corruption"!  That's great you were able to solve this out :)

    Goodluck & Regards,

    Gautam