• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 97 subscribers
  • Views 474 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

C6713 POST not working and wrong Endianness

Jorge Cifuentes
Prodigy 160 points

Hi!

I was working with my C6713 DSK board from Spectrum Digital and suddenly CCS lost USB connection with the board. This had already happened a few times, and it was fixed by closing CCS, unplugging the DSK, plug it again and opening again CCS. I never had problems with that before.

But this time, I was working with the GPIO pins. So I had to pull up Swtich #4 from the SW3 switch which connects to a pulldown resistor the HD14 pin and enables GPIO, disabling HPI pins. Then, my DSK board had this switch up when it suddenly disconnected from CCS. I proceed as usual: Close CCS, unplug the DSK and Plug it again. But this time, the POST didn't run. I thought "Mmm... this isn't normal". Then I opened CCS and when I tried to load a program, CCS studio complained telling me that the processor didn't have the right endianness. I saw in the Technical Reference from Spectrum Digital, that the Endianness of the DSP is controlled by the position of Switch #1 from the SW3. I didn't touch that switch. I just had switch #4 from SW3 (the one that enables de GPIO pins) up and the rest were down. So I don't understand how this change in the Endianness happened.

So I pressed the RESET button on the DSK and hoped everything would be OK but the POST didn't run. I turned switch #4 in SW3 back to its original position (off) and pressed RESET button again, but still the same. Then I closed CCS and reopen it and this time it told me that the Board was not found or something like that (the error that appears when the DSK is not plugged in the USB port and you open CCS). I also ran the "DSK Diagnostics tool" and the third test, "DSP Diagnostics" fails.

I went with my Digital Signal Processing class' professor's office to ask for help. He tried to reprogram the Flash with Flashburn but Flashburn told him the same error about the incorrect endianness of the DSP. He opened CCS ( v3.1) and it told him the same thing about the endianness. By the way, I am using CCS 3.3

 

I think that something went wrong when the DSK lost connection with CCS the first time. I was wondering if this could have potentially damaged the board or what may have happened.

 

These are the symptoms:

-The POST program when you power on the DSK does not work

-CCS and Flashburn keep telling that the DSP has not the right Endianness so no programs can be loaded on it

Edit: I remembered now that my program reads the value of the GPIO pins, saves them in a variable and prints them to the Stdout in CCS. Also, my board has a loose USB connection (my professor told me that when assigning me the board) and actually the USB cable wiggles a little while connected to the board. This could explain why CCS randomly lost connection with the DSK sometimes (even though I didn't move the DSK at all).

Then, this "connection lost" happened while the DSK was sending data to CCS (or maybe the other way). This sudden connection lost while communicating with the DSK must have unconfigured the board or something...

 

Is there a way to fix it? Can I restore - in some way - the DSK to factory settings?

Or do I have to start saving money for buying a new DSK board for the University =S?

 

Thank you very much for your help

 

Jorge Cifuentes

 

 

 

 

  • 0
    TI__Guru 62975 points

    Jorge,

    A bad emulator connection can create a lot of unpredictable problems.  Are you able to resolder the USB connection so that it's not loose anymore?

    -Tommy

  • 0 in reply to tlee
    Prodigy 160 points

    Thanks for the reply Tommy.

     

    I guess I could do that, but do you think there is a way to fix this "wrong endianness" issue? Like some kind of  "master reset" to the board? Because CCS recognized the board but when I try to load a program it tells me about the wrong endianness. I another post about the GPIO I was telling one of your colleagues that  I was working with +5V with the GPIO. And he told me that since the core works with 3.3 V the 5 V may have damaged the board.

    Do you think that working with 5 V may have caused the board  to act that way? With my poor knowledge (I'm a college student) I would tell that if those 5 V damaged the board, then it shouldn't even be recognized by CCS or Flash burn.

     

    Thank you very much

     

    Jorge Cifuentes

  • 0 in reply to Jorge Cifuentes
    TI__Guru 62975 points

    Jorge,

    The endianness is only selected once during reset (determined by the value on HD[8]/GP[8]).  If the mode is changing and you have sudden disconnects, you are probably experiencing power problems.  Please make sure that your device has a clean supply voltage.

    Overdriving your pins with 5V can damage the IO transistors on your device.  The symptoms are difficult to predict.

    -Tommy

  • 0 in reply to tlee
    Prodigy 160 points

    Thanks Tommy,

    Is there a way to fix this endianness issue? I know there is a switch in the DSK that determines this. But the switch has been all the time OFF. But when the board lost connection with CCS, I was working with GPIO [8 - 15] so maybe I was applying voltage or GND to GPIO [8]. Do you think that this may had changed the endianness??

     

    Thank you very much, again.

     

    Jorge Cifuentes

  • 0 in reply to Jorge Cifuentes
    TI__Guru 62975 points

    Jorge,

    Without being able to handle the board in person, I would guess that you could have power supply problems, damage to the device/board, or something overriding your GP[8] switch.  Even with random resets, the device should be able to redetect the desired endian setting if the supplies are clean and the pin is functioning.

    I'm afraid that there isn't much more that I can offer except to investigate these possible problems.

    -Tommy

  • 0 in reply to tlee
    Prodigy 160 points

    Thank you very much for your attention Tommy.

     

    I think that the board must have been damaged by the 5 V I applied to the GPIO pins.

     

    Thanks for everything,

     

    Have a nice day!

     

    Jorge Cifuentes