• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Subscribers 30 subscribers
  • Views 1500 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.

Error -2062 unable to halt device on RM48 after using nowFlash

Andrew Hartnett
Intellectual 890 points
Other Parts Discussed in Thread: UNIFLASH

Hello,

I was trying to get the ethernet lwip example to work, and after having many problems just trying to build the example in CCS, I tried to use nowFlash to program the out file. However, it turns out I used CORTEX_RXX and not CORTEX_RXX_LE. I read in another forum answer that nowFlash was replaced with uniFlash, but that I may have bricked my HDK.

Now, no matter what I do, I get the error -2062 that it's unable to halt device. Any suggestions on how to fix this?

  • 0
    TI__Guru 59540 points

    Hello Andrew,

    It is possible that the incorrect endianness is causing the CPU to enter a very tight error loop preventing the NowFlash tool from halting the CPU. The most effective method we have seen in this case is to hold the nPORRST button while starting the NowFlash programming and release it when you see the "connecting" message. This is hit and miss since you have to hit a very tight timing where the CPU can be halted so it often takes a lot of attempts until the perfect timing is achieved. In some cases, this may not work as it can be very difficult to release the reset at just the right time.

    Let me know if you have any success with this or if you continue to have issues programming the device.

  • 0 in reply to Chuck Davenport
    Intellectual 890 points

    Is it best to use NowFlash or UniFlash? And should I just try to erase it or program it with a new .out file?

  • 0 in reply to Andrew Hartnett
    TI__Guru 59540 points

    Hello Andrew,

    I am glad the solution was able to work for you. I would start by simpy trying to erase the device to clear out the bad code.

    In regard to using UniFlash or NowFlash, there are several factors to consider. The first and foremost is that NowFlash is either end of life or near to that. However, I do find that NowFlash provides a bit more flexibility in the scenario that you were faced with. Also, there are certainly some down sides to UniFlash due to the global nature of the tool causing a particularly large installation file. Functionally, Uniflash offers similar or the same performance as NowFlash and will be the supported tool moving forward. Basically it is the same programming utility offered within CCS 5.x debug but separated out as a stand alone tool for those note using CCS.

  • 0 in reply to Chuck Davenport
    Intellectual 890 points

    I actually ended up erasing the board that was still working after mistaking it for my bricked board. I'm still playing reflash roulette with my bricked board. I will continue trying to erase it on and off throughout the day, but so far I haven't had any luck with it. 

    Should I be trying to release the PORRST at the same time as I hit execute (in NowFlash), slightly before, or slightly after? 

    Thanks,

    Andrew 

  • 0 in reply to Andrew Hartnett
    TI__Guru 59540 points

    Hello Andrew,

    You should release it slightly after Execute when the status window is showing "Connecting......" This is really as you describe sort of a Reflash Roulette since it is very tricky to hit the "window" when it can connect. Keep me posted.

  • 0 in reply to Chuck Davenport
    Intellectual 890 points

    Finally got it. I ended up actually getting it when I got frustrated of not getting the timing right and decided to press and release the S3 RST instead of S4 PORRST. I'm sure it's all just coincidental and that either would have worked, but I got it on my 3rd try using the S3 RST instead of S4. I let very little time pass between hitting execute and releasing the RST. I definitely didn't wait to react to seeing the terminal window because it comes up almost immediately. I also had checked debug info on to try to slow it down a little bit between hitting execute and it connecting. I have no idea printing those extra statements helped my case any. 

  • 0 in reply to Andrew Hartnett
    TI__Guru 59540 points

    Hello Andrew,

    I am glad that you were able to hit that "magic" window and get your device erased. As always, please feel free to let us know if you have any additional issues and we will be glad to help.