• State
  • Locked Locked
  • Replies 7 replies
  • Subscribers 31 subscribers
  • Views 570 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.

Cannot program TM4C129XDev board using LMFlash 1613 in Manual Ethernet mode

Jeff McDevitt
Expert 1775 points


Can someone that has a TM4C129XDev board please try to update the firmware using LM Flash 1613 set to Manual Mode - Ethernet?

I cannot get this to work. I know my setup is good because I can program a Stellaris part with no problem.

[Edit]: The part on the board is XM4C129XNCZADI3.

Thanks,

Jeff

  • 0
    Guru 86765 points

    Hi Jeff,

         Did you set the correct MAC address at LM Flash Programmer?

    - kel

  • 0 in reply to Markel Robregado
    Expert 1775 points

    Yes - I double-checked that.

    Does it work for you?

    Thanks

  • 0 in reply to Jeff McDevitt
    Intellectual 905 points

    Double check the errata and the version of the chip on the board. 

    I believe the first rev of silicon had an errata on the ROM Ethernet bootloader which would be what you are likely trying to use.  A good number of DK boards were produced with the A0 silicon. 

  • 0 in reply to INTEGRIS Dexter
    Expert 1775 points

    Dexter,

    I became aware of that issue when I first ran into this a while back. I received a sample of the latest version of the Engineering part (XM4C129XNCZADI3) and installed it on the board. I still can't get it to work.

    Thanks

  • 0 in reply to Jeff McDevitt
    Intellectual 905 points

    First, I am impressed with your skills and or your lab that installed that BGA yourself.  It does introduce the possibility that a ball did not get perfectly soldered.  I don't think I have tried this with the DK, mine has the A0 silicon.

    I have used LM Flash to program over Ethernet to the ROM bootloader on the TM4C129 launchpad.  So it does work.  The launchpads all had A1 or later chips.  A cheap investment to double check yourself.

    some other thoughts...

    Disable all network interfaces that you are not using especially virtual box interfaces. 

    Put your board and PC on the same switch/hub etc.  Or perhaps even directly connect the two.  The auto ip and auto crossover stuff should work.

    Get wireshark up and going and see what the network traffic looks like.

    The ROM bootloader polls all of the bootloader interfaces (serial, i2c, etc etc) the first one of those with activity wins.  If there is any activity on any of those (or perceived activity aka a short or noise) then the ROM bootloader may latch to that interface.  Once latched to an interface the bootloader does not let go until reset or firmware update is complete.

  • 0 in reply to INTEGRIS Dexter
    Guru 47900 points

    @Dexter,

    Das ist wunderbar!  Great display of logic, awareness, and, "insider detail."  Imho - very well done...

    Love your suggestion of, "Launchpad - as cheap investment to "double check!"  Our small group always buys/develops new items in, "three's" - just to reduce/resolve dreaded, "Single board anomaly..."

  • 0 in reply to INTEGRIS Dexter
    Expert 1775 points

    Dexter,

    Thanks for the reply.

    I am fortunate enough to have a contract mfg that replaced that part for about $20. A bargain in my book.

    I am working on getting a TM4C129 launchpad to test that out.

    I am confident in my network setup as I use the exact same configuration to program Stellaris parts with no problem.

    Using Wireshark while trying to program the Tiva shows the packets going from LM Flash to the board but absolutely nothing going back. I can communicate with the board just fine if I program it via JTAG.

    I tried removing all of the non-critical jumpers that are on the board and that did not help.

    Thanks,

    Jeff