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.

FOE transfer doesn't work well on sysbios sdk 2.1.1.2

Other Parts Discussed in Thread: SYSBIOS

Hello evreybody,

I'm working with the following software configuration:

    • Code Composer Studio version CCS 6.1.1.00022
    • SYS/BIOS 6.42.2.29 Real Time Operating System
    • XDC Tool 3.31.2.38_core
    • Compiler GNU v4.8.4 (Linaro)
    • NDK 2.24.3.35
    • Sysbios sdk 2.1.1.2

Concerning hardware, I make my test on evaluation platform ice v2.

I would like to test FOE tranfert.

So, I'm runnig the full ecat application which is provided by TI,

I use KPA server as ethercat master and I use KPA Studio interface to make the transfer.

I begin my test:

  • start the code on sitara
  • start kpa master,
  • attach TI slave
  • then I test if I can reach operational mode: everything is ok.

Then, I begin to test FOE:

  • I ask bootstrap mode
  • I dowload ECATFW__ (which is the same bianry file that is already running)
  • I have a message: "ECATFW__ dowloading to slave has finished successfully" but the transfer is really fast (really faster than the one tested with sysbios 1.0.0.8)
  • but in fact, the transfer is not ok and if I scan the bus i can't see the TI slave anymore !!!!

Is there anyone who can help me on this subject please ?

Thanks a lot,

Laurence

  • I will forward this to the ISDK team.
  • Hi Laurence,

    Cazaban Laurence said:
    but in fact, the transfer is not ok and if I scan the bus i can't see the TI slave anymore !!!!

    What do you mean with 'the transfer is not ok'? Do you see any communication errors on ESC level?

    Generally online firmware upgrade (assuming this is what you want to achieve) is an application issue and as such outside the scope of our support. There are many ways how this can be done and it certainly depends somewhat on the master in use. We have no way of reproducing your issue at all.

    I assume you will need to debug the issue step by step.

    regards,

  • Hello Frank,

    So, as you said me, I have debugged my issue step by step by myself....

    And I realise that, in fact, the problem was due to the last packet of code which is transfered by FOE and not copied into the flash....

    In fact, my last packet's size is < 256 and you have a bug in the tiescbsp.c file, in the function bsp_boot_2_init_handler().

    It seems that you have copied the same code than in the former function bsp_store_fw_data() and, so, you don't tret the last packet if its size is <256 bit....

    In the former version of sdk (1.0.0.8), this case was taken into account... So I have replace the code of the 2.1.1.2  by the one in the 1.0.0.8 and my FOE transfer is ok !!!!

    regards,

    Laurence

  • Hi Laurence,

    I am sorry for that. I was not aware about any differences in that code. I will inform the engineering team to check on that and make sure it gets fixed for next release. I assume we also need to adapt our test code accordingly to be able to detect such regression issues.

    Best regards,