[SOLVED] Ozone hangs, uses 100% of a CPU core after programming STM32F469BI

This site uses cookies. By continuing to browse this site, you are agreeing to our Cookie Policy.

  • [SOLVED] Ozone hangs, uses 100% of a CPU core after programming STM32F469BI

    I have a J-Link Ultra+ that I'm trying to use to debug a custom STM32F469BI board.

    I tried to use Ozone v2.56i I'm using the JTAG interface with the speed set to Auto.

    Im able to download the elf file, I can see the report in the log. However, after the download and the Ozone UI shows stepping to main, Ozone just hangs, using 100% of a CPU core.

    I tried downloading Embedded Studio to see if it has any issues with running one of the sample programs and it seems to be able to do so.

    I did see this thread - [SOLVED] Embedded Studio 3.34 + Ozone 2.56e + i.MXRT1052 EVK Issue which talks about modifying the .jdebug file.

    I didnt really get anywhere, but thats to be expected, I suppose, different target.

    How can I get Ozone to be able to debug my application without freezing after download/connection?

    Im unable to provide logs from Ozone because the UI is not responsive.

    I've tried the Windows and Linux versions, same result as in me waiting for the UI to be responsive and giving up.

    I've also tried SWD rather than JTAG. No joy, but I did get a screenshot, which is attached . At this point, nothing in the UI is clickable, Ozone is using 100% of a CPU

    The post was edited 1 time, last by paul.r.thompson ().

  • Hello,

    Thank you for your inquiry.
    Such an issue is not known to us.
    Does the load spike happen on project load or only when starting the debug session?
    Could you provide us with the elf file so we can reproduce the described issue?

    Best regards,
    Nino
    Please read the forum rules before posting: Forum Rules

    Keep in mind, this is not a support forum. Its main purpose is user to user interaction.
    Our engineers will try to answer your questions between their projects if possible but this can be delayed by longer periods of time.
    Should you be entitled to support contact us per e-mail.
    Alternatively our support ticketing system can be used as well: segger.com/ticket/
  • Thank you,

    The hang occurs after the debug session starts, when Ozone is apparently trying to stop at the main() function. I replaced my real main function with one that just blinks a LED. When I do that, I see that when Ozone stops at main(), it spends a bit of time "Writing Memory" and a five-bar progress icon going across the bottom status bar, then the UI is available again.

    With my real main function, the "Writing Memory" part is stuck, the five-box progress bar is not moving across the bottom of the screen.

    Is there a secure way for me to provide the elf? The code is private IP.

    Thank you

    -Paul
  • Additional information:
    I was able to use Eclipse Oxygen with the MCU plugin to build/flash/debug my application using the J-Link. The debug session starts just fine within Eclipse. However, I do really want to use Ozone in order to be able to use the peripheral register view (amongst other things). I'm unable to use the MCU plugin's peripheral view functionality because I have a CMake workflow that is incompatible with MCU's project management.
    If it helps at all, this is the console log from Eclipse for the successful flash, connect, and a couple of steps. I'm unable to get the Ozone log other than what is shown in the screenshot because the UI is unresponsive so I can't select and copy the log text.
    So anyway, I have attached the output of a successful Eclipse debug run with my actual application.
    Files
  • More data.
    I moved over to Windows so that I could use the J-Link Control Panel to log the probe communications.
    I used Eclipse and Ozone to load the exactly same binary. Eclipse was able to stop at start of main() and I stepped a couple times. Ozone exhibited the same issue, after connecting and apparently halting at main(), it hangs and uses up a whole CPU core.
    I'm attaching logs for both runs and the jdebug file. Looking at the end of the Ozone log, it seems that interestingly enough the J-Link appears to be fine, it has halted, but there's something that Ozone doesn't like...
    Thank you for your attention to this issue!
    Files
  • Hello,

    Thank you for providing the information.
    We are quite certain that the issue is related to Ozone + that particular elf file and has nothing to do with J-Link.

    Is there a secure way for me to provide the elf? The code is private IP.

    Of course. I will PM you with contact information, so you can talk to our support channel directly.

    EDIT: The issue has been fixed with Ozone V2.56m.

    Best regards,
    Nino
    Please read the forum rules before posting: Forum Rules

    Keep in mind, this is not a support forum. Its main purpose is user to user interaction.
    Our engineers will try to answer your questions between their projects if possible but this can be delayed by longer periods of time.
    Should you be entitled to support contact us per e-mail.
    Alternatively our support ticketing system can be used as well: segger.com/ticket/