Debug problems with 3.2 update

[davidsherman]

Seeing a couple of problems with KDS 3.2, FreeRTOS, and a KE18.  I had KDS do a recent update, and I see a few things.  I see that the debugger is now aware of FreeRTOS threads, which is nice, but I'm experiencing a problem with breakpoints.  I'll stick a breakpoint in, and it will stop in the FreeRTOS idle task instead (as an example).  How do I get rid of these "phantom" breakpoints?

Another problem which is a bit more annoying is I can't restart the debugger without relaunching.  If I pause and try to restart without terminating, it crashes.  If I try to stop the debugger at that point,  I get a screen saying that arm-none-eabi-gdb has stopped working, and I have to bring up task manager to kill off the pegdbserver_console.exe in order to relaunch completely.  How do I get it so I can just reboot again?

I'm using a P&E Multilink Universal FX .

[BlackNight]

Hi David,

are you using 2.8.5 version of the P&E plugins? I ask because I saw a similar thing in an earlier version (2.8.1), but I have not seen it in 2.8.5.

See P&E ARM Cortex-M Debugging with FreeRTOS Thread Awareness and Real Time Expressions for GDB and Ecli... 

I hope this helps,

Erich

[davidsherman]

Hi Erich, I checked, and it reported something much older for GNU ARM Pemicro debugging support, something like 2.0.8.2015.  I went through the "Install new software" to update, and now it reports 2.8.5.  Problem still occurs, though.  The only other thing I noticed is when checking for GNU ARM Eclipse plugins, it balks saying there was a handshake problem with gnuarmeclipse.sourceforge.net.

Thank you for the article though, I've used your -kernel=none trick to turn off the FreeRTOS debugging, that seems to have fixed it.

[kevin1]

David,

As Erich indicated, we have resolved the issue with the FreeRTOS analysis when the device stops. You can download the update from our eclipse update site. Let us know if you have further issues.

Regards,

Kevin

PEmicro

[BlackNight]

Hi David,

I reported the issue to P&E and they have fixed it on their side, and it could be possible that they release an update on Monday, so you might check that update.

Best regards,

Erich

[BlackNight]

I just did some more trials, and yes, sometimes (not always) I see some phantom stops. That's probably you see too.

Not sure how to reproduce it so I can report it to P&E.

If you just stop the application, it is very likely it will stop in the idle task (as I assume the system is mostly idle).

Erich