Stuck in "Hard fault Debug event" after debug session

[vincent_itk]

Hello,

I encountered a strange behaviour on a nrf52832 : 
- I flash a a debug firmware (with debug optim in build config) : I can run debugger and add breakpoints
- I stop debugger, switch to normal build (without debug optim) and flash
- When program reach an old breakpoint (set in debug mode), I got a "hard fault debug event"

Here is a log I got :

[00:00:00.067,291] <err> os: ***** HARD FAULT *****
[00:00:00.067,321] <err> os:   Debug event
[00:00:00.067,321] <err> os: r0/a1:  0x2000351c  r1/a2:  0x00000000  r2/a3:  0x00000000
[00:00:00.067,321] <err> os: r3/a4:  0x00000040 r12/ip:  0x0001863d r14/lr:  0x00034c49
[00:00:00.067,321] <err> os:  xpsr:  0x01000029
[00:00:00.067,352] <err> os: Faulting instruction address (r15/pc): 0x00043dd0
[00:00:00.067,382] <err> os: >>> ZEPHYR FATAL ERROR 0: CPU exception on CPU 0
[00:00:00.067,382] <err> os: Fault during interrupt handling

[00:00:00.067,413] <err> os: Current thread: 0x20001fd0 (unknown)
[00:00:00.164,978] <err> os: Halting system

I tried to get ride of this, but I was not successful : 
- Erase target
- Do a pristine build an a full reflash after erase in normal mode
- Flash a debug firmware without debugger connected
- Erase all breakpoints in debugger part of vscode

It seems like breakpoints where still existing it nrf52832 flash. This is really strange and ennoying.
To get reide of it, I must restart the debugger, at least clear all breakpoints and reflash with a normal build. Sometimes I got it working, sometimes not. What I miss to get it working back normally ?

Thanks by advance.

[Vidar Berg]

Hello,

I need to investigate this to understand why this fault happens. As a test, please try to power-cycle the board (+ disconnect and re-connect debug probe if you are using an external one) and see if this clears the breakpoint at this address.

Best regards,

Vidar

[vincent_itk]

Thanks for the tip.

Gonna try it when the case occurs again. Will let you know in this thread.

Regards

[Vidar Berg]

Thank you. Yes, it would be good to know how to reproduce this.

[vincent_itk]

Hello,

Got the same behaviour and power-cycle the board make it working properly.

I was able to get rid of the hard this way.

Thanks.

[Vidar Berg]

Thank you for the update. It's good to have a workaround for this. I will try to find the root cause for this if I find a way to reproduce this on my end.