NRF54L15 DK soft bricked?

[Elppu]

Hi!

I have somehow softbricked my nrf54L15 dk. This happened by using nrf conncet board configurator. First symptom was no uart logs and now I get this crash via RTT logging. Program is Blink sample

RTT viewer log:

(Connection lost)
00> 7F 85 8B
00> [00:00:00.430,51[00:00:00.430,525] <dbg> os: setup_thread_stack: stack 0x2000b2e8 for thread 0x20002318: obj_size=320 buf_start=0x2000b2e8  buf_size 320 stack_ptr=0x2000b428
00> [00:00:00.430,547] <dbg> clock_control: clkstarted_handle: lfclk: Clock started
00> [00:00:00.430,562] <dbg> os: setup_thread_stack: stack 0x2000a7e8 for thread 0x200001c0: obj_size=768 buf_start=0x2000a7e8  buf_size 768 stack_ptr=0x2000aae8
00> *** Booting nRF Connect SDK v2.7.0-5cb85570ca43 ***
00> *** Using Zephyr OS v3.6.99-100befc70c74 ***
00> [00:00:00.430,611] <dbg> os: k_sched_unlock: scheduler unlocked (0x20002398:0)
00> [00:00:00.430,632] <dbg> os: z_impl_k_mutex_lock: 0x20002398 took mutex 0x20000198, count: 1, orig prio: 0
00> [00:00:00.430,639] <dbg> os: z_impl_k_mutex_unlock: mutex 0x20000198 lock_count: 1
00> [00:00:00.430,648] <dbg> os: z_impl_k_mutex_unlock: new owner of mutex 0x20000198: 0 (prio: -1000)
00> Testtest RTT[00:00:00.432,320] <dbg> os: z_tick_sleep: thread 0x20002398 for 1 ticks
00> [00:00:00.432,387] <err> os: ***** USAGE FAULT *****
00> [00:00:00.432,392] <err> os:   Stack overflow (context area not valid)
00> [00:00:00.432,400] <err> os: r0/a1:  0x2000a8d8  r1/a2:  0x0000001c  r2/a3:  0x00000000
00> [00:00:00.432,407] <err> os: r3/a4:  0x2000a88c r12/ip:  0x00000000 r14/lr:  0x00001c25
00> [00:00:00.432,411] <err> os:  xpsr:  0x61000200
00> [00:00:00.432,415] <err> os: Faulting instruction address (r15/pc): 0x000010c0
00> [00:00:00.432,432] <err> os: >>> ZEPHYR FATAL ERROR 2: Stack overflow on CPU 0
00> [00:00:00.432,446] <err> os: Current thread: 0x200001c0 (unknown)
00> [00:00:02.734,831] <err> os: Halting system
(Connection lost)
00> 7F 85 8B
00> [00:00:00.430,51[00:00:00.430,575] <dbg> os: setup_thread_stack: stack 0x20004828 for thread 0x200023d8: obj_size=320 buf_start=0x20004828  buf_size 320 stack_ptr=0x20004968
00> [00:00:00.430,614] <dbg> clock_control: clkstarted_handle: lfclk: Clock started
00> [00:00:00.430,630] <dbg> os: setup_thread_stack: stack 0x20003d28 for thread 0x20000260: obj_size=768 buf_start=0x20003d28  buf_size 768 stack_ptr=0x20004028
00> *** Booting nRF Connect SDK v2.7.0-5cb85570ca43 ***
00> *** Using Zephyr OS v3.6.99-100befc70c74 ***
00> [00:00:00.430,714] <dbg> os: setup_thread_stack: stack 0x20003928 for thread 0x200001c0: obj_size=1024 buf_start=0x20003928  buf_size 1024 stack_ptr=0x20003d28
00> [00:00:00.430,774] <dbg> os: k_sched_unlock: scheduler unlocked (0x20002478:0)
00> [00:00:00.430,792] <dbg> os: z_impl_k_mutex_lock: 0x20002478 took mutex 0x20000198, count: 1, orig prio: 0
00> [00:00:00.430,800] <dbg> os: z_impl_k_mutex_unlock: mutex 0x20000198 lock_count: 1
00> [00:00:00.430,808] <dbg> os: z_impl_k_mutex_unlock: new owner of mutex 0x20000198: 0 (prio: -1000)
00> Testtest RTT[00:00:00.432,479] <dbg> os: z_tick_sleep: thread 0x20002478 for 1 ticks
00> [00:00:00.432,553] <err> os: ***** USAGE FAULT *****
00> [00:00:00.432,557] <err> os:   Stack overflow (context area not valid)
00> [00:00:00.432,566] <err> os: r0/a1:  0x20003e18  r1/a2:  0x0000001c  r2/a3:  0x00000000
00> [00:00:00.432,573] <err> os: r3/a4:  0x20003dcc r12/ip:  0x00000000 r14/lr:  0x00001d71
00> [00:00:00.432,577] <err> os:  xpsr:  0x61000200
00> [00:00:00.432,581] <err> os: Faulting instruction address (r15/pc): 0x000010c0
00> [00:00:00.432,599] <err> os: >>> ZEPHYR FATAL ERROR 2: Stack overflow on CPU 0
00> [00:00:00.432,613] <err> os: Current thread: 0x20000260 (unknown)
00> [00:00:02.953,158] <err> os: Halting system

What might cause this? I have tried already recovery command.

1. No uart logs

2. Debugger does not reach main - It crashes before my breakpoint.

3. No blink

Only way I got these were using RTT. prj.conf

CONFIG_LOG=y
CONFIG_LOG_BUFFER_SIZE=8192

CONFIG_UART_ASYNC_API=y


CONFIG_UART_CONSOLE=n
CONFIG_LOG_BACKEND_UART=n
CONFIG_LOG_BACKEND_RTT=y
CONFIG_USE_SEGGER_RTT=y
CONFIG_SEGGER_RTT_BUFFER_SIZE_UP=4096 
#32768
CONFIG_SEGGER_RTT_BUFFER_SIZE_DOWN=64
CONFIG_SEGGER_RTT_MODE_NO_BLOCK_SKIP=y


# General config
CONFIG_LOG=y
CONFIG_LOG_DEFAULT_LEVEL=4

CONFIG_EARLY_CONSOLE=y
CONFIG_THREAD_ANALYZER=y
CONFIG_THREAD_ANALYZER_AUTO=y

[Turbo J]

Try CONFIG_LOG_MODE_DEFERRED=y or increase all those stack sizes quite a bit in the config.

Debug logs take tons of stack space for the printf() formatting.

[Elppu]

But blink alone even without logs should work? I tried this conf with no luck:

CONFIG_MPSL_WORK_STACK_SIZE=4096
CONFIG_GPIO=y

CONFIG_USE_SEGGER_RTT=y

CONFIG_LOG_BACKEND_RTT=y

CONFIG_SEGGER_RTT_BUFFER_SIZE_UP= 4096 

CONFIG_SEGGER_RTT_BUFFER_SIZE_DOWN=64
CONFIG_SEGGER_RTT_MODE_NO_BLOCK_SKIP=y


# General config
CONFIG_LOG=y
CONFIG_LOG_DEFAULT_LEVEL=4
CONFIG_LOG_MODE_DEFERRED=y

# Stack 
CONFIG_MAIN_STACK_SIZE=12000
CONFIG_MPSL_WORK_STACK_SIZE=12000

[Einar Thorsrud]

Hi,

With "Blink" sample, do you refer to the basic blinky sample, or another? Can you backtrack and make sure you test the unmodified sample from an unmodified SDK?

Also, as you write that the first symptom was no UART logs, I wonder if you disabled the UART in the interface MCU (onboard debugger) in the board configuration? Can you show the board configurator configuration as it is currently?