stm32/l152: ztimer Crash after initial flashing #21010
Description
Description
When programming the tests/periph/adc test program to the NUCLEO-L152RE board with make flash, the program crashes on the first call of ztimer_set. After a power cycle, the program runs without any issues. Likewise, when the program is flashed using the builtin mass storage driver of the ST-Link, the program will start without any issues.
Programming the chip with the STM32CubeProgrammer leads to the same result as OpenOCD.
Steps to reproduce the issue
The program I used for testing is tests/periph/adc, maybe it can be reproduced with others well. Optionally you can apply the change described in #20780 (comment), but the unmodified master version works just as well.
First the ISR stacksize has to be increased, otherwise the printf in the Hard Fault handler won't print out anything. Just add CFLAGS += -DISR_STACKSIZE=1024 to tests/periph/adc/Makefile.
BOARD=nucleo-l152re make flash term
Usually the output won't be complete, so hit the reset button:
2024-11-19 15:58:25,032 # main(): This is RIOT! (Version: 2025.01-devel-119-gc636f)
2024-11-19 15:58:25,032 #
2024-11-19 15:58:25,035 # RIOT ADC peripheral driver test
2024-11-19 15:58:25,035 #
2024-11-19 15:58:25,041 # This test will sample all available ADC lines once every 100ms with
2024-11-19 15:58:25,046 # a 10-bit resolution and print the sampled results to STDIO
2024-11-19 15:58:25,046 #
2024-11-19 15:58:25,047 #
2024-11-19 15:58:25,050 # Successfully initialized ADC_LINE(0)
2024-11-19 15:58:25,053 # Successfully initialized ADC_LINE(1)
2024-11-19 15:58:25,056 # Successfully initialized ADC_LINE(2)
2024-11-19 15:58:25,060 # Successfully initialized ADC_LINE(3)
2024-11-19 15:58:25,063 # Successfully initialized ADC_LINE(4)
2024-11-19 15:58:25,066 # Successfully initialized ADC_LINE(5)
2024-11-19 15:58:25,069 # ADC_LINE(0): 18 25 28 30 -1 -1
2024-11-19 15:58:25,072 # ADC_LINE(1): 24 28 30 31 -1 -1
2024-11-19 15:58:25,075 # ADC_LINE(2): 63 63 63 63 -1 -1
2024-11-19 15:58:25,078 # ADC_LINE(3): 25 27 29 30 -1 -1
2024-11-19 15:58:25,081 # ADC_LINE(4): 23 27 30 31 -1 -1
2024-11-19 15:58:25,084 # ADC_LINE(5): 25 29 31 31 -1 -1
2024-11-19 15:58:25,184 #
2024-11-19 15:58:25,186 # Context before hardfault:
2024-11-19 15:58:25,188 # r0: 0x00000002
2024-11-19 15:58:25,189 # r1: 0x20000aa8
2024-11-19 15:58:25,191 # r2: 0x00000001
2024-11-19 15:58:25,193 # r3: 0x00000000
2024-11-19 15:58:25,194 # r12: 0x20000408
2024-11-19 15:58:25,196 # lr: 0x0800106b
2024-11-19 15:58:25,197 # pc: 0x0878306a
2024-11-19 15:58:25,199 # psr: 0x6100000e
2024-11-19 15:58:25,199 #
2024-11-19 15:58:25,200 # FSR/FAR:
2024-11-19 15:58:25,202 # CFSR: 0x00000100
2024-11-19 15:58:25,203 # HFSR: 0x40000000
2024-11-19 15:58:25,205 # DFSR: 0x00000008
2024-11-19 15:58:25,206 # AFSR: 0x00000000
2024-11-19 15:58:25,207 # Misc
2024-11-19 15:58:25,209 # EXC_RET: 0xfffffff1
2024-11-19 15:58:25,212 # Hard fault occurred in ISR number -2
2024-11-19 15:58:25,216 # Attempting to reconstruct state for debugging...
2024-11-19 15:58:25,217 # In GDB:
2024-11-19 15:58:25,219 # set $pc=0x878306a
2024-11-19 15:58:25,220 # frame 0
2024-11-19 15:58:25,220 # bt
2024-11-19 15:58:25,222 # *** RIOT kernel panic:
2024-11-19 15:58:25,224 # HARD FAULT HANDLER
2024-11-19 15:58:25,224 #
2024-11-19 15:58:25,225 # *** halted.
2024-11-19 15:58:25,225 #
2024-11-19 15:58:25,227 # Inside isr -13
With the ztimer ENABLE_DEBUG flag enabled in sys/ztimer/core.c, the output becomes the following:
2024-11-19 15:47:21,679 # ztimer_set(): 0x20000408: set 0x20000bd0 at 93 offset 2147483000
2024-11-19 15:47:21,686 # clock 0x20000408: _ztimer_update_head_offset(): diff=5924 old head 0
2024-11-19 15:47:21,690 # _add_entry_to_list() 0x20000bd0 offset 2147483000
2024-11-19 15:47:21,695 # main(): This is RIOT! (Version: 2025.01-devel-119-gc636f)
2024-11-19 15:47:21,695 #
2024-11-19 15:47:21,698 # RIOT ADC peripheral driver test
2024-11-19 15:47:21,698 #
2024-11-19 15:47:21,704 # This test will sample all available ADC lines once every 100ms with
2024-11-19 15:47:21,709 # a 10-bit resolution and print the sampled results to STDIO
2024-11-19 15:47:21,709 #
2024-11-19 15:47:21,709 #
2024-11-19 15:47:21,713 # Successfully initialized ADC_LINE(0)
2024-11-19 15:47:21,716 # Successfully initialized ADC_LINE(1)
2024-11-19 15:47:21,719 # Successfully initialized ADC_LINE(2)
2024-11-19 15:47:21,723 # Successfully initialized ADC_LINE(3)
2024-11-19 15:47:21,726 # Successfully initialized ADC_LINE(4)
2024-11-19 15:47:21,729 # Successfully initialized ADC_LINE(5)
2024-11-19 15:47:21,732 # ADC_LINE(0): 17 25 28 30 -1 -1
2024-11-19 15:47:21,735 # ADC_LINE(1): 24 28 30 31 -1 -1
2024-11-19 15:47:21,738 # ADC_LINE(2): 63 63 63 63 -1 -1
2024-11-19 15:47:21,741 # ADC_LINE(3): 25 27 29 30 -1 -1
2024-11-19 15:47:21,744 # ADC_LINE(4): 22 27 29 31 -1 -1
2024-11-19 15:47:21,747 # ADC_LINE(5): 25 29 30 31 -1 -1
2024-11-19 15:47:21,752 # ztimer_set(): 0x20000ba8: set 0x20000a10 at 73 offset 100
2024-11-19 15:47:21,757 # ztimer_now() checkpoint=0 lower_last=0 lower_now=78 diff=78
2024-11-19 15:47:21,760 # ztimer_now() returning 78
2024-11-19 15:47:21,766 # clock 0x20000ba8: _ztimer_update_head_offset(): diff=78 old head 0
2024-11-19 15:47:21,769 # _add_entry_to_list() 0x20000a10 offset 100
2024-11-19 15:47:21,775 # ztimer_set(): 0x20000408: set 0x20000bd0 at 96262 offset 100000
2024-11-19 15:47:21,782 # clock 0x20000408: _ztimer_update_head_offset(): diff=96075 old head 0x20000bd0
2024-11-19 15:47:21,791 # ztimer 0x20000408: _ztimer_update_head_offset(): now=101999 new head 0x20000bd0 offset 2147386925
2024-11-19 15:47:21,793 # _del_entry_from_list()
2024-11-19 15:47:21,797 # _add_entry_to_list() 0x20000bd0 offset 100000
2024-11-19 15:47:21,897 #
2024-11-19 15:47:21,899 # Context before hardfault:
2024-11-19 15:47:21,901 # r0: 0x00000002
2024-11-19 15:47:21,902 # r1: 0x20000aa8
2024-11-19 15:47:21,904 # r2: 0x00000001
2024-11-19 15:47:21,906 # r3: 0x00000000
2024-11-19 15:47:21,907 # r12: 0x0000002e
2024-11-19 15:47:21,909 # lr: 0x0800106b
2024-11-19 15:47:21,911 # pc: 0x0878306a
2024-11-19 15:47:21,912 # psr: 0x6100000e
2024-11-19 15:47:21,912 #
2024-11-19 15:47:21,913 # FSR/FAR:
2024-11-19 15:47:21,915 # CFSR: 0x00000100
2024-11-19 15:47:21,916 # HFSR: 0x40000000
2024-11-19 15:47:21,918 # DFSR: 0x00000008
2024-11-19 15:47:21,920 # AFSR: 0x00000000
2024-11-19 15:47:21,920 # Misc
2024-11-19 15:47:21,922 # EXC_RET: 0xfffffff1
2024-11-19 15:47:21,925 # Hard fault occurred in ISR number -2
2024-11-19 15:47:21,929 # Attempting to reconstruct state for debugging...
2024-11-19 15:47:21,930 # In GDB:
2024-11-19 15:47:21,932 # set $pc=0x878306a
2024-11-19 15:47:21,933 # frame 0
2024-11-19 15:47:21,933 # bt
2024-11-19 15:47:21,935 # *** RIOT kernel panic:
2024-11-19 15:47:21,937 # HARD FAULT HANDLER
2024-11-19 15:47:21,937 #
2024-11-19 15:47:21,938 # *** halted.
2024-11-19 15:47:21,938 #
2024-11-19 15:47:21,940 # Inside isr -13
After a power cycle, this is the resulting output:
2024-11-19 15:52:09,369 # _add_entry_to_ztimer_set(): 0x20000408: set 0x20000bd0 at 93 offset 2147483000
2024-11-19 15:52:09,375 # clock 0x20000408: _ztimer_update_head_offset(): diff=5924 old head 0
2024-11-19 15:52:09,379 # _add_entry_to_list() 0x20000bd0 offset 2147483000
2024-11-19 15:52:09,384 # main(): This is RIOT! (Version: 2025.01-devel-119-gc636f)
2024-11-19 15:52:09,385 #
2024-11-19 15:52:09,388 # RIOT ADC peripheral driver test
2024-11-19 15:52:09,388 #
2024-11-19 15:52:09,394 # This test will sample all available ADC lines once every 100ms with
2024-11-19 15:52:09,399 # a 10-bit resolution and print the sampled results to STDIO
2024-11-19 15:52:09,399 #
2024-11-19 15:52:09,399 #
2024-11-19 15:52:09,402 # Successfully initialized ADC_LINE(0)
2024-11-19 15:52:09,406 # Successfully initialized ADC_LINE(1)
2024-11-19 15:52:09,409 # Successfully initialized ADC_LINE(2)
2024-11-19 15:52:09,412 # Successfully initialized ADC_LINE(3)
2024-11-19 15:52:09,415 # Successfully initialized ADC_LINE(4)
2024-11-19 15:52:09,418 # Successfully initialized ADC_LINE(5)
2024-11-19 15:52:09,421 # ADC_LINE(0): 17 25 28 30 -1 -1
2024-11-19 15:52:09,424 # ADC_LINE(1): 24 28 30 31 -1 -1
2024-11-19 15:52:09,428 # ADC_LINE(2): 63 63 63 63 -1 -1
2024-11-19 15:52:09,430 # ADC_LINE(3): 26 28 29 30 -1 -1
2024-11-19 15:52:09,433 # ADC_LINE(4): 22 27 29 31 -1 -1
2024-11-19 15:52:09,436 # ADC_LINE(5): 27 29 30 31 -1 -1
2024-11-19 15:52:09,442 # ztimer_set(): 0x20000ba8: set 0x20000a10 at 73 offset 100
2024-11-19 15:52:09,446 # ztimer_now() checkpoint=0 lower_last=0 lower_now=78 diff=78
2024-11-19 15:52:09,449 # ztimer_now() returning 78
2024-11-19 15:52:09,456 # clock 0x20000ba8: _ztimer_update_head_offset(): diff=78 old head 0
2024-11-19 15:52:09,459 # _add_entry_to_list() 0x20000a10 offset 100
2024-11-19 15:52:09,465 # ztimer_set(): 0x20000408: set 0x20000bd0 at 96262 offset 100000
2024-11-19 15:52:09,472 # clock 0x20000408: _ztimer_update_head_offset(): diff=96075 old head 0x20000bd0
2024-11-19 15:52:09,480 # ztimer 0x20000408: _ztimer_update_head_offset(): now=101999 new head 0x20000bd0 offset 2147386925
2024-11-19 15:52:09,482 # _del_entry_from_list()
2024-11-19 15:52:09,487 # _add_entry_to_list() 0x20000bd0 offset 100000
2024-11-19 15:52:09,590 # ztimer_handler(): 0x20000408 now=223913
2024-11-19 15:52:09,594 # 0x20000408:101999(101999)->0x20000bd0:100000(201999)
2024-11-19 15:52:09,597 # ztimer_handler(): no checkpointing
2024-11-19 15:52:09,603 # ztimer_handler(): trigger 0x20000bd0->0 at 235427
2024-11-19 15:52:09,606 # ztimer_handler(): 0x20000ba8 now=239
2024-11-19 15:52:09,609 # 0x20000ba8:78(78)->0x20000a10:100(178)
2024-11-19 15:52:09,614 # ztimer_now() checkpoint=78 lower_last=78 lower_now=246 diff=168
2024-11-19 15:52:09,617 # ztimer_now() returning 246
2024-11-19 15:52:09,621 # ztimer_handler(): 0x20000ba8 diff=-68
2024-11-19 15:52:09,625 # ztimer_handler(): trigger 0x20000a10->0 at 258
2024-11-19 15:52:09,630 # ztimer_now() checkpoint=246 lower_last=246 lower_now=262 diff=16
2024-11-19 15:52:09,633 # ztimer_now() returning 262
2024-11-19 15:52:09,639 # clock 0x20000ba8: _ztimer_update_head_offset(): diff=84 old head 0
2024-11-19 15:52:09,645 # ztimer_set(): 0x20000408: set 0x20000bd0 at 276824 offset 2147483000
2024-11-19 15:52:09,652 # clock 0x20000408: _ztimer_update_head_offset(): diff=81002 old head 0
2024-11-19 15:52:09,656 # _add_entry_to_list() 0x20000bd0 offset 2147483000
2024-11-19 15:52:09,658 # 0x20000ba8:262(262)!
2024-11-19 15:52:09,661 # ztimer_handler(): 0x20000ba8 done.
2024-11-19 15:52:09,668 # clock 0x20000408: _ztimer_update_head_offset(): diff=15783 old head 0x20000bd0
2024-11-19 15:52:09,676 # ztimer 0x20000408: _ztimer_update_head_offset(): now=298784 new head 0x20000bd0 offset 2147467217
2024-11-19 15:52:09,682 # 0x20000408:298784(298784)->0x20000bd0:2147467217(2147766001)
2024-11-19 15:52:09,685 # ztimer_handler(): 0x20000408 done.
2024-11-19 15:52:09,688 # ADC_LINE(0): 25 28 30 31 -1 -1
2024-11-19 15:52:09,692 # ADC_LINE(1): 29 30 31 32 -1 -1
2024-11-19 15:52:09,694 # ADC_LINE(2): 63 63 63 63 -1 -1
2024-11-19 15:52:09,698 # ADC_LINE(3): 32 32 32 32 -1 -1
2024-11-19 15:52:09,700 # ADC_LINE(4): 28 30 31 31 -1 -1
2024-11-19 15:52:09,703 # ADC_LINE(5): 25 29 30 31 -1 -1
...
It looks like the program crashes either in _add_entry_to_list() or in the ztimer_handler() function.
Maybe someone a bit more familiar with ztimer can have a look at this issue. My theory is that some register is not properly reset.
I don't know which hardware timer ztimer uses on the STM32L152ZE, but there is a remark in the reference manual [1] on page 165, section 6.3.14:
Note: The LSEON, LSEBYP, RTCSEL and RTCEN bits in the RCC control and status register
(RCC_CSR) are in the RTC domain. As these bits are write protected after reset, the DBP
bit in the Power control register (PWR_CR) has to be set to be able to modify them. Refer to
Section RTC and RTC backup registers for further information. These bits are only reset
after a RTC domain reset (see RTC and backup registers reset). Any internal or external
reset does not have any effect on them.
Does ztimer use the RTC or relies on the LSE-registers? Trying to access the write protected RTC registers would surely lead to a Hard Fault?
Expected results
Regardless of the flashing method, the program should not crash when using ztimer.
Actual results
The program crashes.
Versions
Operating System Environment
----------------------------
Operating System: "Linux Mint" "21.3 (Virginia)"
Kernel: Linux 5.15.0-125-generic x86_64 x86_64
System shell: /usr/bin/dash (probably dash)
make's shell: /usr/bin/dash (probably dash)
Installed compiler toolchains
-----------------------------
native gcc: gcc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
arm-none-eabi-gcc: arm-none-eabi-gcc (15:10.3-2021.07-4) 10.3.1 20210621 (release)
avr-gcc: avr-gcc (GCC) 5.4.0
msp430-elf-gcc: missing
riscv-none-elf-gcc: missing
riscv64-unknown-elf-gcc: missing
riscv32-esp-elf-gcc: missing
xtensa-esp32-elf-gcc: missing
xtensa-esp32s2-elf-gcc: missing
xtensa-esp32s3-elf-gcc: missing
xtensa-esp8266-elf-gcc: missing
clang: missing
Installed compiler libs
-----------------------
arm-none-eabi-newlib: "3.3.0"
msp430-elf-newlib: missing
riscv-none-elf-newlib: missing
riscv64-unknown-elf-newlib: missing
riscv32-esp-elf-newlib: missing
xtensa-esp32-elf-newlib: missing
xtensa-esp32s2-elf-newlib: missing
xtensa-esp32s3-elf-newlib: missing
xtensa-esp8266-elf-newlib: missing
avr-libc: "2.0.0" ("20150208")
Installed development tools
---------------------------
ccache: missing
cmake: missing
cppcheck: missing
doxygen: 1.9.1
git: git version 2.34.1
make: GNU Make 4.3
openocd: xPack Open On-Chip Debugger 0.12.0+dev-01685-gb9224c0c0-dirty (2024-08-02-19:50)
python: missing
python2: missing
python3: Python 3.10.12
flake8: 4.0.1 (mccabe: 0.6.1, pycodestyle: 2.8.0, pyflakes: 2.4.0) CPython 3.10.12 on
coccinelle: spatch version 1.1.1 compiled with OCaml version 4.13.1
Note: I updated OpenOCD to see if that was the cause of the issue, but the same happens with OpenOCD 0.11 and 0.12.