Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Problems with low power and entropy on STM32WB55 #72640

Open
gumulka opened this issue May 12, 2024 · 4 comments · May be fixed by #72882
Open

Problems with low power and entropy on STM32WB55 #72640

gumulka opened this issue May 12, 2024 · 4 comments · May be fixed by #72882
Assignees
@erwango
Labels
area: Power Management bug The issue is a bug, or the PR is fixing a bug platform: STM32 ST Micro STM32 priority: low Low impact/importance bug

Comments

@gumulka
Copy link
Contributor

gumulka commented May 12, 2024

Describe the bug
When using PM_DEVICE together with ENTROPY_GENERATOR on the STM32WB55 board, I get an BUS fault after a while.

The time until the bug occurs can vary, but it does occur after a few seconds every time.

I have tried the v3.6.0 and also mainline zephyr and both have the same error.

To Reproduce

I tried this with the nucleo_wb55rg devboard. To be precise MB1355D-01.

Enable CONFIG_PM, CONFIG_PM_DEVICE and CONFIG_ENTROPY_GENERATOR in the Bluetooth peripheral sample and wait a bit. I have speed up the process, by replacing the content of the while loop in main with k_sleep(K_MSEC(2));

patch for this

compile and flash with:

west build -p -b nucleo_wb55rg samples/bluetooth/peripheral
west flash

Expected behavior
No bus fault.

Impact
Not sure yet, will have to see. Currently annoyance.

Logs and console output

Log output:

*** Booting Zephyr OS build v3.6.0 ***
[00:00:00.020,000] <inf> fs_nvs: 2 Sectors of 4096 bytes
[00:00:00.020,000] <inf> fs_nvs: alloc wra: 0, f88
[00:00:00.020,000] <inf> fs_nvs: data wra: 0, 90
[00:00:00.042,000] <inf> bt_hci_core: No ID address. App must call settings_load()
Bluetooth initialized
[00:00:00.044,000] <inf> bt_hci_core: Identity: 02:80:E1:00:00:00 (public)
[00:00:00.044,000] <inf> bt_hci_core: HCI: version 1.0b (0x00) revision 0x8077, manufacturer 0x0030
[00:00:00.044,000] <inf> bt_hci_core: LMP: version 1.0b (0x00) subver 0x2177
Advertising successfully started
Indicate VND attr 0x80189a0 (UUID 12345678-1234-5678-1234-56789abcdef1)
[00:00:00.445,000] <err> os: ***** BUS FAULT *****
[00:00:00.445,000] <err> os:   Imprecise data bus error
[00:00:00.445,000] <err> os: r0/a1:  0x08018070  r1/a2:  0x58001000  r2/a3:  0x2000049c
[00:00:00.445,000] <err> os: r3/a4:  0x00000000 r12/ip:  0x00000001 r14/lr:  0x08002a21
[00:00:00.445,000] <err> os:  xpsr:  0x61000000
[00:00:00.445,000] <err> os: Faulting instruction address (r15/pc): 0x0800e0f6
[00:00:00.446,000] <err> os: >>> ZEPHYR FATAL ERROR 26: Unknown error on CPU 0
[00:00:00.446,000] <err> os: Current thread: 0x200015e8 (unknown)
[00:00:00.448,000] <err> os: Halting system

Time until bus fault occurs can vary. With 2ms sleep I have seen between 0.036 and 2.500 seconds in the logs.

Backtrace from gdb with zephyr v3.6.0:

#0  arch_system_halt (reason=reason@entry=0x1a) at /PATH_TO_ZEPHYR/zephyr/kernel/fatal.c:30
#1  0x0800f82c in k_sys_fatal_error_handler (reason=reason@entry=0x1a, esf=esf@entry=0x20003c80 <z_interrupt_stacks+2048>) at /PATH_TO_ZEPHYR/zephyr/kernel/fatal.c:44
#2  0x0800f8de in z_fatal_error (reason=reason@entry=0x1a, esf=esf@entry=0x20003c80 <z_interrupt_stacks+2048>) at /PATH_TO_ZEPHYR/zephyr/kernel/fatal.c:118
#3  0x08003a88 in z_arm_fatal_error (reason=0x1a, esf=0x20003c80 <z_interrupt_stacks+2048>, esf@entry=0x20003c90 <z_interrupt_stacks+2064>) at /PATH_TO_ZEPHYR/zephyr/arch/arm/core/fatal.c:86
#4  0x08003eb8 in z_arm_fault (msp=<optimized out>, psp=<optimized out>, exc_return=<optimized out>, callee_regs=<optimized out>) at /PATH_TO_ZEPHYR/zephyr/arch/arm/core/cortex_m/fault.c:1157
#5  0x08003f88 in z_arm_usage_fault () at /PATH_TO_ZEPHYR/zephyr/arch/arm/core/cortex_m/fault_s.S:102
#6  <signal handler called>
#7  clock_control_off (sys=0x2000050c <pclken_rng>, dev=0x8018238 <__device_dts_ord_5>) at /PATH_TO_ZEPHYR/zephyr/include/zephyr/drivers/clock_control.h:150
#8  entropy_stm32_suspend () at /PATH_TO_ZEPHYR/zephyr/drivers/entropy/entropy_stm32.c:136
#9  0x080029f4 in pm_device_action_run (dev=dev@entry=0x8018310 <__device_dts_ord_58>, action=action@entry=PM_DEVICE_ACTION_SUSPEND) at /PATH_TO_ZEPHYR/zephyr/subsys/pm/device.c:60
#10 0x0800272c in pm_suspend_devices () at /PATH_TO_ZEPHYR/zephyr/subsys/pm/pm.c:74
#11 pm_system_suspend (ticks=0xa) at /PATH_TO_ZEPHYR/zephyr/subsys/pm/pm.c:201
#12 0x0800fc58 in idle (unused1=<optimized out>, unused2=<optimized out>, unused3=<optimized out>) at /PATH_TO_ZEPHYR/zephyr/kernel/idle.c:85

Environment (please complete the following information):

  • OS: Linux
  • Toolchain: zephyr 0.16.3
  • Commit SHA or Version used: v3.6.0 and main (b8c21d4)
@gumulka gumulka added the bug The issue is a bug, or the PR is fixing a bug label May 12, 2024
@github-actions GitHub Actions
Copy link

Hi @gumulka! We appreciate you submitting your first issue for our open-source project. 🌟

Even though I'm a bot, I can assure you that the whole community is genuinely grateful for your time and effort. 🤖💙

gumulka added a commit to gumulka/led-hat that referenced this issue May 12, 2024
@gumulka
Remove random init and activate power management
80f665a 
entropy generation and power management does not work together in
current zephyr. See
zephyrproject-rtos/zephyr#72640
for details.
@henrikbrixandersen henrikbrixandersen added the platform: STM32 ST Micro STM32 label May 13, 2024
@FRASTM
Copy link
Collaborator

FRASTM commented May 14, 2024 •
edited

@gumulka I confirm the ZEPHYR FATAL ERROR 26 with the condition you exposed, even with k_sleep(K_MSEC(10));
Could you please test the following patch on your own.

diff --git a/drivers/entropy/entropy_stm32.c b/drivers/entropy/entropy_stm32.c
index a9cbda2abc7..8bc79fdc898 100644
--- a/drivers/entropy/entropy_stm32.c
+++ b/drivers/entropy/entropy_stm32.c
@@ -112,6 +112,10 @@ static int entropy_stm32_suspend(void)
 	RNG_TypeDef *rng = dev_data->rng;
 	int res;
 
+#if defined(CONFIG_SOC_SERIES_STM32WBX) || defined(CONFIG_STM32H7_DUAL_CORE)
+	/* Prevent concurrent access with PM */
+	z_stm32_hsem_lock(CFG_HW_RNG_SEMID, HSEM_LOCK_WAIT_FOREVER);
+#endif /* CONFIG_SOC_SERIES_STM32WBX || CONFIG_STM32H7_DUAL_CORE */
 	LL_RNG_Disable(rng);
 
 #ifdef CONFIG_SOC_SERIES_STM32WBAX
@@ -136,6 +140,10 @@ static int entropy_stm32_suspend(void)
 	res = clock_control_off(dev_data->clock,
 			(clock_control_subsys_t)&dev_cfg->pclken[0]);
 
+#if defined(CONFIG_SOC_SERIES_STM32WBX) || defined(CONFIG_STM32H7_DUAL_CORE)
+	z_stm32_hsem_unlock(CFG_HW_RNG_SEMID);
+#endif /* CONFIG_SOC_SERIES_STM32WBX || CONFIG_STM32H7_DUAL_CORE */
+
 	return res;
 }

It fixes the error on my side.

*** Booting Zephyr OS build v3.6.0-3809-gf4e65af48c04 ***
[00:00:00.018,000] <inf> fs_nvs: 2 Sectors of 4096 bytes
[00:00:00.018,000] <inf> fs_nvs: alloc wra: 0, f80
[00:00:00.018,000] <inf> fs_nvs: data wra: 0, a0
[00:00:00.042,000] <inf> bt_hci_core: No ID address. App must call settings_load()
Bluetooth initialized
[00:00:00.044,000] <inf> bt_hci_core: Identity: 02:80:E1:00:00:00 (public)
[00:00:00.044,000] <inf> bt_hci_core: HCI: version 1.0b (0x00) revision 0xa072, manufacturer 0x0030
[00:00:00.044,000] <inf> bt_hci_core: LMP: version 1.0b (0x00) subver 0x2172 
Advertising successfully started                                             
Indicate VND attr 0x8018e20 (UUID 12345678-1234-5678-1234-56789abcdef1)

@aescolar aescolar added priority: low Low impact/importance bug area: Power Management labels May 14, 2024
@gumulka
Copy link
Contributor Author

gumulka commented May 15, 2024

@FRASTM It also fixes the error on my side. Happy to see it getting merged into mainline!

@FRASTM
Copy link
Collaborator

FRASTM commented May 16, 2024

@FRASTM It also fixes the error on my side. Happy to see it getting merged into mainline!

Great, I am preparing a PR

@FRASTM FRASTM linked a pull request May 16, 2024 that will close this issue
Open
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@erwango erwango

Labels
area: Power Management bug The issue is a bug, or the PR is fixing a bug platform: STM32 ST Micro STM32 priority: low Low impact/importance bug
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

drivers: entropy: stm32 entropy driver add semaphore protection FRASTM/zephyr
5 participants
@henrikbrixandersen @gumulka @erwango @aescolar @FRASTM