ESP32 crashes if I use vTaskDelay in app_main function

[mdaupu]

Hello everyone,
I have written a simple program using amazon freertos. I am using ESP32-WROOM-32UE module. If I use vTaskDelay in the app_main function, the ESP32 restarts. But if I use 9 or less than 9 value in the vTaskDelay, it works.

My question is, why esp32 restarts if I use more than 9ms delay in the app_main function?


My Code:

Code: Select all

int app_main(void)
{
      printf("Hello World \r\n");
      prvMiscInitialization();
      config_gpio();
      display_task_create();
      while(1){
        vTaskDelay(100 / portTICK_PERIOD_MS);
      }
  return 0;
}

Error Code:
Guru Meditation Error: Core 0 panic'ed (Unhandled debug exception)
Debug exception reason: BREAK instr
Core 0 register dump:
PC : 0x40081258 PS : 0x00000016 A0 : 0x40080306 A1 : 0x3ffc01b0

[ESP_Sprite]

Hard to say. Perhaps the initialization routines you call before the vTaskDelay corrupt something? What happens if you comment those out?

[abansal22]

Hello everyone,
I have written a simple program using amazon freertos. I am using ESP32-WROOM-32UE module. If I use vTaskDelay in the app_main function, the ESP32 restarts. But if I use 9 or less than 9 value in the vTaskDelay, it works.

My question is, why esp32 restarts if I use more than 9ms delay in the app_main function?


My Code:

Code: Select all

int app_main(void)
{
      printf("Hello World \r\n");
      prvMiscInitialization();
      config_gpio();
      display_task_create();
      while(1){
        vTaskDelay(100 / portTICK_PERIOD_MS);
      }
  return 0;
}

Error Code:
Guru Meditation Error: Core 0 panic'ed (Unhandled debug exception)
Debug exception reason: BREAK instr
Core 0 register dump:
PC : 0x40081258 PS : 0x00000016 A0 : 0x40080306 A1 : 0x3ffc01b0

Could you post the entire code and also let me know from where you get the source I could execute and let you know if same problem is with me or not!

[PeterR]

Great offer!

Or try adding 10K to your main & display task. You have a simple program so do 10/20K as a no brainer - you have the ram. Premature optimisation & all....
9 out of 10 crash issues tend to be stack size (assuming you know one end of a pointer from another).
LOG/printf() will burn a lot of stack & stack issues are hard to debug as you cannot diagnostic printf() without dieing (doh!)

If +10K stack doesn't do you then your into daft pointer issues IMHO.

[chegewara]

Just idea, are you sure that display_task_create(); task does not exit?
I see you are using esp-idf v3.3 and im not 100% sure that exiting task will display proper error message.

[mdaupu]

Thank you guys for your replies. I have found the solution of this issue. In the main.c file, there is a define named IDLE_TASK_SIZE. The value was 100. I have changed it to 1000. Now I can use different vTaskDelay in the app_main function. The ESP32 does not reset now.