A little of history.
I was working on a board that had a stm32f407, where it had a partial code and printf function worked, so long.
Soon after, I received an update to stm32f429 at that moment gave $% & #.
Try debugging a third-party code where you do not have to read lines and lines of code to understand the problem.
Add to this a third-party rtos, it will stink, at the very least!
The tip is, I stopped debugging when I realized and I could not understand where the bad smell came from.
To reproduce the problem:
tools produced;
purchased tools;
tools used;
By ST.
Putting it together STM32CubeMX 5.1.0 + Atollic TrueSTUDIO + Freertos.
Put a printf inside a task. Of course with a usart port configured.
It certainly will not work.
First add this in your usart.c without changing even a line in syscalls.c.
/ * USER CODE BEGIN Includes * /
#include <errno.h>
#include <unistd.h>
/ * USER CODE END Includes * /
/ * USER CODE BEGIN 1 * /
int_read (int file, char * ptr, int len)
{
int n;
int num = 0;
uint8_t ch = '\ 0';
HAL_StatusTypeDef status = HAL_TIMEOUT;
switch (file) {
case STD_FILEN:
for (n = 0; n
of {
status = HAL_UART_Receive (& huart1, (uint8_t *) & ch, 1, 1);
HAL_UART_Transmit (& huart1, & ch, 1, 1);
} while (status! = HAL_OK);
* ptr ++ = ch;
num ++;
}
break;
default:
errno = EBADF;
return -1;
}
return num;
}
int _write (int file, char * ptr, int len)
{
int n;
switch (file) {
case STDOUT_FILENO: / * stdout * /
for (n = 0; n
HAL_UART_Transmit (& huart1, (uint8_t *) ptr ++, 1, 1);
}
break;
case STDERR_FILENO: / * stderr * /
for (n = 0; n
HAL_UART_Transmit (& huart1, (uint8_t *) ptr ++, 1, 1);
}
break;
default:
errno = EBADF;
return -1;
}
return len;
}
/ * USER CODE END 1 * /
To solve the problem, now add a printf before MX_FREERTOS_Init ();
printf ("running working ... \ r \ n");
MX_FREERTOS_Init ();
It will certainly work.
I love seeing these **** done by programmers paid by the contractors themselves "bosses", and nothing is checked.
Please disregard words of anguish.
