STM32F107 not sending data to PC. How can I know what's wrong?

[Mr.Resse]

Here is the code:

#include "stm32f10x.h"                  // Device header
void USART2_ini(void);
void USART_write(int ch);
void delayMS(int delay);
int main(void)
{
	USART2_ini();
	while(1)
	{
		USART_write('H');
		USART_write('I');
		delayMS(100);
	}
}
void USART2_ini(void)
{
	RCC->APB1ENR |= (1<<17);   
	RCC->APB2ENR |= (1<<2);
	RCC->APB2ENR |= 1; 				
	AFIO -> MAPR |= (1<<4);
	GPIOA -> CRL |= 0x00000B00;
	USART2 -> BRR = 0x0341; //Baud rate of 9600 @8MHz
	USART2 -> CR1 |= 0x0008;
	USART2 -> CR1 |= 0x2000; //Enabling USART itself!
}
void USART_write(int ch)
{
	while(!(USART2 -> SR & 0x0080))
		USART2 -> DR = (ch & 0xFF);
}
void delayMS(int delay)
{
	int i;
	for(;delay>0;delay--)
	{
		for(i=0;i<3195;i++);
	}
}
 
 
 

I am new to all of this stuff and also am new to this site and progress of asking questions.

The code is written based on "Bare-Metal"

[Tesla DeLorean]

There's the reference manual.

Working code examples.

A Debugger to look at registers involved.

Scope to confirm output and timings.

Delay looks miles from right, and apt to be optimized away.​

[Mr.Resse]

@Community member​ 

Thanks, could you explain a little bit more about what's wrong with the delay?

[Tesla DeLorean]

Likely needs the variables to be volatile so it's not folded away.

Toggle a GPIO to tune.

Perhaps better to watch time advance via free running TIM, or DWT's CYCCNT, as these will be less dependent on address alignments, and more immune from interrupts.