RCC configuration results in HardFault (STM32F411CEU6).

[YvanR]

I have a board that is running well. There are about 50+ PCBs in production operating fine. The firmware is pretty complex, but the hardware relatively simple, just some SPI devices and some PWM outputs.

I revved the board, and the new board doesn't work at all. Well, it works enough to execute some code, but not much. So to debug, I started with a blinky program:

int main(void)
{
 HAL_Init();
 SystemClock_Config();
 MX_GPIO_Init();
 
 while (1)
 {
	 HAL_GPIO_TogglePin(FAULT_GPIO_Port, FAULT_Pin);
	 HAL_Delay(1000);
 }
}

If I configure the RCC to use the HSI, the program gets as far as `HAL_Delay(1000)` at which point it chokes on `HAL_GetTick()`.

If I configure the RCC to use the HSE (a crystal is present), the program only gets as far as `SystemClock_Config()`, where it again chokes on `HAL_GetTick()` or sometimes `HAL_RCC_OscConfig()`.

By "choke" I mean it throws a HardFault. Though for a reason I can't understand, it won't execute my handler.

So, the code runs ok using the HSI at 16MHz until the RCC is configured. Then, whether configured to use the HSI/PLL or HSE/PLL, it continues to execute until the first HAL_Delay().

I don't think there is anything about `HAL_Delay` specifically, since all it does is return a `volatlile int`; but rather execution becomes unstable once RCC is configured.

So, what could be causing RCC to freak out?

Power supply is a smooth 3.300V and there are 4 100nF bypass caps, one per Vdd/Vss pair, and Vcap has a 4.7uF.

The only thing I see that I missed is that BOOT0 doesn't have a pull-down.

A redacted schematic is attached.

[Tesla DeLorean]

Check PLL gearing. Is appropriate for 25 MHz HSE, and adequate wait states on flash.

Check HSE_VALUE in stm32f4xx_hal_conf.h

[YvanR]

Thanks for the suggestion.

I checked the PLL configuration for sanity. I also removed PLL from the equation and ran HSI directly at 16MHz -- same issue.

So I went completely back to backs and created an empty project. The only thing I configured was a GPIO output to flash an LED on. -- It didn't work.

So, I flashed the same code to an off-the-shelf "black pill" and it worked. So I'll take another look at the PCB to see what I can come up with. I'll start with the floating BOOT0.

I'll report back in case I find anything interesting.

[YvanR]

I'm back. As I suspected the issue was that I left BOOT0 floating. After wiring it to ground, the blinky sanity check operates normally.

With BOOT0 left to float, execution does predictably commence from flash at 0x08000000; but further execution is erratic.