Kiss Lorand

You said there's no need for benchmarks, but you just intrigued me. I will make some benchmarks and will come back with the results. It might take an hour or...

I am already on it. ;)

I have the results. :) 

Actually the two codes compiled do perform the same, the compiler does its job quite well. The difference above could be a rounding issue as the timer is counting microseconds...

> May I ask what nationality you are? (I'm Dutch) There's no secret, I am Hungarian. You could already tell that since I made many additions/changes to the Hungarian language...

What speed increase did your test show for your proposed code?

Opinions are always welcome but taken with a grain of salt.

> @kisslorand has seen the light Actually what I've seen is the rookie mistake I made benchmarking the original function and mine. In the benchmark I didn't use the variables...

Up/Down count and comparison to zero results:  Upcount: count from 0 to X Downcount: count from X + 10 down to 10 CompToZero: count from X down to 0...

In many microcontroller architectures, including those that use the STM32 series with the STM32 HAL library, clearing the interrupt flag at the end of the ISR is the recommended and...