I ve seen this mentioned a lot and there

[random forum]

I considered a Schmitt Trigger Inverter after the divider to cleanup the signal but I guess just a cap and a short timer ought to accomplish it enough, depending on the speed of the bounce.

I think the reason the software debounce isn't working is because the call to get the current timestamp and do the math to find the time from the last call to the interrupt is slower than the next interrupt.

[alpine gyro]

Presumably, you could use a cap or a timer, I don't think you'd need both.

Or just a debouncing chip like a MAX811 (there's also an inverting version, the MAX812, available). It's basically a dedicated timer designed for debouncing (the history of this chip is amusing)

There's also the MC14490 hex debouncer chip, but it needs either an external clock or a few components to implement an oscillator.

[random forum]

I read up more on how a Hall Effect sensor works in practice vs in theory. Almost always, a digitally pulsing Hall Effect Sensor is built with a Schmitt Trigger to cleanup the Analog noise that Hall Effect sensors actually are. It stood to reason that my sensor was probably the same thing, so I put it on a scope, and the signals didn't bounce. So I took apart my test equipment, direct wired everything, tightened down all the links, and all the bounce went away.