fix(GPIO): Alternative Function numbers were inverted

[FoniksFox]

Just what the title says + Make inscribes return std::size_t (the result of the add operation) so that they are easier to use from higher domains

[victor-Lopez25]

👍

[jorgesg82]

They are not inverted I think. The AF0 is the LSB. Check the function valid_af:

inline constexpr bool valid_af(const AlternateFunction af) const {
  if (af == AlternateFunction::NO_AF)
    return true;
  return ((1 << static_cast<uint8_t>(af)) & afs) != 0;
}

[Copilot]

Pull request overview

Copilot reviewed 1 out of 1 changed files in this pull request and generated no new comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[FoniksFox]

They are not inverted I think. The AF0 is the LSB. Check the function valid_af:

inline constexpr bool valid_af(const AlternateFunction af) const {
  if (af == AlternateFunction::NO_AF)
    return true;
  return ((1 << static_cast<uint8_t>(af)) & afs) != 0;
}

@jorgesg82

What I meant with inverted is with respect to the casting made in the build method, alternate functions have a 1 to 1 correspondence (AF10 == 10) so currently the casting doesn't interpret correctly the alternate function.
I didn't take into account the valid_af method, we should either invert the bitmaps, the valid_af method or leave it as it was before and only invert the AF in the cast...
Inverting the bitmaps in Pin.hpp would certainly be the better approach.
As for the AF10 being equal to 10 and so on, you can check it out in stm32h7xx_hal_gpio_ex.h (I think the general AF10 and so on were defined somewhere else, but I can't find where; either way, the _ex.h file is example enough).
You can also check PinModel/Pin.hpp (legacy Pin): the alternative functions are defined the same way I am doing here.

[jorgesg82]

From my point of view, the instantiated pins in Inc/HALAL/Models/Pin.hpp, the AlternateFunction enum, and the valid_af function match one with each other. I think the only thing that does NOT match with the rest is the cast in the build function, so changing that to match with the others will be the less disturbing fix, don't you think?

Otherwise you will have to invert the enum, to modify the valid_af function, and to change all the bitmaps (that would be for sure the most painful thing of them)

[FoniksFox]

From my point of view, the instantiated pins in Inc/HALAL/Models/Pin.hpp, the AlternateFunction enum, and the valid_af function match one with each other. I think the only thing that does NOT match with the rest is the cast in the build function, so changing that to match with the others will be the less disturbing fix, don't you think?

Otherwise you will have to invert the enum, to modify the valid_af function, and to change all the bitmaps (that would be for sure the most painful thing of them)

Yeah, that's true, will do that, I'll also add a comment to the enum clarifying that

[victor-Lopez25]

I like returning std::size_t in inscribe since it allows doing more interesting things elsewhere but remember to change the wiki for it

[FoniksFox]

I like returning std::size_t in inscribe since it allows doing more interesting things elsewhere but remember to change the wiki for it

Yeah, will do the moment this pr is approved

[oganigl]

It seems fine to me.

[oganigl]

It seems fine to me.

Already implemented what was asked in the review exactly as it was requested