Re-opening this since I prematurely merged the previous PR #1122 due to pushing the wrong branch (PEBCAK). Addresses the last review. Original description:
Ok, this escalated a little. I was just looking through the code and saw that the Iwdg class was not inlined, so I quickly refactored it and then noticed that the prescaler was manually computed and then I had to program a constexpr algorithm for it.
[x] Inline IWDG functions
[x] Move into modm::platform namespace.
[x] Add prescaler + downcounter algorithm.
[x] Add unittest for prescaler + downcounter algorithm.
[x] Add Rcc::Lsi and Rcc::Hsi clock.
[x] Add modm.digsep filter for nicely formatting long integers.
[x] Add SystemClock::Iwdg to every board.
[x] Add a generic assertBaudrateInTolerance implementation.
[x] Add documentation for prescaler calculators.
[x] Simplify tolerance assertion functions.
[x] Added seconds_t, milliseconds_t, microseconds_t for template parameters.
[x] percent_t can be float since C++20.
[x] Fix minimum prescaler value for STM32 UART (8 or 16).
[x] Rename from_range to from_linear.
[x] Rename from_iterator to from_range and implement it to be compatible with std::ranges::forward_range.
Re-opening this since I prematurely merged the previous PR #1122 due to pushing the wrong branch (PEBCAK). Addresses the last review. Original description:
Ok, this escalated a little. I was just looking through the code and saw that the Iwdg class was not inlined, so I quickly refactored it and then noticed that the prescaler was manually computed and then I had to program a constexpr algorithm for it.
modm::platformnamespace.Rcc::LsiandRcc::Hsiclock.modm.digsepfilter for nicely formatting long integers.SystemClock::Iwdgto every board.assertBaudrateInToleranceimplementation.seconds_t,milliseconds_t,microseconds_tfor template parameters.percent_tcan be float since C++20.from_rangetofrom_linear.from_iteratortofrom_rangeand implement it to be compatible withstd::ranges::forward_range.