Jan Niehusmann

In the mean time I also saw https://github.com/korken89/fugit/issues/35. Which happens to be similar to what hapend to me. I used `u64::MAX` as a placeholder for 'no alarm scheduled' end expected...

Is it really the same situation as `i32`? (BTW, `i32` overflow is not undefined behavior in rust, in contrast to C/C++. But that's just a side note and not relevant...

This could be a valid approach to implement `Ord` correctly but still provide the wrapping-aware comparison: https://github.com/jannic-dev-forks/fugit/commit/a4c78d3e7d8274ffcec4717c8993ad67287778b6

Another option could be to hide the `Ord`/`PartialOrd` implementations behind a feature flag: https://github.com/jannic-dev-forks/fugit/commit/9662eab4ea6edab182adb249bf49a6963156511f That would be a breaking change, so perhaps it should only be announced in the docs...

To stretch the analogy of `UB` a little bit: The way Rust contains `UB` and tries to make sure that no `UB` can pop up in an unexpected situation is...

That marker trait sounds like a good idea. However, deciding that automatically depending on T, NOM and DENOM needs `#![feature(generic_const_exprs)]`, right? Or do you have an idea how such a...

I'd propose to close this ticket due to inactivity. It can always be opened again if desired. https://github.com/rust-embedded/wg/pull/729 was merged, so its not like @reitermarkus is dropped from the WG.

I'd propose to close this ticket due to inactivity. It can always be opened again if desired. https://github.com/rust-embedded/wg/pull/729 was merged, so its not like @reitermarkus is dropped from the WG.

I updated several repos to `dtolnay/rust-toolchain` a while ago, eg.: - https://github.com/rust-embedded/embedded-hal/pull/439 - https://github.com/rp-rs/rp-hal/pull/532 Not saying that this is the best approach, only noting that it seems to work well.

https://github.com/rust-lang/rust/issues/62079#issuecomment-2148180934 suggests closing the ticket for std. For no_std applications, https://crates.io/crates/rand_core provides good support, and implementations can be provided by HALs. So I think this ticket can be closed.