06:40 <jr-oss> I tried embassy on STM32 Discovery board and had problems with usart (with parity) and clock settings. You can find my attempt to fix it on https://github.com/jr-oss/embassy/tree/stm32_fixes (3 commits). Shall I start a PR or is something missing?

06:44 <re_irc> <lulf> jr-oss: the changes look good to me 🚀, feel free to raise a PR.

06:45 <jr-oss> lulf: Thanks, will do

07:06 <jr-oss> lulf: Do you prefer a force push (and a clean history) to fix formatting or just another commit

07:08 <re_irc> <lulf> jr-oss: in this case either case works, but force push is fine.

07:08 <re_irc> <lulf> * works IMHO,

07:08 <re_irc> <lulf> jr-oss: also, you might want to hang out in #embassy-rs:matrix.org (https://matrix.to/#/#embassy-rs:matrix.org)

07:14 <jr-oss> lulf: Have force pushed the changes. Sorry for not having noticed beforehand. I am not used to matrix and my first attempts always required installing a client, so I haven't look into it any further.

07:17 <re_irc> <lulf> No worries, just thought I'd let you know it exists, it takes some time to discover all the places to hang out 😅

09:32 <re_irc> <ryan-summers> How does rust decide which version of a function to use when there's a name conflict? I have an external trait that defines "iter()" on a struct, but that name is identical to a primitive slice's iter() function

09:32 <re_irc> <ryan-summers> And Rust is picking the trait impl when I want it to use the primitive function instead

09:35 <re_irc> <ryan-summers> Huh, Rust docs indicate that the inherent method should always be called first, and you'd have to explicitly override that to treat it as a trait object to get the trait impl. I'll dive a bit deeper on what's happening here

09:37 <re_irc> <ryan-summers> Ah very interesting. Has to do indeed with Rust's coercion order. In this case, it was a native array, which will defer to the trait object before trying to coerce into a slice and looking up functions

09:54 <re_irc> <Robert Jördens> You can coerce explicitly: "x.as_slice().iter()"

09:56 <re_irc> <ryan-summers> Yeah, you can also do x[..].iter() - just a strange oddity

09:56 <re_irc> <ryan-summers> * "x[..].iter()"

11:07 <re_irc> <Chris [pwnOrbitals]> Hey everyone, I’m building for stm32 and using some C code along Rust with cc. Is there any way to control how/which libc is used ? I’d like picolibc but not sure what would be the best way to do that

11:21 <re_irc> <wassasin> Is there a way currently to get to a HardFault in RTIC? I see https://github.com/rtic-rs/cortex-m-rtic/issues/285 but no resolution.

12:14 <re_irc> <9names (@9names:matrix.org)> i think this will "just work" if you use you the gcc from your picolibc toolchain as your linker

17:36 <re_irc> <Phil Markgraf> James Munns: Is there a way to do incremental CRC calculation when receiving a COBS encoded messages? (Thinking of Postcard 1.0 features...) - It seems like this should be possible on the receive side, as you know where the zeroes are going to be, before they are received.

17:38 <re_irc> <James Munns> Maybe, though you'd also need to do incremental cobs decoding

17:39 <re_irc> <James Munns> at least - I'd imagine you'd need to put your CRC inside the COBS framing, since you can't guarantee the CRC doesn't have any zero bytes

21:09 <re_irc> <Jason Qin> i want to use rust to program an arduino mega. Does embassy support avr targets?

21:09 <re_irc> <dirbaio> no :(

21:10 <re_irc> <dirbaio> check out https://github.com/avr-rust and #avr-rust_Lobby:gitter.im (https://matrix.to/#/#avr-rust_Lobby:gitter.im)