00:12 <korbin[m]> <PhilMarkgraf[m]> "In a "Rust is not C" moment, I'..." <- > <@shakencodes:matrix.org> In a "Rust is not C" moment, I've discovered that Rust does not guarantee the order of data in structs (unless you set the repr attribute.) I have deployed hardware with firmware written in Rust that has data in a Rust data... (full message at <https://catircservices.org/_matrix/media/v3/download/catircservices.org/CBEMqyPUQQncspbjrimJihjx>)

00:17 <JamesMunns[m]> At least with deployed devices with concrete formats, you have a specific subset of instances to check against, and likely toolchain versions from CI as well!

00:17 <JamesMunns[m]> s/formats/versions/

11:59 <thejpster[m]> someone just published arrayvec 0.7.5 and now my builds are failing: https://github.com/rp-rs/rp-hal/actions/runs/10431977859/job/28892214430

11:59 <thejpster[m]> looking at the diff, I can't tell why.

12:00 <thejpster[m]> But if you check out github.com/rp-rs/rp-hal and `cargo update && cargo build --target=thumbv6m-none-eabi --all-features --examples` you'll get:... (full message at <https://catircservices.org/_matrix/media/v3/download/catircservices.org/AmZLIvReTTmAJaAqwmKCZzBm>)

12:09 <tamme[m]> looks like they are already working on a fix https://github.com/bluss/arrayvec/issues/273

12:11 <thejpster[m]> ahhh, they do:... (full message at <https://catircservices.org/_matrix/media/v3/download/catircservices.org/CtgZTbxXCIHxyCALmdHQKyvE>)

12:50 <Lumpio[m]> Why would you do that...

17:47 <YuhanLin[m]> how did that get past code review lmao

18:16 <thejpster[m]> Tired: Sending data over QuadSPI

18:16 <thejpster[m]> Wired: Sending data over the 4-bit Trace Port at 150 MTx/sec

19:00 <thejpster[m]> rust-analyzer is killing my 16GB RAM desktop. I need a bigger computer.

20:52 <rmsyn[m]> @mvirkkunen: from what I've seen, `extern crate core as std` is used as a convenience in crates that build under `no-std` and `std`. it's to avoid doing feature-gates everywhere in the code

20:57 <Lumpio[m]> Yeah but like, why not just directly import stuff from core instead of making a confusing alias like that

20:59 <rmsyn[m]> because there are a number of overlaps in `std` and `core`, avoids context-switching, and avoids explicitly importing the `core` crate in the `std` build.

21:00 <rmsyn[m]> it mostly comes down to a matter of preference, though. you can achieve similar results in multiple ways

21:01 <rmsyn[m]> once you've seen it a couple times, it's really not that confusing

21:02 <rmsyn[m]> the first time I saw it with a similar error to above, it definitely tripped me up, too :shrug: