04:55 <re_irc> <@sajattack:matrix.org> newam:matrix.org: Car no do that, car go road

06:54 <re_irc> <@henrik_alser:matrix.org> dbrgn:matrix.coredump.ch: Morning! Yeah do so! What device is it?

06:56 <re_irc> <@henrik_alser:matrix.org> May wanna check the errata of the part for any i2c issues

06:57 <re_irc> <@henrik_alser:matrix.org> You could try inserting a delay between enabling the peripheral clock and before writing and see if there’s a difference

06:59 <re_irc> <@henrik_alser:matrix.org> I think i’ve seen issues in the past with some part in the f4 family where you had to insert a delay between reseting and re-enabling the clock, don’t know if the hal does this already? (This was before my rust days)

07:06 <re_irc> <@henrik_alser:matrix.org> Using stm32f4xx-hal with pin typestates right? So you know the pins are in the right configuration?

07:12 <re_irc> <@dbrgn:matrix.coredump.ch> henrik_alser: it's a MiniF4 devboard ("black pill") with a RTC module. I'll hook up the scope tonight. Thanks for the suggestions!

07:14 <re_irc> <@henrik_alser:matrix.org> 401 or 411 version?

07:15 <re_irc> <@henrik_alser:matrix.org> I’ve seen no issues with the 411 one

11:12 <re_irc> <@jordens:matrix.org> Is there something (example code, macro) that asserts that an `unwrap()` on an infallible Result actually gets optimized away? Maybe using a linker trick? Like a compile-time `unwrap()` that asserts both infallibility and does not generate any unwrap bloat.

11:20 <re_irc> <@henrik_alser:matrix.org> jordens: How about `.unwrap_or_else(|| unsafe { core::hint::unreachable_unchecked() })`

11:23 <re_irc> <@adamgreig:matrix.org> maybe https://github.com/dtolnay/no-panic ?

11:23 <re_irc> <@adamgreig:matrix.org> there's panic-never for crate-wide guarantee but it's quite hard to meet

11:25 <re_irc> <@rahix:matrix.org> jordens: the `void` crate kind of does this, if I understand you correctly?

11:25 <re_irc> <@rahix:matrix.org> if the `E` type of the result is uninhabited, no match arm for it can ever be code-gen'd.

11:28 <re_irc> <@jordens:matrix.org> Thanks all. Lots of good options!

11:29 <re_irc> <@henrik_alser:matrix.org> https://doc.rust-lang.org/core/hint/fn.unreachable_unchecked.html

11:47 <re_irc> <@jordens:matrix.org> henrik_alser:matrix.org: Ack. But that's only the optimization part. Whether it is in fact unreachable remains `unchecked`.

11:48 <re_irc> <@henrik_alser:matrix.org> Ahh sorry i misunderstood your question :)

12:42 <re_irc> <@korken89:matrix.org> jordens:matrix.org: If you use a nightly compiler you can use the following flags to remove all panics `-Zbuild-std=core, -Zbuild-std-features=panic_immediate_abort`

12:46 <re_irc> <@rahix:matrix.org> well, but this does not assert whether unwraps are infallible. it just makes them use less space in the binary (at the expense of not being able to catch any panic at all)

12:57 <re_irc> <@wim:dewith.io> Couldn't you use `unreachable_unchecked!()`?

15:42 <re_irc> <@qthree:matrix.org> rahix: Hello, I'm trying to use avr-hal and it looks like #[derive(PartialEq)] produces miscompilation when applied to struct with const-generics and array inside.

15:44 <re_irc> <@qthree:matrix.org> #[derive(PartialEq)]

15:44 <re_irc> <@qthree:matrix.org> fn main() -> ! {

15:44 <re_irc> <@qthree:matrix.org> struct Foo<const BAR: usize>([bool; BAR]);

15:44 <re_irc> <@qthree:matrix.org> let dp = arduino_hal::Peripherals::take().unwrap();

15:44 <re_irc> ... long message truncated: https://psion.agg.io/_matrix/media/r0/download/psion.agg.io/89a618db6b19d81a6bae6d108991c7fd5f9475e4e0570479c93e90b206c6645e (12 lines)

15:44 <re_irc> <@qthree:matrix.org> Console /dev/ttyUSB0 at 57600 baud

15:44 <re_irc> <@qthree:matrix.org> Ooops!

15:44 <re_irc> <@qthree:matrix.org> Hello World!

15:47 <re_irc> <@qthree:matrix.org> impl<const BAR: usize> core::cmp::PartialEq for Foo<BAR> {

15:47 <re_irc> <@qthree:matrix.org> self.0.iter()

15:47 <re_irc> <@qthree:matrix.org> Manual implementation works fine

15:47 <re_irc> <@qthree:matrix.org> fn eq(&self, other: &Self) -> bool {

15:47 <re_irc> ... long message truncated: https://psion.agg.io/_matrix/media/r0/download/psion.agg.io/5c73ba9eba20e52ca42fa0d128f08a9376f8931544486256902577eff505d6e2 (8 lines)

15:50 <re_irc> <@rahix:matrix.org> qthree: please post this to #avr-rust_Lobby:gitter.im , that's where you'll find the right people for rust on avr :)

17:31 <re_irc> <@emilgardis:matrix.org> can I get someone to r+ https://github.com/rust-embedded/cross/pull/602 :)

19:43 <re_irc> <@gauteh:matrix.org> Hi, I'm trying to get an RTC alarm to work with firing an interrupt every second. Trying to debug this I put panic! in the interrupt handler, this should halt everything right? I have an `wfi`, but once I enable the alarm it just keeps spinning. If I don't enable the specific interrupt in NVIC nothing happens. So now I wonder if the interrupt is hooked up at all..

20:42 <re_irc> <@jamwaffles:matrix.org> Are you clearing the interrupt flag in the handler? The interrupt will keep firing forever if you don't

21:23 <re_irc> <@gauteh:matrix.org> jamwaffles: yes, tha

21:35 <re_irc> <@jamwaffles:matrix.org> Hm, not sure then :/. Have you got some code you could paste in?

23:58 <re_irc> <@firefrommoonlight:matrix.org> What MCU?

23:59 <re_irc> <@firefrommoonlight:matrix.org> Note that on Stm32, it's the wakeup timer you're looking for. Alarm is something different