07:39 <gauteh[m]> Hi, I'm storing a pin in a Mutex<RefCell<Option>>> to access it from an GPIO interrupt. However, before I even get to that the static is being reset to None...... (full message at <https://catircservices.org/_irc/v1/media/download/Aep7j23TdayvBTHfNXWYY0O98xFCqBub3aT5E2CQXhu9woVXH6WRLUVvlA8o7UDsNCbeYaGWiMLD8FRuxG3kEAxCeSmaIR_wAGNhdGlyY3NlcnZpY2VzLm9yZy9SZ3l0WklEa0hVVndTTENKenRrdVlHT0M>)

07:40 <gauteh[m]> * Hi, I'm storing a pin in a Mutex\<RefCell\<Option>>> to access it from an GPIO interrupt. However, before I even get to that the static is being reset to None...... (full message at <https://catircservices.org/_irc/v1/media/download/AdLdTYRkKPiy3ypy0A-2dX8zD2FIv9iI_eSKFmRtbAgaVM9fevUyzzxRtCNIMGhdNVMvxK9GAc8YTft48z8n2wRCeSmaJsuAAGNhdGlyY3NlcnZpY2VzLm9yZy9kYUxxUVlIekJ3anpFV0JLdGJJbXF1ZGo>)

07:41 <gauteh[m]> At this point interrupts are not enabled yet. This is a single core MCU.

07:42 <gauteh[m]> * Hi, I'm storing a pin in a Mutex\<RefCell\<Option>>> to access it from an GPIO interrupt. However, before I even get to that the static is being reset to None...... (full message at <https://catircservices.org/_irc/v1/media/download/AZDPAUbt599vXeDsSnMBif-WaB8BmL2s22Abr2S87AtS99MMlVge4oD_oBQUTD4b2mJFYIn0cVXTY5k9k7pJv-ZCeSmaTCmAAGNhdGlyY3NlcnZpY2VzLm9yZy94UEVyeVVDWkdSYlFQZ05pVEZhWGJVU1Q>)

08:09 <gauteh[m]> Yeah, looks like stack overflow.

09:13 <thejpster[m]> jamesmunns: I think the OOM handler on Linux doesn’t terminate the process that wanted the memory - it terminates a random process weighted by how much memory they’ve allocated. So the odds are good that in fact it’s the program that ate all the memory that gets terminated and not the poor process that just wanted a few more bytes for something.

09:14 <thejpster[m]> I’ve listened to two back to back and I very much enjoyed the telegraph analogy.

13:38 <JamesMunns[m]> <thejpster[m]> "jamesmunns: I think the OOM..." <- TIL!

15:05 <Ralph[m]> <gauteh[m]> "Yeah, looks like stack overflow." <- do you know [flip-link](https://crates.io/crates/flip-link)? it "adds zero-cost stack overflow protection to your embedded programs"

17:50 <dirbaio[m]> rust 1.82 is out https://blog.rust-lang.org/2024/10/17/Rust-1.82.0.html

18:00 <gauteh[m]> <Ralph[m]> "do you know [flip-link](https..." <- I couldn't get it to work

18:07 <JamesMunns[m]> There's a lot of static mut and unsafe in your code gauteh, at least some of it is at least theoretically UB

18:07 <JamesMunns[m]> For example... (full message at <https://catircservices.org/_irc/v1/media/download/ARaQ01WiPuoc2Ob8UM8OGShghwOKdcd8jyFLcAOIb9y74ShUvHbHAo_1T09dmsGzsfGZj-p9yV_XlPetnkE4CG1CeSm-EqOQAGNhdGlyY3NlcnZpY2VzLm9yZy9sQ0JJV0dLWWNVTFZla3pzYlFhWWdLakU>)

18:08 <JamesMunns[m]> you put a pointer to note in log::NOTE and then continue using a mutable reference to it

18:09 <gauteh[m]> yeah 🫣

18:09 <gauteh[m]> I don't think the compiler noticed

18:09 <JamesMunns[m]> I mean, you're asking us about debugging things that "seem impossible"

18:09 <JamesMunns[m]> :/

18:10 <gauteh[m]> fair enough. it did turn out to likely be stack overflow though, i reduced some buffers and queues and it was ok. seems pretty likely that the static A2 was overwritten with those

18:11 <JamesMunns[m]> fair!

18:12 <gauteh[m]> I use the reference to the note(modem) to try and do a last-ditch attempt at sending panics home before resetting: https://github.com/gauteh/sfy/blob/2597c7688655ec2c161ffbf06c504e520b7ec986/sfy-buoy/sfy-artemis/src/main.rs#L536

18:13 <gauteh[m]> I hope I only use the static reference in the panic handler/hardfault, since I have access to it the regular way usually.

18:19 <huayra1[m]> what is the "recommended" way to do c++ bindings? i know of cxxbridge and include macro. btw, is it normal for a file using cxxbridge to take like 15 seconds?