Timo

Yeah I agree catching panics really shouldnt be done for error handling à la try-catch, but panics are also per thread, so you can panic on one thread and the...

Had to rebase because the PR was getting too far behind. The last commit should contain the changes for (most of) the review comments. Also had to split the test...

Sorry for the wait, am a bit busy again, I'll try to get to the rest of this in 2-3 weeks @rustbot author

After having thought about this more for a while, I'm not sure if some of the logic in here is really necessary/worth the complexity. For example, checking if the `.spawn()`...

@xFrednet The FCP thread hasn't gotten comments but 4 upvote reactions and it's been about a month -- do you think we're good? I guess it's also waiting on me...

Did #12040 not also fix this? It added a call to `is_from_proc_macro`, and it already uses `in_external_macro`. Fix isn't on nightly yet though.

@okynos I can't see the linked page, but I ran clippy on your project and I assume you mean clippy linting on this `match`? https://github.com/Achiefs/fim/blob/c3d1f23bf2df51f8b2a8d9a7c1402a2b7a0a65b9/src/monitor.rs#L146-L147 In that case, it doesn't...

It'd be good to have some kind of MCVE for this, because a simple async fn annotated with `#[tracing::instrument]` and a match is apparently not enough to trigger it: https://play.rust-lang.org/?version=nightly&mode=debug&edition=2021&gist=2e011e037ed64ff5003b40b78c61605d...

Yes, the fix will be in 1.81. 1.80 branched on June 7th and the referenced PR was merged on June 8th. [releases.rs](https://releases.rs/docs/1.80.0) can be useful for this.

Looks like I was too slow :) #12287