Julien Cretin

Of course, that would be great. That's a good first issue. You can start by changing those 2 lines: https://github.com/google/wasefire/blob/5da7a5e0de8545825ce6ca023c389f2e35926c02/scripts/sync.sh#L33-L34 into (I'm removing `missing-debug-implementations` because I'm not yet convinced it's...

To debug: - Could you push your working directory to a branch on your fork? - What is your operating system? - What is your version of `sed`? (`sed --version`)...

Could this be related to https://github.com/knurling-rs/defmt/issues/825#issuecomment-2624109853 where probe-rs seems to not handle high RTT throughput?

Also the issue is still present with 0.27.0.

https://github.com/knurling-rs/defmt/issues/825#issuecomment-2124663112 is probably the most simple demonstration of the issue. You can easily modify `app-template` like I did in this branch: https://github.com/ia0/app-template/tree/issue/probe-rs/3045 I get a similar output as in the...

> The only data loss I see happens during boot Weird... I've added a `bug.sh` script to the branch above. Here's my output running `./bug.sh`. This really looks to me...

Ok so this may help corner the issue, it could be specific to any of: - nrf52840 vs nrf52833 - linux vs windows I don't have any other board than...

Thanks for the feature request. Panics should only be used for unrecoverable errors. That's a very good point and [hopefully](https://users.rust-lang.org/t/is-downgrading-a-panic-to-an-error-a-breaking-change/124756) something that can be fixed without a breaking change. I'll...

Sadly, it would be a breaking change so I've added the feature request to #72 and is tracked in #106 for v3. If this is still an issue for you,...

I've released 2.8.0 which provides [`BASE32_NOPAD_NOCASE`](https://docs.rs/data-encoding/latest/data_encoding/constant.BASE32_NOPAD_NOCASE.html), which saves the need for converting to uppercase first. I'm keeping this issue open for the initial concern: returning an error instead of panic...