13:41 <thejpster[m]> ping on https://github.com/rust-embedded/cortex-m/pull/605 again for the arm-team.

13:42 <thejpster[m]> we really should get a 0.7.8 out soon

15:44 <thejpster[m]> don't ask me how much time I lost today because `bkpt 0xAB` is a HardFault when in unprivileged mode....

15:44 <thejpster[m]> https://github.com/rust-embedded/cortex-m/pull/594 has some updates

17:22 <Pete[m]1> Ok, first time really playing with no_std rust.. Design question: I'm trying to design a system where a "provider" type can be implemented outside the library, so using a trait, and the trait needs a function that can return a Stream/AsyncIterator that generates a stream of data/events to be consumed... I can't seem to find a reasonable solution that doesn't drag in Box and require an allocator. Any ideas?

18:55 <diondokter[m]> <Pete[m]1> Ok, first time really playing with no_std rust.. Design question: I'm trying to design a system where a "provider" type can be implemented outside the library, so using a trait, and the trait needs a function that can return a Stream/AsyncIterator that generates a stream of data/events to be...

18:55 <diondokter[m]> The trait can have an associated type. Then the impl decides how it's done

18:56 <diondokter[m]> There are a lot of primitives for streaming/sharing data: https://crates.io/crates/embassy-sync

18:56 <diondokter[m]> In embedded we pretty much always make these with a bounded size

19:29 <vollbrecht[m]> Did anybody had a look at the BlueOS kernel? It seams to have landed couple days ago on [github](https://github.com/vivoblueos/kernel/tree/main) and already got 500 stars.

19:29 <vollbrecht[m]> its a fully in rust written kernal/OS that also sources some dependency like embedded-io heapless smoletcp etc.

19:49 <thejpster[m]> interesting. They use gn rather than cargo.

19:52 <Pete[m]1> diondokterThanks. I'll explore there.

19:53 <thejpster[m]> it's POSIX, but they run tests on the QEMU MPS2-AN385 machine, which uses an Arm Cortex-M3, and the MPS3-AN547 which is a generic Armv8-M Mainline based machine (like a Cortex-M33).

20:14 <Ralph[m]> what's their goal?... (full message at <https://catircservices.org/_irc/v1/media/download/AVJLH9osIW8s7ghyHkPUAnocEeOIi7sOf91CM8ijCNL63f-C_BXpG3V0YbWir7l_zI1KuTu-i76Pl8K_7sTSSM-_8AAAAAAAAGNhdGlyY3NlcnZpY2VzLm9yZy9EenltbWF3cGRJY0FRak56b3VlekRxa1Q>)

20:30 <vollbrecht[m]> they are building smartphones afaik, and since many Chinese firms are needing to ditch Google/Android i think this is one of there attempts here

20:34 <thejpster[m]> I'd guess more like smart watches and smart bulbs rather than mobile phones. Otherwise why target Cortex-M?

20:35 <vollbrecht[m]> yeah that would make more sense, was just a wild guess by me looking at there background

20:36 <thejpster[m]> I read the Wikipedia page and I feel I know even less: https://en.wikipedia.org/wiki/BlueOS

20:36 <thejpster[m]> "Including devices that has low as 32 MB of RAM"

20:43 <thejpster[m]> I'm kind of stuck here: https://github.com/thejpster/pico-term-rs/blob/14be7f77c7dd7a4613f97c4f9bf7838532b4fe50/core0/src/main.rs#L194.

20:43 <thejpster[m]> If anyone has any ideas on how to handle cotton UsbEvents and interrupt endpoint reports from a future::Stream, please let me know

21:05 <dirbaio[m]> You can make two tasks each with it's own loop

21:05 <dirbaio[m]> Or put each loop in an async{} block then join() them

21:05 <dirbaio[m]> Or you can match select(stream1.next(), stream2.next()).await

21:09 <dirbaio[m]> ah well, I see the hid object comes from the "connect" event. maybe you can have a dedicated task for handling the hid device and spawn it when connected, and let it exit when it disconnects so you spawn it again later when reconnected

21:15 <thejpster[m]> I'll try that

21:15 <thejpster[m]> thanks