Le Sat, Dec 09, 2023 at 11:36:48PM -0800, Greg Steuck a écrit : > Greg Steuck <gne...@openbsd.org> writes: > > > So I was wondering which program is responsible (because my previous > > "fix" was to restart the X session). Turns out exiting firefox is > > enough to get the device back. > > I don't understand how or why firefox decides that it should hold on to > fido devices, but hibernate isn't necessary. Right now I'm seeing:
i have absolutely no knowledge about fido nor rust, but the code dealing with this is in https://github.com/mozilla/authenticator-rs/blob/ctap2-2021/src/transport/openbsd/monitor.rs (that's "upstream") - or more actually https://searchfox.org/mozilla-central/source/third_party/rust/authenticator/src/transport/openbsd/monitor.rs for firefox trunk and https://searchfox.org/mozilla-release/source/third_party/rust/authenticator/src/transport/openbsd/monitor.rs for firefox releases. the authenticator-rs crate has had much development upstream for something called 'ctap2' (whatever that is) and im not fully sure corresponding code has been done properly for openbsd, if that matters. Landry
