Intermittent deadlock in OpenHMD when rapidly creating/destroying instances.
In the context of some testing, I'm running on a Vive Pro. It works reasonably well for a few tests, then it will seemingly deadlock with this call stack:
Here are the other threads - I wonder if it's trying to "join" the thread that's waiting on a mutex, since that would explain things:
I suspect this might be an OpenHMD issue rather than a Monado issue, but it's possible that our sequence of calls isn't appropriate at handle cleanup