Here's a question that may sound a little naive but I'm behind on where the Rust async stuff is shaping up and I don't know what to search for to answer it. Yo

Here's a question that may sound a little naive but I'm behind on where the Rust async stuff is shaping up and I don't know what to search for to answer it. You can of course already do async things in Rust the same way you do in C: by passing around function pointers and either passing around a handle to the event loop or, more commonly, using a single global handle. But you theoretically could run multiple event loops (one per core, say) and could pass around a pointer to which event loop you wanted to work with and a lot of libevent examples do this. (IMO it's the only correct way to do it, but I have sympathy for wanting to have the global variable.) This is a little different to Javascript where the event loop is part of the language/runtime. Javascript's "the event loop" model means you can't run more than one event loop per runtime, so it's kind of nonsense to want to pass it around explicitly. Rust is more like C in this regard. So if I wanted to mix threads and events, or run an event loop per core, or have an event loop that my Fancy HTTP Library manages itself but exposes a blocking interface to, do I lose access to the new fancy `.await` syntax? There's nowhere in this syntax to say which event loop I'm registering myself with. Is there just one global one? Does it get globally registered on runtime boot, and everyone references it? Or is there a thread-local variable for the "current" event loop that this thread knows about? Even more concerning is how to make _other_ libraries do this than the one I'm writing. If I'm calling into an async library how can I tell that library that I want it to use "an" event loop instead of "the" event loop? What should I be googling for to answer this? :)