ty

This could be fixed by ``` rust extern crate coio; extern crate env_logger; use coio::Scheduler; use coio::sync::mpsc; fn main() { env_logger::init().unwrap(); Scheduler::new().run(|| { let (tx, rx) = mpsc::sync_channel(1); let h...

I spent some time on reading the source code of the official implementation of mpsc channel, and I think it too complicated to reinvent this wheel. The behavior should have...

Yep. I agree.

But every time the `ready()` is called, you would have to check the parked list and wake one thread up. Please try to implement it and do a simple benchmarking...

Ah, I accidentally merged this branch to master by hand. My fault. We could keep on discussing in the issue.

Well, it seems to be the same problem as #37 .

In this example, I uses `panic::resume_unwind` to trigger a stack unwinding process, which should call all the `Drop::drop()` as I was expected. In the other issue, the program crashes when...

I think those rustc versions in 2017 works.

Yes, my fault. Confirmed not working in: `Linux VM-65-106-debian 4.18.0-1-amd64 #1 SMP Debian 4.18.8-1 (2018-09-18) x86_64 GNU/Linux` `rustc 1.24.1 (d3ae9a9e0 2018-02-27)`

I think so.