Qi

> I wonder whether this is really worth it. @Darksonn This is a reasonable point, the figures in real world should not look like that, let's hold on the merge...

Added the `S-blocked` label, I need to work out a proper benchmark script for the time subsystem to see if there is a significant improvement in a scenario that more...

Converted to draft, waiting benchmark.

> much more volatile than 1.44.2. In

I think closing the fd before dropping the `AsyncFd` breaks the contract mentioned in the docs. > The [AsyncFd](https://docs.rs/tokio/latest/tokio/io/unix/struct.AsyncFd.html) takes ownership of an arbitrary object to represent the IO object....

My gut tells me that this issue involves several classic engineering topic: * “Let it crash" vs. "Keep it running" vs. "Best effort" * "Unrecoverable error" vs. "Recoverable error" We...

> I don't think there are cases where we can't clean up the list on _de_ register. ELOOP is only for registration. You are right, now, I also think we...

> > Propagated the error out of the AsyncFd (panic? or Result?). > > I believe this operations happens on `AsyncFd` drop. I don't think we can propagate the error...

Sorry, I don't fully understand your feature request. Would you mind writing some examples for your proposed feature including your workaround without this feature?. This could help me to understand...

It looks like [`std::net::TcpStream::connect`] has the similar behavior. I think the error message should be improved, but I’m not sure how to do it. ```rust // https://doc.rust-lang.org/1.90.0/src/std/net/tcp.rs.html#169-171 pub fn connect(addr:...