[twen_ty]

So it's 2024 and Linux still doesn't have what Windows NT had 30 years ago?

[jart]

Are you saying io_uring is basically IOCP for Linux? If that's true, then I'm not happy about it being in the kernel. IOCP only goes 2x faster in my experience (vs. threads and blocking i/o) and that isn't worth it for the ugly error prone code you have to write. System calls are also a lot faster on Linux than they are on Windows, so I'd be surprised if io_uring manages even that.

[yencabulator]

Must be weird trying to explain why Windows had to introduce a new API that they've apparently already had for 30 years.

https://learn.microsoft.com/en-us/windows/win32/api/ioringap...

[asveikau]

The async I/O model is one that NT had for all that time.

The idea of using a ring data structure to do faster syscalls is what's new.

io_uring represents both a pivot to an NT style I/O model (true async instead of readiness as in epoll) using a new syscall interface.