Visit idea.popcount.org → RSS feedOn Hacker News Jul 2013 – Sep 2019
Software Engineering 7Systems & Performance 5Security & Privacy 1Networking & Internet 1Programming Languages 1
While working on our Spectrum server , we noticed something weird: the TCP sockets which we thought should have been closed were lingering around. We realized we don't really understand when TCP sockets are supposed to time out!
Let me assure you, the "fluxcapacitor" project is very interesting. Unfortunately, I find it very difficult to describe what it does. For this project I completely fail the elevator pitch .
This time we'll focus on Linux's select(2) successor - the epoll(2) I/O multiplexing syscall.
In a previous blog post we discussed a brief history of the select(2) syscall . The article concludes that some I/O multiplexing was necessary to do console emulation, games and non-trivial TCP/IP applications.
In this post we'll discuss the second argument on why the epoll() is broken. The problem is best described in an LWN comment by Foom :