-
Pieter Noordhuis authored
Every request is now kept in a maximum of two queues. When a request is initially inserted, it is inserted in the `to_write` queue. When at least one write buffer is extracted from that request, it is *moved* to the `wait_write` queue. When at least one byte of the request has been written to the socket, it is *also* placed in the `wait_read` queue. This means that it is possible for a request to be emitting write buffers, while simultaneously receiving read callbacks.
125cf58c