Improved EPOLLRDHUP handling. When it's known that the kernel supports EPOLLRDHUP, there is no need in additional recv() call to get EOF or error when the flag is absent in the event generated by the kernel. A special runtime test is done at startup to detect if EPOLLRDHUP is actually supported by the kernel because epoll_ctl() silently ignores unknown flags. With this knowledge it's now possible to drop the "ready" flag for partial read. Previously, the "ready" flag was kept until the recv() returned EOF or error. In particular, this change allows the lingering close heuristics (which relies on the "ready" flag state) to actually work on Linux, and not wait for more data in most cases. The "available" flag is now used in the read event with the semantics similar to the corresponding counter in kqueue.

--- a/src/event/ngx_event.h
+++ b/src/event/ngx_event.h
@@ -96,6 +96,10 @@ struct ngx_event_s {
      *   write:      available space in buffer when event is ready
      *               or lowat when event is set with NGX_LOWAT_EVENT flag
      *
+     * epoll with EPOLLRDHUP:
+     *   accept:     1 if accept many, 0 otherwise
+     *   read:       1 if there can be data to read, 0 otherwise
+     *
      * iocp: TODO
      *
      * otherwise:
@@ -196,6 +200,9 @@ typedef struct {
 
 
 extern ngx_event_actions_t   ngx_event_actions;
+#if (NGX_HAVE_EPOLLRDHUP)
+extern ngx_uint_t            ngx_use_epoll_rdhup;
+#endif
 
 
 /*