MEDIUM: connection: make the subscribe() call able to wakeup if ready

There's currently an internal API limitation at the connection layer
regarding conn_subscribe(). We must not subscribe if we haven't yet
met EAGAIN or such a condition, so we sometimes force ourselves to read
in order to meet this condition and being allowed to call subscribe.
But reading cannot always be done (e.g. at the end of a loop where we
cannot afford to retrieve new data and start again) so we instead
perform a tasklet_wakeup() of the requester's io_cb. This is what is
done in mux_h1 for example. The problem with this is that it forces
a new receive when we're not necessarily certain we need one. And if
the FD is not ready and was already being polled, it's a useless
wakeup.

The current patch improves the connection-level subscribe() so that
it really manipulates the polling if the FD is marked not-ready, but
instead schedules the argument tasklet for a wakeup if the FD is
ready. This guarantees we'll wake this tasklet up in any case once the
FD is ready, either immediately or after polling.

By doing so, a test on pure close mode shows we cut in half the number
of epoll_ctl() calls and almost eliminate failed recvfrom():

  $ ./h1load -n 100000 -r 1 -t 4 -c 1000 -T 20 -F 127.0.0.1:8001/?s=1k/t=20

  before:
   399464 epoll_ctl 1
   200007 recvfrom 1
   200000 sendto 1
   100000 recvfrom -1
     7508 epoll_wait 1

  after:
   205739 epoll_ctl 1
   200000 sendto 1
   200000 recvfrom 1
     6084 epoll_wait 1
     2651 recvfrom -1

On keep-alive there is no change however.

src/connection.c

@@ -334,26 +334,46 @@ int conn_unsubscribe(struct connection *conn, void *xprt_ctx, int event_type, st
 	return 0;
 }
 
-/* Called from the upper layer, to unsubscribe <es> from events <event_type>
+/* Called from the upper layer, to subscribe <es> to events <event_type>.
- * (undo fcgi_subscribe). The <es> struct is not allowed to differ from the one
+ * The <es> struct is not allowed to differ from the one passed during a
- * passed to the subscribe() call. It always returns zero.
+ * previous call to subscribe(). If the FD is ready, the wait_event is
+ * immediately woken up and the subcription is cancelled. It always
+ * returns zero.
  */
 int conn_subscribe(struct connection *conn, void *xprt_ctx, int event_type, struct wait_event *es)
 {
 	BUG_ON(event_type & ~(SUB_RETRY_SEND|SUB_RETRY_RECV));
-	BUG_ON(conn->subs && conn->subs->events & event_type);
 	BUG_ON(conn->subs && conn->subs != es);
 
+	if (conn->subs && (conn->subs->events & event_type) == event_type)
+		return 0;
+
 	conn->subs = es;
 	es->events |= event_type;
 
 	if (conn_ctrl_ready(conn)) {
-		if (event_type & SUB_RETRY_RECV)
+		if (event_type & SUB_RETRY_RECV) {
+			if (fd_recv_ready(conn->handle.fd)) {
+				tasklet_wakeup(es->tasklet);
+				es->events &= ~SUB_RETRY_RECV;
+				if (!es->events)
+					conn->subs = NULL;
+			}
+			else
 				fd_want_recv(conn->handle.fd);
+		}
 
-		if (event_type & SUB_RETRY_SEND)
+		if (event_type & SUB_RETRY_SEND) {
+			if (fd_send_ready(conn->handle.fd)) {
+				tasklet_wakeup(es->tasklet);
+				es->events &= ~SUB_RETRY_SEND;
+				if (!es->events)
+					conn->subs = NULL;
+			}
+			else
 				fd_want_send(conn->handle.fd);
 		}
+	}
 	return 0;
 }