Fix tight loop on aborted connection (#285)

When the underlying IO returns 0 on read, we must stop polling it since
it's closed. Otherwise we'll be stuck in a tight loop.

this fixes https://github.com/sfackler/rust-openssl/issues/949

src/server.rs

@@ -140,7 +140,7 @@ use proto::{self, Config, Prioritized};
 use bytes::{Buf, Bytes, IntoBuf};
 use futures::{self, Async, Future, Poll};
 use http::{Request, Response};
-use std::{convert, fmt, mem};
+use std::{convert, fmt, io, mem};
 use std::time::Duration;
 use tokio_io::{AsyncRead, AsyncWrite};
 
@@ -1045,6 +1045,12 @@ where
 
         while rem > 0 {
             let n = try_nb!(self.inner_mut().read(&mut buf[..rem]));
+            if n == 0 {
+                return Err(io::Error::new(
+                    io::ErrorKind::ConnectionReset,
+                    "connection closed unexpectedly",
+                ).into());
+            }
 
             if PREFACE[self.pos..self.pos + n] != buf[..n] {
                 // TODO: Should this just write the GO_AWAY frame directly?