Merge remote-tracking branch 'remotes/berrange/tags/pull-qio-2017-10-16-1' into staging

Merge QIO 2017/10/16 v1

# gpg: Signature made Mon 16 Oct 2017 17:10:54 BST
# gpg:                using RSA key 0xBE86EBB415104FDF
# gpg: Good signature from "Daniel P. Berrange <dan@berrange.com>"
# gpg:                 aka "Daniel P. Berrange <berrange@redhat.com>"
# Primary key fingerprint: DAF3 A6FD B26B 6291 2D0E  8E3F BE86 EBB4 1510 4FDF

* remotes/berrange/tags/pull-qio-2017-10-16-1:
  io: fix mem leak in websock error path
  io: add trace points for websocket HTTP protocol headers
  io: cope with websock 'Connection' header having multiple values
  io: get rid of bounce buffering in websock write path
  io: pass a struct iovec into qio_channel_websock_encode
  io: get rid of qio_channel_websock_encode helper method
  io: simplify websocket ping reply handling
  io: monitor encoutput buffer size from websocket GSource
  sockets: Handle race condition between binds to the same port
  sockets: factor out create_fast_reuse_socket
  sockets: factor out a new try_bind() function

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

1 files changed, 90 insertions, 48 deletions

diff --git a/util/qemu-sockets.c b/util/qemu-sockets.c
index d149383bb9..b47fb45885 100644
--- a/util/qemu-sockets.c
+++ b/util/qemu-sockets.c
@@ -149,6 +149,54 @@ int inet_ai_family_from_address(InetSocketAddress *addr,
     return PF_UNSPEC;
 }
 
+static int create_fast_reuse_socket(struct addrinfo *e)
+{
+    int slisten = qemu_socket(e->ai_family, e->ai_socktype, e->ai_protocol);
+    if (slisten < 0) {
+        return -1;
+    }
+    socket_set_fast_reuse(slisten);
+    return slisten;
+}
+
+static int try_bind(int socket, InetSocketAddress *saddr, struct addrinfo *e)
+{
+#ifndef IPV6_V6ONLY
+    return bind(socket, e->ai_addr, e->ai_addrlen);
+#else
+    /*
+     * Deals with first & last cases in matrix in comment
+     * for inet_ai_family_from_address().
+     */
+    int v6only =
+        ((!saddr->has_ipv4 && !saddr->has_ipv6) ||
+         (saddr->has_ipv4 && saddr->ipv4 &&
+          saddr->has_ipv6 && saddr->ipv6)) ? 0 : 1;
+    int stat;
+
+ rebind:
+    if (e->ai_family == PF_INET6) {
+        qemu_setsockopt(socket, IPPROTO_IPV6, IPV6_V6ONLY, &v6only,
+                        sizeof(v6only));
+    }
+
+    stat = bind(socket, e->ai_addr, e->ai_addrlen);
+    if (!stat) {
+        return 0;
+    }
+
+    /* If we got EADDRINUSE from an IPv6 bind & v6only is unset,
+     * it could be that the IPv4 port is already claimed, so retry
+     * with v6only set
+     */
+    if (e->ai_family == PF_INET6 && errno == EADDRINUSE && !v6only) {
+        v6only = 1;
+        goto rebind;
+    }
+    return stat;
+#endif
+}
+
 static int inet_listen_saddr(InetSocketAddress *saddr,
                              int port_offset,
                              bool update_addr,
@@ -158,7 +206,10 @@ static int inet_listen_saddr(InetSocketAddress *saddr,
     char port[33];
     char uaddr[INET6_ADDRSTRLEN+1];
     char uport[33];
-    int slisten, rc, port_min, port_max, p;
+    int rc, port_min, port_max, p;
+    int slisten = 0;
+    int saved_errno = 0;
+    bool socket_created = false;
     Error *err = NULL;
 
     memset(&ai,0, sizeof(ai));
@@ -210,75 +261,66 @@ static int inet_listen_saddr(InetSocketAddress *saddr,
         return -1;
     }
 
-    /* create socket + bind */
+    /* create socket + bind/listen */
     for (e = res; e != NULL; e = e->ai_next) {
         getnameinfo((struct sockaddr*)e->ai_addr,e->ai_addrlen,
 		        uaddr,INET6_ADDRSTRLEN,uport,32,
 		        NI_NUMERICHOST | NI_NUMERICSERV);
-        slisten = qemu_socket(e->ai_family, e->ai_socktype, e->ai_protocol);
+
+        slisten = create_fast_reuse_socket(e);
         if (slisten < 0) {
-            if (!e->ai_next) {
-                error_setg_errno(errp, errno, "Failed to create socket");
-            }
             continue;
         }
 
-        socket_set_fast_reuse(slisten);
-
+        socket_created = true;
         port_min = inet_getport(e);
         port_max = saddr->has_to ? saddr->to + port_offset : port_min;
         for (p = port_min; p <= port_max; p++) {
-#ifdef IPV6_V6ONLY
-            /*
-             * Deals with first & last cases in matrix in comment
-             * for inet_ai_family_from_address().
-             */
-            int v6only =
-                ((!saddr->has_ipv4 && !saddr->has_ipv6) ||
-                 (saddr->has_ipv4 && saddr->ipv4 &&
-                  saddr->has_ipv6 && saddr->ipv6)) ? 0 : 1;
-#endif
             inet_setport(e, p);
-#ifdef IPV6_V6ONLY
-        rebind:
-            if (e->ai_family == PF_INET6) {
-                qemu_setsockopt(slisten, IPPROTO_IPV6, IPV6_V6ONLY, &v6only,
-                                sizeof(v6only));
+            rc = try_bind(slisten, saddr, e);
+            if (rc) {
+                if (errno == EADDRINUSE) {
+                    continue;
+                } else {
+                    error_setg_errno(errp, errno, "Failed to bind socket");
+                    goto listen_failed;
+                }
             }
-#endif
-            if (bind(slisten, e->ai_addr, e->ai_addrlen) == 0) {
-                goto listen;
+            if (!listen(slisten, 1)) {
+                goto listen_ok;
             }
-
-#ifdef IPV6_V6ONLY
-            /* If we got EADDRINUSE from an IPv6 bind & V6ONLY is unset,
-             * it could be that the IPv4 port is already claimed, so retry
-             * with V6ONLY set
-             */
-            if (e->ai_family == PF_INET6 && errno == EADDRINUSE && !v6only) {
-                v6only = 1;
-                goto rebind;
+            if (errno != EADDRINUSE) {
+                error_setg_errno(errp, errno, "Failed to listen on socket");
+                goto listen_failed;
             }
-#endif
-
-            if (p == port_max) {
-                if (!e->ai_next) {
-                    error_setg_errno(errp, errno, "Failed to bind socket");
-                }
+            /* Someone else managed to bind to the same port and beat us
+             * to listen on it! Socket semantics does not allow us to
+             * recover from this situation, so we need to recreate the
+             * socket to allow bind attempts for subsequent ports:
+             */
+            closesocket(slisten);
+            slisten = create_fast_reuse_socket(e);
+            if (slisten < 0) {
+                error_setg_errno(errp, errno,
+                                 "Failed to recreate failed listening socket");
+                goto listen_failed;
             }
         }
+    }
+    error_setg_errno(errp, errno,
+                     socket_created ?
+                     "Failed to find an available port" :
+                     "Failed to create a socket");
+listen_failed:
+    saved_errno = errno;
+    if (slisten >= 0) {
         closesocket(slisten);
     }
     freeaddrinfo(res);
+    errno = saved_errno;
     return -1;
 
-listen:
-    if (listen(slisten,1) != 0) {
-        error_setg_errno(errp, errno, "Failed to listen on socket");
-        closesocket(slisten);
-        freeaddrinfo(res);
-        return -1;
-    }
+listen_ok:
     if (update_addr) {
         g_free(saddr->host);
         saddr->host = g_strdup(uaddr);