NFSv4.1: Don't rebind to the same source port when reconnecting to the server

NFSv2, v3 and NFSv4 servers often have duplicate replay caches that look
at the source port when deciding whether or not an RPC call is a replay
of a previous call. This requires clients to perform strange TCP gymnastics
in order to ensure that when they reconnect to the server, they bind
to the same source port.

NFSv4.1 and NFSv4.2 have sessions that provide proper replay semantics,
that do not look at the source port of the connection. This patch therefore
ensures they can ignore the rebind requirement.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>

fs/lockd/host.c

@@ -464,7 +464,8 @@ nlm_bind_host(struct nlm_host *host)
 			.version	= host->h_version,
 			.authflavor	= RPC_AUTH_UNIX,
 			.flags		= (RPC_CLNT_CREATE_NOPING |
-					   RPC_CLNT_CREATE_AUTOBIND),
+					   RPC_CLNT_CREATE_AUTOBIND |
+					   RPC_CLNT_CREATE_REUSEPORT),
 			.cred		= host->h_cred,
 		};
 

fs/nfs/client.c

@@ -523,6 +523,8 @@ int nfs_create_rpc_client(struct nfs_client *clp,
 		args.flags |= RPC_CLNT_CREATE_INFINITE_SLOTS;
 	if (test_bit(NFS_CS_NOPING, &clp->cl_flags))
 		args.flags |= RPC_CLNT_CREATE_NOPING;
+	if (test_bit(NFS_CS_REUSEPORT, &clp->cl_flags))
+		args.flags |= RPC_CLNT_CREATE_REUSEPORT;
 
 	if (!IS_ERR(clp->cl_rpcclient))
 		return 0;
@@ -670,6 +672,7 @@ static int nfs_init_server(struct nfs_server *server,
 		.timeparms = &timeparms,
 		.cred = server->cred,
 		.nconnect = data->nfs_server.nconnect,
+		.init_flags = (1UL << NFS_CS_REUSEPORT),
 	};
 	struct nfs_client *clp;
 	int error;

fs/nfs/nfs4client.c

@@ -879,8 +879,11 @@ static int nfs4_set_client(struct nfs_server *server,
 	};
 	struct nfs_client *clp;
 
-	if (minorversion > 0 && proto == XPRT_TRANSPORT_TCP)
+	if (minorversion == 0)
+		__set_bit(NFS_CS_REUSEPORT, &cl_init.init_flags);
+	else if (proto == XPRT_TRANSPORT_TCP)
 		cl_init.nconnect = nconnect;
+
 	if (server->flags & NFS_MOUNT_NORESVPORT)
 		__set_bit(NFS_CS_NORESVPORT, &cl_init.init_flags);
 	if (server->options & NFS_OPTION_MIGRATION)

include/linux/nfs_fs_sb.h

@@ -47,6 +47,7 @@ struct nfs_client {
 #define NFS_CS_TSM_POSSIBLE	5		/* - Maybe state migration */
 #define NFS_CS_NOPING		6		/* - don't ping on connect */
 #define NFS_CS_DS		7		/* - Server is a DS */
+#define NFS_CS_REUSEPORT	8		/* - reuse src port on reconnect */
 	struct sockaddr_storage	cl_addr;	/* server identifier */
 	size_t			cl_addrlen;
 	char *			cl_hostname;	/* hostname of server */

include/linux/sunrpc/clnt.h

@@ -149,6 +149,7 @@ struct rpc_add_xprt_test {
 #define RPC_CLNT_CREATE_NO_IDLE_TIMEOUT	(1UL << 8)
 #define RPC_CLNT_CREATE_NO_RETRANS_TIMEOUT	(1UL << 9)
 #define RPC_CLNT_CREATE_SOFTERR		(1UL << 10)
+#define RPC_CLNT_CREATE_REUSEPORT	(1UL << 11)
 
 struct rpc_clnt *rpc_create(struct rpc_create_args *args);
 struct rpc_clnt	*rpc_bind_new_program(struct rpc_clnt *,

include/linux/sunrpc/xprt.h

@@ -207,7 +207,8 @@ struct rpc_xprt {
 	unsigned int		min_reqs;	/* min number of slots */
 	unsigned int		num_reqs;	/* total slots */
 	unsigned long		state;		/* transport state */
-	unsigned char		resvport   : 1; /* use a reserved port */
+	unsigned char		resvport   : 1,	/* use a reserved port */
+				reuseport  : 1; /* reuse port on reconnect */
 	atomic_t		swapper;	/* we're swapping over this
 						   transport */
 	unsigned int		bind_index;	/* bind function index */

net/sunrpc/clnt.c

@@ -591,6 +591,9 @@ struct rpc_clnt *rpc_create(struct rpc_create_args *args)
 	xprt->resvport = 1;
 	if (args->flags & RPC_CLNT_CREATE_NONPRIVPORT)
 		xprt->resvport = 0;
+	xprt->reuseport = 0;
+	if (args->flags & RPC_CLNT_CREATE_REUSEPORT)
+		xprt->reuseport = 1;
 
 	clnt = rpc_create_xprt(args, xprt);
 	if (IS_ERR(clnt) || args->nconnect <= 1)
@@ -2906,7 +2909,7 @@ int rpc_clnt_add_xprt(struct rpc_clnt *clnt,
 	struct rpc_xprt *xprt;
 	unsigned long connect_timeout;
 	unsigned long reconnect_timeout;
-	unsigned char resvport;
+	unsigned char resvport, reuseport;
 	int ret = 0;
 
 	rcu_read_lock();
@@ -2918,6 +2921,7 @@ int rpc_clnt_add_xprt(struct rpc_clnt *clnt,
 		return -EAGAIN;
 	}
 	resvport = xprt->resvport;
+	reuseport = xprt->reuseport;
 	connect_timeout = xprt->connect_timeout;
 	reconnect_timeout = xprt->max_reconnect_timeout;
 	rcu_read_unlock();
@@ -2928,6 +2932,7 @@ int rpc_clnt_add_xprt(struct rpc_clnt *clnt,
 		goto out_put_switch;
 	}
 	xprt->resvport = resvport;
+	xprt->reuseport = reuseport;
 	if (xprt->ops->set_connect_timeout != NULL)
 		xprt->ops->set_connect_timeout(xprt,
 				connect_timeout,

net/sunrpc/xprtsock.c

@@ -1752,7 +1752,7 @@ static void xs_set_port(struct rpc_xprt *xprt, unsigned short port)
 
 static void xs_set_srcport(struct sock_xprt *transport, struct socket *sock)
 {
-	if (transport->srcport == 0)
+	if (transport->srcport == 0 && transport->xprt.reuseport)
 		transport->srcport = xs_sock_getport(sock);
 }