diff mbox series

[4/4] NFSv4.1: Don't rebind to the same source port when reconnecting to the server

Message ID 20221122203914.344894-5-tim.gardner@canonical.com
State New
Headers show
Series NFSv4.1: Don't rebind to the same source port when reconnecting to the server | expand

Commit Message

Tim Gardner Nov. 22, 2022, 8:39 p.m. UTC 
From: Trond Myklebust <trond.myklebust@hammerspace.com>

BugLink: https://bugs.launchpad.net/bugs/1997488

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>
(cherry picked from commit e6237b6feb37582fbd6bd7a8336d1256a6b4b4f9)
Signed-off-by: Tim Gardner <tim.gardner@canonical.com>
---
 fs/lockd/host.c             | 3 ++-
 fs/nfs/client.c             | 3 +++
 fs/nfs/nfs4client.c         | 5 ++++-
 include/linux/nfs_fs_sb.h   | 1 +
 include/linux/sunrpc/clnt.h | 1 +
 include/linux/sunrpc/xprt.h | 3 ++-
 net/sunrpc/clnt.c           | 7 ++++++-
 net/sunrpc/xprtsock.c       | 2 +-
 8 files changed, 20 insertions(+), 5 deletions(-)
diff mbox series

Patch

diff --git a/fs/lockd/host.c b/fs/lockd/host.c
index 584c03e11844..771c289f6df7 100644
--- a/fs/lockd/host.c
+++ b/fs/lockd/host.c
@@ -459,7 +459,8 @@  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,
 		};
 
diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index b43a931edba5..98f791a914d2 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -524,6 +524,8 @@  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;
@@ -671,6 +673,7 @@  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;
diff --git a/fs/nfs/nfs4client.c b/fs/nfs/nfs4client.c
index 3671a51fe5eb..70e770b7dd12 100644
--- a/fs/nfs/nfs4client.c
+++ b/fs/nfs/nfs4client.c
@@ -880,8 +880,11 @@  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)
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index 739cc46412a9..41e6de5c3680 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/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 */
diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h
index 336802acc629..eca293c139b4 100644
--- a/include/linux/sunrpc/clnt.h
+++ b/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 *,
diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h
index 3c6c4b1dbf1a..86fb981adbf0 100644
--- a/include/linux/sunrpc/xprt.h
+++ b/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 */
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 1893203cc94f..a6d579bdefd8 100644
--- a/net/sunrpc/clnt.c
+++ b/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)
@@ -2917,7 +2920,7 @@  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();
@@ -2929,6 +2932,7 @@  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();
@@ -2939,6 +2943,7 @@  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,
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index 91c8249ee818..68fc5539b123 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -1787,7 +1787,7 @@  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);
 }