1 files changed, 470 insertions, 376 deletions

diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index 870929e08e5..b507cd327d9 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -21,6 +21,7 @@
 
 #include <linux/kernel.h>
 #include <linux/sched.h>
+#include <linux/module.h>
 #include <linux/errno.h>
 #include <linux/fcntl.h>
 #include <linux/net.h>
@@ -42,6 +43,7 @@
 #include <net/tcp_states.h>
 #include <asm/uaccess.h>
 #include <asm/ioctls.h>
+#include <trace/events/skb.h>
 
 #include <linux/sunrpc/types.h>
 #include <linux/sunrpc/clnt.h>
@@ -51,12 +53,14 @@
 #include <linux/sunrpc/stats.h>
 #include <linux/sunrpc/xprt.h>
 
+#include "sunrpc.h"
+
 #define RPCDBG_FACILITY	RPCDBG_SVCXPRT
 
 
 static struct svc_sock *svc_setup_socket(struct svc_serv *, struct socket *,
-					 int *errp, int flags);
-static void		svc_udp_data_ready(struct sock *, int);
+					 int flags);
+static void		svc_udp_data_ready(struct sock *);
 static int		svc_udp_recvfrom(struct svc_rqst *);
 static int		svc_udp_sendto(struct svc_rqst *);
 static void		svc_sock_detach(struct svc_xprt *);
@@ -64,7 +68,15 @@ static void		svc_tcp_sock_detach(struct svc_xprt *);
 static void		svc_sock_free(struct svc_xprt *);
 
 static struct svc_xprt *svc_create_socket(struct svc_serv *, int,
-					  struct sockaddr *, int, int);
+					  struct net *, struct sockaddr *,
+					  int, int);
+#if defined(CONFIG_SUNRPC_BACKCHANNEL)
+static struct svc_xprt *svc_bc_create_socket(struct svc_serv *, int,
+					     struct net *, struct sockaddr *,
+					     int, int);
+static void svc_bc_sock_free(struct svc_xprt *xprt);
+#endif /* CONFIG_SUNRPC_BACKCHANNEL */
+
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
 static struct lock_class_key svc_key[2];
 static struct lock_class_key svc_slock_key[2];
@@ -72,7 +84,11 @@ static struct lock_class_key svc_slock_key[2];
 static void svc_reclassify_socket(struct socket *sock)
 {
 	struct sock *sk = sock->sk;
-	BUG_ON(sock_owned_by_user(sk));
+
+	WARN_ON_ONCE(sock_owned_by_user(sk));
+	if (sock_owned_by_user(sk))
+		return;
+
 	switch (sk->sk_family) {
 	case AF_INET:
 		sock_lock_init_class_and_name(sk, "slock-AF_INET-NFSD",
@@ -133,24 +149,24 @@ static void svc_set_cmsg_data(struct svc_rqst *rqstp, struct cmsghdr *cmh)
 			cmh->cmsg_level = SOL_IP;
 			cmh->cmsg_type = IP_PKTINFO;
 			pki->ipi_ifindex = 0;
-			pki->ipi_spec_dst.s_addr = rqstp->rq_daddr.addr.s_addr;
+			pki->ipi_spec_dst.s_addr =
+				 svc_daddr_in(rqstp)->sin_addr.s_addr;
 			cmh->cmsg_len = CMSG_LEN(sizeof(*pki));
 		}
 		break;
 
 	case AF_INET6: {
 			struct in6_pktinfo *pki = CMSG_DATA(cmh);
+			struct sockaddr_in6 *daddr = svc_daddr_in6(rqstp);
 
 			cmh->cmsg_level = SOL_IPV6;
 			cmh->cmsg_type = IPV6_PKTINFO;
-			pki->ipi6_ifindex = 0;
-			ipv6_addr_copy(&pki->ipi6_addr,
-					&rqstp->rq_daddr.addr6);
+			pki->ipi6_ifindex = daddr->sin6_scope_id;
+			pki->ipi6_addr = daddr->sin6_addr;
 			cmh->cmsg_len = CMSG_LEN(sizeof(*pki));
 		}
 		break;
 	}
-	return;
 }
 
 /*
@@ -275,12 +291,14 @@ static int svc_one_sock_name(struct svc_sock *svsk, char *buf, int remaining)
 				&inet_sk(sk)->inet_rcv_saddr,
 				inet_sk(sk)->inet_num);
 		break;
+#if IS_ENABLED(CONFIG_IPV6)
 	case PF_INET6:
 		len = snprintf(buf, remaining, "ipv6 %s %pI6 %d\n",
 				proto_name,
-				&inet6_sk(sk)->rcv_saddr,
+				&sk->sk_v6_rcv_saddr,
 				inet_sk(sk)->inet_num);
 		break;
+#endif
 	default:
 		len = snprintf(buf, remaining, "*unknown-%d*\n",
 				sk->sk_family);
@@ -293,55 +311,6 @@ static int svc_one_sock_name(struct svc_sock *svsk, char *buf, int remaining)
 	return len;
 }
 
-/**
- * svc_sock_names - construct a list of listener names in a string
- * @serv: pointer to RPC service
- * @buf: pointer to a buffer to fill in with socket names
- * @buflen: size of the buffer to be filled
- * @toclose: pointer to '\0'-terminated C string containing the name
- *		of a listener to be closed
- *
- * Fills in @buf with a '\n'-separated list of names of listener
- * sockets.  If @toclose is not NULL, the socket named by @toclose
- * is closed, and is not included in the output list.
- *
- * Returns positive length of the socket name string, or a negative
- * errno value on error.
- */
-int svc_sock_names(struct svc_serv *serv, char *buf, const size_t buflen,
-		   const char *toclose)
-{
-	struct svc_sock *svsk, *closesk = NULL;
-	int len = 0;
-
-	if (!serv)
-		return 0;
-
-	spin_lock_bh(&serv->sv_lock);
-	list_for_each_entry(svsk, &serv->sv_permsocks, sk_xprt.xpt_list) {
-		int onelen = svc_one_sock_name(svsk, buf + len, buflen - len);
-		if (onelen < 0) {
-			len = onelen;
-			break;
-		}
-		if (toclose && strcmp(toclose, buf + len) == 0)
-			closesk = svsk;
-		else
-			len += onelen;
-	}
-	spin_unlock_bh(&serv->sv_lock);
-
-	if (closesk)
-		/* Should unregister with portmap, but you cannot
-		 * unregister just one protocol...
-		 */
-		svc_close_xprt(&closesk->sk_xprt);
-	else if (toclose)
-		return -ENOENT;
-	return len;
-}
-EXPORT_SYMBOL_GPL(svc_sock_names);
-
 /*
  * Check input queue length
  */
@@ -378,6 +347,33 @@ static int svc_recvfrom(struct svc_rqst *rqstp, struct kvec *iov, int nr,
 	return len;
 }
 
+static int svc_partial_recvfrom(struct svc_rqst *rqstp,
+				struct kvec *iov, int nr,
+				int buflen, unsigned int base)
+{
+	size_t save_iovlen;
+	void *save_iovbase;
+	unsigned int i;
+	int ret;
+
+	if (base == 0)
+		return svc_recvfrom(rqstp, iov, nr, buflen);
+
+	for (i = 0; i < nr; i++) {
+		if (iov[i].iov_len > base)
+			break;
+		base -= iov[i].iov_len;
+	}
+	save_iovlen = iov[i].iov_len;
+	save_iovbase = iov[i].iov_base;
+	iov[i].iov_len -= base;
+	iov[i].iov_base += base;
+	ret = svc_recvfrom(rqstp, &iov[i], nr - i, buflen);
+	iov[i].iov_len = save_iovlen;
+	iov[i].iov_base = save_iovbase;
+	return ret;
+}
+
 /*
  * Set socket snd and rcv buffer lengths
  */
@@ -400,27 +396,33 @@ static void svc_sock_setbufsize(struct socket *sock, unsigned int snd,
 	lock_sock(sock->sk);
 	sock->sk->sk_sndbuf = snd * 2;
 	sock->sk->sk_rcvbuf = rcv * 2;
-	sock->sk->sk_userlocks |= SOCK_SNDBUF_LOCK|SOCK_RCVBUF_LOCK;
 	sock->sk->sk_write_space(sock->sk);
 	release_sock(sock->sk);
 #endif
 }
+
+static int svc_sock_secure_port(struct svc_rqst *rqstp)
+{
+	return svc_port_is_privileged(svc_addr(rqstp));
+}
+
 /*
  * INET callback when data has been received on the socket.
  */
-static void svc_udp_data_ready(struct sock *sk, int count)
+static void svc_udp_data_ready(struct sock *sk)
 {
 	struct svc_sock	*svsk = (struct svc_sock *)sk->sk_user_data;
+	wait_queue_head_t *wq = sk_sleep(sk);
 
 	if (svsk) {
-		dprintk("svc: socket %p(inet %p), count=%d, busy=%d\n",
-			svsk, sk, count,
+		dprintk("svc: socket %p(inet %p), busy=%d\n",
+			svsk, sk,
 			test_bit(XPT_BUSY, &svsk->sk_xprt.xpt_flags));
 		set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
 		svc_xprt_enqueue(&svsk->sk_xprt);
 	}
-	if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))
-		wake_up_interruptible(sk->sk_sleep);
+	if (wq && waitqueue_active(wq))
+		wake_up_interruptible(wq);
 }
 
 /*
@@ -429,6 +431,7 @@ static void svc_udp_data_ready(struct sock *sk, int count)
 static void svc_write_space(struct sock *sk)
 {
 	struct svc_sock	*svsk = (struct svc_sock *)(sk->sk_user_data);
+	wait_queue_head_t *wq = sk_sleep(sk);
 
 	if (svsk) {
 		dprintk("svc: socket %p(inet %p), write_space busy=%d\n",
@@ -436,10 +439,10 @@ static void svc_write_space(struct sock *sk)
 		svc_xprt_enqueue(&svsk->sk_xprt);
 	}
 
-	if (sk->sk_sleep && waitqueue_active(sk->sk_sleep)) {
+	if (wq && waitqueue_active(wq)) {
 		dprintk("RPC svc_write_space: someone sleeping on %p\n",
 		       svsk);
-		wake_up_interruptible(sk->sk_sleep);
+		wake_up_interruptible(wq);
 	}
 }
 
@@ -447,7 +450,7 @@ static void svc_tcp_write_space(struct sock *sk)
 {
 	struct socket *sock = sk->sk_socket;
 
-	if (sk_stream_wspace(sk) >= sk_stream_min_wspace(sk) && sock)
+	if (sk_stream_is_writeable(sk) && sock)
 		clear_bit(SOCK_NOSPACE, &sock->flags);
 	svc_write_space(sk);
 }
@@ -459,22 +462,31 @@ static int svc_udp_get_dest_address4(struct svc_rqst *rqstp,
 				     struct cmsghdr *cmh)
 {
 	struct in_pktinfo *pki = CMSG_DATA(cmh);
+	struct sockaddr_in *daddr = svc_daddr_in(rqstp);
+
 	if (cmh->cmsg_type != IP_PKTINFO)
 		return 0;
-	rqstp->rq_daddr.addr.s_addr = pki->ipi_spec_dst.s_addr;
+
+	daddr->sin_family = AF_INET;
+	daddr->sin_addr.s_addr = pki->ipi_spec_dst.s_addr;
 	return 1;
 }
 
 /*
- * See net/ipv6/datagram.c : datagram_recv_ctl
+ * See net/ipv6/datagram.c : ip6_datagram_recv_ctl
  */
 static int svc_udp_get_dest_address6(struct svc_rqst *rqstp,
 				     struct cmsghdr *cmh)
 {
 	struct in6_pktinfo *pki = CMSG_DATA(cmh);
+	struct sockaddr_in6 *daddr = svc_daddr_in6(rqstp);
+
 	if (cmh->cmsg_type != IPV6_PKTINFO)
 		return 0;
-	ipv6_addr_copy(&rqstp->rq_daddr.addr6, &pki->ipi6_addr);
+
+	daddr->sin6_family = AF_INET6;
+	daddr->sin6_addr = pki->ipi6_addr;
+	daddr->sin6_scope_id = pki->ipi6_ifindex;
 	return 1;
 }
 
@@ -547,12 +559,9 @@ static int svc_udp_recvfrom(struct svc_rqst *rqstp)
 			dprintk("svc: recvfrom returned error %d\n", -err);
 			set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
 		}
-		svc_xprt_received(&svsk->sk_xprt);
-		return -EAGAIN;
+		return 0;
 	}
 	len = svc_addr_len(svc_addr(rqstp));
-	if (len == 0)
-		return -EAFNOSUPPORT;
 	rqstp->rq_addrlen = len;
 	if (skb->tstamp.tv64 == 0) {
 		skb->tstamp = ktime_get_real();
@@ -562,25 +571,17 @@ static int svc_udp_recvfrom(struct svc_rqst *rqstp)
 	svsk->sk_sk->sk_stamp = skb->tstamp;
 	set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags); /* there may be more data... */
 
-	/*
-	 * Maybe more packets - kick another thread ASAP.
-	 */
-	svc_xprt_received(&svsk->sk_xprt);
-
 	len  = skb->len - sizeof(struct udphdr);
 	rqstp->rq_arg.len = len;
 
 	rqstp->rq_prot = IPPROTO_UDP;
 
 	if (!svc_udp_get_dest_address(rqstp, cmh)) {
-		if (net_ratelimit())
-			printk(KERN_WARNING
-				"svc: received unknown control message %d/%d; "
-				"dropping RPC reply datagram\n",
-					cmh->cmsg_level, cmh->cmsg_type);
-		skb_free_datagram_locked(svsk->sk_sk, skb);
-		return 0;
+		net_warn_ratelimited("svc: received unknown control message %d/%d; dropping RPC reply datagram\n",
+				     cmh->cmsg_level, cmh->cmsg_type);
+		goto out_free;
 	}
+	rqstp->rq_daddrlen = svc_addr_len(svc_daddr(rqstp));
 
 	if (skb_is_nonlinear(skb)) {
 		/* we have to copy */
@@ -588,8 +589,7 @@ static int svc_udp_recvfrom(struct svc_rqst *rqstp)
 		if (csum_partial_copy_to_xdr(&rqstp->rq_arg, skb)) {
 			local_bh_enable();
 			/* checksum error */
-			skb_free_datagram_locked(svsk->sk_sk, skb);
-			return 0;
+			goto out_free;
 		}
 		local_bh_enable();
 		skb_free_datagram_locked(svsk->sk_sk, skb);
@@ -598,10 +598,8 @@ static int svc_udp_recvfrom(struct svc_rqst *rqstp)
 		rqstp->rq_arg.head[0].iov_base = skb->data +
 			sizeof(struct udphdr);
 		rqstp->rq_arg.head[0].iov_len = len;
-		if (skb_checksum_complete(skb)) {
-			skb_free_datagram_locked(svsk->sk_sk, skb);
-			return 0;
-		}
+		if (skb_checksum_complete(skb))
+			goto out_free;
 		rqstp->rq_xprt_ctxt = skb;
 	}
 
@@ -615,11 +613,16 @@ static int svc_udp_recvfrom(struct svc_rqst *rqstp)
 		rqstp->rq_respages = rqstp->rq_pages + 1 +
 			DIV_ROUND_UP(rqstp->rq_arg.page_len, PAGE_SIZE);
 	}
+	rqstp->rq_next_page = rqstp->rq_respages+1;
 
 	if (serv->sv_stats)
 		serv->sv_stats->netudpcnt++;
 
 	return len;
+out_free:
+	trace_kfree_skb(skb, svc_udp_recvfrom);
+	skb_free_datagram_locked(svsk->sk_sk, skb);
+	return 0;
 }
 
 static int
@@ -664,10 +667,11 @@ static struct svc_xprt *svc_udp_accept(struct svc_xprt *xprt)
 }
 
 static struct svc_xprt *svc_udp_create(struct svc_serv *serv,
+				       struct net *net,
 				       struct sockaddr *sa, int salen,
 				       int flags)
 {
-	return svc_create_socket(serv, IPPROTO_UDP, sa, salen, flags);
+	return svc_create_socket(serv, IPPROTO_UDP, net, sa, salen, flags);
 }
 
 static struct svc_xprt_ops svc_udp_ops = {
@@ -680,6 +684,7 @@ static struct svc_xprt_ops svc_udp_ops = {
 	.xpo_prep_reply_hdr = svc_udp_prep_reply_hdr,
 	.xpo_has_wspace = svc_udp_has_wspace,
 	.xpo_accept = svc_udp_accept,
+	.xpo_secure_port = svc_sock_secure_port,
 };
 
 static struct svc_xprt_class svc_udp_class = {
@@ -693,7 +698,8 @@ static void svc_udp_init(struct svc_sock *svsk, struct svc_serv *serv)
 {
 	int err, level, optname, one = 1;
 
-	svc_xprt_init(&svc_udp_class, &svsk->sk_xprt, serv);
+	svc_xprt_init(sock_net(svsk->sk_sock->sk), &svc_udp_class,
+		      &svsk->sk_xprt, serv);
 	clear_bit(XPT_CACHE_AUTH, &svsk->sk_xprt.xpt_flags);
 	svsk->sk_sk->sk_data_ready = svc_udp_data_ready;
 	svsk->sk_sk->sk_write_space = svc_write_space;
@@ -732,9 +738,10 @@ static void svc_udp_init(struct svc_sock *svsk, struct svc_serv *serv)
  * A data_ready event on a listening socket means there's a connection
  * pending. Do not use state_change as a substitute for it.
  */
-static void svc_tcp_listen_data_ready(struct sock *sk, int count_unused)
+static void svc_tcp_listen_data_ready(struct sock *sk)
 {
 	struct svc_sock	*svsk = (struct svc_sock *)sk->sk_user_data;
+	wait_queue_head_t *wq;
 
 	dprintk("svc: socket %p TCP (listen) state change %d\n",
 		sk, sk->sk_state);
@@ -757,8 +764,9 @@ static void svc_tcp_listen_data_ready(struct sock *sk, int count_unused)
 			printk("svc: socket %p: no user data\n", sk);
 	}
 
-	if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))
-		wake_up_interruptible_all(sk->sk_sleep);
+	wq = sk_sleep(sk);
+	if (wq && waitqueue_active(wq))
+		wake_up_interruptible_all(wq);
 }
 
 /*
@@ -767,6 +775,7 @@ static void svc_tcp_listen_data_ready(struct sock *sk, int count_unused)
 static void svc_tcp_state_change(struct sock *sk)
 {
 	struct svc_sock	*svsk = (struct svc_sock *)sk->sk_user_data;
+	wait_queue_head_t *wq = sk_sleep(sk);
 
 	dprintk("svc: socket %p TCP (connected) state change %d (svsk %p)\n",
 		sk, sk->sk_state, sk->sk_user_data);
@@ -777,13 +786,14 @@ static void svc_tcp_state_change(struct sock *sk)
 		set_bit(XPT_CLOSE, &svsk->sk_xprt.xpt_flags);
 		svc_xprt_enqueue(&svsk->sk_xprt);
 	}
-	if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))
-		wake_up_interruptible_all(sk->sk_sleep);
+	if (wq && waitqueue_active(wq))
+		wake_up_interruptible_all(wq);
 }
 
-static void svc_tcp_data_ready(struct sock *sk, int count)
+static void svc_tcp_data_ready(struct sock *sk)
 {
 	struct svc_sock *svsk = (struct svc_sock *)sk->sk_user_data;
+	wait_queue_head_t *wq = sk_sleep(sk);
 
 	dprintk("svc: socket %p TCP data ready (svsk %p)\n",
 		sk, sk->sk_user_data);
@@ -791,8 +801,8 @@ static void svc_tcp_data_ready(struct sock *sk, int count)
 		set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
 		svc_xprt_enqueue(&svsk->sk_xprt);
 	}
-	if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))
-		wake_up_interruptible(sk->sk_sleep);
+	if (wq && waitqueue_active(wq))
+		wake_up_interruptible(wq);
 }
 
 /*
@@ -820,18 +830,17 @@ static struct svc_xprt *svc_tcp_accept(struct svc_xprt *xprt)
 		if (err == -ENOMEM)
 			printk(KERN_WARNING "%s: no more sockets!\n",
 			       serv->sv_name);
-		else if (err != -EAGAIN && net_ratelimit())
-			printk(KERN_WARNING "%s: accept failed (err %d)!\n",
-				   serv->sv_name, -err);
+		else if (err != -EAGAIN)
+			net_warn_ratelimited("%s: accept failed (err %d)!\n",
+					     serv->sv_name, -err);
 		return NULL;
 	}
 	set_bit(XPT_CONN, &svsk->sk_xprt.xpt_flags);
 
 	err = kernel_getpeername(newsock, sin, &slen);
 	if (err < 0) {
-		if (net_ratelimit())
-			printk(KERN_WARNING "%s: peername failed (err %d)!\n",
-				   serv->sv_name, -err);
+		net_warn_ratelimited("%s: peername failed (err %d)!\n",
+				     serv->sv_name, -err);
 		goto failed;		/* aborted connection or whatever */
 	}
 
@@ -840,8 +849,7 @@ static struct svc_xprt *svc_tcp_accept(struct svc_xprt *xprt)
 	 * tell us anything.  For now just warn about unpriv connections.
 	 */
 	if (!svc_port_is_privileged(sin)) {
-		dprintk(KERN_WARNING
-			"%s: connect from unprivileged port: %s\n",
+		dprintk("%s: connect from unprivileged port: %s\n",
 			serv->sv_name,
 			__svc_print_addr(sin, buf, sizeof(buf)));
 	}
@@ -853,8 +861,9 @@ static struct svc_xprt *svc_tcp_accept(struct svc_xprt *xprt)
 	 */
 	newsock->sk->sk_sndtimeo = HZ*30;
 
-	if (!(newsvsk = svc_setup_socket(serv, newsock, &err,
-				 (SVC_SOCK_ANONYMOUS | SVC_SOCK_TEMPORARY))))
+	newsvsk = svc_setup_socket(serv, newsock,
+				 (SVC_SOCK_ANONYMOUS | SVC_SOCK_TEMPORARY));
+	if (IS_ERR(newsvsk))
 		goto failed;
 	svc_xprt_set_remote(&newsvsk->sk_xprt, sin, slen);
 	err = kernel_getsockname(newsock, sin, &slen);
@@ -864,6 +873,10 @@ static struct svc_xprt *svc_tcp_accept(struct svc_xprt *xprt)
 	}
 	svc_xprt_set_local(&newsvsk->sk_xprt, sin, slen);
 
+	if (sock_is_loopback(newsock->sk))
+		set_bit(XPT_LOCAL, &newsvsk->sk_xprt.xpt_flags);
+	else
+		clear_bit(XPT_LOCAL, &newsvsk->sk_xprt.xpt_flags);
 	if (serv->sv_stats)
 		serv->sv_stats->nettcpconn++;
 
@@ -874,40 +887,76 @@ failed:
 	return NULL;
 }
 
+static unsigned int svc_tcp_restore_pages(struct svc_sock *svsk, struct svc_rqst *rqstp)
+{
+	unsigned int i, len, npages;
+
+	if (svsk->sk_datalen == 0)
+		return 0;
+	len = svsk->sk_datalen;
+	npages = (len + PAGE_SIZE - 1) >> PAGE_SHIFT;
+	for (i = 0; i < npages; i++) {
+		if (rqstp->rq_pages[i] != NULL)
+			put_page(rqstp->rq_pages[i]);
+		BUG_ON(svsk->sk_pages[i] == NULL);
+		rqstp->rq_pages[i] = svsk->sk_pages[i];
+		svsk->sk_pages[i] = NULL;
+	}
+	rqstp->rq_arg.head[0].iov_base = page_address(rqstp->rq_pages[0]);
+	return len;
+}
+
+static void svc_tcp_save_pages(struct svc_sock *svsk, struct svc_rqst *rqstp)
+{
+	unsigned int i, len, npages;
+
+	if (svsk->sk_datalen == 0)
+		return;
+	len = svsk->sk_datalen;
+	npages = (len + PAGE_SIZE - 1) >> PAGE_SHIFT;
+	for (i = 0; i < npages; i++) {
+		svsk->sk_pages[i] = rqstp->rq_pages[i];
+		rqstp->rq_pages[i] = NULL;
+	}
+}
+
+static void svc_tcp_clear_pages(struct svc_sock *svsk)
+{
+	unsigned int i, len, npages;
+
+	if (svsk->sk_datalen == 0)
+		goto out;
+	len = svsk->sk_datalen;
+	npages = (len + PAGE_SIZE - 1) >> PAGE_SHIFT;
+	for (i = 0; i < npages; i++) {
+		if (svsk->sk_pages[i] == NULL) {
+			WARN_ON_ONCE(1);
+			continue;
+		}
+		put_page(svsk->sk_pages[i]);
+		svsk->sk_pages[i] = NULL;
+	}
+out:
+	svsk->sk_tcplen = 0;
+	svsk->sk_datalen = 0;
+}
+
 /*
- * Receive data.
+ * Receive fragment record header.
  * If we haven't gotten the record length yet, get the next four bytes.
- * Otherwise try to gobble up as much as possible up to the complete
- * record length.
  */
 static int svc_tcp_recv_record(struct svc_sock *svsk, struct svc_rqst *rqstp)
 {
 	struct svc_serv	*serv = svsk->sk_xprt.xpt_server;
+	unsigned int want;
 	int len;
 
-	if (test_and_clear_bit(XPT_CHNGBUF, &svsk->sk_xprt.xpt_flags))
-		/* sndbuf needs to have room for one request
-		 * per thread, otherwise we can stall even when the
-		 * network isn't a bottleneck.
-		 *
-		 * We count all threads rather than threads in a
-		 * particular pool, which provides an upper bound
-		 * on the number of threads which will access the socket.
-		 *
-		 * rcvbuf just needs to be able to hold a few requests.
-		 * Normally they will be removed from the queue
-		 * as soon a a complete request arrives.
-		 */
-		svc_sock_setbufsize(svsk->sk_sock,
-				    (serv->sv_nrthreads+3) * serv->sv_max_mesg,
-				    3 * serv->sv_max_mesg);
-
 	clear_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
 
 	if (svsk->sk_tcplen < sizeof(rpc_fraghdr)) {
-		int		want = sizeof(rpc_fraghdr) - svsk->sk_tcplen;
 		struct kvec	iov;
 
+		want = sizeof(rpc_fraghdr) - svsk->sk_tcplen;
 		iov.iov_base = ((char *) &svsk->sk_reclen) + svsk->sk_tcplen;
 		iov.iov_len  = want;
 		if ((len = svc_recvfrom(rqstp, &iov, 1, want)) < 0)
@@ -917,110 +966,91 @@ static int svc_tcp_recv_record(struct svc_sock *svsk, struct svc_rqst *rqstp)
 		if (len < want) {
 			dprintk("svc: short recvfrom while reading record "
 				"length (%d of %d)\n", len, want);
-			svc_xprt_received(&svsk->sk_xprt);
-			goto err_again; /* record header not complete */
+			return -EAGAIN;
 		}
 
-		svsk->sk_reclen = ntohl(svsk->sk_reclen);
-		if (!(svsk->sk_reclen & RPC_LAST_STREAM_FRAGMENT)) {
-			/* FIXME: technically, a record can be fragmented,
-			 *  and non-terminal fragments will not have the top
-			 *  bit set in the fragment length header.
-			 *  But apparently no known nfs clients send fragmented
-			 *  records. */
-			if (net_ratelimit())
-				printk(KERN_NOTICE "RPC: multiple fragments "
-					"per record not supported\n");
+		dprintk("svc: TCP record, %d bytes\n", svc_sock_reclen(svsk));
+		if (svc_sock_reclen(svsk) + svsk->sk_datalen >
+							serv->sv_max_mesg) {
+			net_notice_ratelimited("RPC: fragment too large: %d\n",
+					svc_sock_reclen(svsk));
 			goto err_delete;
 		}
-
-		svsk->sk_reclen &= RPC_FRAGMENT_SIZE_MASK;
-		dprintk("svc: TCP record, %d bytes\n", svsk->sk_reclen);
-		if (svsk->sk_reclen > serv->sv_max_mesg) {
-			if (net_ratelimit())
-				printk(KERN_NOTICE "RPC: "
-					"fragment too large: 0x%08lx\n",
-					(unsigned long)svsk->sk_reclen);
-			goto err_delete;
-		}
-	}
-
-	/* Check whether enough data is available */
-	len = svc_recv_available(svsk);
-	if (len < 0)
-		goto error;
-
-	if (len < svsk->sk_reclen) {
-		dprintk("svc: incomplete TCP record (%d of %d)\n",
-			len, svsk->sk_reclen);
-		svc_xprt_received(&svsk->sk_xprt);
-		goto err_again;	/* record not complete */
 	}
-	len = svsk->sk_reclen;
-	set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
 
+	return svc_sock_reclen(svsk);
+error:
+	dprintk("RPC: TCP recv_record got %d\n", len);
 	return len;
- error:
-	if (len == -EAGAIN) {
-		dprintk("RPC: TCP recv_record got EAGAIN\n");
-		svc_xprt_received(&svsk->sk_xprt);
-	}
-	return len;
- err_delete:
+err_delete:
 	set_bit(XPT_CLOSE, &svsk->sk_xprt.xpt_flags);
- err_again:
 	return -EAGAIN;
 }
 
-static int svc_process_calldir(struct svc_sock *svsk, struct svc_rqst *rqstp,
-			       struct rpc_rqst **reqpp, struct kvec *vec)
+static int receive_cb_reply(struct svc_sock *svsk, struct svc_rqst *rqstp)
 {
+	struct rpc_xprt *bc_xprt = svsk->sk_xprt.xpt_bc_xprt;
 	struct rpc_rqst *req = NULL;
-	u32 *p;
-	u32 xid;
-	u32 calldir;
-	int len;
+	struct kvec *src, *dst;
+	__be32 *p = (__be32 *)rqstp->rq_arg.head[0].iov_base;
+	__be32 xid;
+	__be32 calldir;
 
-	len = svc_recvfrom(rqstp, vec, 1, 8);
-	if (len < 0)
-		goto error;
-
-	p = (u32 *)rqstp->rq_arg.head[0].iov_base;
 	xid = *p++;
 	calldir = *p;
 
-	if (calldir == 0) {
-		/* REQUEST is the most common case */
-		vec[0] = rqstp->rq_arg.head[0];
-	} else {
-		/* REPLY */
-		if (svsk->sk_bc_xprt)
-			req = xprt_lookup_rqst(svsk->sk_bc_xprt, xid);
-
-		if (!req) {
-			printk(KERN_NOTICE
-				"%s: Got unrecognized reply: "
-				"calldir 0x%x sk_bc_xprt %p xid %08x\n",
-				__func__, ntohl(calldir),
-				svsk->sk_bc_xprt, xid);
-			vec[0] = rqstp->rq_arg.head[0];
-			goto out;
-		}
+	if (bc_xprt)
+		req = xprt_lookup_rqst(bc_xprt, xid);
 
-		memcpy(&req->rq_private_buf, &req->rq_rcv_buf,
-		       sizeof(struct xdr_buf));
-		/* copy the xid and call direction */
-		memcpy(req->rq_private_buf.head[0].iov_base,
-		       rqstp->rq_arg.head[0].iov_base, 8);
-		vec[0] = req->rq_private_buf.head[0];
+	if (!req) {
+		printk(KERN_NOTICE
+			"%s: Got unrecognized reply: "
+			"calldir 0x%x xpt_bc_xprt %p xid %08x\n",
+			__func__, ntohl(calldir),
+			bc_xprt, xid);
+		return -EAGAIN;
 	}
- out:
-	vec[0].iov_base += 8;
-	vec[0].iov_len -= 8;
-	len = svsk->sk_reclen - 8;
- error:
-	*reqpp = req;
-	return len;
+
+	memcpy(&req->rq_private_buf, &req->rq_rcv_buf, sizeof(struct xdr_buf));
+	/*
+	 * XXX!: cheating for now!  Only copying HEAD.
+	 * But we know this is good enough for now (in fact, for any
+	 * callback reply in the forseeable future).
+	 */
+	dst = &req->rq_private_buf.head[0];
+	src = &rqstp->rq_arg.head[0];
+	if (dst->iov_len < src->iov_len)
+		return -EAGAIN; /* whatever; just giving up. */
+	memcpy(dst->iov_base, src->iov_base, src->iov_len);
+	xprt_complete_rqst(req->rq_task, rqstp->rq_arg.len);
+	rqstp->rq_arg.len = 0;
+	return 0;
+}
+
+static int copy_pages_to_kvecs(struct kvec *vec, struct page **pages, int len)
+{
+	int i = 0;
+	int t = 0;
+
+	while (t < len) {
+		vec[i].iov_base = page_address(pages[i]);
+		vec[i].iov_len = PAGE_SIZE;
+		i++;
+		t += PAGE_SIZE;
+	}
+	return i;
+}
+
+static void svc_tcp_fragment_received(struct svc_sock *svsk)
+{
+	/* If we have more data, signal svc_xprt_enqueue() to try again */
+	if (svc_recv_available(svsk) > sizeof(rpc_fraghdr))
+		set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
+	dprintk("svc: TCP %s record (%d bytes)\n",
+		svc_sock_final_rec(svsk) ? "final" : "nonfinal",
+		svc_sock_reclen(svsk));
+	svsk->sk_tcplen = 0;
+	svsk->sk_reclen = 0;
 }
 
 /*
@@ -1033,8 +1063,10 @@ static int svc_tcp_recvfrom(struct svc_rqst *rqstp)
 	struct svc_serv	*serv = svsk->sk_xprt.xpt_server;
 	int		len;
 	struct kvec *vec;
-	int pnum, vlen;
-	struct rpc_rqst *req = NULL;
+	unsigned int want, base;
+	__be32 *p;
+	__be32 calldir;
+	int pnum;
 
 	dprintk("svc: tcp_recv %p data %d conn %d close %d\n",
 		svsk, test_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags),
@@ -1045,89 +1077,82 @@ static int svc_tcp_recvfrom(struct svc_rqst *rqstp)
 	if (len < 0)
 		goto error;
 
+	base = svc_tcp_restore_pages(svsk, rqstp);
+	want = svc_sock_reclen(svsk) - (svsk->sk_tcplen - sizeof(rpc_fraghdr));
+
 	vec = rqstp->rq_vec;
-	vec[0] = rqstp->rq_arg.head[0];
-	vlen = PAGE_SIZE;
 
-	/*
-	 * We have enough data for the whole tcp record. Let's try and read the
-	 * first 8 bytes to get the xid and the call direction. We can use this
-	 * to figure out if this is a call or a reply to a callback. If
-	 * sk_reclen is < 8 (xid and calldir), then this is a malformed packet.
-	 * In that case, don't bother with the calldir and just read the data.
-	 * It will be rejected in svc_process.
-	 */
-	if (len >= 8) {
-		len = svc_process_calldir(svsk, rqstp, &req, vec);
-		if (len < 0)
-			goto err_again;
-		vlen -= 8;
-	}
+	pnum = copy_pages_to_kvecs(&vec[0], &rqstp->rq_pages[0],
+						svsk->sk_datalen + want);
 
-	pnum = 1;
-	while (vlen < len) {
-		vec[pnum].iov_base = (req) ?
-			page_address(req->rq_private_buf.pages[pnum - 1]) :
-			page_address(rqstp->rq_pages[pnum]);
-		vec[pnum].iov_len = PAGE_SIZE;
-		pnum++;
-		vlen += PAGE_SIZE;
-	}
 	rqstp->rq_respages = &rqstp->rq_pages[pnum];
+	rqstp->rq_next_page = rqstp->rq_respages + 1;
 
 	/* Now receive data */
-	len = svc_recvfrom(rqstp, vec, pnum, len);
-	if (len < 0)
-		goto err_again;
-
-	/*
-	 * Account for the 8 bytes we read earlier
-	 */
-	len += 8;
+	len = svc_partial_recvfrom(rqstp, vec, pnum, want, base);
+	if (len >= 0) {
+		svsk->sk_tcplen += len;
+		svsk->sk_datalen += len;
+	}
+	if (len != want || !svc_sock_final_rec(svsk)) {
+		svc_tcp_save_pages(svsk, rqstp);
+		if (len < 0 && len != -EAGAIN)
+			goto err_delete;
+		if (len == want)
+			svc_tcp_fragment_received(svsk);
+		else
+			dprintk("svc: incomplete TCP record (%d of %d)\n",
+				(int)(svsk->sk_tcplen - sizeof(rpc_fraghdr)),
+				svc_sock_reclen(svsk));
+		goto err_noclose;
+	}
 
-	if (req) {
-		xprt_complete_rqst(req->rq_task, len);
-		len = 0;
-		goto out;
+	if (svsk->sk_datalen < 8) {
+		svsk->sk_datalen = 0;
+		goto err_delete; /* client is nuts. */
 	}
-	dprintk("svc: TCP complete record (%d bytes)\n", len);
-	rqstp->rq_arg.len = len;
+
+	rqstp->rq_arg.len = svsk->sk_datalen;
 	rqstp->rq_arg.page_base = 0;
-	if (len <= rqstp->rq_arg.head[0].iov_len) {
-		rqstp->rq_arg.head[0].iov_len = len;
+	if (rqstp->rq_arg.len <= rqstp->rq_arg.head[0].iov_len) {
+		rqstp->rq_arg.head[0].iov_len = rqstp->rq_arg.len;
 		rqstp->rq_arg.page_len = 0;
-	} else {
-		rqstp->rq_arg.page_len = len - rqstp->rq_arg.head[0].iov_len;
-	}
+	} else
+		rqstp->rq_arg.page_len = rqstp->rq_arg.len - rqstp->rq_arg.head[0].iov_len;
 
 	rqstp->rq_xprt_ctxt   = NULL;
 	rqstp->rq_prot	      = IPPROTO_TCP;
+	rqstp->rq_local	      = !!test_bit(XPT_LOCAL, &svsk->sk_xprt.xpt_flags);
+
+	p = (__be32 *)rqstp->rq_arg.head[0].iov_base;
+	calldir = p[1];
+	if (calldir)
+		len = receive_cb_reply(svsk, rqstp);
 
-out:
 	/* Reset TCP read info */
-	svsk->sk_reclen = 0;
-	svsk->sk_tcplen = 0;
+	svsk->sk_datalen = 0;
+	svc_tcp_fragment_received(svsk);
+
+	if (len < 0)
+		goto error;
 
 	svc_xprt_copy_addrs(rqstp, &svsk->sk_xprt);
-	svc_xprt_received(&svsk->sk_xprt);
 	if (serv->sv_stats)
 		serv->sv_stats->nettcpcnt++;
 
-	return len;
+	return rqstp->rq_arg.len;
 
-err_again:
-	if (len == -EAGAIN) {
-		dprintk("RPC: TCP recvfrom got EAGAIN\n");
-		svc_xprt_received(&svsk->sk_xprt);
-		return len;
-	}
 error:
-	if (len != -EAGAIN) {
-		printk(KERN_NOTICE "%s: recvfrom returned errno %d\n",
-		       svsk->sk_xprt.xpt_server->sv_name, -len);
-		set_bit(XPT_CLOSE, &svsk->sk_xprt.xpt_flags);
-	}
-	return -EAGAIN;
+	if (len != -EAGAIN)
+		goto err_delete;
+	dprintk("RPC: TCP recvfrom got EAGAIN\n");
+	return 0;
+err_delete:
+	printk(KERN_NOTICE "%s: recvfrom returned errno %d\n",
+	       svsk->sk_xprt.xpt_server->sv_name, -len);
+	set_bit(XPT_CLOSE, &svsk->sk_xprt.xpt_flags);
+err_noclose:
+	return 0;	/* record not complete */
 }
 
 /*
@@ -1146,9 +1171,6 @@ static int svc_tcp_sendto(struct svc_rqst *rqstp)
 	reclen = htonl(0x80000000|((xbufp->len ) - 4));
 	memcpy(xbufp->head[0].iov_base, &reclen, 4);
 
-	if (test_bit(XPT_DEAD, &rqstp->rq_xprt->xpt_flags))
-		return -ENOTCONN;
-
 	sent = svc_sendto(rqstp, &rqstp->rq_res);
 	if (sent != xbufp->len) {
 		printk(KERN_NOTICE
@@ -1184,18 +1206,73 @@ static int svc_tcp_has_wspace(struct svc_xprt *xprt)
 	if (test_bit(XPT_LISTENER, &xprt->xpt_flags))
 		return 1;
 	required = atomic_read(&xprt->xpt_reserved) + serv->sv_max_mesg;
-	if (sk_stream_wspace(svsk->sk_sk) >= required)
+	if (sk_stream_wspace(svsk->sk_sk) >= required ||
+	    (sk_stream_min_wspace(svsk->sk_sk) == 0 &&
+	     atomic_read(&xprt->xpt_reserved) == 0))
 		return 1;
 	set_bit(SOCK_NOSPACE, &svsk->sk_sock->flags);
 	return 0;
 }
 
 static struct svc_xprt *svc_tcp_create(struct svc_serv *serv,
+				       struct net *net,
+				       struct sockaddr *sa, int salen,
+				       int flags)
+{
+	return svc_create_socket(serv, IPPROTO_TCP, net, sa, salen, flags);
+}
+
+#if defined(CONFIG_SUNRPC_BACKCHANNEL)
+static struct svc_xprt *svc_bc_create_socket(struct svc_serv *, int,
+					     struct net *, struct sockaddr *,
+					     int, int);
+static void svc_bc_sock_free(struct svc_xprt *xprt);
+
+static struct svc_xprt *svc_bc_tcp_create(struct svc_serv *serv,
+				       struct net *net,
 				       struct sockaddr *sa, int salen,
 				       int flags)
 {
-	return svc_create_socket(serv, IPPROTO_TCP, sa, salen, flags);
+	return svc_bc_create_socket(serv, IPPROTO_TCP, net, sa, salen, flags);
+}
+
+static void svc_bc_tcp_sock_detach(struct svc_xprt *xprt)
+{
+}
+
+static struct svc_xprt_ops svc_tcp_bc_ops = {
+	.xpo_create = svc_bc_tcp_create,
+	.xpo_detach = svc_bc_tcp_sock_detach,
+	.xpo_free = svc_bc_sock_free,
+	.xpo_prep_reply_hdr = svc_tcp_prep_reply_hdr,
+	.xpo_secure_port = svc_sock_secure_port,
+};
+
+static struct svc_xprt_class svc_tcp_bc_class = {
+	.xcl_name = "tcp-bc",
+	.xcl_owner = THIS_MODULE,
+	.xcl_ops = &svc_tcp_bc_ops,
+	.xcl_max_payload = RPCSVC_MAXPAYLOAD_TCP,
+};
+
+static void svc_init_bc_xprt_sock(void)
+{
+	svc_reg_xprt_class(&svc_tcp_bc_class);
+}
+
+static void svc_cleanup_bc_xprt_sock(void)
+{
+	svc_unreg_xprt_class(&svc_tcp_bc_class);
+}
+#else /* CONFIG_SUNRPC_BACKCHANNEL */
+static void svc_init_bc_xprt_sock(void)
+{
+}
+
+static void svc_cleanup_bc_xprt_sock(void)
+{
 }
+#endif /* CONFIG_SUNRPC_BACKCHANNEL */
 
 static struct svc_xprt_ops svc_tcp_ops = {
 	.xpo_create = svc_tcp_create,
@@ -1207,6 +1284,7 @@ static struct svc_xprt_ops svc_tcp_ops = {
 	.xpo_prep_reply_hdr = svc_tcp_prep_reply_hdr,
 	.xpo_has_wspace = svc_tcp_has_wspace,
 	.xpo_accept = svc_tcp_accept,
+	.xpo_secure_port = svc_sock_secure_port,
 };
 
 static struct svc_xprt_class svc_tcp_class = {
@@ -1220,19 +1298,22 @@ void svc_init_xprt_sock(void)
 {
 	svc_reg_xprt_class(&svc_tcp_class);
 	svc_reg_xprt_class(&svc_udp_class);
+	svc_init_bc_xprt_sock();
 }
 
 void svc_cleanup_xprt_sock(void)
 {
 	svc_unreg_xprt_class(&svc_tcp_class);
 	svc_unreg_xprt_class(&svc_udp_class);
+	svc_cleanup_bc_xprt_sock();
 }
 
 static void svc_tcp_init(struct svc_sock *svsk, struct svc_serv *serv)
 {
 	struct sock	*sk = svsk->sk_sk;
 
-	svc_xprt_init(&svc_tcp_class, &svsk->sk_xprt, serv);
+	svc_xprt_init(sock_net(svsk->sk_sock->sk), &svc_tcp_class,
+		      &svsk->sk_xprt, serv);
 	set_bit(XPT_CACHE_AUTH, &svsk->sk_xprt.xpt_flags);
 	if (sk->sk_state == TCP_LISTEN) {
 		dprintk("setting up TCP socket for listening\n");
@@ -1247,18 +1328,11 @@ static void svc_tcp_init(struct svc_sock *svsk, struct svc_serv *serv)
 
 		svsk->sk_reclen = 0;
 		svsk->sk_tcplen = 0;
+		svsk->sk_datalen = 0;
+		memset(&svsk->sk_pages[0], 0, sizeof(svsk->sk_pages));
 
 		tcp_sk(sk)->nonagle |= TCP_NAGLE_OFF;
 
-		/* initialise setting must have enough space to
-		 * receive and respond to one request.
-		 * svc_tcp_recvfrom will re-adjust if necessary
-		 */
-		svc_sock_setbufsize(svsk->sk_sock,
-				    3 * svsk->sk_xprt.xpt_server->sv_max_mesg,
-				    3 * svsk->sk_xprt.xpt_server->sv_max_mesg);
-
-		set_bit(XPT_CHNGBUF, &svsk->sk_xprt.xpt_flags);
 		set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
 		if (sk->sk_state != TCP_ESTABLISHED)
 			set_bit(XPT_CLOSE, &svsk->sk_xprt.xpt_flags);
@@ -1271,19 +1345,11 @@ void svc_sock_update_bufs(struct svc_serv *serv)
 	 * The number of server threads has changed. Update
 	 * rcvbuf and sndbuf accordingly on all sockets
 	 */
-	struct list_head *le;
+	struct svc_sock *svsk;
 
 	spin_lock_bh(&serv->sv_lock);
-	list_for_each(le, &serv->sv_permsocks) {
-		struct svc_sock *svsk =
-			list_entry(le, struct svc_sock, sk_xprt.xpt_list);
+	list_for_each_entry(svsk, &serv->sv_permsocks, sk_xprt.xpt_list)
 		set_bit(XPT_CHNGBUF, &svsk->sk_xprt.xpt_flags);
-	}
-	list_for_each(le, &serv->sv_tempsocks) {
-		struct svc_sock *svsk =
-			list_entry(le, struct svc_sock, sk_xprt.xpt_list);
-		set_bit(XPT_CHNGBUF, &svsk->sk_xprt.xpt_flags);
-	}
 	spin_unlock_bh(&serv->sv_lock);
 }
 EXPORT_SYMBOL_GPL(svc_sock_update_bufs);
@@ -1294,28 +1360,29 @@ EXPORT_SYMBOL_GPL(svc_sock_update_bufs);
  */
 static struct svc_sock *svc_setup_socket(struct svc_serv *serv,
 						struct socket *sock,
-						int *errp, int flags)
+						int flags)
 {
 	struct svc_sock	*svsk;
 	struct sock	*inet;
 	int		pmap_register = !(flags & SVC_SOCK_ANONYMOUS);
+	int		err = 0;
 
 	dprintk("svc: svc_setup_socket %p\n", sock);
-	if (!(svsk = kzalloc(sizeof(*svsk), GFP_KERNEL))) {
-		*errp = -ENOMEM;
-		return NULL;
-	}
+	svsk = kzalloc(sizeof(*svsk), GFP_KERNEL);
+	if (!svsk)
+		return ERR_PTR(-ENOMEM);
 
 	inet = sock->sk;
 
 	/* Register socket with portmapper */
-	if (*errp >= 0 && pmap_register)
-		*errp = svc_register(serv, inet->sk_family, inet->sk_protocol,
+	if (pmap_register)
+		err = svc_register(serv, sock_net(sock->sk), inet->sk_family,
+				     inet->sk_protocol,
 				     ntohs(inet_sk(inet)->inet_sport));
 
-	if (*errp < 0) {
+	if (err < 0) {
 		kfree(svsk);
-		return NULL;
+		return ERR_PTR(err);
 	}
 
 	inet->sk_user_data = svsk;
@@ -1328,8 +1395,14 @@ static struct svc_sock *svc_setup_socket(struct svc_serv *serv,
 	/* Initialize the socket */
 	if (sock->type == SOCK_DGRAM)
 		svc_udp_init(svsk, serv);
-	else
+	else {
+		/* initialise setting must have enough space to
+		 * receive and respond to one request.
+		 */
+		svc_sock_setbufsize(svsk->sk_sock, 4 * serv->sv_max_mesg,
+					4 * serv->sv_max_mesg);
 		svc_tcp_init(svsk, serv);
+	}
 
 	dprintk("svc: svc_setup_socket created %p (inet %p)\n",
 				svsk, svsk->sk_sk);
@@ -1337,6 +1410,22 @@ static struct svc_sock *svc_setup_socket(struct svc_serv *serv,
 	return svsk;
 }
 
+bool svc_alien_sock(struct net *net, int fd)
+{
+	int err;
+	struct socket *sock = sockfd_lookup(fd, &err);
+	bool ret = false;
+
+	if (!sock)
+		goto out;
+	if (sock_net(sock->sk) != net)
+		ret = true;
+	sockfd_put(sock);
+out:
+	return ret;
+}
+EXPORT_SYMBOL_GPL(svc_alien_sock);
+
 /**
  * svc_addsock - add a listener socket to an RPC service
  * @serv: pointer to RPC service to which to add a new listener
@@ -1354,42 +1443,38 @@ int svc_addsock(struct svc_serv *serv, const int fd, char *name_return,
 	int err = 0;
 	struct socket *so = sockfd_lookup(fd, &err);
 	struct svc_sock *svsk = NULL;
+	struct sockaddr_storage addr;
+	struct sockaddr *sin = (struct sockaddr *)&addr;
+	int salen;
 
 	if (!so)
 		return err;
-	if (so->sk->sk_family != AF_INET)
-		err =  -EAFNOSUPPORT;
-	else if (so->sk->sk_protocol != IPPROTO_TCP &&
+	err = -EAFNOSUPPORT;
+	if ((so->sk->sk_family != PF_INET) && (so->sk->sk_family != PF_INET6))
+		goto out;
+	err =  -EPROTONOSUPPORT;
+	if (so->sk->sk_protocol != IPPROTO_TCP &&
 	    so->sk->sk_protocol != IPPROTO_UDP)
-		err =  -EPROTONOSUPPORT;
-	else if (so->state > SS_UNCONNECTED)
-		err = -EISCONN;
-	else {
-		if (!try_module_get(THIS_MODULE))
-			err = -ENOENT;
-		else
-			svsk = svc_setup_socket(serv, so, &err,
-						SVC_SOCK_DEFAULTS);
-		if (svsk) {
-			struct sockaddr_storage addr;
-			struct sockaddr *sin = (struct sockaddr *)&addr;
-			int salen;
-			if (kernel_getsockname(svsk->sk_sock, sin, &salen) == 0)
-				svc_xprt_set_local(&svsk->sk_xprt, sin, salen);
-			clear_bit(XPT_TEMP, &svsk->sk_xprt.xpt_flags);
-			spin_lock_bh(&serv->sv_lock);
-			list_add(&svsk->sk_xprt.xpt_list, &serv->sv_permsocks);
-			spin_unlock_bh(&serv->sv_lock);
-			svc_xprt_received(&svsk->sk_xprt);
-			err = 0;
-		} else
-			module_put(THIS_MODULE);
-	}
-	if (err) {
-		sockfd_put(so);
-		return err;
+		goto out;
+	err = -EISCONN;
+	if (so->state > SS_UNCONNECTED)
+		goto out;
+	err = -ENOENT;
+	if (!try_module_get(THIS_MODULE))
+		goto out;
+	svsk = svc_setup_socket(serv, so, SVC_SOCK_DEFAULTS);
+	if (IS_ERR(svsk)) {
+		module_put(THIS_MODULE);
+		err = PTR_ERR(svsk);
+		goto out;
 	}
+	if (kernel_getsockname(svsk->sk_sock, sin, &salen) == 0)
+		svc_xprt_set_local(&svsk->sk_xprt, sin, salen);
+	svc_add_new_perm_xprt(serv, &svsk->sk_xprt);
 	return svc_one_sock_name(svsk, name_return, len);
+out:
+	sockfd_put(so);
+	return err;
 }
 EXPORT_SYMBOL_GPL(svc_addsock);
 
@@ -1398,6 +1483,7 @@ EXPORT_SYMBOL_GPL(svc_addsock);
  */
 static struct svc_xprt *svc_create_socket(struct svc_serv *serv,
 					  int protocol,
+					  struct net *net,
 					  struct sockaddr *sin, int len,
 					  int flags)
 {
@@ -1434,7 +1520,7 @@ static struct svc_xprt *svc_create_socket(struct svc_serv *serv,
 		return ERR_PTR(-EINVAL);
 	}
 
-	error = sock_create_kern(family, type, protocol, &sock);
+	error = __sock_create(net, family, type, protocol, &sock, 1);
 	if (error < 0)
 		return ERR_PTR(error);
 
@@ -1451,7 +1537,7 @@ static struct svc_xprt *svc_create_socket(struct svc_serv *serv,
 					(char *)&val, sizeof(val));
 
 	if (type == SOCK_STREAM)
-		sock->sk->sk_reuse = 1;		/* allow address reuse */
+		sock->sk->sk_reuse = SK_CAN_REUSE; /* allow address reuse */
 	error = kernel_bind(sock, sin, len);
 	if (error < 0)
 		goto bummer;
@@ -1466,11 +1552,13 @@ static struct svc_xprt *svc_create_socket(struct svc_serv *serv,
 			goto bummer;
 	}
 
-	if ((svsk = svc_setup_socket(serv, sock, &error, flags)) != NULL) {
-		svc_xprt_set_local(&svsk->sk_xprt, newsin, newlen);
-		return (struct svc_xprt *)svsk;
+	svsk = svc_setup_socket(serv, sock, flags);
+	if (IS_ERR(svsk)) {
+		error = PTR_ERR(svsk);
+		goto bummer;
 	}
-
+	svc_xprt_set_local(&svsk->sk_xprt, newsin, newlen);
+	return (struct svc_xprt *)svsk;
 bummer:
 	dprintk("svc: svc_create_socket error = %d\n", -error);
 	sock_release(sock);
@@ -1485,6 +1573,7 @@ static void svc_sock_detach(struct svc_xprt *xprt)
 {
 	struct svc_sock *svsk = container_of(xprt, struct svc_sock, sk_xprt);
 	struct sock *sk = svsk->sk_sk;
+	wait_queue_head_t *wq;
 
 	dprintk("svc: svc_sock_detach(%p)\n", svsk);
 
@@ -1493,8 +1582,9 @@ static void svc_sock_detach(struct svc_xprt *xprt)
 	sk->sk_data_ready = svsk->sk_odata;
 	sk->sk_write_space = svsk->sk_owspace;
 
-	if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))
-		wake_up_interruptible(sk->sk_sleep);
+	wq = sk_sleep(sk);
+	if (wq && waitqueue_active(wq))
+		wake_up_interruptible(wq);
 }
 
 /*
@@ -1508,8 +1598,10 @@ static void svc_tcp_sock_detach(struct svc_xprt *xprt)
 
 	svc_sock_detach(xprt);
 
-	if (!test_bit(XPT_LISTENER, &xprt->xpt_flags))
+	if (!test_bit(XPT_LISTENER, &xprt->xpt_flags)) {
+		svc_tcp_clear_pages(svsk);
 		kernel_sock_shutdown(svsk->sk_sock, SHUT_RDWR);
+	}
 }
 
 /*
@@ -1527,41 +1619,43 @@ static void svc_sock_free(struct svc_xprt *xprt)
 	kfree(svsk);
 }
 
+#if defined(CONFIG_SUNRPC_BACKCHANNEL)
 /*
- * Create a svc_xprt.
- *
- * For internal use only (e.g. nfsv4.1 backchannel).
- * Callers should typically use the xpo_create() method.
+ * Create a back channel svc_xprt which shares the fore channel socket.
  */
-struct svc_xprt *svc_sock_create(struct svc_serv *serv, int prot)
+static struct svc_xprt *svc_bc_create_socket(struct svc_serv *serv,
+					     int protocol,
+					     struct net *net,
+					     struct sockaddr *sin, int len,
+					     int flags)
 {
 	struct svc_sock *svsk;
-	struct svc_xprt *xprt = NULL;
+	struct svc_xprt *xprt;
+
+	if (protocol != IPPROTO_TCP) {
+		printk(KERN_WARNING "svc: only TCP sockets"
+			" supported on shared back channel\n");
+		return ERR_PTR(-EINVAL);
+	}
 
-	dprintk("svc: %s\n", __func__);
 	svsk = kzalloc(sizeof(*svsk), GFP_KERNEL);
 	if (!svsk)
-		goto out;
+		return ERR_PTR(-ENOMEM);
 
 	xprt = &svsk->sk_xprt;
-	if (prot == IPPROTO_TCP)
-		svc_xprt_init(&svc_tcp_class, xprt, serv);
-	else if (prot == IPPROTO_UDP)
-		svc_xprt_init(&svc_udp_class, xprt, serv);
-	else
-		BUG();
-out:
-	dprintk("svc: %s return %p\n", __func__, xprt);
+	svc_xprt_init(net, &svc_tcp_bc_class, xprt, serv);
+
+	serv->sv_bc_xprt = xprt;
+
 	return xprt;
 }
-EXPORT_SYMBOL_GPL(svc_sock_create);
 
 /*
- * Destroy a svc_sock.
+ * Free a back channel svc_sock.
  */
-void svc_sock_destroy(struct svc_xprt *xprt)
+static void svc_bc_sock_free(struct svc_xprt *xprt)
 {
 	if (xprt)
 		kfree(container_of(xprt, struct svc_sock, sk_xprt));
 }
-EXPORT_SYMBOL_GPL(svc_sock_destroy);
+#endif /* CONFIG_SUNRPC_BACKCHANNEL */