Merge branch 'nfs-for-2.6.33'

30 files changed, 1078 insertions, 469 deletions

diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c
index 293fa0528a6e..73ab220354df 100644
--- a/fs/nfs/callback.c
+++ b/fs/nfs/callback.c
@@ -78,11 +78,6 @@ nfs4_callback_svc(void *vrqstp)
 
 	set_freezable();
 
-	/*
-	 * FIXME: do we really need to run this under the BKL? If so, please
-	 * add a comment about what it's intended to protect.
-	 */
-	lock_kernel();
 	while (!kthread_should_stop()) {
 		/*
 		 * Listen for a request on the socket
@@ -104,7 +99,6 @@ nfs4_callback_svc(void *vrqstp)
 		preverr = err;
 		svc_process(rqstp);
 	}
-	unlock_kernel();
 	return 0;
 }
 
@@ -160,11 +154,6 @@ nfs41_callback_svc(void *vrqstp)
 
 	set_freezable();
 
-	/*
-	 * FIXME: do we really need to run this under the BKL? If so, please
-	 * add a comment about what it's intended to protect.
-	 */
-	lock_kernel();
 	while (!kthread_should_stop()) {
 		prepare_to_wait(&serv->sv_cb_waitq, &wq, TASK_INTERRUPTIBLE);
 		spin_lock_bh(&serv->sv_cb_lock);
@@ -183,7 +172,6 @@ nfs41_callback_svc(void *vrqstp)
 		}
 		finish_wait(&serv->sv_cb_waitq, &wq);
 	}
-	unlock_kernel();
 	return 0;
 }
 
@@ -397,6 +385,7 @@ static int nfs_callback_authenticate(struct svc_rqst *rqstp)
  */
 static struct svc_version *nfs4_callback_version[] = {
 	[1] = &nfs4_callback_version1,
+	[4] = &nfs4_callback_version4,
 };
 
 static struct svc_stat nfs4_callback_stats;
diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h
index 07baa8254ca1..d4036be0b589 100644
--- a/fs/nfs/callback.h
+++ b/fs/nfs/callback.h
@@ -106,6 +106,19 @@ struct cb_sequenceres {
 extern unsigned nfs4_callback_sequence(struct cb_sequenceargs *args,
 				       struct cb_sequenceres *res);
 
+extern int nfs41_validate_delegation_stateid(struct nfs_delegation *delegation,
+					     const nfs4_stateid *stateid);
+
+#define RCA4_TYPE_MASK_RDATA_DLG	0
+#define RCA4_TYPE_MASK_WDATA_DLG	1
+
+struct cb_recallanyargs {
+	struct sockaddr	*craa_addr;
+	uint32_t	craa_objs_to_keep;
+	uint32_t	craa_type_mask;
+};
+
+extern unsigned nfs4_callback_recallany(struct cb_recallanyargs *args, void *dummy);
 #endif /* CONFIG_NFS_V4_1 */
 
 extern __be32 nfs4_callback_getattr(struct cb_getattrargs *args, struct cb_getattrres *res);
@@ -114,8 +127,9 @@ extern __be32 nfs4_callback_recall(struct cb_recallargs *args, void *dummy);
 #ifdef CONFIG_NFS_V4
 extern int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt);
 extern void nfs_callback_down(int minorversion);
+extern int nfs4_validate_delegation_stateid(struct nfs_delegation *delegation,
+					    const nfs4_stateid *stateid);
 #endif /* CONFIG_NFS_V4 */
-
 /*
  * nfs41: Callbacks are expected to not cause substantial latency,
  * so we limit their concurrency to 1 by setting up the maximum number
diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c
index b7da1f54da68..defa9b4c470e 100644
--- a/fs/nfs/callback_proc.c
+++ b/fs/nfs/callback_proc.c
@@ -61,6 +61,16 @@ out:
 	return res->status;
 }
 
+static int (*nfs_validate_delegation_stateid(struct nfs_client *clp))(struct nfs_delegation *, const nfs4_stateid *)
+{
+#if defined(CONFIG_NFS_V4_1)
+	if (clp->cl_minorversion > 0)
+		return nfs41_validate_delegation_stateid;
+#endif
+	return nfs4_validate_delegation_stateid;
+}
+
+
 __be32 nfs4_callback_recall(struct cb_recallargs *args, void *dummy)
 {
 	struct nfs_client *clp;
@@ -81,7 +91,8 @@ __be32 nfs4_callback_recall(struct cb_recallargs *args, void *dummy)
 		inode = nfs_delegation_find_inode(clp, &args->fh);
 		if (inode != NULL) {
 			/* Set up a helper thread to actually return the delegation */
-			switch(nfs_async_inode_return_delegation(inode, &args->stateid)) {
+			switch (nfs_async_inode_return_delegation(inode, &args->stateid,
+								  nfs_validate_delegation_stateid(clp))) {
 				case 0:
 					res = 0;
 					break;
@@ -102,8 +113,31 @@ out:
 	return res;
 }
 
+int nfs4_validate_delegation_stateid(struct nfs_delegation *delegation, const nfs4_stateid *stateid)
+{
+	if (delegation == NULL || memcmp(delegation->stateid.data, stateid->data,
+					 sizeof(delegation->stateid.data)) != 0)
+		return 0;
+	return 1;
+}
+
 #if defined(CONFIG_NFS_V4_1)
 
+int nfs41_validate_delegation_stateid(struct nfs_delegation *delegation, const nfs4_stateid *stateid)
+{
+	if (delegation == NULL)
+		return 0;
+
+	/* seqid is 4-bytes long */
+	if (((u32 *) &stateid->data)[0] != 0)
+		return 0;
+	if (memcmp(&delegation->stateid.data[4], &stateid->data[4],
+		   sizeof(stateid->data)-4))
+		return 0;
+
+	return 1;
+}
+
 /*
  * Validate the sequenceID sent by the server.
  * Return success if the sequenceID is one more than what we last saw on
@@ -227,4 +261,32 @@ out:
 	return res->csr_status;
 }
 
+unsigned nfs4_callback_recallany(struct cb_recallanyargs *args, void *dummy)
+{
+	struct nfs_client *clp;
+	int status;
+	fmode_t flags = 0;
+
+	status = htonl(NFS4ERR_OP_NOT_IN_SESSION);
+	clp = nfs_find_client(args->craa_addr, 4);
+	if (clp == NULL)
+		goto out;
+
+	dprintk("NFS: RECALL_ANY callback request from %s\n",
+		rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_ADDR));
+
+	if (test_bit(RCA4_TYPE_MASK_RDATA_DLG, (const unsigned long *)
+		     &args->craa_type_mask))
+		flags = FMODE_READ;
+	if (test_bit(RCA4_TYPE_MASK_WDATA_DLG, (const unsigned long *)
+		     &args->craa_type_mask))
+		flags |= FMODE_WRITE;
+
+	if (flags)
+		nfs_expire_all_delegation_types(clp, flags);
+	status = htonl(NFS4_OK);
+out:
+	dprintk("%s: exit with status = %d\n", __func__, ntohl(status));
+	return status;
+}
 #endif /* CONFIG_NFS_V4_1 */
diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c
index 76b0aa0f73bf..8e1a2511c8be 100644
--- a/fs/nfs/callback_xdr.c
+++ b/fs/nfs/callback_xdr.c
@@ -23,6 +23,7 @@
 #if defined(CONFIG_NFS_V4_1)
 #define CB_OP_SEQUENCE_RES_MAXSZ	(CB_OP_HDR_RES_MAXSZ + \
 					4 + 1 + 3)
+#define CB_OP_RECALLANY_RES_MAXSZ	(CB_OP_HDR_RES_MAXSZ)
 #endif /* CONFIG_NFS_V4_1 */
 
 #define NFSDBG_FACILITY NFSDBG_CALLBACK
@@ -326,6 +327,25 @@ out_free:
 	goto out;
 }
 
+static unsigned decode_recallany_args(struct svc_rqst *rqstp,
+				      struct xdr_stream *xdr,
+				      struct cb_recallanyargs *args)
+{
+	uint32_t *p;
+
+	args->craa_addr = svc_addr(rqstp);
+	p = read_buf(xdr, 4);
+	if (unlikely(p == NULL))
+		return htonl(NFS4ERR_BADXDR);
+	args->craa_objs_to_keep = ntohl(*p++);
+	p = read_buf(xdr, 4);
+	if (unlikely(p == NULL))
+		return htonl(NFS4ERR_BADXDR);
+	args->craa_type_mask = ntohl(*p);
+
+	return 0;
+}
+
 #endif /* CONFIG_NFS_V4_1 */
 
 static __be32 encode_string(struct xdr_stream *xdr, unsigned int len, const char *str)
@@ -533,6 +553,7 @@ preprocess_nfs41_op(int nop, unsigned int op_nr, struct callback_op **op)
 	case OP_CB_GETATTR:
 	case OP_CB_RECALL:
 	case OP_CB_SEQUENCE:
+	case OP_CB_RECALL_ANY:
 		*op = &callback_ops[op_nr];
 		break;
 
@@ -540,7 +561,6 @@ preprocess_nfs41_op(int nop, unsigned int op_nr, struct callback_op **op)
 	case OP_CB_NOTIFY_DEVICEID:
 	case OP_CB_NOTIFY:
 	case OP_CB_PUSH_DELEG:
-	case OP_CB_RECALL_ANY:
 	case OP_CB_RECALLABLE_OBJ_AVAIL:
 	case OP_CB_RECALL_SLOT:
 	case OP_CB_WANTS_CANCELLED:
@@ -688,6 +708,11 @@ static struct callback_op callback_ops[] = {
 		.encode_res = (callback_encode_res_t)encode_cb_sequence_res,
 		.res_maxsize = CB_OP_SEQUENCE_RES_MAXSZ,
 	},
+	[OP_CB_RECALL_ANY] = {
+		.process_op = (callback_process_op_t)nfs4_callback_recallany,
+		.decode_args = (callback_decode_arg_t)decode_recallany_args,
+		.res_maxsize = CB_OP_RECALLANY_RES_MAXSZ,
+	},
 #endif /* CONFIG_NFS_V4_1 */
 };
 
@@ -718,3 +743,10 @@ struct svc_version nfs4_callback_version1 = {
 	.vs_dispatch = NULL,
 };
 
+struct svc_version nfs4_callback_version4 = {
+	.vs_vers = 4,
+	.vs_nproc = ARRAY_SIZE(nfs4_callback_procedures1),
+	.vs_proc = nfs4_callback_procedures1,
+	.vs_xdrsize = NFS4_CALLBACK_XDRSIZE,
+	.vs_dispatch = NULL,
+};
diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 99ea196f071f..ee77713ce68b 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -1260,10 +1260,20 @@ error:
 static void nfs4_session_set_rwsize(struct nfs_server *server)
 {
 #ifdef CONFIG_NFS_V4_1
+	struct nfs4_session *sess;
+	u32 server_resp_sz;
+	u32 server_rqst_sz;
+
 	if (!nfs4_has_session(server->nfs_client))
 		return;
-	server->rsize = server->nfs_client->cl_session->fc_attrs.max_resp_sz;
-	server->wsize = server->nfs_client->cl_session->fc_attrs.max_rqst_sz;
+	sess = server->nfs_client->cl_session;
+	server_resp_sz = sess->fc_attrs.max_resp_sz - nfs41_maxread_overhead;
+	server_rqst_sz = sess->fc_attrs.max_rqst_sz - nfs41_maxwrite_overhead;
+
+	if (server->rsize > server_resp_sz)
+		server->rsize = server_resp_sz;
+	if (server->wsize > server_rqst_sz)
+		server->wsize = server_rqst_sz;
 #endif /* CONFIG_NFS_V4_1 */
 }
 
diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c
index 6dd48a4405b4..2563bebc4c67 100644
--- a/fs/nfs/delegation.c
+++ b/fs/nfs/delegation.c
@@ -92,7 +92,7 @@ out:
 	return status;
 }
 
-static void nfs_delegation_claim_opens(struct inode *inode, const nfs4_stateid *stateid)
+static int nfs_delegation_claim_opens(struct inode *inode, const nfs4_stateid *stateid)
 {
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs_open_context *ctx;
@@ -116,10 +116,11 @@ again:
 			err = nfs_delegation_claim_locks(ctx, state);
 		put_nfs_open_context(ctx);
 		if (err != 0)
-			return;
+			return err;
 		goto again;
 	}
 	spin_unlock(&inode->i_lock);
+	return 0;
 }
 
 /*
@@ -261,30 +262,34 @@ static void nfs_msync_inode(struct inode *inode)
 /*
  * Basic procedure for returning a delegation to the server
  */
-static int __nfs_inode_return_delegation(struct inode *inode, struct nfs_delegation *delegation)
+static int __nfs_inode_return_delegation(struct inode *inode, struct nfs_delegation *delegation, int issync)
 {
 	struct nfs_inode *nfsi = NFS_I(inode);
+	int err;
 
-	nfs_msync_inode(inode);
 	/*
 	 * Guard against new delegated open/lock/unlock calls and against
 	 * state recovery
 	 */
 	down_write(&nfsi->rwsem);
-	nfs_delegation_claim_opens(inode, &delegation->stateid);
+	err = nfs_delegation_claim_opens(inode, &delegation->stateid);
 	up_write(&nfsi->rwsem);
-	nfs_msync_inode(inode);
+	if (err)
+		goto out;
 
-	return nfs_do_return_delegation(inode, delegation, 1);
+	err = nfs_do_return_delegation(inode, delegation, issync);
+out:
+	return err;
 }
 
 /*
  * Return all delegations that have been marked for return
  */
-void nfs_client_return_marked_delegations(struct nfs_client *clp)
+int nfs_client_return_marked_delegations(struct nfs_client *clp)
 {
 	struct nfs_delegation *delegation;
 	struct inode *inode;
+	int err = 0;
 
 restart:
 	rcu_read_lock();
@@ -298,12 +303,18 @@ restart:
 		delegation = nfs_detach_delegation_locked(NFS_I(inode), NULL);
 		spin_unlock(&clp->cl_lock);
 		rcu_read_unlock();
-		if (delegation != NULL)
-			__nfs_inode_return_delegation(inode, delegation);
+		if (delegation != NULL) {
+			filemap_flush(inode->i_mapping);
+			err = __nfs_inode_return_delegation(inode, delegation, 0);
+		}
 		iput(inode);
-		goto restart;
+		if (!err)
+			goto restart;
+		set_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state);
+		return err;
 	}
 	rcu_read_unlock();
+	return 0;
 }
 
 /*
@@ -338,8 +349,10 @@ int nfs_inode_return_delegation(struct inode *inode)
 		spin_lock(&clp->cl_lock);
 		delegation = nfs_detach_delegation_locked(nfsi, NULL);
 		spin_unlock(&clp->cl_lock);
-		if (delegation != NULL)
-			err = __nfs_inode_return_delegation(inode, delegation);
+		if (delegation != NULL) {
+			nfs_msync_inode(inode);
+			err = __nfs_inode_return_delegation(inode, delegation, 1);
+		}
 	}
 	return err;
 }
@@ -368,33 +381,47 @@ void nfs_super_return_all_delegations(struct super_block *sb)
 		spin_unlock(&delegation->lock);
 	}
 	rcu_read_unlock();
-	nfs_client_return_marked_delegations(clp);
+	if (nfs_client_return_marked_delegations(clp) != 0)
+		nfs4_schedule_state_manager(clp);
 }
 
-static void nfs_client_mark_return_all_delegations(struct nfs_client *clp)
+static
+void nfs_client_mark_return_all_delegation_types(struct nfs_client *clp, fmode_t flags)
 {
 	struct nfs_delegation *delegation;
 
 	rcu_read_lock();
 	list_for_each_entry_rcu(delegation, &clp->cl_delegations, super_list) {
-		set_bit(NFS_DELEGATION_RETURN, &delegation->flags);
-		set_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state);
+		if ((delegation->type == (FMODE_READ|FMODE_WRITE)) && !(flags & FMODE_WRITE))
+			continue;
+		if (delegation->type & flags)
+			nfs_mark_return_delegation(clp, delegation);
 	}
 	rcu_read_unlock();
 }
 
+static void nfs_client_mark_return_all_delegations(struct nfs_client *clp)
+{
+	nfs_client_mark_return_all_delegation_types(clp, FMODE_READ|FMODE_WRITE);
+}
+
 static void nfs_delegation_run_state_manager(struct nfs_client *clp)
 {
 	if (test_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state))
 		nfs4_schedule_state_manager(clp);
 }
 
-void nfs_expire_all_delegations(struct nfs_client *clp)
+void nfs_expire_all_delegation_types(struct nfs_client *clp, fmode_t flags)
 {
-	nfs_client_mark_return_all_delegations(clp);
+	nfs_client_mark_return_all_delegation_types(clp, flags);
 	nfs_delegation_run_state_manager(clp);
 }
 
+void nfs_expire_all_delegations(struct nfs_client *clp)
+{
+	nfs_expire_all_delegation_types(clp, FMODE_READ|FMODE_WRITE);
+}
+
 /*
  * Return all delegations following an NFS4ERR_CB_PATH_DOWN error.
  */
@@ -413,8 +440,7 @@ static void nfs_client_mark_return_unreferenced_delegations(struct nfs_client *c
 	list_for_each_entry_rcu(delegation, &clp->cl_delegations, super_list) {
 		if (test_and_clear_bit(NFS_DELEGATION_REFERENCED, &delegation->flags))
 			continue;
-		set_bit(NFS_DELEGATION_RETURN, &delegation->flags);
-		set_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state);
+		nfs_mark_return_delegation(clp, delegation);
 	}
 	rcu_read_unlock();
 }
@@ -428,18 +454,21 @@ void nfs_expire_unreferenced_delegations(struct nfs_client *clp)
 /*
  * Asynchronous delegation recall!
  */
-int nfs_async_inode_return_delegation(struct inode *inode, const nfs4_stateid *stateid)
+int nfs_async_inode_return_delegation(struct inode *inode, const nfs4_stateid *stateid,
+				      int (*validate_stateid)(struct nfs_delegation *delegation,
+							      const nfs4_stateid *stateid))
 {
 	struct nfs_client *clp = NFS_SERVER(inode)->nfs_client;
 	struct nfs_delegation *delegation;
 
 	rcu_read_lock();
 	delegation = rcu_dereference(NFS_I(inode)->delegation);
-	if (delegation == NULL || memcmp(delegation->stateid.data, stateid->data,
-				sizeof(delegation->stateid.data)) != 0) {
+
+	if (!validate_stateid(delegation, stateid)) {
 		rcu_read_unlock();
 		return -ENOENT;
 	}
+
 	nfs_mark_return_delegation(clp, delegation);
 	rcu_read_unlock();
 	nfs_delegation_run_state_manager(clp);
diff --git a/fs/nfs/delegation.h b/fs/nfs/delegation.h
index 09f383795174..944b627ec6e1 100644
--- a/fs/nfs/delegation.h
+++ b/fs/nfs/delegation.h
@@ -34,15 +34,18 @@ enum {
 int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct nfs_openres *res);
 void nfs_inode_reclaim_delegation(struct inode *inode, struct rpc_cred *cred, struct nfs_openres *res);
 int nfs_inode_return_delegation(struct inode *inode);
-int nfs_async_inode_return_delegation(struct inode *inode, const nfs4_stateid *stateid);
+int nfs_async_inode_return_delegation(struct inode *inode, const nfs4_stateid *stateid,
+				      int (*validate_stateid)(struct nfs_delegation *delegation,
+							      const nfs4_stateid *stateid));
 void nfs_inode_return_delegation_noreclaim(struct inode *inode);
 
 struct inode *nfs_delegation_find_inode(struct nfs_client *clp, const struct nfs_fh *fhandle);
 void nfs_super_return_all_delegations(struct super_block *sb);
 void nfs_expire_all_delegations(struct nfs_client *clp);
+void nfs_expire_all_delegation_types(struct nfs_client *clp, fmode_t flags);
 void nfs_expire_unreferenced_delegations(struct nfs_client *clp);
 void nfs_handle_cb_pathdown(struct nfs_client *clp);
-void nfs_client_return_marked_delegations(struct nfs_client *clp);
+int nfs_client_return_marked_delegations(struct nfs_client *clp);
 
 void nfs_delegation_mark_reclaim(struct nfs_client *clp);
 void nfs_delegation_reap_unclaimed(struct nfs_client *clp);
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 7cb298525eef..2c5ace4f00a7 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -1579,55 +1579,46 @@ static int nfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	struct dentry *dentry = NULL, *rehash = NULL;
 	int error = -EBUSY;
 
-	/*
-	 * To prevent any new references to the target during the rename,
-	 * we unhash the dentry and free the inode in advance.
-	 */
-	if (!d_unhashed(new_dentry)) {
-		d_drop(new_dentry);
-		rehash = new_dentry;
-	}
-
 	dfprintk(VFS, "NFS: rename(%s/%s -> %s/%s, ct=%d)\n",
 		 old_dentry->d_parent->d_name.name, old_dentry->d_name.name,
 		 new_dentry->d_parent->d_name.name, new_dentry->d_name.name,
 		 atomic_read(&new_dentry->d_count));
 
 	/*
-	 * First check whether the target is busy ... we can't
-	 * safely do _any_ rename if the target is in use.
-	 *
-	 * For files, make a copy of the dentry and then do a 
-	 * silly-rename. If the silly-rename succeeds, the
-	 * copied dentry is hashed and becomes the new target.
+	 * For non-directories, check whether the target is busy and if so,
+	 * make a copy of the dentry and then do a silly-rename. If the
+	 * silly-rename succeeds, the copied dentry is hashed and becomes
+	 * the new target.
 	 */
-	if (!new_inode)
-		goto go_ahead;
-	if (S_ISDIR(new_inode->i_mode)) {
-		error = -EISDIR;
-		if (!S_ISDIR(old_inode->i_mode))
-			goto out;
-	} else if (atomic_read(&new_dentry->d_count) > 2) {
-		int err;
-		/* copy the target dentry's name */
-		dentry = d_alloc(new_dentry->d_parent,
-				 &new_dentry->d_name);
-		if (!dentry)
-			goto out;
+	if (new_inode && !S_ISDIR(new_inode->i_mode)) {
+		/*
+		 * To prevent any new references to the target during the
+		 * rename, we unhash the dentry in advance.
+		 */
+		if (!d_unhashed(new_dentry)) {
+			d_drop(new_dentry);
+			rehash = new_dentry;
+		}
+
+		if (atomic_read(&new_dentry->d_count) > 2) {
+			int err;
+
+			/* copy the target dentry's name */
+			dentry = d_alloc(new_dentry->d_parent,
+					 &new_dentry->d_name);
+			if (!dentry)
+				goto out;
 
-		/* silly-rename the existing target ... */
-		err = nfs_sillyrename(new_dir, new_dentry);
-		if (!err) {
-			new_dentry = rehash = dentry;
+			/* silly-rename the existing target ... */
+			err = nfs_sillyrename(new_dir, new_dentry);
+			if (err)
+				goto out;
+
+			new_dentry = dentry;
 			new_inode = NULL;
-			/* instantiate the replacement target */
-			d_instantiate(new_dentry, NULL);
-		} else if (atomic_read(&new_dentry->d_count) > 1)
-			/* dentry still busy? */
-			goto out;
+		}
 	}
 
-go_ahead:
 	/*
 	 * ... prune child dentries and writebacks if needed.
 	 */
diff --git a/fs/nfs/dns_resolve.c b/fs/nfs/dns_resolve.c
index f4d54ba97cc6..95e1ca765d47 100644
--- a/fs/nfs/dns_resolve.c
+++ b/fs/nfs/dns_resolve.c
@@ -146,7 +146,7 @@ static int nfs_dns_show(struct seq_file *m, struct cache_detail *cd,
 	return 0;
 }
 
-struct nfs_dns_ent *nfs_dns_lookup(struct cache_detail *cd,
+static struct nfs_dns_ent *nfs_dns_lookup(struct cache_detail *cd,
 		struct nfs_dns_ent *key)
 {
 	struct cache_head *ch;
@@ -159,7 +159,7 @@ struct nfs_dns_ent *nfs_dns_lookup(struct cache_detail *cd,
 	return container_of(ch, struct nfs_dns_ent, h);
 }
 
-struct nfs_dns_ent *nfs_dns_update(struct cache_detail *cd,
+static struct nfs_dns_ent *nfs_dns_update(struct cache_detail *cd,
 		struct nfs_dns_ent *new,
 		struct nfs_dns_ent *key)
 {
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index e21b1bb9972f..29e464d23b32 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -30,6 +30,15 @@ static inline int nfs4_has_session(const struct nfs_client *clp)
 	return 0;
 }
 
+static inline int nfs4_has_persistent_session(const struct nfs_client *clp)
+{
+#ifdef CONFIG_NFS_V4_1
+	if (nfs4_has_session(clp))
+		return (clp->cl_session->flags & SESSION4_PERSIST);
+#endif /* CONFIG_NFS_V4_1 */
+	return 0;
+}
+
 struct nfs_clone_mount {
 	const struct super_block *sb;
 	const struct dentry *dentry;
@@ -156,6 +165,7 @@ struct vfsmount *nfs_do_refmount(const struct vfsmount *mnt_parent, struct dentr
 
 /* callback_xdr.c */
 extern struct svc_version nfs4_callback_version1;
+extern struct svc_version nfs4_callback_version4;
 
 /* pagelist.c */
 extern int __init nfs_init_nfspagecache(void);
@@ -177,24 +187,14 @@ extern __be32 * nfs_decode_dirent(__be32 *, struct nfs_entry *, int);
 extern struct rpc_procinfo nfs3_procedures[];
 extern __be32 *nfs3_decode_dirent(__be32 *, struct nfs_entry *, int);
 
-/* nfs4proc.c */
-static inline void nfs4_restart_rpc(struct rpc_task *task,
-				    const struct nfs_client *clp)
-{
-#ifdef CONFIG_NFS_V4_1
-	if (nfs4_has_session(clp) &&
-	    test_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state)) {
-		rpc_restart_call_prepare(task);
-		return;
-	}
-#endif /* CONFIG_NFS_V4_1 */
-	rpc_restart_call(task);
-}
-
 /* nfs4xdr.c */
 #ifdef CONFIG_NFS_V4
 extern __be32 *nfs4_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus);
 #endif
+#ifdef CONFIG_NFS_V4_1
+extern const u32 nfs41_maxread_overhead;
+extern const u32 nfs41_maxwrite_overhead;
+#endif
 
 /* nfs4proc.c */
 #ifdef CONFIG_NFS_V4
@@ -273,20 +273,6 @@ extern int _nfs4_call_sync_session(struct nfs_server *server,
 				   struct nfs4_sequence_res *res,
 				   int cache_reply);
 
-#ifdef CONFIG_NFS_V4_1
-extern void nfs41_sequence_free_slot(const struct nfs_client *,
-				     struct nfs4_sequence_res *res);
-#endif /* CONFIG_NFS_V4_1 */
-
-static inline void nfs4_sequence_free_slot(const struct nfs_client *clp,
-					   struct nfs4_sequence_res *res)
-{
-#ifdef CONFIG_NFS_V4_1
-	if (nfs4_has_session(clp))
-		nfs41_sequence_free_slot(clp, res);
-#endif /* CONFIG_NFS_V4_1 */
-}
-
 /*
  * Determine the device name as a string
  */
@@ -380,3 +366,15 @@ unsigned int nfs_page_array_len(unsigned int base, size_t len)
 	return ((unsigned long)len + (unsigned long)base +
 		PAGE_SIZE - 1) >> PAGE_SHIFT;
 }
+
+/*
+ * Helper for restarting RPC calls in the possible presence of NFSv4.1
+ * sessions.
+ */
+static inline void nfs_restart_rpc(struct rpc_task *task, const struct nfs_client *clp)
+{
+	if (nfs4_has_session(clp))
+		rpc_restart_call_prepare(task);
+	else
+		rpc_restart_call(task);
+}
diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index 6ea07a3c75d4..7e57b04e4014 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -44,7 +44,8 @@ enum nfs4_client_state {
 	NFS4CLNT_RECLAIM_REBOOT,
 	NFS4CLNT_RECLAIM_NOGRACE,
 	NFS4CLNT_DELEGRETURN,
-	NFS4CLNT_SESSION_SETUP,
+	NFS4CLNT_SESSION_RESET,
+	NFS4CLNT_SESSION_DRAINING,
 };
 
 /*
@@ -180,6 +181,7 @@ struct nfs4_state_recovery_ops {
 	int (*recover_lock)(struct nfs4_state *, struct file_lock *);
 	int (*establish_clid)(struct nfs_client *, struct rpc_cred *);
 	struct rpc_cred * (*get_clid_cred)(struct nfs_client *);
+	int (*reclaim_complete)(struct nfs_client *);
 };
 
 struct nfs4_state_maintenance_ops {
@@ -200,9 +202,11 @@ extern ssize_t nfs4_listxattr(struct dentry *, char *, size_t);
 /* nfs4proc.c */
 extern int nfs4_proc_setclientid(struct nfs_client *, u32, unsigned short, struct rpc_cred *);
 extern int nfs4_proc_setclientid_confirm(struct nfs_client *, struct rpc_cred *);
+extern int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred);
 extern int nfs4_proc_async_renew(struct nfs_client *, struct rpc_cred *);
 extern int nfs4_proc_renew(struct nfs_client *, struct rpc_cred *);
 extern int nfs4_init_clientid(struct nfs_client *, struct rpc_cred *);
+extern int nfs41_init_clientid(struct nfs_client *, struct rpc_cred *);
 extern int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait);
 extern struct dentry *nfs4_atomic_open(struct inode *, struct dentry *, struct nameidata *);
 extern int nfs4_open_revalidate(struct inode *, struct dentry *, int, struct nameidata *);
@@ -218,9 +222,11 @@ extern int nfs4_setup_sequence(struct nfs_client *clp,
 		int cache_reply, struct rpc_task *task);
 extern void nfs4_destroy_session(struct nfs4_session *session);
 extern struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp);
-extern int nfs4_proc_create_session(struct nfs_client *, int reset);
+extern int nfs4_proc_create_session(struct nfs_client *);
 extern int nfs4_proc_destroy_session(struct nfs4_session *);
 extern int nfs4_init_session(struct nfs_server *server);
+extern int nfs4_proc_get_lease_time(struct nfs_client *clp,
+		struct nfs_fsinfo *fsinfo);
 #else /* CONFIG_NFS_v4_1 */
 static inline int nfs4_setup_sequence(struct nfs_client *clp,
 		struct nfs4_sequence_args *args, struct nfs4_sequence_res *res,
@@ -267,6 +273,7 @@ extern void nfs4_state_set_mode_locked(struct nfs4_state *, fmode_t);
 extern void nfs4_schedule_state_recovery(struct nfs_client *);
 extern void nfs4_schedule_state_manager(struct nfs_client *);
 extern int nfs4_state_mark_reclaim_nograce(struct nfs_client *clp, struct nfs4_state *state);
+extern void nfs41_handle_sequence_flag_errors(struct nfs_client *clp, u32 flags);
 extern void nfs4_put_lock_state(struct nfs4_lock_state *lsp);
 extern int nfs4_set_lock_state(struct nfs4_state *state, struct file_lock *fl);
 extern void nfs4_copy_stateid(nfs4_stateid *, struct nfs4_state *, fl_owner_t);
@@ -287,6 +294,7 @@ struct nfs4_mount_data;
 
 /* callback_xdr.c */
 extern struct svc_version nfs4_callback_version1;
+extern struct svc_version nfs4_callback_version4;
 
 #else
 
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 741a562177fc..9f5f11ecfd93 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -270,11 +270,18 @@ static int nfs4_handle_exception(const struct nfs_server *server, int errorcode,
 		case -NFS4ERR_SEQ_MISORDERED:
 			dprintk("%s ERROR: %d Reset session\n", __func__,
 				errorcode);
-			set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
+			nfs4_schedule_state_recovery(clp);
 			exception->retry = 1;
-			/* FALLTHROUGH */
+			break;
 #endif /* !defined(CONFIG_NFS_V4_1) */
 		case -NFS4ERR_FILE_OPEN:
+			if (exception->timeout > HZ) {
+				/* We have retried a decent amount, time to
+				 * fail
+				 */
+				ret = -EBUSY;
+				break;
+			}
 		case -NFS4ERR_GRACE:
 		case -NFS4ERR_DELAY:
 			ret = nfs4_delay(server->client, &exception->timeout);
@@ -311,48 +318,54 @@ static void renew_lease(const struct nfs_server *server, unsigned long timestamp
  * so we need to scan down from highest_used_slotid to 0 looking for the now
  * highest slotid in use.
  * If none found, highest_used_slotid is set to -1.
+ *
+ * Must be called while holding tbl->slot_tbl_lock
  */
 static void
 nfs4_free_slot(struct nfs4_slot_table *tbl, u8 free_slotid)
 {
 	int slotid = free_slotid;
 
-	spin_lock(&tbl->slot_tbl_lock);
 	/* clear used bit in bitmap */
 	__clear_bit(slotid, tbl->used_slots);
 
 	/* update highest_used_slotid when it is freed */
 	if (slotid == tbl->highest_used_slotid) {
 		slotid = find_last_bit(tbl->used_slots, tbl->max_slots);
-		if (slotid >= 0 && slotid < tbl->max_slots)
+		if (slotid < tbl->max_slots)
 			tbl->highest_used_slotid = slotid;
 		else
 			tbl->highest_used_slotid = -1;
 	}
-	rpc_wake_up_next(&tbl->slot_tbl_waitq);
-	spin_unlock(&tbl->slot_tbl_lock);
 	dprintk("%s: free_slotid %u highest_used_slotid %d\n", __func__,
 		free_slotid, tbl->highest_used_slotid);
 }
 
-void nfs41_sequence_free_slot(const struct nfs_client *clp,
+static void nfs41_sequence_free_slot(const struct nfs_client *clp,
 			      struct nfs4_sequence_res *res)
 {
 	struct nfs4_slot_table *tbl;
 
-	if (!nfs4_has_session(clp)) {
-		dprintk("%s: No session\n", __func__);
-		return;
-	}
 	tbl = &clp->cl_session->fc_slot_table;
 	if (res->sr_slotid == NFS4_MAX_SLOT_TABLE) {
-		dprintk("%s: No slot\n", __func__);
 		/* just wake up the next guy waiting since
 		 * we may have not consumed a slot after all */
-		rpc_wake_up_next(&tbl->slot_tbl_waitq);
+		dprintk("%s: No slot\n", __func__);
 		return;
 	}
+
+	spin_lock(&tbl->slot_tbl_lock);
 	nfs4_free_slot(tbl, res->sr_slotid);
+
+	/* Signal state manager thread if session is drained */
+	if (test_bit(NFS4CLNT_SESSION_DRAINING, &clp->cl_state)) {
+		if (tbl->highest_used_slotid == -1) {
+			dprintk("%s COMPLETE: Session Drained\n", __func__);
+			complete(&clp->cl_session->complete);
+		}
+	} else
+		rpc_wake_up_next(&tbl->slot_tbl_waitq);
+	spin_unlock(&tbl->slot_tbl_lock);
 	res->sr_slotid = NFS4_MAX_SLOT_TABLE;
 }
 
@@ -377,10 +390,10 @@ static void nfs41_sequence_done(struct nfs_client *clp,
 	if (res->sr_slotid == NFS4_MAX_SLOT_TABLE)
 		goto out;
 
-	tbl = &clp->cl_session->fc_slot_table;
-	slot = tbl->slots + res->sr_slotid;
-
+	/* Check the SEQUENCE operation status */
 	if (res->sr_status == 0) {
+		tbl = &clp->cl_session->fc_slot_table;
+		slot = tbl->slots + res->sr_slotid;
 		/* Update the slot's sequence and clientid lease timer */
 		++slot->seq_nr;
 		timestamp = res->sr_renewal_time;
@@ -388,7 +401,8 @@ static void nfs41_sequence_done(struct nfs_client *clp,
 		if (time_before(clp->cl_last_renewal, timestamp))
 			clp->cl_last_renewal = timestamp;
 		spin_unlock(&clp->cl_lock);
-		return;
+		/* Check sequence flags */
+		nfs41_handle_sequence_flag_errors(clp, res->sr_status_flags);
 	}
 out:
 	/* The session may be reset by one of the error handlers. */
@@ -429,24 +443,6 @@ out:
 	return ret_id;
 }
 
-static int nfs4_recover_session(struct nfs4_session *session)
-{
-	struct nfs_client *clp = session->clp;
-	unsigned int loop;
-	int ret;
-
-	for (loop = NFS4_MAX_LOOP_ON_RECOVER; loop != 0; loop--) {
-		ret = nfs4_wait_clnt_recover(clp);
-		if (ret != 0)
-			break;
-		if (!test_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state))
-			break;
-		nfs4_schedule_state_manager(clp);
-		ret = -EIO;
-	}
-	return ret;
-}
-
 static int nfs41_setup_sequence(struct nfs4_session *session,
 				struct nfs4_sequence_args *args,
 				struct nfs4_sequence_res *res,
@@ -455,7 +451,6 @@ static int nfs41_setup_sequence(struct nfs4_session *session,
 {
 	struct nfs4_slot *slot;
 	struct nfs4_slot_table *tbl;
-	int status = 0;
 	u8 slotid;
 
 	dprintk("--> %s\n", __func__);
@@ -468,21 +463,15 @@ static int nfs41_setup_sequence(struct nfs4_session *session,
 	tbl = &session->fc_slot_table;
 
 	spin_lock(&tbl->slot_tbl_lock);
-	if (test_bit(NFS4CLNT_SESSION_SETUP, &session->clp->cl_state)) {
-		if (tbl->highest_used_slotid != -1) {
-			rpc_sleep_on(&tbl->slot_tbl_waitq, task, NULL);
-			spin_unlock(&tbl->slot_tbl_lock);
-			dprintk("<-- %s: Session reset: draining\n", __func__);
-			return -EAGAIN;
-		}
-
-		/* The slot table is empty; start the reset thread */
-		dprintk("%s Session Reset\n", __func__);
+	if (test_bit(NFS4CLNT_SESSION_DRAINING, &session->clp->cl_state)) {
+		/*
+		 * The state manager will wait until the slot table is empty.
+		 * Schedule the reset thread
+		 */
+		rpc_sleep_on(&tbl->slot_tbl_waitq, task, NULL);
 		spin_unlock(&tbl->slot_tbl_lock);
-		status = nfs4_recover_session(session);
-		if (status)
-			return status;
-		spin_lock(&tbl->slot_tbl_lock);
+		dprintk("%s Schedule Session Reset\n", __func__);
+		return -EAGAIN;
 	}
 
 	slotid = nfs4_find_slot(tbl, task);
@@ -527,7 +516,7 @@ int nfs4_setup_sequence(struct nfs_client *clp,
 		goto out;
 	ret = nfs41_setup_sequence(clp->cl_session, args, res, cache_reply,
 				   task);
-	if (ret != -EAGAIN) {
+	if (ret && ret != -EAGAIN) {
 		/* terminate rpc task */
 		task->tk_status = ret;
 		task->tk_action = NULL;
@@ -561,7 +550,6 @@ static void nfs41_call_sync_done(struct rpc_task *task, void *calldata)
 	struct nfs41_call_sync_data *data = calldata;
 
 	nfs41_sequence_done(data->clp, data->seq_res, task->tk_status);
-	nfs41_sequence_free_slot(data->clp, data->seq_res);
 }
 
 struct rpc_call_ops nfs41_call_sync_ops = {
@@ -637,15 +625,6 @@ static void nfs4_sequence_done(const struct nfs_server *server,
 #endif /* CONFIG_NFS_V4_1 */
 }
 
-/* no restart, therefore free slot here */
-static void nfs4_sequence_done_free_slot(const struct nfs_server *server,
-					 struct nfs4_sequence_res *res,
-					 int rpc_status)
-{
-	nfs4_sequence_done(server, res, rpc_status);
-	nfs4_sequence_free_slot(server->nfs_client, res);
-}
-
 static void update_changeattr(struct inode *dir, struct nfs4_change_info *cinfo)
 {
 	struct nfs_inode *nfsi = NFS_I(dir);
@@ -720,9 +699,15 @@ static struct nfs4_opendata *nfs4_opendata_alloc(struct path *path,
 	p->o_arg.bitmask = server->attr_bitmask;
 	p->o_arg.claim = NFS4_OPEN_CLAIM_NULL;
 	if (flags & O_EXCL) {
-		u32 *s = (u32 *) p->o_arg.u.verifier.data;
-		s[0] = jiffies;
-		s[1] = current->pid;
+		if (nfs4_has_persistent_session(server->nfs_client)) {
+			/* GUARDED */
+			p->o_arg.u.attrs = &p->attrs;
+			memcpy(&p->attrs, attrs, sizeof(p->attrs));
+		} else { /* EXCLUSIVE4_1 */
+			u32 *s = (u32 *) p->o_arg.u.verifier.data;
+			s[0] = jiffies;
+			s[1] = current->pid;
+		}
 	} else if (flags & O_CREAT) {
 		p->o_arg.u.attrs = &p->attrs;
 		memcpy(&p->attrs, attrs, sizeof(p->attrs));
@@ -776,13 +761,16 @@ static int can_open_cached(struct nfs4_state *state, fmode_t mode, int open_mode
 		goto out;
 	switch (mode & (FMODE_READ|FMODE_WRITE)) {
 		case FMODE_READ:
-			ret |= test_bit(NFS_O_RDONLY_STATE, &state->flags) != 0;
+			ret |= test_bit(NFS_O_RDONLY_STATE, &state->flags) != 0
+				&& state->n_rdonly != 0;
 			break;
 		case FMODE_WRITE:
-			ret |= test_bit(NFS_O_WRONLY_STATE, &state->flags) != 0;
+			ret |= test_bit(NFS_O_WRONLY_STATE, &state->flags) != 0
+				&& state->n_wronly != 0;
 			break;
 		case FMODE_READ|FMODE_WRITE:
-			ret |= test_bit(NFS_O_RDWR_STATE, &state->flags) != 0;
+			ret |= test_bit(NFS_O_RDWR_STATE, &state->flags) != 0
+				&& state->n_rdwr != 0;
 	}
 out:
 	return ret;
@@ -1183,6 +1171,14 @@ int nfs4_open_delegation_recall(struct nfs_open_context *ctx, struct nfs4_state
 			case -ENOENT:
 			case -ESTALE:
 				goto out;
+			case -NFS4ERR_BADSESSION:
+			case -NFS4ERR_BADSLOT:
+			case -NFS4ERR_BAD_HIGH_SLOT:
+			case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
+			case -NFS4ERR_DEADSESSION:
+				nfs4_schedule_state_recovery(
+					server->nfs_client);
+				goto out;
 			case -NFS4ERR_STALE_CLIENTID:
 			case -NFS4ERR_STALE_STATEID:
 			case -NFS4ERR_EXPIRED:
@@ -1336,8 +1332,8 @@ static void nfs4_open_done(struct rpc_task *task, void *calldata)
 
 	data->rpc_status = task->tk_status;
 
-	nfs4_sequence_done_free_slot(data->o_arg.server, &data->o_res.seq_res,
-				     task->tk_status);
+	nfs4_sequence_done(data->o_arg.server, &data->o_res.seq_res,
+			task->tk_status);
 
 	if (RPC_ASSASSINATED(task))
 		return;
@@ -1488,7 +1484,7 @@ static int _nfs4_open_expired(struct nfs_open_context *ctx, struct nfs4_state *s
 	return ret;
 }
 
-static inline int nfs4_do_open_expired(struct nfs_open_context *ctx, struct nfs4_state *state)
+static int nfs4_do_open_expired(struct nfs_open_context *ctx, struct nfs4_state *state)
 {
 	struct nfs_server *server = NFS_SERVER(state->inode);
 	struct nfs4_exception exception = { };
@@ -1496,10 +1492,16 @@ static inline int nfs4_do_open_expired(struct nfs_open_context *ctx, struct nfs4
 
 	do {
 		err = _nfs4_open_expired(ctx, state);
-		if (err != -NFS4ERR_DELAY)
-			break;
-		nfs4_handle_exception(server, err, &exception);
+		switch (err) {
+		default:
+			goto out;
+		case -NFS4ERR_GRACE:
+		case -NFS4ERR_DELAY:
+			nfs4_handle_exception(server, err, &exception);
+			err = 0;
+		}
 	} while (exception.retry);
+out:
 	return err;
 }
 
@@ -1712,6 +1714,18 @@ static void nfs4_free_closedata(void *data)
 	kfree(calldata);
 }
 
+static void nfs4_close_clear_stateid_flags(struct nfs4_state *state,
+		fmode_t fmode)
+{
+	spin_lock(&state->owner->so_lock);
+	if (!(fmode & FMODE_READ))
+		clear_bit(NFS_O_RDONLY_STATE, &state->flags);
+	if (!(fmode & FMODE_WRITE))
+		clear_bit(NFS_O_WRONLY_STATE, &state->flags);
+	clear_bit(NFS_O_RDWR_STATE, &state->flags);
+	spin_unlock(&state->owner->so_lock);
+}
+
 static void nfs4_close_done(struct rpc_task *task, void *data)
 {
 	struct nfs4_closedata *calldata = data;
@@ -1728,6 +1742,8 @@ static void nfs4_close_done(struct rpc_task *task, void *data)
 		case 0:
 			nfs_set_open_stateid(state, &calldata->res.stateid, 0);
 			renew_lease(server, calldata->timestamp);
+			nfs4_close_clear_stateid_flags(state,
+					calldata->arg.fmode);
 			break;
 		case -NFS4ERR_STALE_STATEID:
 		case -NFS4ERR_OLD_STATEID:
@@ -1737,11 +1753,10 @@ static void nfs4_close_done(struct rpc_task *task, void *data)
 				break;
 		default:
 			if (nfs4_async_handle_error(task, server, state) == -EAGAIN) {
-				nfs4_restart_rpc(task, server->nfs_client);
+				nfs_restart_rpc(task, server->nfs_client);
 				return;
 			}
 	}
-	nfs4_sequence_free_slot(server->nfs_client, &calldata->res.seq_res);
 	nfs_refresh_inode(calldata->inode, calldata->res.fattr);
 }
 
@@ -1749,38 +1764,39 @@ static void nfs4_close_prepare(struct rpc_task *task, void *data)
 {
 	struct nfs4_closedata *calldata = data;
 	struct nfs4_state *state = calldata->state;
-	int clear_rd, clear_wr, clear_rdwr;
+	int call_close = 0;
 
 	if (nfs_wait_on_sequence(calldata->arg.seqid, task) != 0)
 		return;
 
-	clear_rd = clear_wr = clear_rdwr = 0;
+	task->tk_msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_DOWNGRADE];
+	calldata->arg.fmode = FMODE_READ|FMODE_WRITE;
 	spin_lock(&state->owner->so_lock);
 	/* Calculate the change in open mode */
 	if (state->n_rdwr == 0) {
 		if (state->n_rdonly == 0) {
-			clear_rd |= test_and_clear_bit(NFS_O_RDONLY_STATE, &state->flags);
-			clear_rdwr |= test_and_clear_bit(NFS_O_RDWR_STATE, &state->flags);
+			call_close |= test_bit(NFS_O_RDONLY_STATE, &state->flags);
+			call_close |= test_bit(NFS_O_RDWR_STATE, &state->flags);
+			calldata->arg.fmode &= ~FMODE_READ;
 		}
 		if (state->n_wronly == 0) {
-			clear_wr |= test_and_clear_bit(NFS_O_WRONLY_STATE, &state->flags);
-			clear_rdwr |= test_and_clear_bit(NFS_O_RDWR_STATE, &state->flags);
+			call_close |= test_bit(NFS_O_WRONLY_STATE, &state->flags);
+			call_close |= test_bit(NFS_O_RDWR_STATE, &state->flags);
+			calldata->arg.fmode &= ~FMODE_WRITE;
 		}
 	}
 	spin_unlock(&state->owner->so_lock);
-	if (!clear_rd && !clear_wr && !clear_rdwr) {
+
+	if (!call_close) {
 		/* Note: exit _without_ calling nfs4_close_done */
 		task->tk_action = NULL;
 		return;
 	}
+
+	if (calldata->arg.fmode == 0)
+		task->tk_msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CLOSE];
+
 	nfs_fattr_init(calldata->res.fattr);
-	if (test_bit(NFS_O_RDONLY_STATE, &state->flags) != 0) {
-		task->tk_msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_DOWNGRADE];
-		calldata->arg.fmode = FMODE_READ;
-	} else if (test_bit(NFS_O_WRONLY_STATE, &state->flags) != 0) {
-		task->tk_msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_DOWNGRADE];
-		calldata->arg.fmode = FMODE_WRITE;
-	}
 	calldata->timestamp = jiffies;
 	if (nfs4_setup_sequence((NFS_SERVER(calldata->inode))->nfs_client,
 				&calldata->arg.seq_args, &calldata->res.seq_res,
@@ -1981,7 +1997,7 @@ out_drop:
 	return 0;
 }
 
-void nfs4_close_context(struct nfs_open_context *ctx, int is_sync)
+static void nfs4_close_context(struct nfs_open_context *ctx, int is_sync)
 {
 	if (ctx->state == NULL)
 		return;
@@ -2532,7 +2548,6 @@ static int nfs4_proc_unlink_done(struct rpc_task *task, struct inode *dir)
 	nfs4_sequence_done(res->server, &res->seq_res, task->tk_status);
 	if (nfs4_async_handle_error(task, res->server, NULL) == -EAGAIN)
 		return 0;
-	nfs4_sequence_free_slot(res->server->nfs_client, &res->seq_res);
 	update_changeattr(dir, &res->cinfo);
 	nfs_post_op_update_inode(dir, &res->dir_attr);
 	return 1;
@@ -2971,11 +2986,10 @@ static int nfs4_read_done(struct rpc_task *task, struct nfs_read_data *data)
 
 	dprintk("--> %s\n", __func__);
 
-	/* nfs4_sequence_free_slot called in the read rpc_call_done */
 	nfs4_sequence_done(server, &data->res.seq_res, task->tk_status);
 
 	if (nfs4_async_handle_error(task, server, data->args.context->state) == -EAGAIN) {
-		nfs4_restart_rpc(task, server->nfs_client);
+		nfs_restart_rpc(task, server->nfs_client);
 		return -EAGAIN;
 	}
 
@@ -2995,12 +3009,11 @@ static int nfs4_write_done(struct rpc_task *task, struct nfs_write_data *data)
 {
 	struct inode *inode = data->inode;
 	
-	/* slot is freed in nfs_writeback_done */
 	nfs4_sequence_done(NFS_SERVER(inode), &data->res.seq_res,
 			   task->tk_status);
 
 	if (nfs4_async_handle_error(task, NFS_SERVER(inode), data->args.context->state) == -EAGAIN) {
-		nfs4_restart_rpc(task, NFS_SERVER(inode)->nfs_client);
+		nfs_restart_rpc(task, NFS_SERVER(inode)->nfs_client);
 		return -EAGAIN;
 	}
 	if (task->tk_status >= 0) {
@@ -3028,11 +3041,9 @@ static int nfs4_commit_done(struct rpc_task *task, struct nfs_write_data *data)
 	nfs4_sequence_done(NFS_SERVER(inode), &data->res.seq_res,
 			   task->tk_status);
 	if (nfs4_async_handle_error(task, NFS_SERVER(inode), NULL) == -EAGAIN) {
-		nfs4_restart_rpc(task, NFS_SERVER(inode)->nfs_client);
+		nfs_restart_rpc(task, NFS_SERVER(inode)->nfs_client);
 		return -EAGAIN;
 	}
-	nfs4_sequence_free_slot(NFS_SERVER(inode)->nfs_client,
-				&data->res.seq_res);
 	nfs_refresh_inode(inode, data->res.fattr);
 	return 0;
 }
@@ -3350,7 +3361,7 @@ _nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server,
 		case -NFS4ERR_SEQ_MISORDERED:
 			dprintk("%s ERROR %d, Reset session\n", __func__,
 				task->tk_status);
-			set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
+			nfs4_schedule_state_recovery(clp);
 			task->tk_status = 0;
 			return -EAGAIN;
 #endif /* CONFIG_NFS_V4_1 */
@@ -3483,12 +3494,23 @@ static void nfs4_delegreturn_done(struct rpc_task *task, void *calldata)
 {
 	struct nfs4_delegreturndata *data = calldata;
 
-	nfs4_sequence_done_free_slot(data->res.server, &data->res.seq_res,
-				     task->tk_status);
+	nfs4_sequence_done(data->res.server, &data->res.seq_res,
+			task->tk_status);
 
-	data->rpc_status = task->tk_status;
-	if (data->rpc_status == 0)
+	switch (task->tk_status) {
+	case -NFS4ERR_STALE_STATEID:
+	case -NFS4ERR_EXPIRED:
+	case 0:
 		renew_lease(data->res.server, data->timestamp);
+		break;
+	default:
+		if (nfs4_async_handle_error(task, data->res.server, NULL) ==
+				-EAGAIN) {
+			nfs_restart_rpc(task, data->res.server->nfs_client);
+			return;
+		}
+	}
+	data->rpc_status = task->tk_status;
 }
 
 static void nfs4_delegreturn_release(void *calldata)
@@ -3741,11 +3763,9 @@ static void nfs4_locku_done(struct rpc_task *task, void *data)
 			break;
 		default:
 			if (nfs4_async_handle_error(task, calldata->server, NULL) == -EAGAIN)
-				nfs4_restart_rpc(task,
-						calldata->server->nfs_client);
+				nfs_restart_rpc(task,
+						 calldata->server->nfs_client);
 	}
-	nfs4_sequence_free_slot(calldata->server->nfs_client,
-				&calldata->res.seq_res);
 }
 
 static void nfs4_locku_prepare(struct rpc_task *task, void *data)
@@ -3927,8 +3947,8 @@ static void nfs4_lock_done(struct rpc_task *task, void *calldata)
 
 	dprintk("%s: begin!\n", __func__);
 
-	nfs4_sequence_done_free_slot(data->server, &data->res.seq_res,
-				     task->tk_status);
+	nfs4_sequence_done(data->server, &data->res.seq_res,
+			task->tk_status);
 
 	data->rpc_status = task->tk_status;
 	if (RPC_ASSASSINATED(task))
@@ -4049,10 +4069,16 @@ static int nfs4_lock_expired(struct nfs4_state *state, struct file_lock *request
 		if (test_bit(NFS_DELEGATED_STATE, &state->flags) != 0)
 			return 0;
 		err = _nfs4_do_setlk(state, F_SETLK, request, 0);
-		if (err != -NFS4ERR_DELAY)
-			break;
-		nfs4_handle_exception(server, err, &exception);
+		switch (err) {
+		default:
+			goto out;
+		case -NFS4ERR_GRACE:
+		case -NFS4ERR_DELAY:
+			nfs4_handle_exception(server, err, &exception);
+			err = 0;
+		}
 	} while (exception.retry);
+out:
 	return err;
 }
 
@@ -4172,6 +4198,11 @@ int nfs4_lock_delegation_recall(struct nfs4_state *state, struct file_lock *fl)
 			case -NFS4ERR_EXPIRED:
 			case -NFS4ERR_STALE_CLIENTID:
 			case -NFS4ERR_STALE_STATEID:
+			case -NFS4ERR_BADSESSION:
+			case -NFS4ERR_BADSLOT:
+			case -NFS4ERR_BAD_HIGH_SLOT:
+			case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
+			case -NFS4ERR_DEADSESSION:
 				nfs4_schedule_state_recovery(server->nfs_client);
 				goto out;
 			case -ERESTARTSYS:
@@ -4296,7 +4327,7 @@ int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name,
  * NFS4ERR_BADSESSION in the sequence operation, and will therefore
  * be in some phase of session reset.
  */
-static int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred)
+int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred)
 {
 	nfs4_verifier verifier;
 	struct nfs41_exchange_id_args args = {
@@ -4318,6 +4349,9 @@ static int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred)
 	dprintk("--> %s\n", __func__);
 	BUG_ON(clp == NULL);
 
+	/* Remove server-only flags */
+	args.flags &= ~EXCHGID4_FLAG_CONFIRMED_R;
+
 	p = (u32 *)verifier.data;
 	*p++ = htonl((u32)clp->cl_boot_time.tv_sec);
 	*p = htonl((u32)clp->cl_boot_time.tv_nsec);
@@ -4389,10 +4423,9 @@ static void nfs4_get_lease_time_done(struct rpc_task *task, void *calldata)
 		dprintk("%s Retry: tk_status %d\n", __func__, task->tk_status);
 		rpc_delay(task, NFS4_POLL_RETRY_MIN);
 		task->tk_status = 0;
-		nfs4_restart_rpc(task, data->clp);
+		nfs_restart_rpc(task, data->clp);
 		return;
 	}
-	nfs41_sequence_free_slot(data->clp, &data->res->lr_seq_res);
 	dprintk("<-- %s\n", __func__);
 }
 
@@ -4465,7 +4498,6 @@ static int nfs4_reset_slot_table(struct nfs4_slot_table *tbl, int max_slots,
 	spin_lock(&tbl->slot_tbl_lock);
 	for (i = 0; i < max_slots; ++i)
 		tbl->slots[i].seq_nr = ivalue;
-	tbl->highest_used_slotid = -1;
 	spin_unlock(&tbl->slot_tbl_lock);
 	dprintk("%s: tbl=%p slots=%p max_slots=%d\n", __func__,
 		tbl, tbl->slots, tbl->max_slots);
@@ -4515,7 +4547,6 @@ static void nfs4_destroy_slot_tables(struct nfs4_session *session)
 static int nfs4_init_slot_table(struct nfs4_slot_table *tbl,
 		int max_slots, int ivalue)
 {
-	int i;
 	struct nfs4_slot *slot;
 	int ret = -ENOMEM;
 
@@ -4526,18 +4557,9 @@ static int nfs4_init_slot_table(struct nfs4_slot_table *tbl,
 	slot = kcalloc(max_slots, sizeof(struct nfs4_slot), GFP_KERNEL);
 	if (!slot)
 		goto out;
-	for (i = 0; i < max_slots; ++i)
-		slot[i].seq_nr = ivalue;
 	ret = 0;
 
 	spin_lock(&tbl->slot_tbl_lock);
-	if (tbl->slots != NULL) {
-		spin_unlock(&tbl->slot_tbl_lock);
-		dprintk("%s: slot table already initialized. tbl=%p slots=%p\n",
-			__func__, tbl, tbl->slots);
-		WARN_ON(1);
-		goto out_free;
-	}
 	tbl->max_slots = max_slots;
 	tbl->slots = slot;
 	tbl->highest_used_slotid = -1;  /* no slot is currently used */
@@ -4547,10 +4569,6 @@ static int nfs4_init_slot_table(struct nfs4_slot_table *tbl,
 out:
 	dprintk("<-- %s: return %d\n", __func__, ret);
 	return ret;
-
-out_free:
-	kfree(slot);
-	goto out;
 }
 
 /*
@@ -4558,17 +4576,24 @@ out_free:
  */
 static int nfs4_init_slot_tables(struct nfs4_session *session)
 {
-	int status;
+	struct nfs4_slot_table *tbl;
+	int status = 0;
 
-	status = nfs4_init_slot_table(&session->fc_slot_table,
-			session->fc_attrs.max_reqs, 1);
-	if (status)
-		return status;
+	tbl = &session->fc_slot_table;
+	if (tbl->slots == NULL) {
+		status = nfs4_init_slot_table(tbl,
+				session->fc_attrs.max_reqs, 1);
+		if (status)
+			return status;
+	}
 
-	status = nfs4_init_slot_table(&session->bc_slot_table,
-			session->bc_attrs.max_reqs, 0);
-	if (status)
-		nfs4_destroy_slot_tables(session);
+	tbl = &session->bc_slot_table;
+	if (tbl->slots == NULL) {
+		status = nfs4_init_slot_table(tbl,
+				session->bc_attrs.max_reqs, 0);
+		if (status)
+			nfs4_destroy_slot_tables(session);
+	}
 
 	return status;
 }
@@ -4582,7 +4607,6 @@ struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp)
 	if (!session)
 		return NULL;
 
-	set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
 	/*
 	 * The create session reply races with the server back
 	 * channel probe. Mark the client NFS_CS_SESSION_INITING
@@ -4590,12 +4614,15 @@ struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp)
 	 * nfs_client struct
 	 */
 	clp->cl_cons_state = NFS_CS_SESSION_INITING;
+	init_completion(&session->complete);
 
 	tbl = &session->fc_slot_table;
+	tbl->highest_used_slotid = -1;
 	spin_lock_init(&tbl->slot_tbl_lock);
 	rpc_init_wait_queue(&tbl->slot_tbl_waitq, "ForeChannel Slot table");
 
 	tbl = &session->bc_slot_table;
+	tbl->highest_used_slotid = -1;
 	spin_lock_init(&tbl->slot_tbl_lock);
 	rpc_init_wait_queue(&tbl->slot_tbl_waitq, "BackChannel Slot table");
 
@@ -4747,11 +4774,10 @@ static int _nfs4_proc_create_session(struct nfs_client *clp)
  * It is the responsibility of the caller to verify the session is
  * expired before calling this routine.
  */
-int nfs4_proc_create_session(struct nfs_client *clp, int reset)
+int nfs4_proc_create_session(struct nfs_client *clp)
 {
 	int status;
 	unsigned *ptr;
-	struct nfs_fsinfo fsinfo;
 	struct nfs4_session *session = clp->cl_session;
 
 	dprintk("--> %s clp=%p session=%p\n", __func__, clp, session);
@@ -4760,35 +4786,19 @@ int nfs4_proc_create_session(struct nfs_client *clp, int reset)
 	if (status)
 		goto out;
 
-	/* Init or reset the fore channel */
-	if (reset)
-		status = nfs4_reset_slot_tables(session);
-	else
-		status = nfs4_init_slot_tables(session);
-	dprintk("fore channel slot table initialization returned %d\n", status);
+	/* Init and reset the fore channel */
+	status = nfs4_init_slot_tables(session);
+	dprintk("slot table initialization returned %d\n", status);
+	if (status)
+		goto out;
+	status = nfs4_reset_slot_tables(session);
+	dprintk("slot table reset returned %d\n", status);
 	if (status)
 		goto out;
 
 	ptr = (unsigned *)&session->sess_id.data[0];
 	dprintk("%s client>seqid %d sessionid %u:%u:%u:%u\n", __func__,
 		clp->cl_seqid, ptr[0], ptr[1], ptr[2], ptr[3]);
-
-	if (reset)
-		/* Lease time is aleady set */
-		goto out;
-
-	/* Get the lease time */
-	status = nfs4_proc_get_lease_time(clp, &fsinfo);
-	if (status == 0) {
-		/* Update lease time and schedule renewal */
-		spin_lock(&clp->cl_lock);
-		clp->cl_lease_time = fsinfo.lease_time * HZ;
-		clp->cl_last_renewal = jiffies;
-		clear_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
-		spin_unlock(&clp->cl_lock);
-
-		nfs4_schedule_state_renewal(clp);
-	}
 out:
 	dprintk("<-- %s\n", __func__);
 	return status;
@@ -4827,13 +4837,16 @@ int nfs4_proc_destroy_session(struct nfs4_session *session)
 int nfs4_init_session(struct nfs_server *server)
 {
 	struct nfs_client *clp = server->nfs_client;
+	struct nfs4_session *session;
 	int ret;
 
 	if (!nfs4_has_session(clp))
 		return 0;
 
-	clp->cl_session->fc_attrs.max_rqst_sz = server->wsize;
-	clp->cl_session->fc_attrs.max_resp_sz = server->rsize;
+	session = clp->cl_session;
+	session->fc_attrs.max_rqst_sz = server->wsize + nfs41_maxwrite_overhead;
+	session->fc_attrs.max_resp_sz = server->rsize + nfs41_maxread_overhead;
+
 	ret = nfs4_recover_expired_lease(server);
 	if (!ret)
 		ret = nfs4_check_client_ready(clp);
@@ -4872,11 +4885,10 @@ void nfs41_sequence_call_done(struct rpc_task *task, void *data)
 
 		if (_nfs4_async_handle_error(task, NULL, clp, NULL)
 								== -EAGAIN) {
-			nfs4_restart_rpc(task, clp);
+			nfs_restart_rpc(task, clp);
 			return;
 		}
 	}
-	nfs41_sequence_free_slot(clp, task->tk_msg.rpc_resp);
 	dprintk("%s rpc_cred %p\n", __func__, task->tk_msg.rpc_cred);
 
 	kfree(task->tk_msg.rpc_argp);
@@ -4931,6 +4943,109 @@ static int nfs41_proc_async_sequence(struct nfs_client *clp,
 			      &nfs41_sequence_ops, (void *)clp);
 }
 
+struct nfs4_reclaim_complete_data {
+	struct nfs_client *clp;
+	struct nfs41_reclaim_complete_args arg;
+	struct nfs41_reclaim_complete_res res;
+};
+
+static void nfs4_reclaim_complete_prepare(struct rpc_task *task, void *data)
+{
+	struct nfs4_reclaim_complete_data *calldata = data;
+
+	if (nfs4_setup_sequence(calldata->clp, &calldata->arg.seq_args,
+				&calldata->res.seq_res, 0, task))
+		return;
+
+	rpc_call_start(task);
+}
+
+static void nfs4_reclaim_complete_done(struct rpc_task *task, void *data)
+{
+	struct nfs4_reclaim_complete_data *calldata = data;
+	struct nfs_client *clp = calldata->clp;
+	struct nfs4_sequence_res *res = &calldata->res.seq_res;
+
+	dprintk("--> %s\n", __func__);
+	nfs41_sequence_done(clp, res, task->tk_status);
+	switch (task->tk_status) {
+	case 0:
+	case -NFS4ERR_COMPLETE_ALREADY:
+		break;
+	case -NFS4ERR_BADSESSION:
+	case -NFS4ERR_DEADSESSION:
+		/*
+		 * Handle the session error, but do not retry the operation, as
+		 * we have no way of telling whether the clientid had to be
+		 * reset before we got our reply.  If reset, a new wave of
+		 * reclaim operations will follow, containing their own reclaim
+		 * complete.  We don't want our retry to get on the way of
+		 * recovery by incorrectly indicating to the server that we're
+		 * done reclaiming state since the process had to be restarted.
+		 */
+		_nfs4_async_handle_error(task, NULL, clp, NULL);
+		break;
+	default:
+		if (_nfs4_async_handle_error(
+				task, NULL, clp, NULL) == -EAGAIN) {
+			rpc_restart_call_prepare(task);
+			return;
+		}
+	}
+
+	dprintk("<-- %s\n", __func__);
+}
+
+static void nfs4_free_reclaim_complete_data(void *data)
+{
+	struct nfs4_reclaim_complete_data *calldata = data;
+
+	kfree(calldata);
+}
+
+static const struct rpc_call_ops nfs4_reclaim_complete_call_ops = {
+	.rpc_call_prepare = nfs4_reclaim_complete_prepare,
+	.rpc_call_done = nfs4_reclaim_complete_done,
+	.rpc_release = nfs4_free_reclaim_complete_data,
+};
+
+/*
+ * Issue a global reclaim complete.
+ */
+static int nfs41_proc_reclaim_complete(struct nfs_client *clp)
+{
+	struct nfs4_reclaim_complete_data *calldata;
+	struct rpc_task *task;
+	struct rpc_message msg = {
+		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_RECLAIM_COMPLETE],
+	};
+	struct rpc_task_setup task_setup_data = {
+		.rpc_client = clp->cl_rpcclient,
+		.rpc_message = &msg,
+		.callback_ops = &nfs4_reclaim_complete_call_ops,
+		.flags = RPC_TASK_ASYNC,
+	};
+	int status = -ENOMEM;
+
+	dprintk("--> %s\n", __func__);
+	calldata = kzalloc(sizeof(*calldata), GFP_KERNEL);
+	if (calldata == NULL)
+		goto out;
+	calldata->clp = clp;
+	calldata->arg.one_fs = 0;
+	calldata->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
+
+	msg.rpc_argp = &calldata->arg;
+	msg.rpc_resp = &calldata->res;
+	task_setup_data.callback_data = calldata;
+	task = rpc_run_task(&task_setup_data);
+	if (IS_ERR(task))
+		status = PTR_ERR(task);
+	rpc_put_task(task);
+out:
+	dprintk("<-- %s status=%d\n", __func__, status);
+	return status;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 struct nfs4_state_recovery_ops nfs40_reboot_recovery_ops = {
@@ -4948,8 +5063,9 @@ struct nfs4_state_recovery_ops nfs41_reboot_recovery_ops = {
 	.state_flag_bit	= NFS_STATE_RECLAIM_REBOOT,
 	.recover_open	= nfs4_open_reclaim,
 	.recover_lock	= nfs4_lock_reclaim,
-	.establish_clid = nfs4_proc_exchange_id,
+	.establish_clid = nfs41_init_clientid,
 	.get_clid_cred	= nfs4_get_exchange_id_cred,
+	.reclaim_complete = nfs41_proc_reclaim_complete,
 };
 #endif /* CONFIG_NFS_V4_1 */
 
@@ -4968,7 +5084,7 @@ struct nfs4_state_recovery_ops nfs41_nograce_recovery_ops = {
 	.state_flag_bit	= NFS_STATE_RECLAIM_NOGRACE,
 	.recover_open	= nfs4_open_expired,
 	.recover_lock	= nfs4_lock_expired,
-	.establish_clid = nfs4_proc_exchange_id,
+	.establish_clid = nfs41_init_clientid,
 	.get_clid_cred	= nfs4_get_exchange_id_cred,
 };
 #endif /* CONFIG_NFS_V4_1 */
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 2ef4fecf3984..e76427e6346f 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -116,6 +116,68 @@ struct rpc_cred *nfs4_get_renew_cred_locked(struct nfs_client *clp)
 
 #if defined(CONFIG_NFS_V4_1)
 
+static int nfs41_setup_state_renewal(struct nfs_client *clp)
+{
+	int status;
+	struct nfs_fsinfo fsinfo;
+
+	status = nfs4_proc_get_lease_time(clp, &fsinfo);
+	if (status == 0) {
+		/* Update lease time and schedule renewal */
+		spin_lock(&clp->cl_lock);
+		clp->cl_lease_time = fsinfo.lease_time * HZ;
+		clp->cl_last_renewal = jiffies;
+		spin_unlock(&clp->cl_lock);
+
+		nfs4_schedule_state_renewal(clp);
+	}
+
+	return status;
+}
+
+static void nfs41_end_drain_session(struct nfs_client *clp,
+		struct nfs4_session *ses)
+{
+	if (test_and_clear_bit(NFS4CLNT_SESSION_DRAINING, &clp->cl_state))
+		rpc_wake_up(&ses->fc_slot_table.slot_tbl_waitq);
+}
+
+static int nfs41_begin_drain_session(struct nfs_client *clp,
+		struct nfs4_session *ses)
+{
+	struct nfs4_slot_table *tbl = &ses->fc_slot_table;
+
+	spin_lock(&tbl->slot_tbl_lock);
+	set_bit(NFS4CLNT_SESSION_DRAINING, &clp->cl_state);
+	if (tbl->highest_used_slotid != -1) {
+		INIT_COMPLETION(ses->complete);
+		spin_unlock(&tbl->slot_tbl_lock);
+		return wait_for_completion_interruptible(&ses->complete);
+	}
+	spin_unlock(&tbl->slot_tbl_lock);
+	return 0;
+}
+
+int nfs41_init_clientid(struct nfs_client *clp, struct rpc_cred *cred)
+{
+	int status;
+
+	status = nfs41_begin_drain_session(clp, clp->cl_session);
+	if (status != 0)
+		goto out;
+	status = nfs4_proc_exchange_id(clp, cred);
+	if (status != 0)
+		goto out;
+	status = nfs4_proc_create_session(clp);
+	if (status != 0)
+		goto out;
+	nfs41_end_drain_session(clp, clp->cl_session);
+	nfs41_setup_state_renewal(clp);
+	nfs_mark_client_ready(clp, NFS_CS_READY);
+out:
+	return status;
+}
+
 struct rpc_cred *nfs4_get_exchange_id_cred(struct nfs_client *clp)
 {
 	struct rpc_cred *cred;
@@ -877,6 +939,10 @@ static int nfs4_reclaim_locks(struct nfs4_state *state, const struct nfs4_state_
 			case -NFS4ERR_EXPIRED:
 			case -NFS4ERR_NO_GRACE:
 			case -NFS4ERR_STALE_CLIENTID:
+			case -NFS4ERR_BADSESSION:
+			case -NFS4ERR_BADSLOT:
+			case -NFS4ERR_BAD_HIGH_SLOT:
+			case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
 				goto out;
 			default:
 				printk(KERN_ERR "%s: unhandled error %d. Zeroing state\n",
@@ -959,6 +1025,10 @@ restart:
 			case -NFS4ERR_NO_GRACE:
 				nfs4_state_mark_reclaim_nograce(sp->so_client, state);
 			case -NFS4ERR_STALE_CLIENTID:
+			case -NFS4ERR_BADSESSION:
+			case -NFS4ERR_BADSLOT:
+			case -NFS4ERR_BAD_HIGH_SLOT:
+			case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
 				goto out_err;
 		}
 		nfs4_put_open_state(state);
@@ -1011,6 +1081,14 @@ static void nfs4_state_start_reclaim_reboot(struct nfs_client *clp)
 	nfs4_state_mark_reclaim_helper(clp, nfs4_state_mark_reclaim_reboot);
 }
 
+static void nfs4_reclaim_complete(struct nfs_client *clp,
+				 const struct nfs4_state_recovery_ops *ops)
+{
+	/* Notify the server we're done reclaiming our state */
+	if (ops->reclaim_complete)
+		(void)ops->reclaim_complete(clp);
+}
+
 static void nfs4_state_end_reclaim_reboot(struct nfs_client *clp)
 {
 	struct nfs4_state_owner *sp;
@@ -1020,6 +1098,9 @@ static void nfs4_state_end_reclaim_reboot(struct nfs_client *clp)
 	if (!test_and_clear_bit(NFS4CLNT_RECLAIM_REBOOT, &clp->cl_state))
 		return;
 
+	nfs4_reclaim_complete(clp,
+		nfs4_reboot_recovery_ops[clp->cl_minorversion]);
+
 	for (pos = rb_first(&clp->cl_state_owners); pos != NULL; pos = rb_next(pos)) {
 		sp = rb_entry(pos, struct nfs4_state_owner, so_client_node);
 		spin_lock(&sp->so_lock);
@@ -1046,25 +1127,25 @@ static void nfs4_state_start_reclaim_nograce(struct nfs_client *clp)
 	nfs4_state_mark_reclaim_helper(clp, nfs4_state_mark_reclaim_nograce);
 }
 
-static void nfs4_state_end_reclaim_nograce(struct nfs_client *clp)
-{
-	clear_bit(NFS4CLNT_RECLAIM_NOGRACE, &clp->cl_state);
-}
-
-static void nfs4_recovery_handle_error(struct nfs_client *clp, int error)
+static int nfs4_recovery_handle_error(struct nfs_client *clp, int error)
 {
 	switch (error) {
 		case -NFS4ERR_CB_PATH_DOWN:
 			nfs_handle_cb_pathdown(clp);
-			break;
+			return 0;
+		case -NFS4ERR_NO_GRACE:
+			nfs4_state_end_reclaim_reboot(clp);
+			return 0;
 		case -NFS4ERR_STALE_CLIENTID:
 		case -NFS4ERR_LEASE_MOVED:
 			set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
+			nfs4_state_end_reclaim_reboot(clp);
 			nfs4_state_start_reclaim_reboot(clp);
 			break;
 		case -NFS4ERR_EXPIRED:
 			set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
 			nfs4_state_start_reclaim_nograce(clp);
+			break;
 		case -NFS4ERR_BADSESSION:
 		case -NFS4ERR_BADSLOT:
 		case -NFS4ERR_BAD_HIGH_SLOT:
@@ -1072,8 +1153,11 @@ static void nfs4_recovery_handle_error(struct nfs_client *clp, int error)
 		case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
 		case -NFS4ERR_SEQ_FALSE_RETRY:
 		case -NFS4ERR_SEQ_MISORDERED:
-			set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
+			set_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state);
+			/* Zero session reset errors */
+			return 0;
 	}
+	return error;
 }
 
 static int nfs4_do_reclaim(struct nfs_client *clp, const struct nfs4_state_recovery_ops *ops)
@@ -1093,8 +1177,7 @@ restart:
 		if (status < 0) {
 			set_bit(ops->owner_flag_bit, &sp->so_flags);
 			nfs4_put_state_owner(sp);
-			nfs4_recovery_handle_error(clp, status);
-			return status;
+			return nfs4_recovery_handle_error(clp, status);
 		}
 		nfs4_put_state_owner(sp);
 		goto restart;
@@ -1124,8 +1207,7 @@ static int nfs4_check_lease(struct nfs_client *clp)
 	status = ops->renew_lease(clp, cred);
 	put_rpccred(cred);
 out:
-	nfs4_recovery_handle_error(clp, status);
-	return status;
+	return nfs4_recovery_handle_error(clp, status);
 }
 
 static int nfs4_reclaim_lease(struct nfs_client *clp)
@@ -1151,55 +1233,65 @@ static int nfs4_reclaim_lease(struct nfs_client *clp)
 }
 
 #ifdef CONFIG_NFS_V4_1
-static void nfs4_session_recovery_handle_error(struct nfs_client *clp, int err)
+void nfs41_handle_sequence_flag_errors(struct nfs_client *clp, u32 flags)
 {
-	switch (err) {
-	case -NFS4ERR_STALE_CLIENTID:
+	if (!flags)
+		return;
+	else if (flags & SEQ4_STATUS_RESTART_RECLAIM_NEEDED) {
 		set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
-		set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
-	}
+		nfs4_state_start_reclaim_reboot(clp);
+		nfs4_schedule_state_recovery(clp);
+	} else if (flags & (SEQ4_STATUS_EXPIRED_ALL_STATE_REVOKED |
+			    SEQ4_STATUS_EXPIRED_SOME_STATE_REVOKED |
+			    SEQ4_STATUS_ADMIN_STATE_REVOKED |
+			    SEQ4_STATUS_RECALLABLE_STATE_REVOKED |
+			    SEQ4_STATUS_LEASE_MOVED)) {
+		set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
+		nfs4_state_start_reclaim_nograce(clp);
+		nfs4_schedule_state_recovery(clp);
+	} else if (flags & (SEQ4_STATUS_CB_PATH_DOWN |
+			    SEQ4_STATUS_BACKCHANNEL_FAULT |
+			    SEQ4_STATUS_CB_PATH_DOWN_SESSION))
+		nfs_expire_all_delegations(clp);
 }
 
 static int nfs4_reset_session(struct nfs_client *clp)
 {
+	struct nfs4_session *ses = clp->cl_session;
 	int status;
 
+	status = nfs41_begin_drain_session(clp, ses);
+	if (status != 0)
+		return status;
+
 	status = nfs4_proc_destroy_session(clp->cl_session);
 	if (status && status != -NFS4ERR_BADSESSION &&
 	    status != -NFS4ERR_DEADSESSION) {
-		nfs4_session_recovery_handle_error(clp, status);
+		status = nfs4_recovery_handle_error(clp, status);
 		goto out;
 	}
 
 	memset(clp->cl_session->sess_id.data, 0, NFS4_MAX_SESSIONID_LEN);
-	status = nfs4_proc_create_session(clp, 1);
+	status = nfs4_proc_create_session(clp);
 	if (status)
-		nfs4_session_recovery_handle_error(clp, status);
-		/* fall through*/
-out:
-	/* Wake up the next rpc task even on error */
-	rpc_wake_up_next(&clp->cl_session->fc_slot_table.slot_tbl_waitq);
-	return status;
-}
-
-static int nfs4_initialize_session(struct nfs_client *clp)
-{
-	int status;
+		status = nfs4_recovery_handle_error(clp, status);
 
-	status = nfs4_proc_create_session(clp, 0);
-	if (!status) {
-		nfs_mark_client_ready(clp, NFS_CS_READY);
-	} else if (status == -NFS4ERR_STALE_CLIENTID) {
-		set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
-		set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
-	} else {
-		nfs_mark_client_ready(clp, status);
+out:
+	/*
+	 * Let the state manager reestablish state
+	 * without waking other tasks yet.
+	 */
+	if (!test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state)) {
+		/* Wake up the next rpc task */
+		nfs41_end_drain_session(clp, ses);
+		if (status == 0)
+			nfs41_setup_state_renewal(clp);
 	}
 	return status;
 }
+
 #else /* CONFIG_NFS_V4_1 */
 static int nfs4_reset_session(struct nfs_client *clp) { return 0; }
-static int nfs4_initialize_session(struct nfs_client *clp) { return 0; }
 #endif /* CONFIG_NFS_V4_1 */
 
 /* Set NFS4CLNT_LEASE_EXPIRED for all v4.0 errors and for recoverable errors
@@ -1234,7 +1326,8 @@ static void nfs4_state_manager(struct nfs_client *clp)
 			status = nfs4_reclaim_lease(clp);
 			if (status) {
 				nfs4_set_lease_expired(clp, status);
-				if (status == -EAGAIN)
+				if (test_bit(NFS4CLNT_LEASE_EXPIRED,
+							&clp->cl_state))
 					continue;
 				if (clp->cl_cons_state ==
 							NFS_CS_SESSION_INITING)
@@ -1242,55 +1335,51 @@ static void nfs4_state_manager(struct nfs_client *clp)
 				goto out_error;
 			}
 			clear_bit(NFS4CLNT_CHECK_LEASE, &clp->cl_state);
+			set_bit(NFS4CLNT_RECLAIM_REBOOT, &clp->cl_state);
 		}
 
 		if (test_and_clear_bit(NFS4CLNT_CHECK_LEASE, &clp->cl_state)) {
 			status = nfs4_check_lease(clp);
-			if (status != 0)
+			if (test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state))
 				continue;
+			if (status < 0 && status != -NFS4ERR_CB_PATH_DOWN)
+				goto out_error;
 		}
+
 		/* Initialize or reset the session */
-		if (test_and_clear_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state)
+		if (test_and_clear_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state)
 		   && nfs4_has_session(clp)) {
-			if (clp->cl_cons_state == NFS_CS_SESSION_INITING)
-				status = nfs4_initialize_session(clp);
-			else
-				status = nfs4_reset_session(clp);
-			if (status) {
-				if (status == -NFS4ERR_STALE_CLIENTID)
-					continue;
+			status = nfs4_reset_session(clp);
+			if (test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state))
+				continue;
+			if (status < 0)
 				goto out_error;
-			}
 		}
+
 		/* First recover reboot state... */
-		if (test_and_clear_bit(NFS4CLNT_RECLAIM_REBOOT, &clp->cl_state)) {
+		if (test_bit(NFS4CLNT_RECLAIM_REBOOT, &clp->cl_state)) {
 			status = nfs4_do_reclaim(clp,
 				nfs4_reboot_recovery_ops[clp->cl_minorversion]);
-			if (status == -NFS4ERR_STALE_CLIENTID)
-				continue;
-			if (test_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state))
+			if (test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state) ||
+			    test_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state))
 				continue;
 			nfs4_state_end_reclaim_reboot(clp);
-			continue;
+			if (test_bit(NFS4CLNT_RECLAIM_NOGRACE, &clp->cl_state))
+				continue;
+			if (status < 0)
+				goto out_error;
 		}
 
 		/* Now recover expired state... */
 		if (test_and_clear_bit(NFS4CLNT_RECLAIM_NOGRACE, &clp->cl_state)) {
 			status = nfs4_do_reclaim(clp,
 				nfs4_nograce_recovery_ops[clp->cl_minorversion]);
-			if (status < 0) {
-				set_bit(NFS4CLNT_RECLAIM_NOGRACE, &clp->cl_state);
-				if (status == -NFS4ERR_STALE_CLIENTID)
-					continue;
-				if (status == -NFS4ERR_EXPIRED)
-					continue;
-				if (test_bit(NFS4CLNT_SESSION_SETUP,
-								&clp->cl_state))
-					continue;
+			if (test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state) ||
+			    test_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state) ||
+			    test_bit(NFS4CLNT_RECLAIM_REBOOT, &clp->cl_state))
+				continue;
+			if (status < 0)
 				goto out_error;
-			} else
-				nfs4_state_end_reclaim_nograce(clp);
-			continue;
 		}
 
 		if (test_and_clear_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state)) {
@@ -1309,8 +1398,6 @@ static void nfs4_state_manager(struct nfs_client *clp)
 out_error:
 	printk(KERN_WARNING "Error: state manager failed on NFSv4 server %s"
 			" with error %d\n", clp->cl_hostname, -status);
-	if (test_bit(NFS4CLNT_RECLAIM_REBOOT, &clp->cl_state))
-		nfs4_state_end_reclaim_reboot(clp);
 	nfs4_clear_state_manager_bit(clp);
 }
 
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 20b4e30e6c82..e437fd6a819f 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -46,11 +46,13 @@
 #include <linux/proc_fs.h>
 #include <linux/kdev_t.h>
 #include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/msg_prot.h>
 #include <linux/nfs.h>
 #include <linux/nfs4.h>
 #include <linux/nfs_fs.h>
 #include <linux/nfs_idmap.h>
 #include "nfs4_fs.h"
+#include "internal.h"
 
 #define NFSDBG_FACILITY		NFSDBG_XDR
 
@@ -134,7 +136,7 @@ static int nfs4_stat_to_errno(int);
 #define decode_lookup_maxsz	(op_decode_hdr_maxsz)
 #define encode_share_access_maxsz \
 				(2)
-#define encode_createmode_maxsz	(1 + encode_attrs_maxsz)
+#define encode_createmode_maxsz	(1 + encode_attrs_maxsz + encode_verifier_maxsz)
 #define encode_opentype_maxsz	(1 + encode_createmode_maxsz)
 #define encode_claim_null_maxsz	(1 + nfs4_name_maxsz)
 #define encode_open_maxsz	(op_encode_hdr_maxsz + \
@@ -299,6 +301,8 @@ static int nfs4_stat_to_errno(int);
 				XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + 4)
 #define decode_sequence_maxsz	(op_decode_hdr_maxsz + \
 				XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + 5)
+#define encode_reclaim_complete_maxsz	(op_encode_hdr_maxsz + 4)
+#define decode_reclaim_complete_maxsz	(op_decode_hdr_maxsz + 4)
 #else /* CONFIG_NFS_V4_1 */
 #define encode_sequence_maxsz	0
 #define decode_sequence_maxsz	0
@@ -676,6 +680,25 @@ static int nfs4_stat_to_errno(int);
 					 decode_sequence_maxsz + \
 					 decode_putrootfh_maxsz + \
 					 decode_fsinfo_maxsz)
+#define NFS4_enc_reclaim_complete_sz	(compound_encode_hdr_maxsz + \
+					 encode_sequence_maxsz + \
+					 encode_reclaim_complete_maxsz)
+#define NFS4_dec_reclaim_complete_sz	(compound_decode_hdr_maxsz + \
+					 decode_sequence_maxsz + \
+					 decode_reclaim_complete_maxsz)
+
+const u32 nfs41_maxwrite_overhead = ((RPC_MAX_HEADER_WITH_AUTH +
+				      compound_encode_hdr_maxsz +
+				      encode_sequence_maxsz +
+				      encode_putfh_maxsz +
+				      encode_getattr_maxsz) *
+				     XDR_UNIT);
+
+const u32 nfs41_maxread_overhead = ((RPC_MAX_HEADER_WITH_AUTH +
+				     compound_decode_hdr_maxsz +
+				     decode_sequence_maxsz +
+				     decode_putfh_maxsz) *
+				    XDR_UNIT);
 #endif /* CONFIG_NFS_V4_1 */
 
 static const umode_t nfs_type2fmt[] = {
@@ -1140,6 +1163,7 @@ static inline void encode_openhdr(struct xdr_stream *xdr, const struct nfs_opena
 static inline void encode_createmode(struct xdr_stream *xdr, const struct nfs_openargs *arg)
 {
 	__be32 *p;
+	struct nfs_client *clp;
 
 	p = reserve_space(xdr, 4);
 	switch(arg->open_flags & O_EXCL) {
@@ -1148,8 +1172,23 @@ static inline void encode_createmode(struct xdr_stream *xdr, const struct nfs_op
 		encode_attrs(xdr, arg->u.attrs, arg->server);
 		break;
 	default:
-		*p = cpu_to_be32(NFS4_CREATE_EXCLUSIVE);
-		encode_nfs4_verifier(xdr, &arg->u.verifier);
+		clp = arg->server->nfs_client;
+		if (clp->cl_minorversion > 0) {
+			if (nfs4_has_persistent_session(clp)) {
+				*p = cpu_to_be32(NFS4_CREATE_GUARDED);
+				encode_attrs(xdr, arg->u.attrs, arg->server);
+			} else {
+				struct iattr dummy;
+
+				*p = cpu_to_be32(NFS4_CREATE_EXCLUSIVE4_1);
+				encode_nfs4_verifier(xdr, &arg->u.verifier);
+				dummy.ia_valid = 0;
+				encode_attrs(xdr, &dummy, arg->server);
+			}
+		} else {
+			*p = cpu_to_be32(NFS4_CREATE_EXCLUSIVE);
+			encode_nfs4_verifier(xdr, &arg->u.verifier);
+		}
 	}
 }
 
@@ -1592,6 +1631,19 @@ static void encode_destroy_session(struct xdr_stream *xdr,
 	hdr->nops++;
 	hdr->replen += decode_destroy_session_maxsz;
 }
+
+static void encode_reclaim_complete(struct xdr_stream *xdr,
+				    struct nfs41_reclaim_complete_args *args,
+				    struct compound_hdr *hdr)
+{
+	__be32 *p;
+
+	p = reserve_space(xdr, 8);
+	*p++ = cpu_to_be32(OP_RECLAIM_COMPLETE);
+	*p++ = cpu_to_be32(args->one_fs);
+	hdr->nops++;
+	hdr->replen += decode_reclaim_complete_maxsz;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 static void encode_sequence(struct xdr_stream *xdr,
@@ -2096,7 +2148,7 @@ nfs4_xdr_enc_getacl(struct rpc_rqst *req, __be32 *p,
 	encode_compound_hdr(&xdr, req, &hdr);
 	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
-	replen = hdr.replen + nfs4_fattr_bitmap_maxsz + 1;
+	replen = hdr.replen + op_decode_hdr_maxsz + nfs4_fattr_bitmap_maxsz + 1;
 	encode_getattr_two(&xdr, FATTR4_WORD0_ACL, 0, &hdr);
 
 	xdr_inline_pages(&req->rq_rcv_buf, replen << 2,
@@ -2420,6 +2472,26 @@ static int nfs4_xdr_enc_get_lease_time(struct rpc_rqst *req, uint32_t *p,
 	encode_nops(&hdr);
 	return 0;
 }
+
+/*
+ * a RECLAIM_COMPLETE request
+ */
+static int nfs4_xdr_enc_reclaim_complete(struct rpc_rqst *req, uint32_t *p,
+				     struct nfs41_reclaim_complete_args *args)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr = {
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args)
+	};
+
+	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
+	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
+	encode_reclaim_complete(&xdr, args, &hdr);
+	encode_nops(&hdr);
+	return 0;
+}
+
 #endif /* CONFIG_NFS_V4_1 */
 
 static void print_overflow_msg(const char *func, const struct xdr_stream *xdr)
@@ -4528,6 +4600,11 @@ static int decode_destroy_session(struct xdr_stream *xdr, void *dummy)
 {
 	return decode_op_hdr(xdr, OP_DESTROY_SESSION);
 }
+
+static int decode_reclaim_complete(struct xdr_stream *xdr, void *dummy)
+{
+	return decode_op_hdr(xdr, OP_RECLAIM_COMPLETE);
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 static int decode_sequence(struct xdr_stream *xdr,
@@ -4583,8 +4660,8 @@ static int decode_sequence(struct xdr_stream *xdr,
 	dummy = be32_to_cpup(p++);
 	/* target highest slot id - currently not processed */
 	dummy = be32_to_cpup(p++);
-	/* result flags - currently not processed */
-	dummy = be32_to_cpup(p);
+	/* result flags */
+	res->sr_status_flags = be32_to_cpup(p);
 	status = 0;
 out_err:
 	res->sr_status = status;
@@ -5309,7 +5386,7 @@ out:
 }
 
 /*
- * FSINFO request
+ * Decode FSINFO response
  */
 static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, __be32 *p,
 			       struct nfs4_fsinfo_res *res)
@@ -5330,7 +5407,7 @@ static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, __be32 *p,
 }
 
 /*
- * PATHCONF request
+ * Decode PATHCONF response
  */
 static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, __be32 *p,
 				 struct nfs4_pathconf_res *res)
@@ -5351,7 +5428,7 @@ static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, __be32 *p,
 }
 
 /*
- * STATFS request
+ * Decode STATFS response
  */
 static int nfs4_xdr_dec_statfs(struct rpc_rqst *req, __be32 *p,
 			       struct nfs4_statfs_res *res)
@@ -5372,7 +5449,7 @@ static int nfs4_xdr_dec_statfs(struct rpc_rqst *req, __be32 *p,
 }
 
 /*
- * GETATTR_BITMAP request
+ * Decode GETATTR_BITMAP response
  */
 static int nfs4_xdr_dec_server_caps(struct rpc_rqst *req, __be32 *p, struct nfs4_server_caps_res *res)
 {
@@ -5411,7 +5488,7 @@ static int nfs4_xdr_dec_renew(struct rpc_rqst *rqstp, __be32 *p, void *dummy)
 }
 
 /*
- * a SETCLIENTID request
+ * Decode SETCLIENTID response
  */
 static int nfs4_xdr_dec_setclientid(struct rpc_rqst *req, __be32 *p,
 		struct nfs_client *clp)
@@ -5428,7 +5505,7 @@ static int nfs4_xdr_dec_setclientid(struct rpc_rqst *req, __be32 *p,
 }
 
 /*
- * a SETCLIENTID_CONFIRM request
+ * Decode SETCLIENTID_CONFIRM response
  */
 static int nfs4_xdr_dec_setclientid_confirm(struct rpc_rqst *req, __be32 *p, struct nfs_fsinfo *fsinfo)
 {
@@ -5448,7 +5525,7 @@ static int nfs4_xdr_dec_setclientid_confirm(struct rpc_rqst *req, __be32 *p, str
 }
 
 /*
- * DELEGRETURN request
+ * Decode DELEGRETURN response
  */
 static int nfs4_xdr_dec_delegreturn(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_delegreturnres *res)
 {
@@ -5474,7 +5551,7 @@ out:
 }
 
 /*
- * FS_LOCATIONS request
+ * Decode FS_LOCATIONS response
  */
 static int nfs4_xdr_dec_fs_locations(struct rpc_rqst *req, __be32 *p,
 				     struct nfs4_fs_locations_res *res)
@@ -5504,7 +5581,7 @@ out:
 
 #if defined(CONFIG_NFS_V4_1)
 /*
- * EXCHANGE_ID request
+ * Decode EXCHANGE_ID response
  */
 static int nfs4_xdr_dec_exchange_id(struct rpc_rqst *rqstp, uint32_t *p,
 				    void *res)
@@ -5521,7 +5598,7 @@ static int nfs4_xdr_dec_exchange_id(struct rpc_rqst *rqstp, uint32_t *p,
 }
 
 /*
- * a CREATE_SESSION request
+ * Decode CREATE_SESSION response
  */
 static int nfs4_xdr_dec_create_session(struct rpc_rqst *rqstp, uint32_t *p,
 				       struct nfs41_create_session_res *res)
@@ -5538,7 +5615,7 @@ static int nfs4_xdr_dec_create_session(struct rpc_rqst *rqstp, uint32_t *p,
 }
 
 /*
- * a DESTROY_SESSION request
+ * Decode DESTROY_SESSION response
  */
 static int nfs4_xdr_dec_destroy_session(struct rpc_rqst *rqstp, uint32_t *p,
 					void *dummy)
@@ -5555,7 +5632,7 @@ static int nfs4_xdr_dec_destroy_session(struct rpc_rqst *rqstp, uint32_t *p,
 }
 
 /*
- * a SEQUENCE request
+ * Decode SEQUENCE response
  */
 static int nfs4_xdr_dec_sequence(struct rpc_rqst *rqstp, uint32_t *p,
 				 struct nfs4_sequence_res *res)
@@ -5572,7 +5649,7 @@ static int nfs4_xdr_dec_sequence(struct rpc_rqst *rqstp, uint32_t *p,
 }
 
 /*
- * a GET_LEASE_TIME request
+ * Decode GET_LEASE_TIME response
  */
 static int nfs4_xdr_dec_get_lease_time(struct rpc_rqst *rqstp, uint32_t *p,
 				       struct nfs4_get_lease_time_res *res)
@@ -5591,6 +5668,25 @@ static int nfs4_xdr_dec_get_lease_time(struct rpc_rqst *rqstp, uint32_t *p,
 		status = decode_fsinfo(&xdr, res->lr_fsinfo);
 	return status;
 }
+
+/*
+ * Decode RECLAIM_COMPLETE response
+ */
+static int nfs4_xdr_dec_reclaim_complete(struct rpc_rqst *rqstp, uint32_t *p,
+					 struct nfs41_reclaim_complete_res *res)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr;
+	int status;
+
+	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (!status)
+		status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (!status)
+		status = decode_reclaim_complete(&xdr, (void *)NULL);
+	return status;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 __be32 *nfs4_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus)
@@ -5767,6 +5863,7 @@ struct rpc_procinfo	nfs4_procedures[] = {
   PROC(DESTROY_SESSION,	enc_destroy_session,	dec_destroy_session),
   PROC(SEQUENCE,	enc_sequence,	dec_sequence),
   PROC(GET_LEASE_TIME,	enc_get_lease_time,	dec_get_lease_time),
+  PROC(RECLAIM_COMPLETE, enc_reclaim_complete,  dec_reclaim_complete),
 #endif /* CONFIG_NFS_V4_1 */
 };
 
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index 12c9e66d3f1d..db9b360ae19d 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -356,25 +356,19 @@ static void nfs_readpage_retry(struct rpc_task *task, struct nfs_read_data *data
 	struct nfs_readres *resp = &data->res;
 
 	if (resp->eof || resp->count == argp->count)
-		goto out;
+		return;
 
 	/* This is a short read! */
 	nfs_inc_stats(data->inode, NFSIOS_SHORTREAD);
 	/* Has the server at least made some progress? */
 	if (resp->count == 0)
-		goto out;
+		return;
 
 	/* Yes, so retry the read at the end of the data */
 	argp->offset += resp->count;
 	argp->pgbase += resp->count;
 	argp->count -= resp->count;
-	nfs4_restart_rpc(task, NFS_SERVER(data->inode)->nfs_client);
-	return;
-out:
-	nfs4_sequence_free_slot(NFS_SERVER(data->inode)->nfs_client,
-				&data->res.seq_res);
-	return;
-
+	nfs_restart_rpc(task, NFS_SERVER(data->inode)->nfs_client);
 }
 
 /*
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 90be551b80c1..ce907efc5508 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -175,14 +175,16 @@ static const match_table_t nfs_mount_option_tokens = {
 };
 
 enum {
-	Opt_xprt_udp, Opt_xprt_tcp, Opt_xprt_rdma,
+	Opt_xprt_udp, Opt_xprt_udp6, Opt_xprt_tcp, Opt_xprt_tcp6, Opt_xprt_rdma,
 
 	Opt_xprt_err
 };
 
 static const match_table_t nfs_xprt_protocol_tokens = {
 	{ Opt_xprt_udp, "udp" },
+	{ Opt_xprt_udp6, "udp6" },
 	{ Opt_xprt_tcp, "tcp" },
+	{ Opt_xprt_tcp6, "tcp6" },
 	{ Opt_xprt_rdma, "rdma" },
 
 	{ Opt_xprt_err, NULL }
@@ -492,6 +494,45 @@ static const char *nfs_pseudoflavour_to_name(rpc_authflavor_t flavour)
 	return sec_flavours[i].str;
 }
 
+static void nfs_show_mountd_netid(struct seq_file *m, struct nfs_server *nfss,
+				  int showdefaults)
+{
+	struct sockaddr *sap = (struct sockaddr *) &nfss->mountd_address;
+
+	seq_printf(m, ",mountproto=");
+	switch (sap->sa_family) {
+	case AF_INET:
+		switch (nfss->mountd_protocol) {
+		case IPPROTO_UDP:
+			seq_printf(m, RPCBIND_NETID_UDP);
+			break;
+		case IPPROTO_TCP:
+			seq_printf(m, RPCBIND_NETID_TCP);
+			break;
+		default:
+			if (showdefaults)
+				seq_printf(m, "auto");
+		}
+		break;
+	case AF_INET6:
+		switch (nfss->mountd_protocol) {
+		case IPPROTO_UDP:
+			seq_printf(m, RPCBIND_NETID_UDP6);
+			break;
+		case IPPROTO_TCP:
+			seq_printf(m, RPCBIND_NETID_TCP6);
+			break;
+		default:
+			if (showdefaults)
+				seq_printf(m, "auto");
+		}
+		break;
+	default:
+		if (showdefaults)
+			seq_printf(m, "auto");
+	}
+}
+
 static void nfs_show_mountd_options(struct seq_file *m, struct nfs_server *nfss,
 				    int showdefaults)
 {
@@ -505,7 +546,7 @@ static void nfs_show_mountd_options(struct seq_file *m, struct nfs_server *nfss,
 	}
 	case AF_INET6: {
 		struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)sap;
-		seq_printf(m, ",mountaddr=%pI6", &sin6->sin6_addr);
+		seq_printf(m, ",mountaddr=%pI6c", &sin6->sin6_addr);
 		break;
 	}
 	default:
@@ -518,17 +559,7 @@ static void nfs_show_mountd_options(struct seq_file *m, struct nfs_server *nfss,
 	if (nfss->mountd_port || showdefaults)
 		seq_printf(m, ",mountport=%u", nfss->mountd_port);
 
-	switch (nfss->mountd_protocol) {
-	case IPPROTO_UDP:
-		seq_printf(m, ",mountproto=udp");
-		break;
-	case IPPROTO_TCP:
-		seq_printf(m, ",mountproto=tcp");
-		break;
-	default:
-		if (showdefaults)
-			seq_printf(m, ",mountproto=auto");
-	}
+	nfs_show_mountd_netid(m, nfss, showdefaults);
 }
 
 /*
@@ -578,7 +609,7 @@ static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss,
 			seq_puts(m, nfs_infop->nostr);
 	}
 	seq_printf(m, ",proto=%s",
-		   rpc_peeraddr2str(nfss->client, RPC_DISPLAY_PROTO));
+		   rpc_peeraddr2str(nfss->client, RPC_DISPLAY_NETID));
 	if (version == 4) {
 		if (nfss->port != NFS_PORT)
 			seq_printf(m, ",port=%u", nfss->port);
@@ -714,8 +745,6 @@ static void nfs_umount_begin(struct super_block *sb)
 	struct nfs_server *server;
 	struct rpc_clnt *rpc;
 
-	lock_kernel();
-
 	server = NFS_SB(sb);
 	/* -EIO all pending I/O */
 	rpc = server->client_acl;
@@ -724,8 +753,6 @@ static void nfs_umount_begin(struct super_block *sb)
 	rpc = server->client;
 	if (!IS_ERR(rpc))
 		rpc_killall_tasks(rpc);
-
-	unlock_kernel();
 }
 
 static struct nfs_parsed_mount_data *nfs_alloc_parsed_mount_data(unsigned int version)
@@ -734,8 +761,6 @@ static struct nfs_parsed_mount_data *nfs_alloc_parsed_mount_data(unsigned int ve
 
 	data = kzalloc(sizeof(*data), GFP_KERNEL);
 	if (data) {
-		data->rsize		= NFS_MAX_FILE_IO_SIZE;
-		data->wsize		= NFS_MAX_FILE_IO_SIZE;
 		data->acregmin		= NFS_DEF_ACREGMIN;
 		data->acregmax		= NFS_DEF_ACREGMAX;
 		data->acdirmin		= NFS_DEF_ACDIRMIN;
@@ -887,6 +912,8 @@ static int nfs_parse_mount_options(char *raw,
 {
 	char *p, *string, *secdata;
 	int rc, sloppy = 0, invalid_option = 0;
+	unsigned short protofamily = AF_UNSPEC;
+	unsigned short mountfamily = AF_UNSPEC;
 
 	if (!raw) {
 		dfprintk(MOUNT, "NFS: mount options string was NULL.\n");
@@ -1232,12 +1259,17 @@ static int nfs_parse_mount_options(char *raw,
 			token = match_token(string,
 					    nfs_xprt_protocol_tokens, args);
 
+			protofamily = AF_INET;
 			switch (token) {
+			case Opt_xprt_udp6:
+				protofamily = AF_INET6;
 			case Opt_xprt_udp:
 				mnt->flags &= ~NFS_MOUNT_TCP;
 				mnt->nfs_server.protocol = XPRT_TRANSPORT_UDP;
 				kfree(string);
 				break;
+			case Opt_xprt_tcp6:
+				protofamily = AF_INET6;
 			case Opt_xprt_tcp:
 				mnt->flags |= NFS_MOUNT_TCP;
 				mnt->nfs_server.protocol = XPRT_TRANSPORT_TCP;
@@ -1265,10 +1297,15 @@ static int nfs_parse_mount_options(char *raw,
 					    nfs_xprt_protocol_tokens, args);
 			kfree(string);
 
+			mountfamily = AF_INET;
 			switch (token) {
+			case Opt_xprt_udp6:
+				mountfamily = AF_INET6;
 			case Opt_xprt_udp:
 				mnt->mount_server.protocol = XPRT_TRANSPORT_UDP;
 				break;
+			case Opt_xprt_tcp6:
+				mountfamily = AF_INET6;
 			case Opt_xprt_tcp:
 				mnt->mount_server.protocol = XPRT_TRANSPORT_TCP;
 				break;
@@ -1367,8 +1404,33 @@ static int nfs_parse_mount_options(char *raw,
 	if (!sloppy && invalid_option)
 		return 0;
 
+	/*
+	 * verify that any proto=/mountproto= options match the address
+	 * familiies in the addr=/mountaddr= options.
+	 */
+	if (protofamily != AF_UNSPEC &&
+	    protofamily != mnt->nfs_server.address.ss_family)
+		goto out_proto_mismatch;
+
+	if (mountfamily != AF_UNSPEC) {
+		if (mnt->mount_server.addrlen) {
+			if (mountfamily != mnt->mount_server.address.ss_family)
+				goto out_mountproto_mismatch;
+		} else {
+			if (mountfamily != mnt->nfs_server.address.ss_family)
+				goto out_mountproto_mismatch;
+		}
+	}
+
 	return 1;
 
+out_mountproto_mismatch:
+	printk(KERN_INFO "NFS: mount server address does not match mountproto= "
+			 "option\n");
+	return 0;
+out_proto_mismatch:
+	printk(KERN_INFO "NFS: server address does not match proto= option\n");
+	return 0;
 out_invalid_address:
 	printk(KERN_INFO "NFS: bad IP address specified: %s\n", p);
 	return 0;
@@ -1881,7 +1943,6 @@ nfs_remount(struct super_block *sb, int *flags, char *raw_data)
 	if (data == NULL)
 		return -ENOMEM;
 
-	lock_kernel();
 	/* fill out struct with values from existing mount */
 	data->flags = nfss->flags;
 	data->rsize = nfss->rsize;
@@ -1907,7 +1968,6 @@ nfs_remount(struct super_block *sb, int *flags, char *raw_data)
 	error = nfs_compare_remount_data(nfss, data);
 out:
 	kfree(data);
-	unlock_kernel();
 	return error;
 }
 
diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c
index 1064c91ae810..6da3d3ff6edd 100644
--- a/fs/nfs/unlink.c
+++ b/fs/nfs/unlink.c
@@ -83,7 +83,7 @@ static void nfs_async_unlink_done(struct rpc_task *task, void *calldata)
 	struct inode *dir = data->dir;
 
 	if (!NFS_PROTO(dir)->unlink_done(task, dir))
-		nfs4_restart_rpc(task, NFS_SERVER(dir)->nfs_client);
+		nfs_restart_rpc(task, NFS_SERVER(dir)->nfs_client);
 }
 
 /**
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index b1ce2ea9b93b..d171696017f4 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -1216,7 +1216,7 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
 				 */
 				argp->stable = NFS_FILE_SYNC;
 			}
-			nfs4_restart_rpc(task, server->nfs_client);
+			nfs_restart_rpc(task, server->nfs_client);
 			return -EAGAIN;
 		}
 		if (time_before(complain, jiffies)) {
@@ -1228,7 +1228,6 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
 		/* Can't do anything about it except throw an error. */
 		task->tk_status = -EIO;
 	}
-	nfs4_sequence_free_slot(server->nfs_client, &data->res.seq_res);
 	return 0;
 }
 
@@ -1612,15 +1611,16 @@ int nfs_migrate_page(struct address_space *mapping, struct page *newpage,
 	if (ret)
 		goto out_unlock;
 	page_cache_get(newpage);
+	spin_lock(&mapping->host->i_lock);
 	req->wb_page = newpage;
 	SetPagePrivate(newpage);
-	set_page_private(newpage, page_private(page));
+	set_page_private(newpage, (unsigned long)req);
 	ClearPagePrivate(page);
 	set_page_private(page, 0);
+	spin_unlock(&mapping->host->i_lock);
 	page_cache_release(page);
 out_unlock:
 	nfs_clear_page_tag_locked(req);
-	nfs_release_request(req);
 out:
 	return ret;
 }
diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h
index c4c060208109..9b8299af3741 100644
--- a/include/linux/nfs4.h
+++ b/include/linux/nfs4.h
@@ -128,6 +128,8 @@
 #define SEQ4_STATUS_RECALLABLE_STATE_REVOKED	0x00000040
 #define SEQ4_STATUS_LEASE_MOVED			0x00000080
 #define SEQ4_STATUS_RESTART_RECLAIM_NEEDED	0x00000100
+#define SEQ4_STATUS_CB_PATH_DOWN_SESSION	0x00000200
+#define SEQ4_STATUS_BACKCHANNEL_FAULT		0x00000400
 
 #define NFS4_MAX_UINT64	(~(u64)0)
 
@@ -528,6 +530,7 @@ enum {
 	NFSPROC4_CLNT_DESTROY_SESSION,
 	NFSPROC4_CLNT_SEQUENCE,
 	NFSPROC4_CLNT_GET_LEASE_TIME,
+	NFSPROC4_CLNT_RECLAIM_COMPLETE,
 };
 
 /* nfs41 types */
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index 320569eabe3b..34fc6be5bfcf 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -209,6 +209,7 @@ struct nfs4_session {
 	unsigned long			session_state;
 	u32				hash_alg;
 	u32				ssv_len;
+	struct completion		complete;
 
 	/* The fore and back channel */
 	struct nfs4_channel_attrs	fc_attrs;
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 62f63fb0c4c8..51071b335751 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -170,8 +170,9 @@ struct nfs4_sequence_args {
 struct nfs4_sequence_res {
 	struct nfs4_session	*sr_session;
 	u8			sr_slotid;	/* slot used to send request */
-	unsigned long		sr_renewal_time;
 	int			sr_status;	/* sequence operation status */
+	unsigned long		sr_renewal_time;
+	u32			sr_status_flags;
 };
 
 struct nfs4_get_lease_time_args {
@@ -938,6 +939,16 @@ struct nfs41_create_session_args {
 struct nfs41_create_session_res {
 	struct nfs_client	       *client;
 };
+
+struct nfs41_reclaim_complete_args {
+	/* In the future extend to include curr_fh for use with migration */
+	unsigned char			one_fs:1;
+	struct nfs4_sequence_args	seq_args;
+};
+
+struct nfs41_reclaim_complete_res {
+	struct nfs4_sequence_res	seq_res;
+};
 #endif /* CONFIG_NFS_V4_1 */
 
 struct nfs_page;
diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h
index 401097781fc0..1906782ec86b 100644
--- a/include/linux/sunrpc/sched.h
+++ b/include/linux/sunrpc/sched.h
@@ -130,12 +130,14 @@ struct rpc_task_setup {
 #define RPC_TASK_DYNAMIC	0x0080		/* task was kmalloc'ed */
 #define RPC_TASK_KILLED		0x0100		/* task was killed */
 #define RPC_TASK_SOFT		0x0200		/* Use soft timeouts */
+#define RPC_TASK_SOFTCONN	0x0400		/* Fail if can't connect */
 
 #define RPC_IS_ASYNC(t)		((t)->tk_flags & RPC_TASK_ASYNC)
 #define RPC_IS_SWAPPER(t)	((t)->tk_flags & RPC_TASK_SWAPPER)
 #define RPC_DO_ROOTOVERRIDE(t)	((t)->tk_flags & RPC_TASK_ROOTCREDS)
 #define RPC_ASSASSINATED(t)	((t)->tk_flags & RPC_TASK_KILLED)
 #define RPC_IS_SOFT(t)		((t)->tk_flags & RPC_TASK_SOFT)
+#define RPC_IS_SOFTCONN(t)	((t)->tk_flags & RPC_TASK_SOFTCONN)
 
 #define RPC_TASK_RUNNING	0
 #define RPC_TASK_QUEUED		1
diff --git a/net/sunrpc/addr.c b/net/sunrpc/addr.c
index c7450c8f0a7c..6dcdd2517819 100644
--- a/net/sunrpc/addr.c
+++ b/net/sunrpc/addr.c
@@ -55,16 +55,8 @@ static size_t rpc_ntop6_noscopeid(const struct sockaddr *sap,
 
 	/*
 	 * RFC 4291, Section 2.2.1
-	 *
-	 * To keep the result as short as possible, especially
-	 * since we don't shorthand, we don't want leading zeros
-	 * in each halfword, so avoid %pI6.
 	 */
-	return snprintf(buf, buflen, "%x:%x:%x:%x:%x:%x:%x:%x",
-		ntohs(addr->s6_addr16[0]), ntohs(addr->s6_addr16[1]),
-		ntohs(addr->s6_addr16[2]), ntohs(addr->s6_addr16[3]),
-		ntohs(addr->s6_addr16[4]), ntohs(addr->s6_addr16[5]),
-		ntohs(addr->s6_addr16[6]), ntohs(addr->s6_addr16[7]));
+	return snprintf(buf, buflen, "%pI6c", addr);
 }
 
 static size_t rpc_ntop6(const struct sockaddr *sap,
diff --git a/net/sunrpc/auth.c b/net/sunrpc/auth.c
index 7535a7bed2fa..f394fc190a49 100644
--- a/net/sunrpc/auth.c
+++ b/net/sunrpc/auth.c
@@ -123,16 +123,19 @@ rpcauth_unhash_cred_locked(struct rpc_cred *cred)
 	clear_bit(RPCAUTH_CRED_HASHED, &cred->cr_flags);
 }
 
-static void
+static int
 rpcauth_unhash_cred(struct rpc_cred *cred)
 {
 	spinlock_t *cache_lock;
+	int ret;
 
 	cache_lock = &cred->cr_auth->au_credcache->lock;
 	spin_lock(cache_lock);
-	if (atomic_read(&cred->cr_count) == 0)
+	ret = atomic_read(&cred->cr_count) == 0;
+	if (ret)
 		rpcauth_unhash_cred_locked(cred);
 	spin_unlock(cache_lock);
+	return ret;
 }
 
 /*
@@ -446,31 +449,35 @@ void
 put_rpccred(struct rpc_cred *cred)
 {
 	/* Fast path for unhashed credentials */
-	if (test_bit(RPCAUTH_CRED_HASHED, &cred->cr_flags) != 0)
-		goto need_lock;
-
-	if (!atomic_dec_and_test(&cred->cr_count))
+	if (test_bit(RPCAUTH_CRED_HASHED, &cred->cr_flags) == 0) {
+		if (atomic_dec_and_test(&cred->cr_count))
+			cred->cr_ops->crdestroy(cred);
 		return;
-	goto out_destroy;
-need_lock:
+	}
+
 	if (!atomic_dec_and_lock(&cred->cr_count, &rpc_credcache_lock))
 		return;
 	if (!list_empty(&cred->cr_lru)) {
 		number_cred_unused--;
 		list_del_init(&cred->cr_lru);
 	}
-	if (test_bit(RPCAUTH_CRED_UPTODATE, &cred->cr_flags) == 0)
-		rpcauth_unhash_cred(cred);
 	if (test_bit(RPCAUTH_CRED_HASHED, &cred->cr_flags) != 0) {
-		cred->cr_expire = jiffies;
-		list_add_tail(&cred->cr_lru, &cred_unused);
-		number_cred_unused++;
-		spin_unlock(&rpc_credcache_lock);
-		return;
+		if (test_bit(RPCAUTH_CRED_UPTODATE, &cred->cr_flags) != 0) {
+			cred->cr_expire = jiffies;
+			list_add_tail(&cred->cr_lru, &cred_unused);
+			number_cred_unused++;
+			goto out_nodestroy;
+		}
+		if (!rpcauth_unhash_cred(cred)) {
+			/* We were hashed and someone looked us up... */
+			goto out_nodestroy;
+		}
 	}
 	spin_unlock(&rpc_credcache_lock);
-out_destroy:
 	cred->cr_ops->crdestroy(cred);
+	return;
+out_nodestroy:
+	spin_unlock(&rpc_credcache_lock);
 }
 EXPORT_SYMBOL_GPL(put_rpccred);
 
diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c
index fc6a43ccd950..3c3c50f38a1c 100644
--- a/net/sunrpc/auth_gss/auth_gss.c
+++ b/net/sunrpc/auth_gss/auth_gss.c
@@ -304,7 +304,7 @@ __gss_find_upcall(struct rpc_inode *rpci, uid_t uid)
  * to that upcall instead of adding the new upcall.
  */
 static inline struct gss_upcall_msg *
-gss_add_msg(struct gss_auth *gss_auth, struct gss_upcall_msg *gss_msg)
+gss_add_msg(struct gss_upcall_msg *gss_msg)
 {
 	struct rpc_inode *rpci = gss_msg->inode;
 	struct inode *inode = &rpci->vfs_inode;
@@ -445,7 +445,7 @@ gss_setup_upcall(struct rpc_clnt *clnt, struct gss_auth *gss_auth, struct rpc_cr
 	gss_new = gss_alloc_msg(gss_auth, uid, clnt, gss_cred->gc_machine_cred);
 	if (IS_ERR(gss_new))
 		return gss_new;
-	gss_msg = gss_add_msg(gss_auth, gss_new);
+	gss_msg = gss_add_msg(gss_new);
 	if (gss_msg == gss_new) {
 		struct inode *inode = &gss_new->inode->vfs_inode;
 		int res = rpc_queue_upcall(inode, &gss_new->msg);
@@ -485,7 +485,7 @@ gss_refresh_upcall(struct rpc_task *task)
 	dprintk("RPC: %5u gss_refresh_upcall for uid %u\n", task->tk_pid,
 								cred->cr_uid);
 	gss_msg = gss_setup_upcall(task->tk_client, gss_auth, cred);
-	if (IS_ERR(gss_msg) == -EAGAIN) {
+	if (PTR_ERR(gss_msg) == -EAGAIN) {
 		/* XXX: warning on the first, under the assumption we
 		 * shouldn't normally hit this case on a refresh. */
 		warn_gssd();
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 38829e20500b..154034b675bd 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -79,7 +79,7 @@ static void	call_connect_status(struct rpc_task *task);
 
 static __be32	*rpc_encode_header(struct rpc_task *task);
 static __be32	*rpc_verify_header(struct rpc_task *task);
-static int	rpc_ping(struct rpc_clnt *clnt, int flags);
+static int	rpc_ping(struct rpc_clnt *clnt);
 
 static void rpc_register_client(struct rpc_clnt *clnt)
 {
@@ -340,7 +340,7 @@ struct rpc_clnt *rpc_create(struct rpc_create_args *args)
 		return clnt;
 
 	if (!(args->flags & RPC_CLNT_CREATE_NOPING)) {
-		int err = rpc_ping(clnt, RPC_TASK_SOFT);
+		int err = rpc_ping(clnt);
 		if (err != 0) {
 			rpc_shutdown_client(clnt);
 			return ERR_PTR(err);
@@ -528,7 +528,7 @@ struct rpc_clnt *rpc_bind_new_program(struct rpc_clnt *old,
 	clnt->cl_prog     = program->number;
 	clnt->cl_vers     = version->number;
 	clnt->cl_stats    = program->stats;
-	err = rpc_ping(clnt, RPC_TASK_SOFT);
+	err = rpc_ping(clnt);
 	if (err != 0) {
 		rpc_shutdown_client(clnt);
 		clnt = ERR_PTR(err);
@@ -1060,7 +1060,7 @@ call_bind_status(struct rpc_task *task)
 		goto retry_timeout;
 	case -EPFNOSUPPORT:
 		/* server doesn't support any rpcbind version we know of */
-		dprintk("RPC: %5u remote rpcbind service unavailable\n",
+		dprintk("RPC: %5u unrecognized remote rpcbind service\n",
 				task->tk_pid);
 		break;
 	case -EPROTONOSUPPORT:
@@ -1069,6 +1069,21 @@ call_bind_status(struct rpc_task *task)
 		task->tk_status = 0;
 		task->tk_action = call_bind;
 		return;
+	case -ECONNREFUSED:		/* connection problems */
+	case -ECONNRESET:
+	case -ENOTCONN:
+	case -EHOSTDOWN:
+	case -EHOSTUNREACH:
+	case -ENETUNREACH:
+	case -EPIPE:
+		dprintk("RPC: %5u remote rpcbind unreachable: %d\n",
+				task->tk_pid, task->tk_status);
+		if (!RPC_IS_SOFTCONN(task)) {
+			rpc_delay(task, 5*HZ);
+			goto retry_timeout;
+		}
+		status = task->tk_status;
+		break;
 	default:
 		dprintk("RPC: %5u unrecognized rpcbind error (%d)\n",
 				task->tk_pid, -task->tk_status);
@@ -1180,11 +1195,25 @@ static void
 call_transmit_status(struct rpc_task *task)
 {
 	task->tk_action = call_status;
+
+	/*
+	 * Common case: success.  Force the compiler to put this
+	 * test first.
+	 */
+	if (task->tk_status == 0) {
+		xprt_end_transmit(task);
+		rpc_task_force_reencode(task);
+		return;
+	}
+
 	switch (task->tk_status) {
 	case -EAGAIN:
 		break;
 	default:
+		dprint_status(task);
 		xprt_end_transmit(task);
+		rpc_task_force_reencode(task);
+		break;
 		/*
 		 * Special cases: if we've been waiting on the
 		 * socket's write_space() callback, or if the
@@ -1192,11 +1221,16 @@ call_transmit_status(struct rpc_task *task)
 		 * then hold onto the transport lock.
 		 */
 	case -ECONNREFUSED:
-	case -ECONNRESET:
-	case -ENOTCONN:
 	case -EHOSTDOWN:
 	case -EHOSTUNREACH:
 	case -ENETUNREACH:
+		if (RPC_IS_SOFTCONN(task)) {
+			xprt_end_transmit(task);
+			rpc_exit(task, task->tk_status);
+			break;
+		}
+	case -ECONNRESET:
+	case -ENOTCONN:
 	case -EPIPE:
 		rpc_task_force_reencode(task);
 	}
@@ -1346,6 +1380,10 @@ call_timeout(struct rpc_task *task)
 	dprintk("RPC: %5u call_timeout (major)\n", task->tk_pid);
 	task->tk_timeouts++;
 
+	if (RPC_IS_SOFTCONN(task)) {
+		rpc_exit(task, -ETIMEDOUT);
+		return;
+	}
 	if (RPC_IS_SOFT(task)) {
 		if (clnt->cl_chatty)
 			printk(KERN_NOTICE "%s: server %s not responding, timed out\n",
@@ -1675,14 +1713,14 @@ static struct rpc_procinfo rpcproc_null = {
 	.p_decode = rpcproc_decode_null,
 };
 
-static int rpc_ping(struct rpc_clnt *clnt, int flags)
+static int rpc_ping(struct rpc_clnt *clnt)
 {
 	struct rpc_message msg = {
 		.rpc_proc = &rpcproc_null,
 	};
 	int err;
 	msg.rpc_cred = authnull_ops.lookup_cred(NULL, NULL, 0);
-	err = rpc_call_sync(clnt, &msg, flags);
+	err = rpc_call_sync(clnt, &msg, RPC_TASK_SOFT | RPC_TASK_SOFTCONN);
 	put_rpccred(msg.rpc_cred);
 	return err;
 }
diff --git a/net/sunrpc/rpcb_clnt.c b/net/sunrpc/rpcb_clnt.c
index 830faf4d9997..3e3772d8eb92 100644
--- a/net/sunrpc/rpcb_clnt.c
+++ b/net/sunrpc/rpcb_clnt.c
@@ -20,6 +20,7 @@
 #include <linux/in6.h>
 #include <linux/kernel.h>
 #include <linux/errno.h>
+#include <linux/mutex.h>
 #include <net/ipv6.h>
 
 #include <linux/sunrpc/clnt.h>
@@ -110,6 +111,9 @@ static void			rpcb_getport_done(struct rpc_task *, void *);
 static void			rpcb_map_release(void *data);
 static struct rpc_program	rpcb_program;
 
+static struct rpc_clnt *	rpcb_local_clnt;
+static struct rpc_clnt *	rpcb_local_clnt4;
+
 struct rpcbind_args {
 	struct rpc_xprt *	r_xprt;
 
@@ -163,21 +167,60 @@ static const struct sockaddr_in rpcb_inaddr_loopback = {
 	.sin_port		= htons(RPCBIND_PORT),
 };
 
-static struct rpc_clnt *rpcb_create_local(struct sockaddr *addr,
-					  size_t addrlen, u32 version)
+static DEFINE_MUTEX(rpcb_create_local_mutex);
+
+/*
+ * Returns zero on success, otherwise a negative errno value
+ * is returned.
+ */
+static int rpcb_create_local(void)
 {
 	struct rpc_create_args args = {
-		.protocol	= XPRT_TRANSPORT_UDP,
-		.address	= addr,
-		.addrsize	= addrlen,
+		.protocol	= XPRT_TRANSPORT_TCP,
+		.address	= (struct sockaddr *)&rpcb_inaddr_loopback,
+		.addrsize	= sizeof(rpcb_inaddr_loopback),
 		.servername	= "localhost",
 		.program	= &rpcb_program,
-		.version	= version,
+		.version	= RPCBVERS_2,
 		.authflavor	= RPC_AUTH_UNIX,
 		.flags		= RPC_CLNT_CREATE_NOPING,
 	};
+	struct rpc_clnt *clnt, *clnt4;
+	int result = 0;
+
+	if (rpcb_local_clnt)
+		return result;
+
+	mutex_lock(&rpcb_create_local_mutex);
+	if (rpcb_local_clnt)
+		goto out;
+
+	clnt = rpc_create(&args);
+	if (IS_ERR(clnt)) {
+		dprintk("RPC:       failed to create local rpcbind "
+				"client (errno %ld).\n", PTR_ERR(clnt));
+		result = -PTR_ERR(clnt);
+		goto out;
+	}
 
-	return rpc_create(&args);
+	/*
+	 * This results in an RPC ping.  On systems running portmapper,
+	 * the v4 ping will fail.  Proceed anyway, but disallow rpcb
+	 * v4 upcalls.
+	 */
+	clnt4 = rpc_bind_new_program(clnt, &rpcb_program, RPCBVERS_4);
+	if (IS_ERR(clnt4)) {
+		dprintk("RPC:       failed to create local rpcbind v4 "
+				"cleint (errno %ld).\n", PTR_ERR(clnt4));
+		clnt4 = NULL;
+	}
+
+	rpcb_local_clnt = clnt;
+	rpcb_local_clnt4 = clnt4;
+
+out:
+	mutex_unlock(&rpcb_create_local_mutex);
+	return result;
 }
 
 static struct rpc_clnt *rpcb_create(char *hostname, struct sockaddr *srvaddr,
@@ -209,22 +252,13 @@ static struct rpc_clnt *rpcb_create(char *hostname, struct sockaddr *srvaddr,
 	return rpc_create(&args);
 }
 
-static int rpcb_register_call(const u32 version, struct rpc_message *msg)
+static int rpcb_register_call(struct rpc_clnt *clnt, struct rpc_message *msg)
 {
-	struct sockaddr *addr = (struct sockaddr *)&rpcb_inaddr_loopback;
-	size_t addrlen = sizeof(rpcb_inaddr_loopback);
-	struct rpc_clnt *rpcb_clnt;
 	int result, error = 0;
 
 	msg->rpc_resp = &result;
 
-	rpcb_clnt = rpcb_create_local(addr, addrlen, version);
-	if (!IS_ERR(rpcb_clnt)) {
-		error = rpc_call_sync(rpcb_clnt, msg, 0);
-		rpc_shutdown_client(rpcb_clnt);
-	} else
-		error = PTR_ERR(rpcb_clnt);
-
+	error = rpc_call_sync(clnt, msg, RPC_TASK_SOFTCONN);
 	if (error < 0) {
 		dprintk("RPC:       failed to contact local rpcbind "
 				"server (errno %d).\n", -error);
@@ -279,6 +313,11 @@ int rpcb_register(u32 prog, u32 vers, int prot, unsigned short port)
 	struct rpc_message msg = {
 		.rpc_argp	= &map,
 	};
+	int error;
+
+	error = rpcb_create_local();
+	if (error)
+		return error;
 
 	dprintk("RPC:       %sregistering (%u, %u, %d, %u) with local "
 			"rpcbind\n", (port ? "" : "un"),
@@ -288,7 +327,7 @@ int rpcb_register(u32 prog, u32 vers, int prot, unsigned short port)
 	if (port)
 		msg.rpc_proc = &rpcb_procedures2[RPCBPROC_SET];
 
-	return rpcb_register_call(RPCBVERS_2, &msg);
+	return rpcb_register_call(rpcb_local_clnt, &msg);
 }
 
 /*
@@ -313,7 +352,7 @@ static int rpcb_register_inet4(const struct sockaddr *sap,
 	if (port)
 		msg->rpc_proc = &rpcb_procedures4[RPCBPROC_SET];
 
-	result = rpcb_register_call(RPCBVERS_4, msg);
+	result = rpcb_register_call(rpcb_local_clnt4, msg);
 	kfree(map->r_addr);
 	return result;
 }
@@ -340,7 +379,7 @@ static int rpcb_register_inet6(const struct sockaddr *sap,
 	if (port)
 		msg->rpc_proc = &rpcb_procedures4[RPCBPROC_SET];
 
-	result = rpcb_register_call(RPCBVERS_4, msg);
+	result = rpcb_register_call(rpcb_local_clnt4, msg);
 	kfree(map->r_addr);
 	return result;
 }
@@ -356,7 +395,7 @@ static int rpcb_unregister_all_protofamilies(struct rpc_message *msg)
 	map->r_addr = "";
 	msg->rpc_proc = &rpcb_procedures4[RPCBPROC_UNSET];
 
-	return rpcb_register_call(RPCBVERS_4, msg);
+	return rpcb_register_call(rpcb_local_clnt4, msg);
 }
 
 /**
@@ -414,6 +453,13 @@ int rpcb_v4_register(const u32 program, const u32 version,
 	struct rpc_message msg = {
 		.rpc_argp	= &map,
 	};
+	int error;
+
+	error = rpcb_create_local();
+	if (error)
+		return error;
+	if (rpcb_local_clnt4 == NULL)
+		return -EPROTONOSUPPORT;
 
 	if (address == NULL)
 		return rpcb_unregister_all_protofamilies(&msg);
@@ -491,7 +537,7 @@ static struct rpc_task *rpcb_call_async(struct rpc_clnt *rpcb_clnt, struct rpcbi
 		.rpc_message = &msg,
 		.callback_ops = &rpcb_getport_ops,
 		.callback_data = map,
-		.flags = RPC_TASK_ASYNC,
+		.flags = RPC_TASK_ASYNC | RPC_TASK_SOFTCONN,
 	};
 
 	return rpc_run_task(&task_setup_data);
@@ -1027,3 +1073,15 @@ static struct rpc_program rpcb_program = {
 	.version	= rpcb_version,
 	.stats		= &rpcb_stats,
 };
+
+/**
+ * cleanup_rpcb_clnt - remove xprtsock's sysctls, unregister
+ *
+ */
+void cleanup_rpcb_clnt(void)
+{
+	if (rpcb_local_clnt4)
+		rpc_shutdown_client(rpcb_local_clnt4);
+	if (rpcb_local_clnt)
+		rpc_shutdown_client(rpcb_local_clnt);
+}
diff --git a/net/sunrpc/sunrpc_syms.c b/net/sunrpc/sunrpc_syms.c
index 8cce92189019..f438347d817b 100644
--- a/net/sunrpc/sunrpc_syms.c
+++ b/net/sunrpc/sunrpc_syms.c
@@ -24,6 +24,8 @@
 
 extern struct cache_detail ip_map_cache, unix_gid_cache;
 
+extern void cleanup_rpcb_clnt(void);
+
 static int __init
 init_sunrpc(void)
 {
@@ -53,6 +55,7 @@ out:
 static void __exit
 cleanup_sunrpc(void)
 {
+	cleanup_rpcb_clnt();
 	rpcauth_remove_module();
 	cleanup_socket_xprt();
 	svc_cleanup_xprt_sock();
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index fd46d42afa89..469de292c23c 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -700,6 +700,10 @@ void xprt_connect(struct rpc_task *task)
 	}
 	if (!xprt_lock_write(xprt, task))
 		return;
+
+	if (test_and_clear_bit(XPRT_CLOSE_WAIT, &xprt->state))
+		xprt->ops->close(xprt);
+
 	if (xprt_connected(xprt))
 		xprt_release_write(xprt, task);
 	else {
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index 04732d09013e..3d739e5d15d8 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -2019,7 +2019,7 @@ static void xs_connect(struct rpc_task *task)
 	if (xprt_test_and_set_connecting(xprt))
 		return;
 
-	if (transport->sock != NULL) {
+	if (transport->sock != NULL && !RPC_IS_SOFTCONN(task)) {
 		dprintk("RPC:       xs_connect delayed xprt %p for %lu "
 				"seconds\n",
 				xprt, xprt->reestablish_timeout / HZ);