Merge branch 'devel' into nfs-for-next

19 files changed, 1187 insertions, 294 deletions

diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c
index e374050a911c..8392cb85bd54 100644
--- a/fs/lockd/clntproc.c
+++ b/fs/lockd/clntproc.c
@@ -302,7 +302,8 @@ nlmclnt_call(struct rpc_cred *cred, struct nlm_rqst *req, u32 proc)
 				/* We appear to be out of the grace period */
 				wake_up_all(&host->h_gracewait);
 			}
-			dprintk("lockd: server returns status %d\n", resp->status);
+			dprintk("lockd: server returns status %d\n",
+				ntohl(resp->status));
 			return 0;	/* Okay, call complete */
 		}
 
@@ -690,7 +691,8 @@ nlmclnt_unlock(struct nlm_rqst *req, struct file_lock *fl)
 		goto out;
 
 	if (resp->status != nlm_lck_denied_nolocks)
-		printk("lockd: unexpected unlock status: %d\n", resp->status);
+		printk("lockd: unexpected unlock status: %d\n",
+			ntohl(resp->status));
 	/* What to do now? I'm out of my depth... */
 	status = -ENOLCK;
 out:
@@ -843,6 +845,7 @@ nlm_stat_to_errno(__be32 status)
 		return -ENOLCK;
 #endif
 	}
-	printk(KERN_NOTICE "lockd: unexpected server status %d\n", status);
+	printk(KERN_NOTICE "lockd: unexpected server status %d\n",
+		 ntohl(status));
 	return -ENOLCK;
 }
diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c
index d4d1954e9bb9..74780f9f852c 100644
--- a/fs/nfs/callback_proc.c
+++ b/fs/nfs/callback_proc.c
@@ -111,6 +111,7 @@ int nfs4_validate_delegation_stateid(struct nfs_delegation *delegation, const nf
 static u32 initiate_file_draining(struct nfs_client *clp,
 				  struct cb_layoutrecallargs *args)
 {
+	struct nfs_server *server;
 	struct pnfs_layout_hdr *lo;
 	struct inode *ino;
 	bool found = false;
@@ -118,21 +119,28 @@ static u32 initiate_file_draining(struct nfs_client *clp,
 	LIST_HEAD(free_me_list);
 
 	spin_lock(&clp->cl_lock);
-	list_for_each_entry(lo, &clp->cl_layouts, plh_layouts) {
-		if (nfs_compare_fh(&args->cbl_fh,
-				   &NFS_I(lo->plh_inode)->fh))
-			continue;
-		ino = igrab(lo->plh_inode);
-		if (!ino)
-			continue;
-		found = true;
-		/* Without this, layout can be freed as soon
-		 * as we release cl_lock.
-		 */
-		get_layout_hdr(lo);
-		break;
+	rcu_read_lock();
+	list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) {
+		list_for_each_entry(lo, &server->layouts, plh_layouts) {
+			if (nfs_compare_fh(&args->cbl_fh,
+					   &NFS_I(lo->plh_inode)->fh))
+				continue;
+			ino = igrab(lo->plh_inode);
+			if (!ino)
+				continue;
+			found = true;
+			/* Without this, layout can be freed as soon
+			 * as we release cl_lock.
+			 */
+			get_layout_hdr(lo);
+			break;
+		}
+		if (found)
+			break;
 	}
+	rcu_read_unlock();
 	spin_unlock(&clp->cl_lock);
+
 	if (!found)
 		return NFS4ERR_NOMATCHING_LAYOUT;
 
@@ -154,6 +162,7 @@ static u32 initiate_file_draining(struct nfs_client *clp,
 static u32 initiate_bulk_draining(struct nfs_client *clp,
 				  struct cb_layoutrecallargs *args)
 {
+	struct nfs_server *server;
 	struct pnfs_layout_hdr *lo;
 	struct inode *ino;
 	u32 rv = NFS4ERR_NOMATCHING_LAYOUT;
@@ -167,18 +176,24 @@ static u32 initiate_bulk_draining(struct nfs_client *clp,
 	};
 
 	spin_lock(&clp->cl_lock);
-	list_for_each_entry(lo, &clp->cl_layouts, plh_layouts) {
+	rcu_read_lock();
+	list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) {
 		if ((args->cbl_recall_type == RETURN_FSID) &&
-		    memcmp(&NFS_SERVER(lo->plh_inode)->fsid,
-			   &args->cbl_fsid, sizeof(struct nfs_fsid)))
-			continue;
-		if (!igrab(lo->plh_inode))
+		    memcmp(&server->fsid, &args->cbl_fsid,
+			   sizeof(struct nfs_fsid)))
 			continue;
-		get_layout_hdr(lo);
-		BUG_ON(!list_empty(&lo->plh_bulk_recall));
-		list_add(&lo->plh_bulk_recall, &recall_list);
+
+		list_for_each_entry(lo, &server->layouts, plh_layouts) {
+			if (!igrab(lo->plh_inode))
+				continue;
+			get_layout_hdr(lo);
+			BUG_ON(!list_empty(&lo->plh_bulk_recall));
+			list_add(&lo->plh_bulk_recall, &recall_list);
+		}
 	}
+	rcu_read_unlock();
 	spin_unlock(&clp->cl_lock);
+
 	list_for_each_entry_safe(lo, tmp,
 				 &recall_list, plh_bulk_recall) {
 		ino = lo->plh_inode;
diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index b3dc2b88b65b..5452ada59461 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -188,9 +188,6 @@ static struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_
 	cred = rpc_lookup_machine_cred();
 	if (!IS_ERR(cred))
 		clp->cl_machine_cred = cred;
-#if defined(CONFIG_NFS_V4_1)
-	INIT_LIST_HEAD(&clp->cl_layouts);
-#endif
 	nfs_fscache_get_client_cookie(clp);
 
 	return clp;
@@ -293,6 +290,7 @@ static void nfs_free_client(struct nfs_client *clp)
 	nfs4_deviceid_purge_client(clp);
 
 	kfree(clp->cl_hostname);
+	kfree(clp->server_scope);
 	kfree(clp);
 
 	dprintk("<-- nfs_free_client()\n");
@@ -1062,6 +1060,7 @@ static struct nfs_server *nfs_alloc_server(void)
 	INIT_LIST_HEAD(&server->client_link);
 	INIT_LIST_HEAD(&server->master_link);
 	INIT_LIST_HEAD(&server->delegations);
+	INIT_LIST_HEAD(&server->layouts);
 
 	atomic_set(&server->active, 0);
 
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index 2a55347a2daa..31e8b50011af 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -277,6 +277,9 @@ extern void nfs_sb_deactive(struct super_block *sb);
 extern char *nfs_path(char **p, struct dentry *dentry,
 		      char *buffer, ssize_t buflen);
 extern struct vfsmount *nfs_d_automount(struct path *path);
+#ifdef CONFIG_NFS_V4
+rpc_authflavor_t nfs_find_best_sec(struct nfs4_secinfo_flavors *);
+#endif
 
 /* getroot.c */
 extern struct dentry *nfs_get_root(struct super_block *, struct nfs_fh *,
@@ -293,7 +296,13 @@ extern int nfs_initiate_read(struct nfs_read_data *data, struct rpc_clnt *clnt,
 			     const struct rpc_call_ops *call_ops);
 extern void nfs_read_prepare(struct rpc_task *task, void *calldata);
 
+struct nfs_pageio_descriptor;
+extern void nfs_pageio_init_read_mds(struct nfs_pageio_descriptor *pgio,
+		struct inode *inode);
+
 /* write.c */
+extern void nfs_pageio_init_write_mds(struct nfs_pageio_descriptor *pgio,
+				  struct inode *inode, int ioflags);
 extern void nfs_commit_free(struct nfs_write_data *p);
 extern int nfs_initiate_write(struct nfs_write_data *data,
 			      struct rpc_clnt *clnt,
diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c
index 1f063bacd285..8102391bb374 100644
--- a/fs/nfs/namespace.c
+++ b/fs/nfs/namespace.c
@@ -119,7 +119,7 @@ Elong:
 }
 
 #ifdef CONFIG_NFS_V4
-static rpc_authflavor_t nfs_find_best_sec(struct nfs4_secinfo_flavors *flavors)
+rpc_authflavor_t nfs_find_best_sec(struct nfs4_secinfo_flavors *flavors)
 {
 	struct gss_api_mech *mech;
 	struct xdr_netobj oid;
diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index c4a69833dd0d..c30aed2c70f4 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -48,6 +48,7 @@ enum nfs4_client_state {
 	NFS4CLNT_SESSION_RESET,
 	NFS4CLNT_RECALL_SLOT,
 	NFS4CLNT_LEASE_CONFIRM,
+	NFS4CLNT_SERVER_SCOPE_MISMATCH,
 };
 
 enum nfs4_session_state {
@@ -66,6 +67,8 @@ struct nfs4_minor_version_ops {
 			int cache_reply);
 	int	(*validate_stateid)(struct nfs_delegation *,
 			const nfs4_stateid *);
+	int	(*find_root_sec)(struct nfs_server *, struct nfs_fh *,
+			struct nfs_fsinfo *);
 	const struct nfs4_state_recovery_ops *reboot_recovery_ops;
 	const struct nfs4_state_recovery_ops *nograce_recovery_ops;
 	const struct nfs4_state_maintenance_ops *state_renewal_ops;
@@ -349,6 +352,8 @@ extern void nfs4_schedule_state_manager(struct nfs_client *);
 extern void nfs4_schedule_stateid_recovery(const struct nfs_server *, struct nfs4_state *);
 extern void nfs41_handle_sequence_flag_errors(struct nfs_client *clp, u32 flags);
 extern void nfs41_handle_recall_slot(struct nfs_client *clp);
+extern void nfs41_handle_server_scope(struct nfs_client *,
+				      struct server_scope **);
 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, pid_t);
diff --git a/fs/nfs/nfs4filelayout.c b/fs/nfs/nfs4filelayout.c
index 0bafcc91c27f..d26c2297a466 100644
--- a/fs/nfs/nfs4filelayout.c
+++ b/fs/nfs/nfs4filelayout.c
@@ -334,6 +334,9 @@ filelayout_read_pagelist(struct nfs_read_data *data)
 		__func__, data->inode->i_ino,
 		data->args.pgbase, (size_t)data->args.count, offset);
 
+	if (test_bit(NFS_DEVICEID_INVALID, &FILELAYOUT_DEVID_NODE(lseg)->flags))
+		return PNFS_NOT_ATTEMPTED;
+
 	/* Retrieve the correct rpc_client for the byte range */
 	j = nfs4_fl_calc_j_index(lseg, offset);
 	idx = nfs4_fl_calc_ds_index(lseg, j);
@@ -344,8 +347,7 @@ filelayout_read_pagelist(struct nfs_read_data *data)
 		set_bit(lo_fail_bit(IOMODE_READ), &lseg->pls_layout->plh_flags);
 		return PNFS_NOT_ATTEMPTED;
 	}
-	dprintk("%s USE DS:ip %x %hu\n", __func__,
-		ntohl(ds->ds_ip_addr), ntohs(ds->ds_port));
+	dprintk("%s USE DS: %s\n", __func__, ds->ds_remotestr);
 
 	/* No multipath support. Use first DS */
 	data->ds_clp = ds->ds_clp;
@@ -374,6 +376,9 @@ filelayout_write_pagelist(struct nfs_write_data *data, int sync)
 	struct nfs_fh *fh;
 	int status;
 
+	if (test_bit(NFS_DEVICEID_INVALID, &FILELAYOUT_DEVID_NODE(lseg)->flags))
+		return PNFS_NOT_ATTEMPTED;
+
 	/* Retrieve the correct rpc_client for the byte range */
 	j = nfs4_fl_calc_j_index(lseg, offset);
 	idx = nfs4_fl_calc_ds_index(lseg, j);
@@ -384,9 +389,9 @@ filelayout_write_pagelist(struct nfs_write_data *data, int sync)
 		set_bit(lo_fail_bit(IOMODE_READ), &lseg->pls_layout->plh_flags);
 		return PNFS_NOT_ATTEMPTED;
 	}
-	dprintk("%s ino %lu sync %d req %Zu@%llu DS:%x:%hu\n", __func__,
+	dprintk("%s ino %lu sync %d req %Zu@%llu DS: %s\n", __func__,
 		data->inode->i_ino, sync, (size_t) data->args.count, offset,
-		ntohl(ds->ds_ip_addr), ntohs(ds->ds_port));
+		ds->ds_remotestr);
 
 	data->write_done_cb = filelayout_write_done_cb;
 	data->ds_clp = ds->ds_clp;
@@ -429,6 +434,14 @@ filelayout_check_layout(struct pnfs_layout_hdr *lo,
 
 	dprintk("--> %s\n", __func__);
 
+	/* FIXME: remove this check when layout segment support is added */
+	if (lgr->range.offset != 0 ||
+	    lgr->range.length != NFS4_MAX_UINT64) {
+		dprintk("%s Only whole file layouts supported. Use MDS i/o\n",
+			__func__);
+		goto out;
+	}
+
 	if (fl->pattern_offset > lgr->range.offset) {
 		dprintk("%s pattern_offset %lld too large\n",
 				__func__, fl->pattern_offset);
@@ -450,6 +463,10 @@ filelayout_check_layout(struct pnfs_layout_hdr *lo,
 			goto out;
 	} else
 		dsaddr = container_of(d, struct nfs4_file_layout_dsaddr, id_node);
+	/* Found deviceid is being reaped */
+	if (test_bit(NFS_DEVICEID_INVALID, &dsaddr->id_node.flags))
+			goto out_put;
+
 	fl->dsaddr = dsaddr;
 
 	if (fl->first_stripe_index < 0 ||
@@ -660,7 +677,7 @@ filelayout_alloc_lseg(struct pnfs_layout_hdr *layoutid,
  * return true  : coalesce page
  * return false : don't coalesce page
  */
-bool
+static bool
 filelayout_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev,
 		   struct nfs_page *req)
 {
@@ -671,8 +688,6 @@ filelayout_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev,
 	    !nfs_generic_pg_test(pgio, prev, req))
 		return false;
 
-	if (!pgio->pg_lseg)
-		return 1;
 	p_stripe = (u64)prev->wb_index << PAGE_CACHE_SHIFT;
 	r_stripe = (u64)req->wb_index << PAGE_CACHE_SHIFT;
 	stripe_unit = FILELAYOUT_LSEG(pgio->pg_lseg)->stripe_unit;
@@ -683,6 +698,53 @@ filelayout_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev,
 	return (p_stripe == r_stripe);
 }
 
+void
+filelayout_pg_init_read(struct nfs_pageio_descriptor *pgio,
+			struct nfs_page *req)
+{
+	BUG_ON(pgio->pg_lseg != NULL);
+
+	pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
+					   req->wb_context,
+					   0,
+					   NFS4_MAX_UINT64,
+					   IOMODE_READ,
+					   GFP_KERNEL);
+	/* If no lseg, fall back to read through mds */
+	if (pgio->pg_lseg == NULL)
+		nfs_pageio_init_read_mds(pgio, pgio->pg_inode);
+}
+
+void
+filelayout_pg_init_write(struct nfs_pageio_descriptor *pgio,
+			 struct nfs_page *req)
+{
+	BUG_ON(pgio->pg_lseg != NULL);
+
+	pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
+					   req->wb_context,
+					   0,
+					   NFS4_MAX_UINT64,
+					   IOMODE_RW,
+					   GFP_NOFS);
+	/* If no lseg, fall back to write through mds */
+	if (pgio->pg_lseg == NULL)
+		nfs_pageio_init_write_mds(pgio, pgio->pg_inode,
+					  pgio->pg_ioflags);
+}
+
+static const struct nfs_pageio_ops filelayout_pg_read_ops = {
+	.pg_init = filelayout_pg_init_read,
+	.pg_test = filelayout_pg_test,
+	.pg_doio = nfs_generic_pg_readpages,
+};
+
+static const struct nfs_pageio_ops filelayout_pg_write_ops = {
+	.pg_init = filelayout_pg_init_write,
+	.pg_test = filelayout_pg_test,
+	.pg_doio = nfs_generic_pg_writepages,
+};
+
 static bool filelayout_mark_pnfs_commit(struct pnfs_layout_segment *lseg)
 {
 	return !FILELAYOUT_LSEG(lseg)->commit_through_mds;
@@ -880,7 +942,8 @@ static struct pnfs_layoutdriver_type filelayout_type = {
 	.owner			= THIS_MODULE,
 	.alloc_lseg		= filelayout_alloc_lseg,
 	.free_lseg		= filelayout_free_lseg,
-	.pg_test		= filelayout_pg_test,
+	.pg_read_ops		= &filelayout_pg_read_ops,
+	.pg_write_ops		= &filelayout_pg_write_ops,
 	.mark_pnfs_commit	= filelayout_mark_pnfs_commit,
 	.choose_commit_list	= filelayout_choose_commit_list,
 	.commit_pagelist	= filelayout_commit_pagelist,
diff --git a/fs/nfs/nfs4filelayout.h b/fs/nfs/nfs4filelayout.h
index cebe01e3795e..2e42284253fa 100644
--- a/fs/nfs/nfs4filelayout.h
+++ b/fs/nfs/nfs4filelayout.h
@@ -47,10 +47,17 @@ enum stripetype4 {
 };
 
 /* Individual ip address */
+struct nfs4_pnfs_ds_addr {
+	struct sockaddr_storage	da_addr;
+	size_t			da_addrlen;
+	struct list_head	da_node;  /* nfs4_pnfs_dev_hlist dev_dslist */
+	char			*da_remotestr;	/* human readable addr+port */
+};
+
 struct nfs4_pnfs_ds {
 	struct list_head	ds_node;  /* nfs4_pnfs_dev_hlist dev_dslist */
-	u32			ds_ip_addr;
-	u32			ds_port;
+	char			*ds_remotestr;	/* comma sep list of addrs */
+	struct list_head	ds_addrs;
 	struct nfs_client	*ds_clp;
 	atomic_t		ds_count;
 };
@@ -89,6 +96,12 @@ FILELAYOUT_LSEG(struct pnfs_layout_segment *lseg)
 			    generic_hdr);
 }
 
+static inline struct nfs4_deviceid_node *
+FILELAYOUT_DEVID_NODE(struct pnfs_layout_segment *lseg)
+{
+	return &FILELAYOUT_LSEG(lseg)->dsaddr->id_node;
+}
+
 extern struct nfs_fh *
 nfs4_fl_select_ds_fh(struct pnfs_layout_segment *lseg, u32 j);
 
diff --git a/fs/nfs/nfs4filelayoutdev.c b/fs/nfs/nfs4filelayoutdev.c
index 3b7bf1377264..ed388aae9689 100644
--- a/fs/nfs/nfs4filelayoutdev.c
+++ b/fs/nfs/nfs4filelayoutdev.c
@@ -56,54 +56,139 @@ print_ds(struct nfs4_pnfs_ds *ds)
 		printk("%s NULL device\n", __func__);
 		return;
 	}
-	printk("        ip_addr %x port %hu\n"
+	printk("        ds %s\n"
 		"        ref count %d\n"
 		"        client %p\n"
 		"        cl_exchange_flags %x\n",
-		ntohl(ds->ds_ip_addr), ntohs(ds->ds_port),
+		ds->ds_remotestr,
 		atomic_read(&ds->ds_count), ds->ds_clp,
 		ds->ds_clp ? ds->ds_clp->cl_exchange_flags : 0);
 }
 
-/* nfs4_ds_cache_lock is held */
-static struct nfs4_pnfs_ds *
-_data_server_lookup_locked(u32 ip_addr, u32 port)
+static bool
+same_sockaddr(struct sockaddr *addr1, struct sockaddr *addr2)
 {
-	struct nfs4_pnfs_ds *ds;
+	struct sockaddr_in *a, *b;
+	struct sockaddr_in6 *a6, *b6;
+
+	if (addr1->sa_family != addr2->sa_family)
+		return false;
+
+	switch (addr1->sa_family) {
+	case AF_INET:
+		a = (struct sockaddr_in *)addr1;
+		b = (struct sockaddr_in *)addr2;
+
+		if (a->sin_addr.s_addr == b->sin_addr.s_addr &&
+		    a->sin_port == b->sin_port)
+			return true;
+		break;
+
+	case AF_INET6:
+		a6 = (struct sockaddr_in6 *)addr1;
+		b6 = (struct sockaddr_in6 *)addr2;
+
+		/* LINKLOCAL addresses must have matching scope_id */
+		if (ipv6_addr_scope(&a6->sin6_addr) ==
+		    IPV6_ADDR_SCOPE_LINKLOCAL &&
+		    a6->sin6_scope_id != b6->sin6_scope_id)
+			return false;
+
+		if (ipv6_addr_equal(&a6->sin6_addr, &b6->sin6_addr) &&
+		    a6->sin6_port == b6->sin6_port)
+			return true;
+		break;
+
+	default:
+		dprintk("%s: unhandled address family: %u\n",
+			__func__, addr1->sa_family);
+		return false;
+	}
 
-	dprintk("_data_server_lookup: ip_addr=%x port=%hu\n",
-			ntohl(ip_addr), ntohs(port));
+	return false;
+}
 
-	list_for_each_entry(ds, &nfs4_data_server_cache, ds_node) {
-		if (ds->ds_ip_addr == ip_addr &&
-		    ds->ds_port == port) {
-			return ds;
+/*
+ * Lookup DS by addresses.  The first matching address returns true.
+ * nfs4_ds_cache_lock is held
+ */
+static struct nfs4_pnfs_ds *
+_data_server_lookup_locked(struct list_head *dsaddrs)
+{
+	struct nfs4_pnfs_ds *ds;
+	struct nfs4_pnfs_ds_addr *da1, *da2;
+
+	list_for_each_entry(da1, dsaddrs, da_node) {
+		list_for_each_entry(ds, &nfs4_data_server_cache, ds_node) {
+			list_for_each_entry(da2, &ds->ds_addrs, da_node) {
+				if (same_sockaddr(
+					(struct sockaddr *)&da1->da_addr,
+					(struct sockaddr *)&da2->da_addr))
+					return ds;
+			}
 		}
 	}
 	return NULL;
 }
 
 /*
+ * Compare two lists of addresses.
+ */
+static bool
+_data_server_match_all_addrs_locked(struct list_head *dsaddrs1,
+				    struct list_head *dsaddrs2)
+{
+	struct nfs4_pnfs_ds_addr *da1, *da2;
+	size_t count1 = 0,
+	       count2 = 0;
+
+	list_for_each_entry(da1, dsaddrs1, da_node)
+		count1++;
+
+	list_for_each_entry(da2, dsaddrs2, da_node) {
+		bool found = false;
+		count2++;
+		list_for_each_entry(da1, dsaddrs1, da_node) {
+			if (same_sockaddr((struct sockaddr *)&da1->da_addr,
+				(struct sockaddr *)&da2->da_addr)) {
+				found = true;
+				break;
+			}
+		}
+		if (!found)
+			return false;
+	}
+
+	return (count1 == count2);
+}
+
+/*
  * Create an rpc connection to the nfs4_pnfs_ds data server
- * Currently only support IPv4
+ * Currently only supports IPv4 and IPv6 addresses
  */
 static int
 nfs4_ds_connect(struct nfs_server *mds_srv, struct nfs4_pnfs_ds *ds)
 {
-	struct nfs_client *clp;
-	struct sockaddr_in sin;
+	struct nfs_client *clp = ERR_PTR(-EIO);
+	struct nfs4_pnfs_ds_addr *da;
 	int status = 0;
 
-	dprintk("--> %s ip:port %x:%hu au_flavor %d\n", __func__,
-		ntohl(ds->ds_ip_addr), ntohs(ds->ds_port),
+	dprintk("--> %s DS %s au_flavor %d\n", __func__, ds->ds_remotestr,
 		mds_srv->nfs_client->cl_rpcclient->cl_auth->au_flavor);
 
-	sin.sin_family = AF_INET;
-	sin.sin_addr.s_addr = ds->ds_ip_addr;
-	sin.sin_port = ds->ds_port;
+	BUG_ON(list_empty(&ds->ds_addrs));
+
+	list_for_each_entry(da, &ds->ds_addrs, da_node) {
+		dprintk("%s: DS %s: trying address %s\n",
+			__func__, ds->ds_remotestr, da->da_remotestr);
+
+		clp = nfs4_set_ds_client(mds_srv->nfs_client,
+				 (struct sockaddr *)&da->da_addr,
+				 da->da_addrlen, IPPROTO_TCP);
+		if (!IS_ERR(clp))
+			break;
+	}
 
-	clp = nfs4_set_ds_client(mds_srv->nfs_client, (struct sockaddr *)&sin,
-				 sizeof(sin), IPPROTO_TCP);
 	if (IS_ERR(clp)) {
 		status = PTR_ERR(clp);
 		goto out;
@@ -115,8 +200,8 @@ nfs4_ds_connect(struct nfs_server *mds_srv, struct nfs4_pnfs_ds *ds)
 			goto out_put;
 		}
 		ds->ds_clp = clp;
-		dprintk("%s [existing] ip=%x, port=%hu\n", __func__,
-			ntohl(ds->ds_ip_addr), ntohs(ds->ds_port));
+		dprintk("%s [existing] server=%s\n", __func__,
+			ds->ds_remotestr);
 		goto out;
 	}
 
@@ -135,8 +220,7 @@ nfs4_ds_connect(struct nfs_server *mds_srv, struct nfs4_pnfs_ds *ds)
 		goto out_put;
 
 	ds->ds_clp = clp;
-	dprintk("%s [new] ip=%x, port=%hu\n", __func__, ntohl(ds->ds_ip_addr),
-		ntohs(ds->ds_port));
+	dprintk("%s [new] addr: %s\n", __func__, ds->ds_remotestr);
 out:
 	return status;
 out_put:
@@ -147,12 +231,25 @@ out_put:
 static void
 destroy_ds(struct nfs4_pnfs_ds *ds)
 {
+	struct nfs4_pnfs_ds_addr *da;
+
 	dprintk("--> %s\n", __func__);
 	ifdebug(FACILITY)
 		print_ds(ds);
 
 	if (ds->ds_clp)
 		nfs_put_client(ds->ds_clp);
+
+	while (!list_empty(&ds->ds_addrs)) {
+		da = list_first_entry(&ds->ds_addrs,
+				      struct nfs4_pnfs_ds_addr,
+				      da_node);
+		list_del_init(&da->da_node);
+		kfree(da->da_remotestr);
+		kfree(da);
+	}
+
+	kfree(ds->ds_remotestr);
 	kfree(ds);
 }
 
@@ -179,31 +276,96 @@ nfs4_fl_free_deviceid(struct nfs4_file_layout_dsaddr *dsaddr)
 	kfree(dsaddr);
 }
 
+/*
+ * Create a string with a human readable address and port to avoid
+ * complicated setup around many dprinks.
+ */
+static char *
+nfs4_pnfs_remotestr(struct list_head *dsaddrs, gfp_t gfp_flags)
+{
+	struct nfs4_pnfs_ds_addr *da;
+	char *remotestr;
+	size_t len;
+	char *p;
+
+	len = 3;        /* '{', '}' and eol */
+	list_for_each_entry(da, dsaddrs, da_node) {
+		len += strlen(da->da_remotestr) + 1;    /* string plus comma */
+	}
+
+	remotestr = kzalloc(len, gfp_flags);
+	if (!remotestr)
+		return NULL;
+
+	p = remotestr;
+	*(p++) = '{';
+	len--;
+	list_for_each_entry(da, dsaddrs, da_node) {
+		size_t ll = strlen(da->da_remotestr);
+
+		if (ll > len)
+			goto out_err;
+
+		memcpy(p, da->da_remotestr, ll);
+		p += ll;
+		len -= ll;
+
+		if (len < 1)
+			goto out_err;
+		(*p++) = ',';
+		len--;
+	}
+	if (len < 2)
+		goto out_err;
+	*(p++) = '}';
+	*p = '\0';
+	return remotestr;
+out_err:
+	kfree(remotestr);
+	return NULL;
+}
+
 static struct nfs4_pnfs_ds *
-nfs4_pnfs_ds_add(struct inode *inode, u32 ip_addr, u32 port, gfp_t gfp_flags)
+nfs4_pnfs_ds_add(struct list_head *dsaddrs, gfp_t gfp_flags)
 {
-	struct nfs4_pnfs_ds *tmp_ds, *ds;
+	struct nfs4_pnfs_ds *tmp_ds, *ds = NULL;
+	char *remotestr;
 
-	ds = kzalloc(sizeof(*tmp_ds), gfp_flags);
+	if (list_empty(dsaddrs)) {
+		dprintk("%s: no addresses defined\n", __func__);
+		goto out;
+	}
+
+	ds = kzalloc(sizeof(*ds), gfp_flags);
 	if (!ds)
 		goto out;
 
+	/* this is only used for debugging, so it's ok if its NULL */
+	remotestr = nfs4_pnfs_remotestr(dsaddrs, gfp_flags);
+
 	spin_lock(&nfs4_ds_cache_lock);
-	tmp_ds = _data_server_lookup_locked(ip_addr, port);
+	tmp_ds = _data_server_lookup_locked(dsaddrs);
 	if (tmp_ds == NULL) {
-		ds->ds_ip_addr = ip_addr;
-		ds->ds_port = port;
+		INIT_LIST_HEAD(&ds->ds_addrs);
+		list_splice_init(dsaddrs, &ds->ds_addrs);
+		ds->ds_remotestr = remotestr;
 		atomic_set(&ds->ds_count, 1);
 		INIT_LIST_HEAD(&ds->ds_node);
 		ds->ds_clp = NULL;
 		list_add(&ds->ds_node, &nfs4_data_server_cache);
-		dprintk("%s add new data server ip 0x%x\n", __func__,
-			ds->ds_ip_addr);
+		dprintk("%s add new data server %s\n", __func__,
+			ds->ds_remotestr);
 	} else {
+		if (!_data_server_match_all_addrs_locked(&tmp_ds->ds_addrs,
+							 dsaddrs)) {
+			dprintk("%s:  multipath address mismatch: %s != %s",
+				__func__, tmp_ds->ds_remotestr, remotestr);
+		}
+		kfree(remotestr);
 		kfree(ds);
 		atomic_inc(&tmp_ds->ds_count);
-		dprintk("%s data server found ip 0x%x, inc'ed ds_count to %d\n",
-			__func__, tmp_ds->ds_ip_addr,
+		dprintk("%s data server %s found, inc'ed ds_count to %d\n",
+			__func__, tmp_ds->ds_remotestr,
 			atomic_read(&tmp_ds->ds_count));
 		ds = tmp_ds;
 	}
@@ -213,18 +375,22 @@ out:
 }
 
 /*
- * Currently only support ipv4, and one multi-path address.
+ * Currently only supports ipv4, ipv6 and one multi-path address.
  */
-static struct nfs4_pnfs_ds *
-decode_and_add_ds(struct xdr_stream *streamp, struct inode *inode, gfp_t gfp_flags)
+static struct nfs4_pnfs_ds_addr *
+decode_ds_addr(struct xdr_stream *streamp, gfp_t gfp_flags)
 {
-	struct nfs4_pnfs_ds *ds = NULL;
-	char *buf;
-	const char *ipend, *pstr;
-	u32 ip_addr, port;
-	int nlen, rlen, i;
+	struct nfs4_pnfs_ds_addr *da = NULL;
+	char *buf, *portstr;
+	u32 port;
+	int nlen, rlen;
 	int tmp[2];
 	__be32 *p;
+	char *netid, *match_netid;
+	size_t len, match_netid_len;
+	char *startsep = "";
+	char *endsep = "";
+
 
 	/* r_netid */
 	p = xdr_inline_decode(streamp, 4);
@@ -236,64 +402,123 @@ decode_and_add_ds(struct xdr_stream *streamp, struct inode *inode, gfp_t gfp_fla
 	if (unlikely(!p))
 		goto out_err;
 
-	/* Check that netid is "tcp" */
-	if (nlen != 3 ||  memcmp((char *)p, "tcp", 3)) {
-		dprintk("%s: ERROR: non ipv4 TCP r_netid\n", __func__);
+	netid = kmalloc(nlen+1, gfp_flags);
+	if (unlikely(!netid))
 		goto out_err;
-	}
 
-	/* r_addr */
+	netid[nlen] = '\0';
+	memcpy(netid, p, nlen);
+
+	/* r_addr: ip/ip6addr with port in dec octets - see RFC 5665 */
 	p = xdr_inline_decode(streamp, 4);
 	if (unlikely(!p))
-		goto out_err;
+		goto out_free_netid;
 	rlen = be32_to_cpup(p);
 
 	p = xdr_inline_decode(streamp, rlen);
 	if (unlikely(!p))
-		goto out_err;
+		goto out_free_netid;
 
-	/* ipv6 length plus port is legal */
-	if (rlen > INET6_ADDRSTRLEN + 8) {
+	/* port is ".ABC.DEF", 8 chars max */
+	if (rlen > INET6_ADDRSTRLEN + IPV6_SCOPE_ID_LEN + 8) {
 		dprintk("%s: Invalid address, length %d\n", __func__,
 			rlen);
-		goto out_err;
+		goto out_free_netid;
 	}
 	buf = kmalloc(rlen + 1, gfp_flags);
 	if (!buf) {
 		dprintk("%s: Not enough memory\n", __func__);
-		goto out_err;
+		goto out_free_netid;
 	}
 	buf[rlen] = '\0';
 	memcpy(buf, p, rlen);
 
-	/* replace the port dots with dashes for the in4_pton() delimiter*/
-	for (i = 0; i < 2; i++) {
-		char *res = strrchr(buf, '.');
-		if (!res) {
-			dprintk("%s: Failed finding expected dots in port\n",
-				__func__);
-			goto out_free;
-		}
-		*res = '-';
+	/* replace port '.' with '-' */
+	portstr = strrchr(buf, '.');
+	if (!portstr) {
+		dprintk("%s: Failed finding expected dot in port\n",
+			__func__);
+		goto out_free_buf;
+	}
+	*portstr = '-';
+
+	/* find '.' between address and port */
+	portstr = strrchr(buf, '.');
+	if (!portstr) {
+		dprintk("%s: Failed finding expected dot between address and "
+			"port\n", __func__);
+		goto out_free_buf;
 	}
+	*portstr = '\0';
 
-	/* Currently only support ipv4 address */
-	if (in4_pton(buf, rlen, (u8 *)&ip_addr, '-', &ipend) == 0) {
-		dprintk("%s: Only ipv4 addresses supported\n", __func__);
-		goto out_free;
+	da = kzalloc(sizeof(*da), gfp_flags);
+	if (unlikely(!da))
+		goto out_free_buf;
+
+	INIT_LIST_HEAD(&da->da_node);
+
+	if (!rpc_pton(buf, portstr-buf, (struct sockaddr *)&da->da_addr,
+		      sizeof(da->da_addr))) {
+		dprintk("%s: error parsing address %s\n", __func__, buf);
+		goto out_free_da;
 	}
 
-	/* port */
-	pstr = ipend;
-	sscanf(pstr, "-%d-%d", &tmp[0], &tmp[1]);
+	portstr++;
+	sscanf(portstr, "%d-%d", &tmp[0], &tmp[1]);
 	port = htons((tmp[0] << 8) | (tmp[1]));
 
-	ds = nfs4_pnfs_ds_add(inode, ip_addr, port, gfp_flags);
-	dprintk("%s: Decoded address and port %s\n", __func__, buf);
-out_free:
+	switch (da->da_addr.ss_family) {
+	case AF_INET:
+		((struct sockaddr_in *)&da->da_addr)->sin_port = port;
+		da->da_addrlen = sizeof(struct sockaddr_in);
+		match_netid = "tcp";
+		match_netid_len = 3;
+		break;
+
+	case AF_INET6:
+		((struct sockaddr_in6 *)&da->da_addr)->sin6_port = port;
+		da->da_addrlen = sizeof(struct sockaddr_in6);
+		match_netid = "tcp6";
+		match_netid_len = 4;
+		startsep = "[";
+		endsep = "]";
+		break;
+
+	default:
+		dprintk("%s: unsupported address family: %u\n",
+			__func__, da->da_addr.ss_family);
+		goto out_free_da;
+	}
+
+	if (nlen != match_netid_len || strncmp(netid, match_netid, nlen)) {
+		dprintk("%s: ERROR: r_netid \"%s\" != \"%s\"\n",
+			__func__, netid, match_netid);
+		goto out_free_da;
+	}
+
+	/* save human readable address */
+	len = strlen(startsep) + strlen(buf) + strlen(endsep) + 7;
+	da->da_remotestr = kzalloc(len, gfp_flags);
+
+	/* NULL is ok, only used for dprintk */
+	if (da->da_remotestr)
+		snprintf(da->da_remotestr, len, "%s%s%s:%u", startsep,
+			 buf, endsep, ntohs(port));
+
+	dprintk("%s: Parsed DS addr %s\n", __func__, da->da_remotestr);
 	kfree(buf);
+	kfree(netid);
+	return da;
+
+out_free_da:
+	kfree(da);
+out_free_buf:
+	dprintk("%s: Error parsing DS addr: %s\n", __func__, buf);
+	kfree(buf);
+out_free_netid:
+	kfree(netid);
 out_err:
-	return ds;
+	return NULL;
 }
 
 /* Decode opaque device data and return the result */
@@ -310,6 +535,8 @@ decode_device(struct inode *ino, struct pnfs_device *pdev, gfp_t gfp_flags)
 	struct xdr_stream stream;
 	struct xdr_buf buf;
 	struct page *scratch;
+	struct list_head dsaddrs;
+	struct nfs4_pnfs_ds_addr *da;
 
 	/* set up xdr stream */
 	scratch = alloc_page(gfp_flags);
@@ -386,6 +613,8 @@ decode_device(struct inode *ino, struct pnfs_device *pdev, gfp_t gfp_flags)
 				NFS_SERVER(ino)->nfs_client,
 				&pdev->dev_id);
 
+	INIT_LIST_HEAD(&dsaddrs);
+
 	for (i = 0; i < dsaddr->ds_num; i++) {
 		int j;
 		u32 mp_count;
@@ -395,48 +624,43 @@ decode_device(struct inode *ino, struct pnfs_device *pdev, gfp_t gfp_flags)
 			goto out_err_free_deviceid;
 
 		mp_count = be32_to_cpup(p); /* multipath count */
-		if (mp_count > 1) {
-			printk(KERN_WARNING
-			       "%s: Multipath count %d not supported, "
-			       "skipping all greater than 1\n", __func__,
-				mp_count);
-		}
 		for (j = 0; j < mp_count; j++) {
-			if (j == 0) {
-				dsaddr->ds_list[i] = decode_and_add_ds(&stream,
-					ino, gfp_flags);
-				if (dsaddr->ds_list[i] == NULL)
-					goto out_err_free_deviceid;
-			} else {
-				u32 len;
-				/* skip extra multipath */
-
-				/* read len, skip */
-				p = xdr_inline_decode(&stream, 4);
-				if (unlikely(!p))
-					goto out_err_free_deviceid;
-				len = be32_to_cpup(p);
-
-				p = xdr_inline_decode(&stream, len);
-				if (unlikely(!p))
-					goto out_err_free_deviceid;
-
-				/* read len, skip */
-				p = xdr_inline_decode(&stream, 4);
-				if (unlikely(!p))
-					goto out_err_free_deviceid;
-				len = be32_to_cpup(p);
-
-				p = xdr_inline_decode(&stream, len);
-				if (unlikely(!p))
-					goto out_err_free_deviceid;
-			}
+			da = decode_ds_addr(&stream, gfp_flags);
+			if (da)
+				list_add_tail(&da->da_node, &dsaddrs);
+		}
+		if (list_empty(&dsaddrs)) {
+			dprintk("%s: no suitable DS addresses found\n",
+				__func__);
+			goto out_err_free_deviceid;
+		}
+
+		dsaddr->ds_list[i] = nfs4_pnfs_ds_add(&dsaddrs, gfp_flags);
+		if (!dsaddr->ds_list[i])
+			goto out_err_drain_dsaddrs;
+
+		/* If DS was already in cache, free ds addrs */
+		while (!list_empty(&dsaddrs)) {
+			da = list_first_entry(&dsaddrs,
+					      struct nfs4_pnfs_ds_addr,
+					      da_node);
+			list_del_init(&da->da_node);
+			kfree(da->da_remotestr);
+			kfree(da);
 		}
 	}
 
 	__free_page(scratch);
 	return dsaddr;
 
+out_err_drain_dsaddrs:
+	while (!list_empty(&dsaddrs)) {
+		da = list_first_entry(&dsaddrs, struct nfs4_pnfs_ds_addr,
+				      da_node);
+		list_del_init(&da->da_node);
+		kfree(da->da_remotestr);
+		kfree(da);
+	}
 out_err_free_deviceid:
 	nfs4_fl_free_deviceid(dsaddr);
 	/* stripe_indicies was part of dsaddr */
@@ -591,13 +815,13 @@ nfs4_fl_select_ds_fh(struct pnfs_layout_segment *lseg, u32 j)
 
 static void
 filelayout_mark_devid_negative(struct nfs4_file_layout_dsaddr *dsaddr,
-			       int err, u32 ds_addr)
+			       int err, const char *ds_remotestr)
 {
 	u32 *p = (u32 *)&dsaddr->id_node.deviceid;
 
-	printk(KERN_ERR "NFS: data server %x connection error %d."
+	printk(KERN_ERR "NFS: data server %s connection error %d."
 		" Deviceid [%x%x%x%x] marked out of use.\n",
-		ds_addr, err, p[0], p[1], p[2], p[3]);
+		ds_remotestr, err, p[0], p[1], p[2], p[3]);
 
 	spin_lock(&nfs4_ds_cache_lock);
 	dsaddr->flags |= NFS4_DEVICE_ID_NEG_ENTRY;
@@ -628,7 +852,7 @@ nfs4_fl_prepare_ds(struct pnfs_layout_segment *lseg, u32 ds_idx)
 		err = nfs4_ds_connect(s, ds);
 		if (err) {
 			filelayout_mark_devid_negative(dsaddr, err,
-						       ntohl(ds->ds_ip_addr));
+						       ds->ds_remotestr);
 			return NULL;
 		}
 	}
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 5879b23e0c99..93ef77666efc 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -80,7 +80,10 @@ static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle,
 static int nfs4_do_setattr(struct inode *inode, struct rpc_cred *cred,
 			    struct nfs_fattr *fattr, struct iattr *sattr,
 			    struct nfs4_state *state);
-
+#ifdef CONFIG_NFS_V4_1
+static int nfs41_test_stateid(struct nfs_server *, struct nfs4_state *);
+static int nfs41_free_stateid(struct nfs_server *, struct nfs4_state *);
+#endif
 /* Prevent leaks of NFSv4 errors into userland */
 static int nfs4_map_errors(int err)
 {
@@ -1687,6 +1690,20 @@ static int nfs4_open_expired(struct nfs4_state_owner *sp, struct nfs4_state *sta
 	return ret;
 }
 
+#if defined(CONFIG_NFS_V4_1)
+static int nfs41_open_expired(struct nfs4_state_owner *sp, struct nfs4_state *state)
+{
+	int status;
+	struct nfs_server *server = NFS_SERVER(state->inode);
+
+	status = nfs41_test_stateid(server, state);
+	if (status == NFS_OK)
+		return 0;
+	nfs41_free_stateid(server, state);
+	return nfs4_open_expired(sp, state);
+}
+#endif
+
 /*
  * on an EXCLUSIVE create, the server should send back a bitmask with FATTR4-*
  * fields corresponding to attributes that were used to store the verifier.
@@ -2251,13 +2268,14 @@ static int nfs4_find_root_sec(struct nfs_server *server, struct nfs_fh *fhandle,
 static int nfs4_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle,
 			      struct nfs_fsinfo *info)
 {
+	int minor_version = server->nfs_client->cl_minorversion;
 	int status = nfs4_lookup_root(server, fhandle, info);
 	if ((status == -NFS4ERR_WRONGSEC) && !(server->flags & NFS_MOUNT_SECFLAVOUR))
 		/*
 		 * A status of -NFS4ERR_WRONGSEC will be mapped to -EPERM
 		 * by nfs4_map_errors() as this function exits.
 		 */
-		status = nfs4_find_root_sec(server, fhandle, info);
+		status = nfs_v4_minor_ops[minor_version]->find_root_sec(server, fhandle, info);
 	if (status == 0)
 		status = nfs4_server_capabilities(server, fhandle);
 	if (status == 0)
@@ -4443,6 +4461,20 @@ out:
 	return err;
 }
 
+#if defined(CONFIG_NFS_V4_1)
+static int nfs41_lock_expired(struct nfs4_state *state, struct file_lock *request)
+{
+	int status;
+	struct nfs_server *server = NFS_SERVER(state->inode);
+
+	status = nfs41_test_stateid(server, state);
+	if (status == NFS_OK)
+		return 0;
+	nfs41_free_stateid(server, state);
+	return nfs4_lock_expired(state, request);
+}
+#endif
+
 static int _nfs4_proc_setlk(struct nfs4_state *state, int cmd, struct file_lock *request)
 {
 	struct nfs_inode *nfsi = NFS_I(state->inode);
@@ -4781,6 +4813,16 @@ out_inval:
 	return -NFS4ERR_INVAL;
 }
 
+static bool
+nfs41_same_server_scope(struct server_scope *a, struct server_scope *b)
+{
+	if (a->server_scope_sz == b->server_scope_sz &&
+	    memcmp(a->server_scope, b->server_scope, a->server_scope_sz) == 0)
+		return true;
+
+	return false;
+}
+
 /*
  * nfs4_proc_exchange_id()
  *
@@ -4823,9 +4865,31 @@ int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred)
 				init_utsname()->domainname,
 				clp->cl_rpcclient->cl_auth->au_flavor);
 
+	res.server_scope = kzalloc(sizeof(struct server_scope), GFP_KERNEL);
+	if (unlikely(!res.server_scope))
+		return -ENOMEM;
+
 	status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT);
 	if (!status)
 		status = nfs4_check_cl_exchange_flags(clp->cl_exchange_flags);
+
+	if (!status) {
+		if (clp->server_scope &&
+		    !nfs41_same_server_scope(clp->server_scope,
+					     res.server_scope)) {
+			dprintk("%s: server_scope mismatch detected\n",
+				__func__);
+			set_bit(NFS4CLNT_SERVER_SCOPE_MISMATCH, &clp->cl_state);
+			kfree(clp->server_scope);
+			clp->server_scope = NULL;
+		}
+
+		if (!clp->server_scope)
+			clp->server_scope = res.server_scope;
+		else
+			kfree(res.server_scope);
+	}
+
 	dprintk("<-- %s status= %d\n", __func__, status);
 	return status;
 }
@@ -5706,7 +5770,7 @@ static void nfs4_layoutreturn_done(struct rpc_task *task, void *calldata)
 {
 	struct nfs4_layoutreturn *lrp = calldata;
 	struct nfs_server *server;
-	struct pnfs_layout_hdr *lo = NFS_I(lrp->args.inode)->layout;
+	struct pnfs_layout_hdr *lo = lrp->args.layout;
 
 	dprintk("--> %s\n", __func__);
 
@@ -5735,7 +5799,7 @@ static void nfs4_layoutreturn_release(void *calldata)
 	struct nfs4_layoutreturn *lrp = calldata;
 
 	dprintk("--> %s\n", __func__);
-	put_layout_hdr(NFS_I(lrp->args.inode)->layout);
+	put_layout_hdr(lrp->args.layout);
 	kfree(calldata);
 	dprintk("<-- %s\n", __func__);
 }
@@ -5903,6 +5967,143 @@ out:
 	rpc_put_task(task);
 	return status;
 }
+
+static int
+_nfs41_proc_secinfo_no_name(struct nfs_server *server, struct nfs_fh *fhandle,
+		    struct nfs_fsinfo *info, struct nfs4_secinfo_flavors *flavors)
+{
+	struct nfs41_secinfo_no_name_args args = {
+		.style = SECINFO_STYLE_CURRENT_FH,
+	};
+	struct nfs4_secinfo_res res = {
+		.flavors = flavors,
+	};
+	struct rpc_message msg = {
+		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SECINFO_NO_NAME],
+		.rpc_argp = &args,
+		.rpc_resp = &res,
+	};
+	return nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0);
+}
+
+static int
+nfs41_proc_secinfo_no_name(struct nfs_server *server, struct nfs_fh *fhandle,
+			   struct nfs_fsinfo *info, struct nfs4_secinfo_flavors *flavors)
+{
+	struct nfs4_exception exception = { };
+	int err;
+	do {
+		err = _nfs41_proc_secinfo_no_name(server, fhandle, info, flavors);
+		switch (err) {
+		case 0:
+		case -NFS4ERR_WRONGSEC:
+		case -NFS4ERR_NOTSUPP:
+			break;
+		default:
+			err = nfs4_handle_exception(server, err, &exception);
+		}
+	} while (exception.retry);
+	return err;
+}
+
+static int
+nfs41_find_root_sec(struct nfs_server *server, struct nfs_fh *fhandle,
+		    struct nfs_fsinfo *info)
+{
+	int err;
+	struct page *page;
+	rpc_authflavor_t flavor;
+	struct nfs4_secinfo_flavors *flavors;
+
+	page = alloc_page(GFP_KERNEL);
+	if (!page) {
+		err = -ENOMEM;
+		goto out;
+	}
+
+	flavors = page_address(page);
+	err = nfs41_proc_secinfo_no_name(server, fhandle, info, flavors);
+
+	/*
+	 * Fall back on "guess and check" method if
+	 * the server doesn't support SECINFO_NO_NAME
+	 */
+	if (err == -NFS4ERR_WRONGSEC || err == -NFS4ERR_NOTSUPP) {
+		err = nfs4_find_root_sec(server, fhandle, info);
+		goto out_freepage;
+	}
+	if (err)
+		goto out_freepage;
+
+	flavor = nfs_find_best_sec(flavors);
+	if (err == 0)
+		err = nfs4_lookup_root_sec(server, fhandle, info, flavor);
+
+out_freepage:
+	put_page(page);
+	if (err == -EACCES)
+		return -EPERM;
+out:
+	return err;
+}
+static int _nfs41_test_stateid(struct nfs_server *server, struct nfs4_state *state)
+{
+	int status;
+	struct nfs41_test_stateid_args args = {
+		.stateid = &state->stateid,
+	};
+	struct nfs41_test_stateid_res res;
+	struct rpc_message msg = {
+		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_TEST_STATEID],
+		.rpc_argp = &args,
+		.rpc_resp = &res,
+	};
+	args.seq_args.sa_session = res.seq_res.sr_session = NULL;
+	status = nfs4_call_sync_sequence(server->client, server, &msg, &args.seq_args, &res.seq_res, 0, 1);
+	return status;
+}
+
+static int nfs41_test_stateid(struct nfs_server *server, struct nfs4_state *state)
+{
+	struct nfs4_exception exception = { };
+	int err;
+	do {
+		err = nfs4_handle_exception(server,
+				_nfs41_test_stateid(server, state),
+				&exception);
+	} while (exception.retry);
+	return err;
+}
+
+static int _nfs4_free_stateid(struct nfs_server *server, struct nfs4_state *state)
+{
+	int status;
+	struct nfs41_free_stateid_args args = {
+		.stateid = &state->stateid,
+	};
+	struct nfs41_free_stateid_res res;
+	struct rpc_message msg = {
+		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_FREE_STATEID],
+		.rpc_argp = &args,
+		.rpc_resp = &res,
+	};
+
+	args.seq_args.sa_session = res.seq_res.sr_session = NULL;
+	status = nfs4_call_sync_sequence(server->client, server, &msg, &args.seq_args, &res.seq_res, 0, 1);
+	return status;
+}
+
+static int nfs41_free_stateid(struct nfs_server *server, struct nfs4_state *state)
+{
+	struct nfs4_exception exception = { };
+	int err;
+	do {
+		err = nfs4_handle_exception(server,
+				_nfs4_free_stateid(server, state),
+				&exception);
+	} while (exception.retry);
+	return err;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 struct nfs4_state_recovery_ops nfs40_reboot_recovery_ops = {
@@ -5939,8 +6140,8 @@ struct nfs4_state_recovery_ops nfs40_nograce_recovery_ops = {
 struct nfs4_state_recovery_ops nfs41_nograce_recovery_ops = {
 	.owner_flag_bit = NFS_OWNER_RECLAIM_NOGRACE,
 	.state_flag_bit	= NFS_STATE_RECLAIM_NOGRACE,
-	.recover_open	= nfs4_open_expired,
-	.recover_lock	= nfs4_lock_expired,
+	.recover_open	= nfs41_open_expired,
+	.recover_lock	= nfs41_lock_expired,
 	.establish_clid = nfs41_init_clientid,
 	.get_clid_cred	= nfs4_get_exchange_id_cred,
 };
@@ -5964,6 +6165,7 @@ static const struct nfs4_minor_version_ops nfs_v4_0_minor_ops = {
 	.minor_version = 0,
 	.call_sync = _nfs4_call_sync,
 	.validate_stateid = nfs4_validate_delegation_stateid,
+	.find_root_sec = nfs4_find_root_sec,
 	.reboot_recovery_ops = &nfs40_reboot_recovery_ops,
 	.nograce_recovery_ops = &nfs40_nograce_recovery_ops,
 	.state_renewal_ops = &nfs40_state_renewal_ops,
@@ -5974,6 +6176,7 @@ static const struct nfs4_minor_version_ops nfs_v4_1_minor_ops = {
 	.minor_version = 1,
 	.call_sync = _nfs4_call_sync_session,
 	.validate_stateid = nfs41_validate_delegation_stateid,
+	.find_root_sec = nfs41_find_root_sec,
 	.reboot_recovery_ops = &nfs41_reboot_recovery_ops,
 	.nograce_recovery_ops = &nfs41_nograce_recovery_ops,
 	.state_renewal_ops = &nfs41_state_renewal_ops,
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index e97dd219f84f..5d744a52b4e1 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -1643,7 +1643,14 @@ 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_SERVER_SCOPE_MISMATCH,
+					       &clp->cl_state))
+				nfs4_state_start_reclaim_nograce(clp);
+			else
+				set_bit(NFS4CLNT_RECLAIM_REBOOT,
+					&clp->cl_state);
+
 			pnfs_destroy_all_layouts(clp);
 		}
 
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 5db44a8dddf9..aa83aa7820e7 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -255,7 +255,7 @@ static int nfs4_stat_to_errno(int);
 #define decode_fs_locations_maxsz \
 				(0)
 #define encode_secinfo_maxsz	(op_encode_hdr_maxsz + nfs4_name_maxsz)
-#define decode_secinfo_maxsz	(op_decode_hdr_maxsz + 4 + (NFS_MAX_SECFLAVORS * (16 + GSS_OID_MAX_LEN)))
+#define decode_secinfo_maxsz	(op_decode_hdr_maxsz + 1 + ((NFS_MAX_SECFLAVORS * (16 + GSS_OID_MAX_LEN)) / 4))
 
 #if defined(CONFIG_NFS_V4_1)
 #define NFS4_MAX_MACHINE_NAME_LEN (64)
@@ -343,6 +343,14 @@ static int nfs4_stat_to_errno(int);
 				1 /* FIXME: opaque lrf_body always empty at the moment */)
 #define decode_layoutreturn_maxsz (op_decode_hdr_maxsz + \
 				1 + decode_stateid_maxsz)
+#define encode_secinfo_no_name_maxsz (op_encode_hdr_maxsz + 1)
+#define decode_secinfo_no_name_maxsz decode_secinfo_maxsz
+#define encode_test_stateid_maxsz	(op_encode_hdr_maxsz + 2 + \
+					 XDR_QUADLEN(NFS4_STATEID_SIZE))
+#define decode_test_stateid_maxsz	(op_decode_hdr_maxsz + 2 + 1)
+#define encode_free_stateid_maxsz	(op_encode_hdr_maxsz + 1 + \
+					 XDR_QUADLEN(NFS4_STATEID_SIZE))
+#define decode_free_stateid_maxsz	(op_decode_hdr_maxsz + 1)
 #else /* CONFIG_NFS_V4_1 */
 #define encode_sequence_maxsz	0
 #define decode_sequence_maxsz	0
@@ -772,6 +780,26 @@ static int nfs4_stat_to_errno(int);
 				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_layoutreturn_maxsz)
+#define NFS4_enc_secinfo_no_name_sz	(compound_encode_hdr_maxsz + \
+					encode_sequence_maxsz + \
+					encode_putrootfh_maxsz +\
+					encode_secinfo_no_name_maxsz)
+#define NFS4_dec_secinfo_no_name_sz	(compound_decode_hdr_maxsz + \
+					decode_sequence_maxsz + \
+					decode_putrootfh_maxsz + \
+					decode_secinfo_no_name_maxsz)
+#define NFS4_enc_test_stateid_sz	(compound_encode_hdr_maxsz + \
+					 encode_sequence_maxsz + \
+					 encode_test_stateid_maxsz)
+#define NFS4_dec_test_stateid_sz	(compound_decode_hdr_maxsz + \
+					 decode_sequence_maxsz + \
+					 decode_test_stateid_maxsz)
+#define NFS4_enc_free_stateid_sz	(compound_encode_hdr_maxsz + \
+					 encode_sequence_maxsz + \
+					 encode_free_stateid_maxsz)
+#define NFS4_dec_free_stateid_sz	(compound_decode_hdr_maxsz + \
+					 decode_sequence_maxsz + \
+					 decode_free_stateid_maxsz)
 
 const u32 nfs41_maxwrite_overhead = ((RPC_MAX_HEADER_WITH_AUTH +
 				      compound_encode_hdr_maxsz +
@@ -1938,6 +1966,46 @@ encode_layoutreturn(struct xdr_stream *xdr,
 	hdr->nops++;
 	hdr->replen += decode_layoutreturn_maxsz;
 }
+
+static int
+encode_secinfo_no_name(struct xdr_stream *xdr,
+		       const struct nfs41_secinfo_no_name_args *args,
+		       struct compound_hdr *hdr)
+{
+	__be32 *p;
+	p = reserve_space(xdr, 8);
+	*p++ = cpu_to_be32(OP_SECINFO_NO_NAME);
+	*p++ = cpu_to_be32(args->style);
+	hdr->nops++;
+	hdr->replen += decode_secinfo_no_name_maxsz;
+	return 0;
+}
+
+static void encode_test_stateid(struct xdr_stream *xdr,
+				struct nfs41_test_stateid_args *args,
+				struct compound_hdr *hdr)
+{
+	__be32 *p;
+
+	p = reserve_space(xdr, 8 + NFS4_STATEID_SIZE);
+	*p++ = cpu_to_be32(OP_TEST_STATEID);
+	*p++ = cpu_to_be32(1);
+	xdr_encode_opaque_fixed(p, args->stateid->data, NFS4_STATEID_SIZE);
+	hdr->nops++;
+	hdr->replen += decode_test_stateid_maxsz;
+}
+
+static void encode_free_stateid(struct xdr_stream *xdr,
+				struct nfs41_free_stateid_args *args,
+				struct compound_hdr *hdr)
+{
+	__be32 *p;
+	p = reserve_space(xdr, 4 + NFS4_STATEID_SIZE);
+	*p++ = cpu_to_be32(OP_FREE_STATEID);
+	xdr_encode_opaque_fixed(p, args->stateid->data, NFS4_STATEID_SIZE);
+	hdr->nops++;
+	hdr->replen += decode_free_stateid_maxsz;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 /*
@@ -2790,6 +2858,59 @@ static void nfs4_xdr_enc_layoutreturn(struct rpc_rqst *req,
 	encode_layoutreturn(xdr, args, &hdr);
 	encode_nops(&hdr);
 }
+
+/*
+ * Encode SECINFO_NO_NAME request
+ */
+static int nfs4_xdr_enc_secinfo_no_name(struct rpc_rqst *req,
+					struct xdr_stream *xdr,
+					struct nfs41_secinfo_no_name_args *args)
+{
+	struct compound_hdr hdr = {
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
+	};
+
+	encode_compound_hdr(xdr, req, &hdr);
+	encode_sequence(xdr, &args->seq_args, &hdr);
+	encode_putrootfh(xdr, &hdr);
+	encode_secinfo_no_name(xdr, args, &hdr);
+	encode_nops(&hdr);
+	return 0;
+}
+
+/*
+ *  Encode TEST_STATEID request
+ */
+static void nfs4_xdr_enc_test_stateid(struct rpc_rqst *req,
+				      struct xdr_stream *xdr,
+				      struct nfs41_test_stateid_args *args)
+{
+	struct compound_hdr hdr = {
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
+	};
+
+	encode_compound_hdr(xdr, req, &hdr);
+	encode_sequence(xdr, &args->seq_args, &hdr);
+	encode_test_stateid(xdr, args, &hdr);
+	encode_nops(&hdr);
+}
+
+/*
+ *  Encode FREE_STATEID request
+ */
+static void nfs4_xdr_enc_free_stateid(struct rpc_rqst *req,
+				     struct xdr_stream *xdr,
+				     struct nfs41_free_stateid_args *args)
+{
+	struct compound_hdr hdr = {
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
+	};
+
+	encode_compound_hdr(xdr, req, &hdr);
+	encode_sequence(xdr, &args->seq_args, &hdr);
+	encode_free_stateid(xdr, args, &hdr);
+	encode_nops(&hdr);
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 static void print_overflow_msg(const char *func, const struct xdr_stream *xdr)
@@ -4977,11 +5098,17 @@ static int decode_exchange_id(struct xdr_stream *xdr,
 	if (unlikely(status))
 		return status;
 
-	/* Throw away server_scope */
+	/* Save server_scope */
 	status = decode_opaque_inline(xdr, &dummy, &dummy_str);
 	if (unlikely(status))
 		return status;
 
+	if (unlikely(dummy > NFS4_OPAQUE_LIMIT))
+		return -EIO;
+
+	memcpy(res->server_scope->server_scope, dummy_str, dummy);
+	res->server_scope->server_scope_sz = dummy;
+
 	/* Throw away Implementation id array */
 	status = decode_opaque_inline(xdr, &dummy, &dummy_str);
 	if (unlikely(status))
@@ -5322,6 +5449,55 @@ out_overflow:
 	print_overflow_msg(__func__, xdr);
 	return -EIO;
 }
+
+static int decode_test_stateid(struct xdr_stream *xdr,
+			       struct nfs41_test_stateid_res *res)
+{
+	__be32 *p;
+	int status;
+	int num_res;
+
+	status = decode_op_hdr(xdr, OP_TEST_STATEID);
+	if (status)
+		return status;
+
+	p = xdr_inline_decode(xdr, 4);
+	if (unlikely(!p))
+		goto out_overflow;
+	num_res = be32_to_cpup(p++);
+	if (num_res != 1)
+		goto out;
+
+	p = xdr_inline_decode(xdr, 4);
+	if (unlikely(!p))
+		goto out_overflow;
+	res->status = be32_to_cpup(p++);
+	return res->status;
+out_overflow:
+	print_overflow_msg(__func__, xdr);
+out:
+	return -EIO;
+}
+
+static int decode_free_stateid(struct xdr_stream *xdr,
+			       struct nfs41_free_stateid_res *res)
+{
+	__be32 *p;
+	int status;
+
+	status = decode_op_hdr(xdr, OP_FREE_STATEID);
+	if (status)
+		return status;
+
+	p = xdr_inline_decode(xdr, 4);
+	if (unlikely(!p))
+		goto out_overflow;
+	res->status = be32_to_cpup(p++);
+	return res->status;
+out_overflow:
+	print_overflow_msg(__func__, xdr);
+	return -EIO;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 /*
@@ -6461,6 +6637,72 @@ static int nfs4_xdr_dec_layoutcommit(struct rpc_rqst *rqstp,
 out:
 	return status;
 }
+
+/*
+ * Decode SECINFO_NO_NAME response
+ */
+static int nfs4_xdr_dec_secinfo_no_name(struct rpc_rqst *rqstp,
+					struct xdr_stream *xdr,
+					struct nfs4_secinfo_res *res)
+{
+	struct compound_hdr hdr;
+	int status;
+
+	status = decode_compound_hdr(xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(xdr, &res->seq_res, rqstp);
+	if (status)
+		goto out;
+	status = decode_putrootfh(xdr);
+	if (status)
+		goto out;
+	status = decode_secinfo(xdr, res);
+out:
+	return status;
+}
+
+/*
+ * Decode TEST_STATEID response
+ */
+static int nfs4_xdr_dec_test_stateid(struct rpc_rqst *rqstp,
+				     struct xdr_stream *xdr,
+				     struct nfs41_test_stateid_res *res)
+{
+	struct compound_hdr hdr;
+	int status;
+
+	status = decode_compound_hdr(xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(xdr, &res->seq_res, rqstp);
+	if (status)
+		goto out;
+	status = decode_test_stateid(xdr, res);
+out:
+	return status;
+}
+
+/*
+ * Decode FREE_STATEID response
+ */
+static int nfs4_xdr_dec_free_stateid(struct rpc_rqst *rqstp,
+				     struct xdr_stream *xdr,
+				     struct nfs41_free_stateid_res *res)
+{
+	struct compound_hdr hdr;
+	int status;
+
+	status = decode_compound_hdr(xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(xdr, &res->seq_res, rqstp);
+	if (status)
+		goto out;
+	status = decode_free_stateid(xdr, res);
+out:
+	return status;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 /**
@@ -6663,6 +6905,9 @@ struct rpc_procinfo	nfs4_procedures[] = {
 	PROC(LAYOUTGET,		enc_layoutget,		dec_layoutget),
 	PROC(LAYOUTCOMMIT,	enc_layoutcommit,	dec_layoutcommit),
 	PROC(LAYOUTRETURN,	enc_layoutreturn,	dec_layoutreturn),
+	PROC(SECINFO_NO_NAME,	enc_secinfo_no_name,	dec_secinfo_no_name),
+	PROC(TEST_STATEID,	enc_test_stateid,	dec_test_stateid),
+	PROC(FREE_STATEID,	enc_free_stateid,	dec_free_stateid),
 #endif /* CONFIG_NFS_V4_1 */
 };
 
diff --git a/fs/nfs/objlayout/objio_osd.c b/fs/nfs/objlayout/objio_osd.c
index eb4aafa9f521..70272d5355b2 100644
--- a/fs/nfs/objlayout/objio_osd.c
+++ b/fs/nfs/objlayout/objio_osd.c
@@ -1004,6 +1004,18 @@ static bool objio_pg_test(struct nfs_pageio_descriptor *pgio,
 			OBJIO_LSEG(pgio->pg_lseg)->max_io_size;
 }
 
+static const struct nfs_pageio_ops objio_pg_read_ops = {
+	.pg_init = pnfs_generic_pg_init_read,
+	.pg_test = objio_pg_test,
+	.pg_doio = nfs_generic_pg_readpages,
+};
+
+static const struct nfs_pageio_ops objio_pg_write_ops = {
+	.pg_init = pnfs_generic_pg_init_write,
+	.pg_test = objio_pg_test,
+	.pg_doio = nfs_generic_pg_writepages,
+};
+
 static struct pnfs_layoutdriver_type objlayout_type = {
 	.id = LAYOUT_OSD2_OBJECTS,
 	.name = "LAYOUT_OSD2_OBJECTS",
@@ -1017,7 +1029,8 @@ static struct pnfs_layoutdriver_type objlayout_type = {
 
 	.read_pagelist           = objlayout_read_pagelist,
 	.write_pagelist          = objlayout_write_pagelist,
-	.pg_test                 = objio_pg_test,
+	.pg_read_ops             = &objio_pg_read_ops,
+	.pg_write_ops            = &objio_pg_write_ops,
 
 	.free_deviceid_node	 = objio_free_deviceid_node,
 
diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c
index 009855716286..d421e19557a5 100644
--- a/fs/nfs/pagelist.c
+++ b/fs/nfs/pagelist.c
@@ -230,7 +230,7 @@ EXPORT_SYMBOL_GPL(nfs_generic_pg_test);
  */
 void nfs_pageio_init(struct nfs_pageio_descriptor *desc,
 		     struct inode *inode,
-		     int (*doio)(struct nfs_pageio_descriptor *),
+		     const struct nfs_pageio_ops *pg_ops,
 		     size_t bsize,
 		     int io_flags)
 {
@@ -241,12 +241,10 @@ void nfs_pageio_init(struct nfs_pageio_descriptor *desc,
 	desc->pg_base = 0;
 	desc->pg_moreio = 0;
 	desc->pg_inode = inode;
-	desc->pg_doio = doio;
+	desc->pg_ops = pg_ops;
 	desc->pg_ioflags = io_flags;
 	desc->pg_error = 0;
 	desc->pg_lseg = NULL;
-	desc->pg_test = nfs_generic_pg_test;
-	pnfs_pageio_init(desc, inode);
 }
 
 /**
@@ -276,7 +274,7 @@ static bool nfs_can_coalesce_requests(struct nfs_page *prev,
 		return false;
 	if (prev->wb_pgbase + prev->wb_bytes != PAGE_CACHE_SIZE)
 		return false;
-	return pgio->pg_test(pgio, prev, req);
+	return pgio->pg_ops->pg_test(pgio, prev, req);
 }
 
 /**
@@ -297,6 +295,8 @@ static int nfs_pageio_do_add_request(struct nfs_pageio_descriptor *desc,
 		if (!nfs_can_coalesce_requests(prev, req, desc))
 			return 0;
 	} else {
+		if (desc->pg_ops->pg_init)
+			desc->pg_ops->pg_init(desc, req);
 		desc->pg_base = req->wb_pgbase;
 	}
 	nfs_list_remove_request(req);
@@ -311,7 +311,7 @@ static int nfs_pageio_do_add_request(struct nfs_pageio_descriptor *desc,
 static void nfs_pageio_doio(struct nfs_pageio_descriptor *desc)
 {
 	if (!list_empty(&desc->pg_list)) {
-		int error = desc->pg_doio(desc);
+		int error = desc->pg_ops->pg_doio(desc);
 		if (error < 0)
 			desc->pg_error = error;
 		else
diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c
index 539b94cb6c0f..5b3cc3f4bb39 100644
--- a/fs/nfs/pnfs.c
+++ b/fs/nfs/pnfs.c
@@ -448,11 +448,20 @@ pnfs_destroy_layout(struct nfs_inode *nfsi)
 void
 pnfs_destroy_all_layouts(struct nfs_client *clp)
 {
+	struct nfs_server *server;
 	struct pnfs_layout_hdr *lo;
 	LIST_HEAD(tmp_list);
 
+	nfs4_deviceid_mark_client_invalid(clp);
+	nfs4_deviceid_purge_client(clp);
+
 	spin_lock(&clp->cl_lock);
-	list_splice_init(&clp->cl_layouts, &tmp_list);
+	rcu_read_lock();
+	list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) {
+		if (!list_empty(&server->layouts))
+			list_splice_init(&server->layouts, &tmp_list);
+	}
+	rcu_read_unlock();
 	spin_unlock(&clp->cl_lock);
 
 	while (!list_empty(&tmp_list)) {
@@ -661,6 +670,7 @@ _pnfs_return_layout(struct inode *ino)
 	lrp->args.stateid = stateid;
 	lrp->args.layout_type = NFS_SERVER(ino)->pnfs_curr_ld->id;
 	lrp->args.inode = ino;
+	lrp->args.layout = lo;
 	lrp->clp = NFS_SERVER(ino)->nfs_client;
 
 	status = nfs4_proc_layoutreturn(lrp);
@@ -920,7 +930,8 @@ pnfs_update_layout(struct inode *ino,
 	};
 	unsigned pg_offset;
 	struct nfs_inode *nfsi = NFS_I(ino);
-	struct nfs_client *clp = NFS_SERVER(ino)->nfs_client;
+	struct nfs_server *server = NFS_SERVER(ino);
+	struct nfs_client *clp = server->nfs_client;
 	struct pnfs_layout_hdr *lo;
 	struct pnfs_layout_segment *lseg = NULL;
 	bool first = false;
@@ -964,7 +975,7 @@ pnfs_update_layout(struct inode *ino,
 		 */
 		spin_lock(&clp->cl_lock);
 		BUG_ON(!list_empty(&lo->plh_layouts));
-		list_add_tail(&lo->plh_layouts, &clp->cl_layouts);
+		list_add_tail(&lo->plh_layouts, &server->layouts);
 		spin_unlock(&clp->cl_lock);
 	}
 
@@ -973,7 +984,8 @@ pnfs_update_layout(struct inode *ino,
 		arg.offset -= pg_offset;
 		arg.length += pg_offset;
 	}
-	arg.length = PAGE_CACHE_ALIGN(arg.length);
+	if (arg.length != NFS4_MAX_UINT64)
+		arg.length = PAGE_CACHE_ALIGN(arg.length);
 
 	lseg = send_layoutget(lo, ctx, &arg, gfp_flags);
 	if (!lseg && first) {
@@ -991,6 +1003,7 @@ out_unlock:
 	spin_unlock(&ino->i_lock);
 	goto out;
 }
+EXPORT_SYMBOL_GPL(pnfs_update_layout);
 
 int
 pnfs_layout_process(struct nfs4_layoutget *lgp)
@@ -1048,40 +1061,89 @@ out_forget_reply:
 	goto out;
 }
 
-bool
-pnfs_generic_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev,
-		     struct nfs_page *req)
+void
+pnfs_generic_pg_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *req)
 {
-	enum pnfs_iomode access_type;
-	gfp_t gfp_flags;
+	BUG_ON(pgio->pg_lseg != NULL);
+
+	pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
+					   req->wb_context,
+					   req_offset(req),
+					   req->wb_bytes,
+					   IOMODE_READ,
+					   GFP_KERNEL);
+	/* If no lseg, fall back to read through mds */
+	if (pgio->pg_lseg == NULL)
+		nfs_pageio_init_read_mds(pgio, pgio->pg_inode);
 
-	/* We assume that pg_ioflags == 0 iff we're reading a page */
-	if (pgio->pg_ioflags == 0) {
-		access_type = IOMODE_READ;
-		gfp_flags = GFP_KERNEL;
-	} else {
-		access_type = IOMODE_RW;
-		gfp_flags = GFP_NOFS;
-	}
+}
+EXPORT_SYMBOL_GPL(pnfs_generic_pg_init_read);
 
-	if (pgio->pg_count == prev->wb_bytes) {
-		/* This is first coelesce call for a series of nfs_pages */
-		pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
-						   prev->wb_context,
-						   req_offset(req),
-						   pgio->pg_count,
-						   access_type,
-						   gfp_flags);
-		return true;
-	}
+void
+pnfs_generic_pg_init_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *req)
+{
+	BUG_ON(pgio->pg_lseg != NULL);
+
+	pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
+					   req->wb_context,
+					   req_offset(req),
+					   req->wb_bytes,
+					   IOMODE_RW,
+					   GFP_NOFS);
+	/* If no lseg, fall back to write through mds */
+	if (pgio->pg_lseg == NULL)
+		nfs_pageio_init_write_mds(pgio, pgio->pg_inode, pgio->pg_ioflags);
+}
+EXPORT_SYMBOL_GPL(pnfs_generic_pg_init_write);
+
+bool
+pnfs_pageio_init_read(struct nfs_pageio_descriptor *pgio, struct inode *inode)
+{
+	struct nfs_server *server = NFS_SERVER(inode);
+	struct pnfs_layoutdriver_type *ld = server->pnfs_curr_ld;
 
-	if (pgio->pg_lseg &&
-	    req_offset(req) > end_offset(pgio->pg_lseg->pls_range.offset,
-					 pgio->pg_lseg->pls_range.length))
+	if (ld == NULL)
 		return false;
+	nfs_pageio_init(pgio, inode, ld->pg_read_ops, server->rsize, 0);
+	return true;
+}
 
+bool
+pnfs_pageio_init_write(struct nfs_pageio_descriptor *pgio, struct inode *inode, int ioflags)
+{
+	struct nfs_server *server = NFS_SERVER(inode);
+	struct pnfs_layoutdriver_type *ld = server->pnfs_curr_ld;
+
+	if (ld == NULL)
+		return false;
+	nfs_pageio_init(pgio, inode, ld->pg_write_ops, server->wsize, ioflags);
 	return true;
 }
+
+bool
+pnfs_generic_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev,
+		     struct nfs_page *req)
+{
+	if (pgio->pg_lseg == NULL)
+		return nfs_generic_pg_test(pgio, prev, req);
+
+	/*
+	 * Test if a nfs_page is fully contained in the pnfs_layout_range.
+	 * Note that this test makes several assumptions:
+	 * - that the previous nfs_page in the struct nfs_pageio_descriptor
+	 *   is known to lie within the range.
+	 *   - that the nfs_page being tested is known to be contiguous with the
+	 *   previous nfs_page.
+	 *   - Layout ranges are page aligned, so we only have to test the
+	 *   start offset of the request.
+	 *
+	 * Please also note that 'end_offset' is actually the offset of the
+	 * first byte that lies outside the pnfs_layout_range. FIXME?
+	 *
+	 */
+	return req_offset(req) < end_offset(pgio->pg_lseg->pls_range.offset,
+					 pgio->pg_lseg->pls_range.length);
+}
 EXPORT_SYMBOL_GPL(pnfs_generic_pg_test);
 
 /*
diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h
index 96bf4e6f45be..a59736eae6ec 100644
--- a/fs/nfs/pnfs.h
+++ b/fs/nfs/pnfs.h
@@ -87,7 +87,8 @@ struct pnfs_layoutdriver_type {
 	void (*free_lseg) (struct pnfs_layout_segment *lseg);
 
 	/* test for nfs page cache coalescing */
-	bool (*pg_test)(struct nfs_pageio_descriptor *, struct nfs_page *, struct nfs_page *);
+	const struct nfs_pageio_ops *pg_read_ops;
+	const struct nfs_pageio_ops *pg_write_ops;
 
 	/* Returns true if layoutdriver wants to divert this request to
 	 * driver's commit routine.
@@ -148,16 +149,18 @@ extern int nfs4_proc_layoutreturn(struct nfs4_layoutreturn *lrp);
 /* pnfs.c */
 void get_layout_hdr(struct pnfs_layout_hdr *lo);
 void put_lseg(struct pnfs_layout_segment *lseg);
-struct pnfs_layout_segment *
-pnfs_update_layout(struct inode *ino, struct nfs_open_context *ctx,
-		   loff_t pos, u64 count, enum pnfs_iomode access_type,
-		   gfp_t gfp_flags);
+
+bool pnfs_pageio_init_read(struct nfs_pageio_descriptor *, struct inode *);
+bool pnfs_pageio_init_write(struct nfs_pageio_descriptor *, struct inode *, int);
+
 void set_pnfs_layoutdriver(struct nfs_server *, u32 id);
 void unset_pnfs_layoutdriver(struct nfs_server *);
 enum pnfs_try_status pnfs_try_to_write_data(struct nfs_write_data *,
 					     const struct rpc_call_ops *, int);
 enum pnfs_try_status pnfs_try_to_read_data(struct nfs_read_data *,
 					    const struct rpc_call_ops *);
+void pnfs_generic_pg_init_read(struct nfs_pageio_descriptor *, struct nfs_page *);
+void pnfs_generic_pg_init_write(struct nfs_pageio_descriptor *, struct nfs_page *);
 bool pnfs_generic_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, struct nfs_page *req);
 int pnfs_layout_process(struct nfs4_layoutget *lgp);
 void pnfs_free_lseg_list(struct list_head *tmp_list);
@@ -182,6 +185,19 @@ int pnfs_layoutcommit_inode(struct inode *inode, bool sync);
 int _pnfs_return_layout(struct inode *);
 int pnfs_ld_write_done(struct nfs_write_data *);
 int pnfs_ld_read_done(struct nfs_read_data *);
+struct pnfs_layout_segment *pnfs_update_layout(struct inode *ino,
+					       struct nfs_open_context *ctx,
+					       loff_t pos,
+					       u64 count,
+					       enum pnfs_iomode iomode,
+					       gfp_t gfp_flags);
+
+void nfs4_deviceid_mark_client_invalid(struct nfs_client *clp);
+
+/* nfs4_deviceid_flags */
+enum {
+	NFS_DEVICEID_INVALID = 0,       /* set when MDS clientid recalled */
+};
 
 /* pnfs_dev.c */
 struct nfs4_deviceid_node {
@@ -189,13 +205,13 @@ struct nfs4_deviceid_node {
 	struct hlist_node		tmpnode;
 	const struct pnfs_layoutdriver_type *ld;
 	const struct nfs_client		*nfs_client;
+	unsigned long 			flags;
 	struct nfs4_deviceid		deviceid;
 	atomic_t			ref;
 };
 
 void nfs4_print_deviceid(const struct nfs4_deviceid *dev_id);
 struct nfs4_deviceid_node *nfs4_find_get_deviceid(const struct pnfs_layoutdriver_type *, const struct nfs_client *, const struct nfs4_deviceid *);
-struct nfs4_deviceid_node *nfs4_unhash_put_deviceid(const struct pnfs_layoutdriver_type *, const struct nfs_client *, const struct nfs4_deviceid *);
 void nfs4_delete_deviceid(const struct pnfs_layoutdriver_type *, const struct nfs_client *, const struct nfs4_deviceid *);
 void nfs4_init_deviceid_node(struct nfs4_deviceid_node *,
 			     const struct pnfs_layoutdriver_type *,
@@ -293,15 +309,6 @@ static inline int pnfs_return_layout(struct inode *ino)
 	return 0;
 }
 
-static inline void pnfs_pageio_init(struct nfs_pageio_descriptor *pgio,
-				    struct inode *inode)
-{
-	struct pnfs_layoutdriver_type *ld = NFS_SERVER(inode)->pnfs_curr_ld;
-
-	if (ld)
-		pgio->pg_test = ld->pg_test;
-}
-
 #else  /* CONFIG_NFS_V4_1 */
 
 static inline void pnfs_destroy_all_layouts(struct nfs_client *clp)
@@ -322,14 +329,6 @@ static inline void put_lseg(struct pnfs_layout_segment *lseg)
 {
 }
 
-static inline struct pnfs_layout_segment *
-pnfs_update_layout(struct inode *ino, struct nfs_open_context *ctx,
-		   loff_t pos, u64 count, enum pnfs_iomode access_type,
-		   gfp_t gfp_flags)
-{
-	return NULL;
-}
-
 static inline enum pnfs_try_status
 pnfs_try_to_read_data(struct nfs_read_data *data,
 		      const struct rpc_call_ops *call_ops)
@@ -385,9 +384,14 @@ static inline void unset_pnfs_layoutdriver(struct nfs_server *s)
 {
 }
 
-static inline void pnfs_pageio_init(struct nfs_pageio_descriptor *pgio,
-				    struct inode *inode)
+static inline bool pnfs_pageio_init_read(struct nfs_pageio_descriptor *pgio, struct inode *inode)
 {
+	return false;
+}
+
+static inline bool pnfs_pageio_init_write(struct nfs_pageio_descriptor *pgio, struct inode *inode, int ioflags)
+{
+	return false;
 }
 
 static inline void
diff --git a/fs/nfs/pnfs_dev.c b/fs/nfs/pnfs_dev.c
index f0f8e1e22f6c..6fda5228ef56 100644
--- a/fs/nfs/pnfs_dev.c
+++ b/fs/nfs/pnfs_dev.c
@@ -100,8 +100,8 @@ _find_get_deviceid(const struct pnfs_layoutdriver_type *ld,
 
 	rcu_read_lock();
 	d = _lookup_deviceid(ld, clp, id, hash);
-	if (d && !atomic_inc_not_zero(&d->ref))
-		d = NULL;
+	if (d != NULL)
+		atomic_inc(&d->ref);
 	rcu_read_unlock();
 	return d;
 }
@@ -115,15 +115,15 @@ nfs4_find_get_deviceid(const struct pnfs_layoutdriver_type *ld,
 EXPORT_SYMBOL_GPL(nfs4_find_get_deviceid);
 
 /*
- * Unhash and put deviceid
+ * Remove a deviceid from cache
  *
  * @clp nfs_client associated with deviceid
  * @id the deviceid to unhash
  *
  * @ret the unhashed node, if found and dereferenced to zero, NULL otherwise.
  */
-struct nfs4_deviceid_node *
-nfs4_unhash_put_deviceid(const struct pnfs_layoutdriver_type *ld,
+void
+nfs4_delete_deviceid(const struct pnfs_layoutdriver_type *ld,
 			 const struct nfs_client *clp, const struct nfs4_deviceid *id)
 {
 	struct nfs4_deviceid_node *d;
@@ -134,7 +134,7 @@ nfs4_unhash_put_deviceid(const struct pnfs_layoutdriver_type *ld,
 	rcu_read_unlock();
 	if (!d) {
 		spin_unlock(&nfs4_deviceid_lock);
-		return NULL;
+		return;
 	}
 	hlist_del_init_rcu(&d->node);
 	spin_unlock(&nfs4_deviceid_lock);
@@ -142,28 +142,7 @@ nfs4_unhash_put_deviceid(const struct pnfs_layoutdriver_type *ld,
 
 	/* balance the initial ref set in pnfs_insert_deviceid */
 	if (atomic_dec_and_test(&d->ref))
-		return d;
-
-	return NULL;
-}
-EXPORT_SYMBOL_GPL(nfs4_unhash_put_deviceid);
-
-/*
- * Delete a deviceid from cache
- *
- * @clp struct nfs_client qualifying the deviceid
- * @id deviceid to delete
- */
-void
-nfs4_delete_deviceid(const struct pnfs_layoutdriver_type *ld,
-		     const struct nfs_client *clp, const struct nfs4_deviceid *id)
-{
-	struct nfs4_deviceid_node *d;
-
-	d = nfs4_unhash_put_deviceid(ld, clp, id);
-	if (!d)
-		return;
-	d->ld->free_deviceid_node(d);
+		d->ld->free_deviceid_node(d);
 }
 EXPORT_SYMBOL_GPL(nfs4_delete_deviceid);
 
@@ -177,6 +156,7 @@ nfs4_init_deviceid_node(struct nfs4_deviceid_node *d,
 	INIT_HLIST_NODE(&d->tmpnode);
 	d->ld = ld;
 	d->nfs_client = nfs_client;
+	d->flags = 0;
 	d->deviceid = *id;
 	atomic_set(&d->ref, 1);
 }
@@ -221,16 +201,15 @@ EXPORT_SYMBOL_GPL(nfs4_insert_deviceid_node);
  *
  * @d deviceid node to put
  *
- * @ret true iff the node was deleted
+ * return true iff the node was deleted
+ * Note that since the test for d->ref == 0 is sufficient to establish
+ * that the node is no longer hashed in the global device id cache.
  */
 bool
 nfs4_put_deviceid_node(struct nfs4_deviceid_node *d)
 {
-	if (!atomic_dec_and_lock(&d->ref, &nfs4_deviceid_lock))
+	if (!atomic_dec_and_test(&d->ref))
 		return false;
-	hlist_del_init_rcu(&d->node);
-	spin_unlock(&nfs4_deviceid_lock);
-	synchronize_rcu();
 	d->ld->free_deviceid_node(d);
 	return true;
 }
@@ -275,3 +254,22 @@ nfs4_deviceid_purge_client(const struct nfs_client *clp)
 	for (h = 0; h < NFS4_DEVICE_ID_HASH_SIZE; h++)
 		_deviceid_purge_client(clp, h);
 }
+
+/*
+ * Stop use of all deviceids associated with an nfs_client
+ */
+void
+nfs4_deviceid_mark_client_invalid(struct nfs_client *clp)
+{
+	struct nfs4_deviceid_node *d;
+	struct hlist_node *n;
+	int i;
+
+	rcu_read_lock();
+	for (i = 0; i < NFS4_DEVICE_ID_HASH_SIZE; i ++){
+		hlist_for_each_entry_rcu(d, n, &nfs4_deviceid_cache[i], node)
+			if (d->nfs_client == clp)
+				set_bit(NFS_DEVICEID_INVALID, &d->flags);
+	}
+	rcu_read_unlock();
+}
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index 20a7f952e244..c394662c790e 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -32,6 +32,7 @@
 
 static int nfs_pagein_multi(struct nfs_pageio_descriptor *desc);
 static int nfs_pagein_one(struct nfs_pageio_descriptor *desc);
+static const struct nfs_pageio_ops nfs_pageio_read_ops;
 static const struct rpc_call_ops nfs_read_partial_ops;
 static const struct rpc_call_ops nfs_read_full_ops;
 
@@ -113,6 +114,21 @@ static void nfs_readpage_truncate_uninitialised_page(struct nfs_read_data *data)
 	}
 }
 
+void nfs_pageio_init_read_mds(struct nfs_pageio_descriptor *pgio,
+		struct inode *inode)
+{
+	nfs_pageio_init(pgio, inode, &nfs_pageio_read_ops,
+			NFS_SERVER(inode)->rsize, 0);
+}
+EXPORT_SYMBOL_GPL(nfs_pageio_init_read_mds);
+
+static void nfs_pageio_init_read(struct nfs_pageio_descriptor *pgio,
+		struct inode *inode)
+{
+	if (!pnfs_pageio_init_read(pgio, inode))
+		nfs_pageio_init_read_mds(pgio, inode);
+}
+
 int nfs_readpage_async(struct nfs_open_context *ctx, struct inode *inode,
 		       struct page *page)
 {
@@ -131,14 +147,9 @@ int nfs_readpage_async(struct nfs_open_context *ctx, struct inode *inode,
 	if (len < PAGE_CACHE_SIZE)
 		zero_user_segment(page, len, PAGE_CACHE_SIZE);
 
-	nfs_pageio_init(&pgio, inode, NULL, 0, 0);
-	nfs_list_add_request(new, &pgio.pg_list);
-	pgio.pg_count = len;
-
-	if (NFS_SERVER(inode)->rsize < PAGE_CACHE_SIZE)
-		nfs_pagein_multi(&pgio);
-	else
-		nfs_pagein_one(&pgio);
+	nfs_pageio_init_read(&pgio, inode);
+	nfs_pageio_add_request(&pgio, new);
+	nfs_pageio_complete(&pgio);
 	return 0;
 }
 
@@ -269,7 +280,7 @@ static int nfs_pagein_multi(struct nfs_pageio_descriptor *desc)
 	unsigned int offset;
 	int requests = 0;
 	int ret = 0;
-	struct pnfs_layout_segment *lseg;
+	struct pnfs_layout_segment *lseg = desc->pg_lseg;
 	LIST_HEAD(list);
 
 	nfs_list_remove_request(req);
@@ -287,10 +298,6 @@ static int nfs_pagein_multi(struct nfs_pageio_descriptor *desc)
 	} while(nbytes != 0);
 	atomic_set(&req->wb_complete, requests);
 
-	BUG_ON(desc->pg_lseg != NULL);
-	lseg = pnfs_update_layout(desc->pg_inode, req->wb_context,
-				  req_offset(req), desc->pg_count,
-				  IOMODE_READ, GFP_KERNEL);
 	ClearPageError(page);
 	offset = 0;
 	nbytes = desc->pg_count;
@@ -324,6 +331,8 @@ out_bad:
 	}
 	SetPageError(page);
 	nfs_readpage_release(req);
+	put_lseg(lseg);
+	desc->pg_lseg = NULL;
 	return -ENOMEM;
 }
 
@@ -352,10 +361,6 @@ static int nfs_pagein_one(struct nfs_pageio_descriptor *desc)
 		*pages++ = req->wb_page;
 	}
 	req = nfs_list_entry(data->pages.next);
-	if ((!lseg) && list_is_singular(&data->pages))
-		lseg = pnfs_update_layout(desc->pg_inode, req->wb_context,
-					  req_offset(req), desc->pg_count,
-					  IOMODE_READ, GFP_KERNEL);
 
 	ret = nfs_read_rpcsetup(req, data, &nfs_read_full_ops, desc->pg_count,
 				0, lseg);
@@ -365,6 +370,20 @@ out:
 	return ret;
 }
 
+int nfs_generic_pg_readpages(struct nfs_pageio_descriptor *desc)
+{
+	if (desc->pg_bsize < PAGE_CACHE_SIZE)
+		return nfs_pagein_multi(desc);
+	return nfs_pagein_one(desc);
+}
+EXPORT_SYMBOL_GPL(nfs_generic_pg_readpages);
+
+
+static const struct nfs_pageio_ops nfs_pageio_read_ops = {
+	.pg_test = nfs_generic_pg_test,
+	.pg_doio = nfs_generic_pg_readpages,
+};
+
 /*
  * This is the callback from RPC telling us whether a reply was
  * received or some error occurred (timeout or socket shutdown).
@@ -635,8 +654,6 @@ int nfs_readpages(struct file *filp, struct address_space *mapping,
 		.pgio = &pgio,
 	};
 	struct inode *inode = mapping->host;
-	struct nfs_server *server = NFS_SERVER(inode);
-	size_t rsize = server->rsize;
 	unsigned long npages;
 	int ret = -ESTALE;
 
@@ -664,10 +681,7 @@ int nfs_readpages(struct file *filp, struct address_space *mapping,
 	if (ret == 0)
 		goto read_complete; /* all pages were read */
 
-	if (rsize < PAGE_CACHE_SIZE)
-		nfs_pageio_init(&pgio, inode, nfs_pagein_multi, rsize, 0);
-	else
-		nfs_pageio_init(&pgio, inode, nfs_pagein_one, rsize, 0);
+	nfs_pageio_init_read(&pgio, inode);
 
 	ret = read_cache_pages(mapping, pages, readpage_async_filler, &desc);
 
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index e268e3b23497..75e7f0e5c1cc 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -914,7 +914,7 @@ static int nfs_flush_multi(struct nfs_pageio_descriptor *desc)
 	unsigned int offset;
 	int requests = 0;
 	int ret = 0;
-	struct pnfs_layout_segment *lseg;
+	struct pnfs_layout_segment *lseg = desc->pg_lseg;
 	LIST_HEAD(list);
 
 	nfs_list_remove_request(req);
@@ -938,10 +938,6 @@ static int nfs_flush_multi(struct nfs_pageio_descriptor *desc)
 	} while (nbytes != 0);
 	atomic_set(&req->wb_complete, requests);
 
-	BUG_ON(desc->pg_lseg);
-	lseg = pnfs_update_layout(desc->pg_inode, req->wb_context,
-				  req_offset(req), desc->pg_count,
-				  IOMODE_RW, GFP_NOFS);
 	ClearPageError(page);
 	offset = 0;
 	nbytes = desc->pg_count;
@@ -974,9 +970,16 @@ out_bad:
 		nfs_writedata_free(data);
 	}
 	nfs_redirty_request(req);
+	put_lseg(lseg);
+	desc->pg_lseg = NULL;
 	return -ENOMEM;
 }
 
+static const struct nfs_pageio_ops nfs_flush_multi_ops = {
+	.pg_test = nfs_generic_pg_test,
+	.pg_doio = nfs_flush_multi,
+};
+
 /*
  * Create an RPC task for the given write request and kick it.
  * The page must have been locked by the caller.
@@ -1014,10 +1017,6 @@ static int nfs_flush_one(struct nfs_pageio_descriptor *desc)
 		*pages++ = req->wb_page;
 	}
 	req = nfs_list_entry(data->pages.next);
-	if ((!lseg) && list_is_singular(&data->pages))
-		lseg = pnfs_update_layout(desc->pg_inode, req->wb_context,
-					  req_offset(req), desc->pg_count,
-					  IOMODE_RW, GFP_NOFS);
 
 	if ((desc->pg_ioflags & FLUSH_COND_STABLE) &&
 	    (desc->pg_moreio || NFS_I(desc->pg_inode)->ncommit))
@@ -1031,15 +1030,32 @@ out:
 	return ret;
 }
 
-static void nfs_pageio_init_write(struct nfs_pageio_descriptor *pgio,
+int nfs_generic_pg_writepages(struct nfs_pageio_descriptor *desc)
+{
+	if (desc->pg_bsize < PAGE_CACHE_SIZE)
+		return nfs_flush_multi(desc);
+	return nfs_flush_one(desc);
+}
+EXPORT_SYMBOL_GPL(nfs_generic_pg_writepages);
+
+static const struct nfs_pageio_ops nfs_pageio_write_ops = {
+	.pg_test = nfs_generic_pg_test,
+	.pg_doio = nfs_generic_pg_writepages,
+};
+
+void nfs_pageio_init_write_mds(struct nfs_pageio_descriptor *pgio,
 				  struct inode *inode, int ioflags)
 {
-	size_t wsize = NFS_SERVER(inode)->wsize;
+	nfs_pageio_init(pgio, inode, &nfs_pageio_write_ops,
+				NFS_SERVER(inode)->wsize, ioflags);
+}
+EXPORT_SYMBOL_GPL(nfs_pageio_init_write_mds);
 
-	if (wsize < PAGE_CACHE_SIZE)
-		nfs_pageio_init(pgio, inode, nfs_flush_multi, wsize, ioflags);
-	else
-		nfs_pageio_init(pgio, inode, nfs_flush_one, wsize, ioflags);
+static void nfs_pageio_init_write(struct nfs_pageio_descriptor *pgio,
+				  struct inode *inode, int ioflags)
+{
+	if (!pnfs_pageio_init_write(pgio, inode, ioflags))
+		nfs_pageio_init_write_mds(pgio, inode, ioflags);
 }
 
 /*