Merge branch 'consolidate-clksrc-i8253' of master.kernel.org:~rmk/linux-2.6-arm into timers/clocksource

Conflicts:
	arch/ia64/kernel/cyclone.c
	arch/mips/kernel/i8253.c
	arch/x86/kernel/i8253.c

Reason: Resolve conflicts so further cleanups do not conflict further

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

30 files changed, 3519 insertions, 1251 deletions

diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c
index 89587573fe50..2f41dccea18e 100644
--- a/fs/nfs/callback_proc.c
+++ b/fs/nfs/callback_proc.c
@@ -188,10 +188,10 @@ static u32 initiate_bulk_draining(struct nfs_client *clp,
                         rv = NFS4ERR_DELAY;
                 list_del_init(&lo->plh_bulk_recall);
                 spin_unlock(&ino->i_lock);
+                pnfs_free_lseg_list(&free_me_list);
                 put_layout_hdr(lo);
                 iput(ino);
         }
-        pnfs_free_lseg_list(&free_me_list);
         return rv;
 }
 
diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c
index 14e0f9371d14..00ecf62ce7c1 100644
--- a/fs/nfs/callback_xdr.c
+++ b/fs/nfs/callback_xdr.c
@@ -241,7 +241,7 @@ static __be32 decode_layoutrecall_args(struct svc_rqst *rqstp,
 
         args->cbl_layout_type = ntohl(*p++);
         /* Depite the spec's xdr, iomode really belongs in the FILE switch,
-         * as it is unuseable and ignored with the other types.
+         * as it is unusable and ignored with the other types.
          */
         iomode = ntohl(*p++);
         args->cbl_layoutchanged = ntohl(*p++);
diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index bd3ca32879e7..139be9647d80 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -82,6 +82,11 @@ retry:
 #endif /* CONFIG_NFS_V4 */
 
 /*
+ * Turn off NFSv4 uid/gid mapping when using AUTH_SYS
+ */
+static int nfs4_disable_idmapping = 0;
+
+/*
  * RPC cruft for NFS
  */
 static struct rpc_version *nfs_version[5] = {
@@ -481,7 +486,12 @@ static struct nfs_client *nfs_match_client(const struct nfs_client_initdata *dat
  * Look up a client by IP address and protocol version
  * - creates a new record if one doesn't yet exist
  */
-static struct nfs_client *nfs_get_client(const struct nfs_client_initdata *cl_init)
+static struct nfs_client *
+nfs_get_client(const struct nfs_client_initdata *cl_init,
+               const struct rpc_timeout *timeparms,
+               const char *ip_addr,
+               rpc_authflavor_t authflavour,
+               int noresvport)
 {
         struct nfs_client *clp, *new = NULL;
         int error;
@@ -512,6 +522,13 @@ install_client:
         clp = new;
         list_add(&clp->cl_share_link, &nfs_client_list);
         spin_unlock(&nfs_client_lock);
+
+        error = cl_init->rpc_ops->init_client(clp, timeparms, ip_addr,
+                                              authflavour, noresvport);
+        if (error < 0) {
+                nfs_put_client(clp);
+                return ERR_PTR(error);
+        }
         dprintk("--> nfs_get_client() = %p [new]\n", clp);
         return clp;
 
@@ -767,9 +784,9 @@ static int nfs_init_server_rpcclient(struct nfs_server *server,
 /*
  * Initialise an NFS2 or NFS3 client
  */
-static int nfs_init_client(struct nfs_client *clp,
-                           const struct rpc_timeout *timeparms,
-                           const struct nfs_parsed_mount_data *data)
+int nfs_init_client(struct nfs_client *clp, const struct rpc_timeout *timeparms,
+                    const char *ip_addr, rpc_authflavor_t authflavour,
+                    int noresvport)
 {
         int error;
 
@@ -784,7 +801,7 @@ static int nfs_init_client(struct nfs_client *clp,
          * - RFC 2623, sec 2.3.2
          */
         error = nfs_create_rpc_client(clp, timeparms, RPC_AUTH_UNIX,
-                                      0, data->flags & NFS_MOUNT_NORESVPORT);
+                                      0, noresvport);
         if (error < 0)
                 goto error;
         nfs_mark_client_ready(clp, NFS_CS_READY);
@@ -820,19 +837,17 @@ static int nfs_init_server(struct nfs_server *server,
                 cl_init.rpc_ops = &nfs_v3_clientops;
 #endif
 
+        nfs_init_timeout_values(&timeparms, data->nfs_server.protocol,
+                        data->timeo, data->retrans);
+
         /* Allocate or find a client reference we can use */
-        clp = nfs_get_client(&cl_init);
+        clp = nfs_get_client(&cl_init, &timeparms, NULL, RPC_AUTH_UNIX,
+                             data->flags & NFS_MOUNT_NORESVPORT);
         if (IS_ERR(clp)) {
                 dprintk("<-- nfs_init_server() = error %ld\n", PTR_ERR(clp));
                 return PTR_ERR(clp);
         }
 
-        nfs_init_timeout_values(&timeparms, data->nfs_server.protocol,
-                        data->timeo, data->retrans);
-        error = nfs_init_client(clp, &timeparms, data);
-        if (error < 0)
-                goto error;
-
         server->nfs_client = clp;
 
         /* Initialise the client representation from the mount data */
@@ -1009,14 +1024,19 @@ static void nfs_server_insert_lists(struct nfs_server *server)
         spin_lock(&nfs_client_lock);
         list_add_tail_rcu(&server->client_link, &clp->cl_superblocks);
         list_add_tail(&server->master_link, &nfs_volume_list);
+        clear_bit(NFS_CS_STOP_RENEW, &clp->cl_res_state);
         spin_unlock(&nfs_client_lock);
 
 }
 
 static void nfs_server_remove_lists(struct nfs_server *server)
 {
+        struct nfs_client *clp = server->nfs_client;
+
         spin_lock(&nfs_client_lock);
         list_del_rcu(&server->client_link);
+        if (clp && list_empty(&clp->cl_superblocks))
+                set_bit(NFS_CS_STOP_RENEW, &clp->cl_res_state);
         list_del(&server->master_link);
         spin_unlock(&nfs_client_lock);
 
@@ -1307,11 +1327,11 @@ static int nfs4_init_client_minor_version(struct nfs_client *clp)
 /*
  * Initialise an NFS4 client record
  */
-static int nfs4_init_client(struct nfs_client *clp,
-                const struct rpc_timeout *timeparms,
-                const char *ip_addr,
-                rpc_authflavor_t authflavour,
-                int flags)
+int nfs4_init_client(struct nfs_client *clp,
+                     const struct rpc_timeout *timeparms,
+                     const char *ip_addr,
+                     rpc_authflavor_t authflavour,
+                     int noresvport)
 {
         int error;
 
@@ -1325,7 +1345,7 @@ static int nfs4_init_client(struct nfs_client *clp,
         clp->rpc_ops = &nfs_v4_clientops;
 
         error = nfs_create_rpc_client(clp, timeparms, authflavour,
-                                      1, flags & NFS_MOUNT_NORESVPORT);
+                                      1, noresvport);
         if (error < 0)
                 goto error;
         strlcpy(clp->cl_ipaddr, ip_addr, sizeof(clp->cl_ipaddr));
@@ -1378,27 +1398,71 @@ static int nfs4_set_client(struct nfs_server *server,
         dprintk("--> nfs4_set_client()\n");
 
         /* Allocate or find a client reference we can use */
-        clp = nfs_get_client(&cl_init);
+        clp = nfs_get_client(&cl_init, timeparms, ip_addr, authflavour,
+                             server->flags & NFS_MOUNT_NORESVPORT);
         if (IS_ERR(clp)) {
                 error = PTR_ERR(clp);
                 goto error;
         }
-        error = nfs4_init_client(clp, timeparms, ip_addr, authflavour,
-                                        server->flags);
-        if (error < 0)
-                goto error_put;
+
+        /*
+         * Query for the lease time on clientid setup or renewal
+         *
+         * Note that this will be set on nfs_clients that were created
+         * only for the DS role and did not set this bit, but now will
+         * serve a dual role.
+         */
+        set_bit(NFS_CS_CHECK_LEASE_TIME, &clp->cl_res_state);
 
         server->nfs_client = clp;
         dprintk("<-- nfs4_set_client() = 0 [new %p]\n", clp);
         return 0;
-
-error_put:
-        nfs_put_client(clp);
 error:
         dprintk("<-- nfs4_set_client() = xerror %d\n", error);
         return error;
 }
 
+/*
+ * Set up a pNFS Data Server client.
+ *
+ * Return any existing nfs_client that matches server address,port,version
+ * and minorversion.
+ *
+ * For a new nfs_client, use a soft mount (default), a low retrans and a
+ * low timeout interval so that if a connection is lost, we retry through
+ * the MDS.
+ */
+struct nfs_client *nfs4_set_ds_client(struct nfs_client* mds_clp,
+                const struct sockaddr *ds_addr,
+                int ds_addrlen, int ds_proto)
+{
+        struct nfs_client_initdata cl_init = {
+                .addr = ds_addr,
+                .addrlen = ds_addrlen,
+                .rpc_ops = &nfs_v4_clientops,
+                .proto = ds_proto,
+                .minorversion = mds_clp->cl_minorversion,
+        };
+        struct rpc_timeout ds_timeout = {
+                .to_initval = 15 * HZ,
+                .to_maxval = 15 * HZ,
+                .to_retries = 1,
+                .to_exponential = 1,
+        };
+        struct nfs_client *clp;
+
+        /*
+         * Set an authflavor equual to the MDS value. Use the MDS nfs_client
+         * cl_ipaddr so as to use the same EXCHANGE_ID co_ownerid as the MDS
+         * (section 13.1 RFC 5661).
+         */
+        clp = nfs_get_client(&cl_init, &ds_timeout, mds_clp->cl_ipaddr,
+                             mds_clp->cl_rpcclient->cl_auth->au_flavor, 0);
+
+        dprintk("<-- %s %p\n", __func__, clp);
+        return clp;
+}
+EXPORT_SYMBOL(nfs4_set_ds_client);
 
 /*
  * Session has been established, and the client marked ready.
@@ -1435,6 +1499,10 @@ static int nfs4_server_common_setup(struct nfs_server *server,
         BUG_ON(!server->nfs_client->rpc_ops);
         BUG_ON(!server->nfs_client->rpc_ops->file_inode_ops);
 
+        /* data servers support only a subset of NFSv4.1 */
+        if (is_ds_only_client(server->nfs_client))
+                return -EPROTONOSUPPORT;
+
         fattr = nfs_alloc_fattr();
         if (fattr == NULL)
                 return -ENOMEM;
@@ -1504,6 +1572,13 @@ static int nfs4_init_server(struct nfs_server *server,
         if (error < 0)
                 goto error;
 
+        /*
+         * Don't use NFS uid/gid mapping if we're using AUTH_SYS or lower
+         * authentication.
+         */
+        if (nfs4_disable_idmapping && data->auth_flavors[0] == RPC_AUTH_UNIX)
+                server->caps |= NFS_CAP_UIDGID_NOMAP;
+
         if (data->rsize)
                 server->rsize = nfs_block_size(data->rsize, NULL);
         if (data->wsize)
@@ -1921,3 +1996,7 @@ void nfs_fs_proc_exit(void)
 }
 
 #endif /* CONFIG_PROC_FS */
+
+module_param(nfs4_disable_idmapping, bool, 0644);
+MODULE_PARM_DESC(nfs4_disable_idmapping,
+                "Turn off NFSv4 idmapping when using 'sec=sys'");
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 2c3eb33b904d..7237672216c8 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -44,6 +44,7 @@
 /* #define NFS_DEBUG_VERBOSE 1 */
 
 static int nfs_opendir(struct inode *, struct file *);
+static int nfs_closedir(struct inode *, struct file *);
 static int nfs_readdir(struct file *, void *, filldir_t);
 static struct dentry *nfs_lookup(struct inode *, struct dentry *, struct nameidata *);
 static int nfs_create(struct inode *, struct dentry *, int, struct nameidata *);
@@ -64,7 +65,7 @@ const struct file_operations nfs_dir_operations = {
         .read           = generic_read_dir,
         .readdir        = nfs_readdir,
         .open           = nfs_opendir,
-        .release        = nfs_release,
+        .release        = nfs_closedir,
         .fsync          = nfs_fsync_dir,
 };
 
@@ -133,13 +134,35 @@ const struct inode_operations nfs4_dir_inode_operations = {
 
 #endif /* CONFIG_NFS_V4 */
 
+static struct nfs_open_dir_context *alloc_nfs_open_dir_context(struct rpc_cred *cred)
+{
+        struct nfs_open_dir_context *ctx;
+        ctx = kmalloc(sizeof(*ctx), GFP_KERNEL);
+        if (ctx != NULL) {
+                ctx->duped = 0;
+                ctx->dir_cookie = 0;
+                ctx->dup_cookie = 0;
+                ctx->cred = get_rpccred(cred);
+        } else
+                ctx = ERR_PTR(-ENOMEM);
+        return ctx;
+}
+
+static void put_nfs_open_dir_context(struct nfs_open_dir_context *ctx)
+{
+        put_rpccred(ctx->cred);
+        kfree(ctx);
+}
+
 /*
  * Open file
  */
 static int
 nfs_opendir(struct inode *inode, struct file *filp)
 {
-        int res;
+        int res = 0;
+        struct nfs_open_dir_context *ctx;
+        struct rpc_cred *cred;
 
         dfprintk(FILE, "NFS: open dir(%s/%s)\n",
                         filp->f_path.dentry->d_parent->d_name.name,
@@ -147,8 +170,15 @@ nfs_opendir(struct inode *inode, struct file *filp)
 
         nfs_inc_stats(inode, NFSIOS_VFSOPEN);
 
-        /* Call generic open code in order to cache credentials */
-        res = nfs_open(inode, filp);
+        cred = rpc_lookup_cred();
+        if (IS_ERR(cred))
+                return PTR_ERR(cred);
+        ctx = alloc_nfs_open_dir_context(cred);
+        if (IS_ERR(ctx)) {
+                res = PTR_ERR(ctx);
+                goto out;
+        }
+        filp->private_data = ctx;
         if (filp->f_path.dentry == filp->f_path.mnt->mnt_root) {
                 /* This is a mountpoint, so d_revalidate will never
                  * have been called, so we need to refresh the
@@ -156,9 +186,18 @@ nfs_opendir(struct inode *inode, struct file *filp)
                  */
                 __nfs_revalidate_inode(NFS_SERVER(inode), inode);
         }
+out:
+        put_rpccred(cred);
         return res;
 }
 
+static int
+nfs_closedir(struct inode *inode, struct file *filp)
+{
+        put_nfs_open_dir_context(filp->private_data);
+        return 0;
+}
+
 struct nfs_cache_array_entry {
         u64 cookie;
         u64 ino;
@@ -284,19 +323,20 @@ int nfs_readdir_search_for_pos(struct nfs_cache_array *array, nfs_readdir_descri
 {
         loff_t diff = desc->file->f_pos - desc->current_index;
         unsigned int index;
+        struct nfs_open_dir_context *ctx = desc->file->private_data;
 
         if (diff < 0)
                 goto out_eof;
         if (diff >= array->size) {
                 if (array->eof_index >= 0)
                         goto out_eof;
-                desc->current_index += array->size;
                 return -EAGAIN;
         }
 
         index = (unsigned int)diff;
         *desc->dir_cookie = array->array[index].cookie;
         desc->cache_entry_index = index;
+        ctx->duped = 0;
         return 0;
 out_eof:
         desc->eof = 1;
@@ -307,10 +347,18 @@ static
 int nfs_readdir_search_for_cookie(struct nfs_cache_array *array, nfs_readdir_descriptor_t *desc)
 {
         int i;
+        loff_t new_pos;
         int status = -EAGAIN;
+        struct nfs_open_dir_context *ctx = desc->file->private_data;
 
         for (i = 0; i < array->size; i++) {
                 if (array->array[i].cookie == *desc->dir_cookie) {
+                        new_pos = desc->current_index + i;
+                        if (new_pos < desc->file->f_pos) {
+                                ctx->dup_cookie = *desc->dir_cookie;
+                                ctx->duped = 1;
+                        }
+                        desc->file->f_pos = new_pos;
                         desc->cache_entry_index = i;
                         return 0;
                 }
@@ -342,6 +390,7 @@ int nfs_readdir_search_array(nfs_readdir_descriptor_t *desc)
 
         if (status == -EAGAIN) {
                 desc->last_cookie = array->last_cookie;
+                desc->current_index += array->size;
                 desc->page_index++;
         }
         nfs_readdir_release_array(desc->page);
@@ -354,7 +403,8 @@ static
 int nfs_readdir_xdr_filler(struct page **pages, nfs_readdir_descriptor_t *desc,
                         struct nfs_entry *entry, struct file *file, struct inode *inode)
 {
-        struct rpc_cred *cred = nfs_file_cred(file);
+        struct nfs_open_dir_context *ctx = file->private_data;
+        struct rpc_cred *cred = ctx->cred;
         unsigned long   timestamp, gencount;
         int             error;
 
@@ -693,6 +743,20 @@ int nfs_do_filldir(nfs_readdir_descriptor_t *desc, void *dirent,
         int i = 0;
         int res = 0;
         struct nfs_cache_array *array = NULL;
+        struct nfs_open_dir_context *ctx = file->private_data;
+
+        if (ctx->duped != 0 && ctx->dup_cookie == *desc->dir_cookie) {
+                if (printk_ratelimit()) {
+                        pr_notice("NFS: directory %s/%s contains a readdir loop.  "
+                                "Please contact your server vendor.  "
+                                "Offending cookie: %llu\n",
+                                file->f_dentry->d_parent->d_name.name,
+                                file->f_dentry->d_name.name,
+                                *desc->dir_cookie);
+                }
+                res = -ELOOP;
+                goto out;
+        }
 
         array = nfs_readdir_get_array(desc->page);
         if (IS_ERR(array)) {
@@ -785,6 +849,7 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
         struct inode    *inode = dentry->d_inode;
         nfs_readdir_descriptor_t my_desc,
                         *desc = &my_desc;
+        struct nfs_open_dir_context *dir_ctx = filp->private_data;
         int res;
 
         dfprintk(FILE, "NFS: readdir(%s/%s) starting at cookie %llu\n",
@@ -801,7 +866,7 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
         memset(desc, 0, sizeof(*desc));
 
         desc->file = filp;
-        desc->dir_cookie = &nfs_file_open_context(filp)->dir_cookie;
+        desc->dir_cookie = &dir_ctx->dir_cookie;
         desc->decode = NFS_PROTO(inode)->decode_dirent;
         desc->plus = NFS_USE_READDIRPLUS(inode);
 
@@ -853,6 +918,7 @@ static loff_t nfs_llseek_dir(struct file *filp, loff_t offset, int origin)
 {
         struct dentry *dentry = filp->f_path.dentry;
         struct inode *inode = dentry->d_inode;
+        struct nfs_open_dir_context *dir_ctx = filp->private_data;
 
         dfprintk(FILE, "NFS: llseek dir(%s/%s, %lld, %d)\n",
                         dentry->d_parent->d_name.name,
@@ -872,7 +938,8 @@ static loff_t nfs_llseek_dir(struct file *filp, loff_t offset, int origin)
         }
         if (offset != filp->f_pos) {
                 filp->f_pos = offset;
-                nfs_file_open_context(filp)->dir_cookie = 0;
+                dir_ctx->dir_cookie = 0;
+                dir_ctx->duped = 0;
         }
 out:
         mutex_unlock(&inode->i_mutex);
@@ -1068,7 +1135,7 @@ static int nfs_lookup_revalidate(struct dentry *dentry, struct nameidata *nd)
         if (fhandle == NULL || fattr == NULL)
                 goto out_error;
 
-        error = NFS_PROTO(dir)->lookup(dir, &dentry->d_name, fhandle, fattr);
+        error = NFS_PROTO(dir)->lookup(NFS_SERVER(dir)->client, dir, &dentry->d_name, fhandle, fattr);
         if (error)
                 goto out_bad;
         if (nfs_compare_fh(NFS_FH(inode), fhandle))
@@ -1169,11 +1236,23 @@ static void nfs_dentry_iput(struct dentry *dentry, struct inode *inode)
         iput(inode);
 }
 
+static void nfs_d_release(struct dentry *dentry)
+{
+        /* free cached devname value, if it survived that far */
+        if (unlikely(dentry->d_fsdata)) {
+                if (dentry->d_flags & DCACHE_NFSFS_RENAMED)
+                        WARN_ON(1);
+                else
+                        kfree(dentry->d_fsdata);
+        }
+}
+
 const struct dentry_operations nfs_dentry_operations = {
         .d_revalidate   = nfs_lookup_revalidate,
         .d_delete       = nfs_dentry_delete,
         .d_iput         = nfs_dentry_iput,
         .d_automount    = nfs_d_automount,
+        .d_release      = nfs_d_release,
 };
 
 static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, struct nameidata *nd)
@@ -1212,7 +1291,7 @@ static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, stru
         parent = dentry->d_parent;
         /* Protect against concurrent sillydeletes */
         nfs_block_sillyrename(parent);
-        error = NFS_PROTO(dir)->lookup(dir, &dentry->d_name, fhandle, fattr);
+        error = NFS_PROTO(dir)->lookup(NFS_SERVER(dir)->client, dir, &dentry->d_name, fhandle, fattr);
         if (error == -ENOENT)
                 goto no_entry;
         if (error < 0) {
@@ -1248,6 +1327,7 @@ const struct dentry_operations nfs4_dentry_operations = {
         .d_delete       = nfs_dentry_delete,
         .d_iput         = nfs_dentry_iput,
         .d_automount    = nfs_d_automount,
+        .d_release      = nfs_d_release,
 };
 
 /*
@@ -1549,7 +1629,7 @@ int nfs_instantiate(struct dentry *dentry, struct nfs_fh *fhandle,
         if (dentry->d_inode)
                 goto out;
         if (fhandle->size == 0) {
-                error = NFS_PROTO(dir)->lookup(dir, &dentry->d_name, fhandle, fattr);
+                error = NFS_PROTO(dir)->lookup(NFS_SERVER(dir)->client, dir, &dentry->d_name, fhandle, fattr);
                 if (error)
                         goto out_error;
         }
diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 9943a75bb6d1..8eea25366717 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -45,6 +45,7 @@
 #include <linux/pagemap.h>
 #include <linux/kref.h>
 #include <linux/slab.h>
+#include <linux/task_io_accounting_ops.h>
 
 #include <linux/nfs_fs.h>
 #include <linux/nfs_page.h>
@@ -649,8 +650,7 @@ static void nfs_direct_write_result(struct rpc_task *task, void *calldata)
 {
         struct nfs_write_data *data = calldata;
 
-        if (nfs_writeback_done(task, data) != 0)
-                return;
+        nfs_writeback_done(task, data);
 }
 
 /*
@@ -938,6 +938,8 @@ ssize_t nfs_file_direct_read(struct kiocb *iocb, const struct iovec *iov,
         if (retval)
                 goto out;
 
+        task_io_account_read(count);
+
         retval = nfs_direct_read(iocb, iov, nr_segs, pos);
         if (retval > 0)
                 iocb->ki_pos = pos + retval;
@@ -999,6 +1001,8 @@ ssize_t nfs_file_direct_write(struct kiocb *iocb, const struct iovec *iov,
         if (retval)
                 goto out;
 
+        task_io_account_write(count);
+
         retval = nfs_direct_write(iocb, iov, nr_segs, pos, count);
 
         if (retval > 0)
diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index 7bf029ef4084..2f093ed16980 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -301,7 +301,7 @@ nfs_file_mmap(struct file * file, struct vm_area_struct * vma)
  * disk, but it retrieves and clears ctx->error after synching, despite
  * the two being set at the same time in nfs_context_set_write_error().
  * This is because the former is used to notify the _next_ call to
- * nfs_file_write() that a write error occured, and hence cause it to
+ * nfs_file_write() that a write error occurred, and hence cause it to
  * fall back to doing a synchronous write.
  */
 static int
@@ -326,6 +326,9 @@ nfs_file_fsync(struct file *file, int datasync)
                 ret = xchg(&ctx->error, 0);
         if (!ret && status < 0)
                 ret = status;
+        if (!ret && !datasync)
+                /* application has asked for meta-data sync */
+                ret = pnfs_layoutcommit_inode(inode, true);
         return ret;
 }
 
@@ -387,10 +390,6 @@ static int nfs_write_begin(struct file *file, struct address_space *mapping,
                 file->f_path.dentry->d_name.name,
                 mapping->host->i_ino, len, (long long) pos);
 
-        pnfs_update_layout(mapping->host,
-                           nfs_file_open_context(file),
-                           IOMODE_RW);
-
 start:
         /*
          * Prevent starvation issues if someone is doing a consistency
diff --git a/fs/nfs/getroot.c b/fs/nfs/getroot.c
index b5ffe8fa291f..dcb61548887f 100644
--- a/fs/nfs/getroot.c
+++ b/fs/nfs/getroot.c
@@ -75,18 +75,25 @@ static int nfs_superblock_set_dummy_root(struct super_block *sb, struct inode *i
 /*
  * get an NFS2/NFS3 root dentry from the root filehandle
  */
-struct dentry *nfs_get_root(struct super_block *sb, struct nfs_fh *mntfh)
+struct dentry *nfs_get_root(struct super_block *sb, struct nfs_fh *mntfh,
+                            const char *devname)
 {
         struct nfs_server *server = NFS_SB(sb);
         struct nfs_fsinfo fsinfo;
         struct dentry *ret;
         struct inode *inode;
+        void *name = kstrdup(devname, GFP_KERNEL);
         int error;
 
+        if (!name)
+                return ERR_PTR(-ENOMEM);
+
         /* get the actual root for this mount */
         fsinfo.fattr = nfs_alloc_fattr();
-        if (fsinfo.fattr == NULL)
+        if (fsinfo.fattr == NULL) {
+                kfree(name);
                 return ERR_PTR(-ENOMEM);
+        }
 
         error = server->nfs_client->rpc_ops->getroot(server, mntfh, &fsinfo);
         if (error < 0) {
@@ -119,7 +126,15 @@ struct dentry *nfs_get_root(struct super_block *sb, struct nfs_fh *mntfh)
         }
 
         security_d_instantiate(ret, inode);
+        spin_lock(&ret->d_lock);
+        if (IS_ROOT(ret) && !(ret->d_flags & DCACHE_NFSFS_RENAMED)) {
+                ret->d_fsdata = name;
+                name = NULL;
+        }
+        spin_unlock(&ret->d_lock);
 out:
+        if (name)
+                kfree(name);
         nfs_free_fattr(fsinfo.fattr);
         return ret;
 }
@@ -169,27 +184,35 @@ out:
 /*
  * get an NFS4 root dentry from the root filehandle
  */
-struct dentry *nfs4_get_root(struct super_block *sb, struct nfs_fh *mntfh)
+struct dentry *nfs4_get_root(struct super_block *sb, struct nfs_fh *mntfh,
+                             const char *devname)
 {
         struct nfs_server *server = NFS_SB(sb);
         struct nfs_fattr *fattr = NULL;
         struct dentry *ret;
         struct inode *inode;
+        void *name = kstrdup(devname, GFP_KERNEL);
         int error;
 
         dprintk("--> nfs4_get_root()\n");
 
+        if (!name)
+                return ERR_PTR(-ENOMEM);
+
         /* get the info about the server and filesystem */
         error = nfs4_server_capabilities(server, mntfh);
         if (error < 0) {
                 dprintk("nfs_get_root: getcaps error = %d\n",
                         -error);
+                kfree(name);
                 return ERR_PTR(error);
         }
 
         fattr = nfs_alloc_fattr();
-        if (fattr == NULL)
-                return ERR_PTR(-ENOMEM);;
+        if (fattr == NULL) {
+                kfree(name);
+                return ERR_PTR(-ENOMEM);
+        }
 
         /* get the actual root for this mount */
         error = server->nfs_client->rpc_ops->getattr(server, mntfh, fattr);
@@ -199,6 +222,10 @@ struct dentry *nfs4_get_root(struct super_block *sb, struct nfs_fh *mntfh)
                 goto out;
         }
 
+        if (fattr->valid & NFS_ATTR_FATTR_FSID &&
+            !nfs_fsid_equal(&server->fsid, &fattr->fsid))
+                memcpy(&server->fsid, &fattr->fsid, sizeof(server->fsid));
+
         inode = nfs_fhget(sb, mntfh, fattr);
         if (IS_ERR(inode)) {
                 dprintk("nfs_get_root: get root inode failed\n");
@@ -223,8 +250,15 @@ struct dentry *nfs4_get_root(struct super_block *sb, struct nfs_fh *mntfh)
         }
 
         security_d_instantiate(ret, inode);
-
+        spin_lock(&ret->d_lock);
+        if (IS_ROOT(ret) && !(ret->d_flags & DCACHE_NFSFS_RENAMED)) {
+                ret->d_fsdata = name;
+                name = NULL;
+        }
+        spin_unlock(&ret->d_lock);
 out:
+        if (name)
+                kfree(name);
         nfs_free_fattr(fattr);
         dprintk("<-- nfs4_get_root()\n");
         return ret;
diff --git a/fs/nfs/idmap.c b/fs/nfs/idmap.c
index 18696882f1c6..79664a1025af 100644
--- a/fs/nfs/idmap.c
+++ b/fs/nfs/idmap.c
@@ -33,16 +33,41 @@
  *  NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
  *  SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  */
+#include <linux/types.h>
+#include <linux/string.h>
+#include <linux/kernel.h>
+
+static int nfs_map_string_to_numeric(const char *name, size_t namelen, __u32 *res)
+{
+        unsigned long val;
+        char buf[16];
+
+        if (memchr(name, '@', namelen) != NULL || namelen >= sizeof(buf))
+                return 0;
+        memcpy(buf, name, namelen);
+        buf[namelen] = '\0';
+        if (strict_strtoul(buf, 0, &val) != 0)
+                return 0;
+        *res = val;
+        return 1;
+}
+
+static int nfs_map_numeric_to_string(__u32 id, char *buf, size_t buflen)
+{
+        return snprintf(buf, buflen, "%u", id);
+}
 
 #ifdef CONFIG_NFS_USE_NEW_IDMAPPER
 
 #include <linux/slab.h>
 #include <linux/cred.h>
+#include <linux/sunrpc/sched.h>
+#include <linux/nfs4.h>
+#include <linux/nfs_fs_sb.h>
 #include <linux/nfs_idmap.h>
 #include <linux/keyctl.h>
 #include <linux/key-type.h>
 #include <linux/rcupdate.h>
-#include <linux/kernel.h>
 #include <linux/err.h>
 
 #include <keys/user-type.h>
@@ -219,23 +244,39 @@ static int nfs_idmap_lookup_id(const char *name, size_t namelen,
         return ret;
 }
 
-int nfs_map_name_to_uid(struct nfs_client *clp, const char *name, size_t namelen, __u32 *uid)
+int nfs_map_name_to_uid(const struct nfs_server *server, const char *name, size_t namelen, __u32 *uid)
 {
+        if (nfs_map_string_to_numeric(name, namelen, uid))
+                return 0;
         return nfs_idmap_lookup_id(name, namelen, "uid", uid);
 }
 
-int nfs_map_group_to_gid(struct nfs_client *clp, const char *name, size_t namelen, __u32 *gid)
+int nfs_map_group_to_gid(const struct nfs_server *server, const char *name, size_t namelen, __u32 *gid)
 {
+        if (nfs_map_string_to_numeric(name, namelen, gid))
+                return 0;
         return nfs_idmap_lookup_id(name, namelen, "gid", gid);
 }
 
-int nfs_map_uid_to_name(struct nfs_client *clp, __u32 uid, char *buf, size_t buflen)
+int nfs_map_uid_to_name(const struct nfs_server *server, __u32 uid, char *buf, size_t buflen)
 {
-        return nfs_idmap_lookup_name(uid, "user", buf, buflen);
+        int ret = -EINVAL;
+
+        if (!(server->caps & NFS_CAP_UIDGID_NOMAP))
+                ret = nfs_idmap_lookup_name(uid, "user", buf, buflen);
+        if (ret < 0)
+                ret = nfs_map_numeric_to_string(uid, buf, buflen);
+        return ret;
 }
-int nfs_map_gid_to_group(struct nfs_client *clp, __u32 gid, char *buf, size_t buflen)
+int nfs_map_gid_to_group(const struct nfs_server *server, __u32 gid, char *buf, size_t buflen)
 {
-        return nfs_idmap_lookup_name(gid, "group", buf, buflen);
+        int ret = -EINVAL;
+
+        if (!(server->caps & NFS_CAP_UIDGID_NOMAP))
+                ret = nfs_idmap_lookup_name(gid, "group", buf, buflen);
+        if (ret < 0)
+                ret = nfs_map_numeric_to_string(gid, buf, buflen);
+        return ret;
 }
 
 #else  /* CONFIG_NFS_USE_NEW_IDMAPPER not defined */
@@ -243,7 +284,6 @@ int nfs_map_gid_to_group(struct nfs_client *clp, __u32 gid, char *buf, size_t bu
 #include <linux/module.h>
 #include <linux/mutex.h>
 #include <linux/init.h>
-#include <linux/types.h>
 #include <linux/slab.h>
 #include <linux/socket.h>
 #include <linux/in.h>
@@ -695,31 +735,45 @@ static unsigned int fnvhash32(const void *buf, size_t buflen)
         return hash;
 }
 
-int nfs_map_name_to_uid(struct nfs_client *clp, const char *name, size_t namelen, __u32 *uid)
+int nfs_map_name_to_uid(const struct nfs_server *server, const char *name, size_t namelen, __u32 *uid)
 {
-        struct idmap *idmap = clp->cl_idmap;
+        struct idmap *idmap = server->nfs_client->cl_idmap;
 
+        if (nfs_map_string_to_numeric(name, namelen, uid))
+                return 0;
         return nfs_idmap_id(idmap, &idmap->idmap_user_hash, name, namelen, uid);
 }
 
-int nfs_map_group_to_gid(struct nfs_client *clp, const char *name, size_t namelen, __u32 *uid)
+int nfs_map_group_to_gid(const struct nfs_server *server, const char *name, size_t namelen, __u32 *uid)
 {
-        struct idmap *idmap = clp->cl_idmap;
+        struct idmap *idmap = server->nfs_client->cl_idmap;
 
+        if (nfs_map_string_to_numeric(name, namelen, uid))
+                return 0;
         return nfs_idmap_id(idmap, &idmap->idmap_group_hash, name, namelen, uid);
 }
 
-int nfs_map_uid_to_name(struct nfs_client *clp, __u32 uid, char *buf, size_t buflen)
+int nfs_map_uid_to_name(const struct nfs_server *server, __u32 uid, char *buf, size_t buflen)
 {
-        struct idmap *idmap = clp->cl_idmap;
+        struct idmap *idmap = server->nfs_client->cl_idmap;
+        int ret = -EINVAL;
 
-        return nfs_idmap_name(idmap, &idmap->idmap_user_hash, uid, buf);
+        if (!(server->caps & NFS_CAP_UIDGID_NOMAP))
+                ret = nfs_idmap_name(idmap, &idmap->idmap_user_hash, uid, buf);
+        if (ret < 0)
+                ret = nfs_map_numeric_to_string(uid, buf, buflen);
+        return ret;
 }
-int nfs_map_gid_to_group(struct nfs_client *clp, __u32 uid, char *buf, size_t buflen)
+int nfs_map_gid_to_group(const struct nfs_server *server, __u32 uid, char *buf, size_t buflen)
 {
-        struct idmap *idmap = clp->cl_idmap;
+        struct idmap *idmap = server->nfs_client->cl_idmap;
+        int ret = -EINVAL;
 
-        return nfs_idmap_name(idmap, &idmap->idmap_group_hash, uid, buf);
+        if (!(server->caps & NFS_CAP_UIDGID_NOMAP))
+                ret = nfs_idmap_name(idmap, &idmap->idmap_group_hash, uid, buf);
+        if (ret < 0)
+                ret = nfs_map_numeric_to_string(uid, buf, buflen);
+        return ret;
 }
 
 #endif /* CONFIG_NFS_USE_NEW_IDMAPPER */
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 1cc600e77bb4..57bb31ad7a5e 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -37,6 +37,7 @@
 #include <linux/inet.h>
 #include <linux/nfs_xdr.h>
 #include <linux/slab.h>
+#include <linux/compat.h>
 
 #include <asm/system.h>
 #include <asm/uaccess.h>
@@ -89,7 +90,11 @@ int nfs_wait_bit_killable(void *word)
  */
 u64 nfs_compat_user_ino64(u64 fileid)
 {
-        int ino;
+#ifdef CONFIG_COMPAT
+        compat_ulong_t ino;
+#else   
+        unsigned long ino;
+#endif
 
         if (enable_ino64)
                 return fileid;
@@ -249,7 +254,9 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr)
         struct inode *inode = ERR_PTR(-ENOENT);
         unsigned long hash;
 
-        if ((fattr->valid & NFS_ATTR_FATTR_FILEID) == 0)
+        nfs_attr_check_mountpoint(sb, fattr);
+
+        if ((fattr->valid & NFS_ATTR_FATTR_FILEID) == 0 && (fattr->valid & NFS_ATTR_FATTR_MOUNTPOINT) == 0)
                 goto out_no_inode;
         if ((fattr->valid & NFS_ATTR_FATTR_TYPE) == 0)
                 goto out_no_inode;
@@ -293,8 +300,8 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr)
                         if (nfs_server_capable(inode, NFS_CAP_READDIRPLUS))
                                 set_bit(NFS_INO_ADVISE_RDPLUS, &NFS_I(inode)->flags);
                         /* Deal with crossing mountpoints */
-                        if ((fattr->valid & NFS_ATTR_FATTR_FSID)
-                                        && !nfs_fsid_equal(&NFS_SB(sb)->fsid, &fattr->fsid)) {
+                        if (fattr->valid & NFS_ATTR_FATTR_MOUNTPOINT ||
+                                        fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL) {
                                 if (fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL)
                                         inode->i_op = &nfs_referral_inode_operations;
                                 else
@@ -634,7 +641,6 @@ struct nfs_open_context *alloc_nfs_open_context(struct path *path, struct rpc_cr
                 ctx->mode = f_mode;
                 ctx->flags = 0;
                 ctx->error = 0;
-                ctx->dir_cookie = 0;
                 nfs_init_lock_context(&ctx->lock_context);
                 ctx->lock_context.open_context = ctx;
                 INIT_LIST_HEAD(&ctx->list);
@@ -1466,6 +1472,7 @@ static inline void nfs4_init_once(struct nfs_inode *nfsi)
         nfsi->delegation_state = 0;
         init_rwsem(&nfsi->rwsem);
         nfsi->layout = NULL;
+        atomic_set(&nfsi->commits_outstanding, 0);
 #endif
 }
 
@@ -1513,7 +1520,7 @@ static int nfsiod_start(void)
 {
         struct workqueue_struct *wq;
         dprintk("RPC:       creating workqueue nfsiod\n");
-        wq = alloc_workqueue("nfsiod", WQ_RESCUER, 0);
+        wq = alloc_workqueue("nfsiod", WQ_MEM_RECLAIM, 0);
         if (wq == NULL)
                 return -ENOMEM;
         nfsiod_workqueue = wq;
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index cf9fdbdabc67..ce118ce885dd 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -39,6 +39,12 @@ static inline int nfs4_has_persistent_session(const struct nfs_client *clp)
         return 0;
 }
 
+static inline void nfs_attr_check_mountpoint(struct super_block *parent, struct nfs_fattr *fattr)
+{
+        if (!nfs_fsid_equal(&NFS_SB(parent)->fsid, &fattr->fsid))
+                fattr->valid |= NFS_ATTR_FATTR_MOUNTPOINT;
+}
+
 struct nfs_clone_mount {
         const struct super_block *sb;
         const struct dentry *dentry;
@@ -148,6 +154,9 @@ extern struct nfs_server *nfs_clone_server(struct nfs_server *,
                                            struct nfs_fattr *);
 extern void nfs_mark_client_ready(struct nfs_client *clp, int state);
 extern int nfs4_check_client_ready(struct nfs_client *clp);
+extern struct nfs_client *nfs4_set_ds_client(struct nfs_client* mds_clp,
+                                             const struct sockaddr *ds_addr,
+                                             int ds_addrlen, int ds_proto);
 #ifdef CONFIG_PROC_FS
 extern int __init nfs_fs_proc_init(void);
 extern void nfs_fs_proc_exit(void);
@@ -163,10 +172,10 @@ static inline void nfs_fs_proc_exit(void)
 
 /* nfs4namespace.c */
 #ifdef CONFIG_NFS_V4
-extern struct vfsmount *nfs_do_refmount(const struct vfsmount *mnt_parent, struct dentry *dentry);
+extern struct vfsmount *nfs_do_refmount(struct dentry *dentry);
 #else
 static inline
-struct vfsmount *nfs_do_refmount(const struct vfsmount *mnt_parent, struct dentry *dentry)
+struct vfsmount *nfs_do_refmount(struct dentry *dentry)
 {
         return ERR_PTR(-ENOENT);
 }
@@ -211,10 +220,17 @@ extern const u32 nfs41_maxwrite_overhead;
 /* nfs4proc.c */
 #ifdef CONFIG_NFS_V4
 extern struct rpc_procinfo nfs4_procedures[];
+void nfs_fixup_secinfo_attributes(struct nfs_fattr *, struct nfs_fh *);
 #endif
 
+extern int nfs4_init_ds_session(struct nfs_client *clp);
+
 /* proc.c */
 void nfs_close_context(struct nfs_open_context *ctx, int is_sync);
+extern int nfs_init_client(struct nfs_client *clp,
+                           const struct rpc_timeout *timeparms,
+                           const char *ip_addr, rpc_authflavor_t authflavour,
+                           int noresvport);
 
 /* dir.c */
 extern int nfs_access_cache_shrinker(struct shrinker *shrink,
@@ -247,25 +263,45 @@ extern void nfs_sb_active(struct super_block *sb);
 extern void nfs_sb_deactive(struct super_block *sb);
 
 /* namespace.c */
-extern char *nfs_path(const char *base,
-                      const struct dentry *droot,
-                      const struct dentry *dentry,
+extern char *nfs_path(char **p, struct dentry *dentry,
                       char *buffer, ssize_t buflen);
 extern struct vfsmount *nfs_d_automount(struct path *path);
 
 /* getroot.c */
-extern struct dentry *nfs_get_root(struct super_block *, struct nfs_fh *);
+extern struct dentry *nfs_get_root(struct super_block *, struct nfs_fh *,
+                                   const char *);
 #ifdef CONFIG_NFS_V4
-extern struct dentry *nfs4_get_root(struct super_block *, struct nfs_fh *);
+extern struct dentry *nfs4_get_root(struct super_block *, struct nfs_fh *,
+                                    const char *);
 
 extern int nfs4_get_rootfh(struct nfs_server *server, struct nfs_fh *mntfh);
 #endif
 
 /* read.c */
+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);
 
 /* write.c */
+extern void nfs_commit_free(struct nfs_write_data *p);
+extern int nfs_initiate_write(struct nfs_write_data *data,
+                              struct rpc_clnt *clnt,
+                              const struct rpc_call_ops *call_ops,
+                              int how);
 extern void nfs_write_prepare(struct rpc_task *task, void *calldata);
+extern int nfs_initiate_commit(struct nfs_write_data *data,
+                               struct rpc_clnt *clnt,
+                               const struct rpc_call_ops *call_ops,
+                               int how);
+extern void nfs_init_commit(struct nfs_write_data *data,
+                            struct list_head *head,
+                            struct pnfs_layout_segment *lseg);
+void nfs_retry_commit(struct list_head *page_list,
+                      struct pnfs_layout_segment *lseg);
+void nfs_commit_clear_lock(struct nfs_inode *nfsi);
+void nfs_commitdata_release(void *data);
+void nfs_commit_release_pages(struct nfs_write_data *data);
+
 #ifdef CONFIG_MIGRATION
 extern int nfs_migrate_page(struct address_space *,
                 struct page *, struct page *);
@@ -274,12 +310,21 @@ extern int nfs_migrate_page(struct address_space *,
 #endif
 
 /* nfs4proc.c */
-extern int _nfs4_call_sync(struct nfs_server *server,
+extern void nfs4_reset_read(struct rpc_task *task, struct nfs_read_data *data);
+extern int nfs4_init_client(struct nfs_client *clp,
+                            const struct rpc_timeout *timeparms,
+                            const char *ip_addr,
+                            rpc_authflavor_t authflavour,
+                            int noresvport);
+extern void nfs4_reset_write(struct rpc_task *task, struct nfs_write_data *data);
+extern int _nfs4_call_sync(struct rpc_clnt *clnt,
+                           struct nfs_server *server,
                            struct rpc_message *msg,
                            struct nfs4_sequence_args *args,
                            struct nfs4_sequence_res *res,
                            int cache_reply);
-extern int _nfs4_call_sync_session(struct nfs_server *server,
+extern int _nfs4_call_sync_session(struct rpc_clnt *clnt,
+                                   struct nfs_server *server,
                                    struct rpc_message *msg,
                                    struct nfs4_sequence_args *args,
                                    struct nfs4_sequence_res *res,
@@ -288,12 +333,11 @@ extern int _nfs4_call_sync_session(struct nfs_server *server,
 /*
  * Determine the device name as a string
  */
-static inline char *nfs_devname(const struct vfsmount *mnt_parent,
-                                const struct dentry *dentry,
+static inline char *nfs_devname(struct dentry *dentry,
                                 char *buffer, ssize_t buflen)
 {
-        return nfs_path(mnt_parent->mnt_devname, mnt_parent->mnt_root,
-                        dentry, buffer, buflen);
+        char *dummy;
+        return nfs_path(&dummy, dentry, buffer, buflen);
 }
 
 /*
diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c
index f32b8603dca8..1f063bacd285 100644
--- a/fs/nfs/namespace.c
+++ b/fs/nfs/namespace.c
@@ -15,6 +15,7 @@
 #include <linux/string.h>
 #include <linux/sunrpc/clnt.h>
 #include <linux/vfs.h>
+#include <linux/sunrpc/gss_api.h>
 #include "internal.h"
 
 #define NFSDBG_FACILITY         NFSDBG_VFS
@@ -25,33 +26,31 @@ static LIST_HEAD(nfs_automount_list);
 static DECLARE_DELAYED_WORK(nfs_automount_task, nfs_expire_automounts);
 int nfs_mountpoint_expiry_timeout = 500 * HZ;
 
-static struct vfsmount *nfs_do_submount(const struct vfsmount *mnt_parent,
-                                        const struct dentry *dentry,
+static struct vfsmount *nfs_do_submount(struct dentry *dentry,
                                         struct nfs_fh *fh,
-                                        struct nfs_fattr *fattr);
+                                        struct nfs_fattr *fattr,
+                                        rpc_authflavor_t authflavor);
 
 /*
  * nfs_path - reconstruct the path given an arbitrary dentry
- * @base - arbitrary string to prepend to the path
- * @droot - pointer to root dentry for mountpoint
+ * @base - used to return pointer to the end of devname part of path
  * @dentry - pointer to dentry
  * @buffer - result buffer
  * @buflen - length of buffer
  *
- * Helper function for constructing the path from the
- * root dentry to an arbitrary hashed dentry.
+ * Helper function for constructing the server pathname
+ * by arbitrary hashed dentry.
  *
  * This is mainly for use in figuring out the path on the
- * server side when automounting on top of an existing partition.
+ * server side when automounting on top of an existing partition
+ * and in generating /proc/mounts and friends.
  */
-char *nfs_path(const char *base,
-               const struct dentry *droot,
-               const struct dentry *dentry,
-               char *buffer, ssize_t buflen)
+char *nfs_path(char **p, struct dentry *dentry, char *buffer, ssize_t buflen)
 {
         char *end;
         int namelen;
         unsigned seq;
+        const char *base;
 
 rename_retry:
         end = buffer+buflen;
@@ -60,7 +59,10 @@ rename_retry:
 
         seq = read_seqbegin(&rename_lock);
         rcu_read_lock();
-        while (!IS_ROOT(dentry) && dentry != droot) {
+        while (1) {
+                spin_lock(&dentry->d_lock);
+                if (IS_ROOT(dentry))
+                        break;
                 namelen = dentry->d_name.len;
                 buflen -= namelen + 1;
                 if (buflen < 0)
@@ -68,27 +70,47 @@ rename_retry:
                 end -= namelen;
                 memcpy(end, dentry->d_name.name, namelen);
                 *--end = '/';
+                spin_unlock(&dentry->d_lock);
                 dentry = dentry->d_parent;
         }
-        rcu_read_unlock();
-        if (read_seqretry(&rename_lock, seq))
+        if (read_seqretry(&rename_lock, seq)) {
+                spin_unlock(&dentry->d_lock);
+                rcu_read_unlock();
                 goto rename_retry;
+        }
         if (*end != '/') {
-                if (--buflen < 0)
+                if (--buflen < 0) {
+                        spin_unlock(&dentry->d_lock);
+                        rcu_read_unlock();
                         goto Elong;
+                }
                 *--end = '/';
         }
+        *p = end;
+        base = dentry->d_fsdata;
+        if (!base) {
+                spin_unlock(&dentry->d_lock);
+                rcu_read_unlock();
+                WARN_ON(1);
+                return end;
+        }
         namelen = strlen(base);
         /* Strip off excess slashes in base string */
         while (namelen > 0 && base[namelen - 1] == '/')
                 namelen--;
         buflen -= namelen;
-        if (buflen < 0)
+        if (buflen < 0) {
+                spin_unlock(&dentry->d_lock);
+                rcu_read_unlock();
                 goto Elong;
+        }
         end -= namelen;
         memcpy(end, base, namelen);
+        spin_unlock(&dentry->d_lock);
+        rcu_read_unlock();
         return end;
 Elong_unlock:
+        spin_unlock(&dentry->d_lock);
         rcu_read_unlock();
         if (read_seqretry(&rename_lock, seq))
                 goto rename_retry;
@@ -96,6 +118,99 @@ Elong:
         return ERR_PTR(-ENAMETOOLONG);
 }
 
+#ifdef CONFIG_NFS_V4
+static rpc_authflavor_t nfs_find_best_sec(struct nfs4_secinfo_flavors *flavors)
+{
+        struct gss_api_mech *mech;
+        struct xdr_netobj oid;
+        int i;
+        rpc_authflavor_t pseudoflavor = RPC_AUTH_UNIX;
+
+        for (i = 0; i < flavors->num_flavors; i++) {
+                struct nfs4_secinfo_flavor *flavor;
+                flavor = &flavors->flavors[i];
+
+                if (flavor->flavor == RPC_AUTH_NULL || flavor->flavor == RPC_AUTH_UNIX) {
+                        pseudoflavor = flavor->flavor;
+                        break;
+                } else if (flavor->flavor == RPC_AUTH_GSS) {
+                        oid.len  = flavor->gss.sec_oid4.len;
+                        oid.data = flavor->gss.sec_oid4.data;
+                        mech = gss_mech_get_by_OID(&oid);
+                        if (!mech)
+                                continue;
+                        pseudoflavor = gss_svc_to_pseudoflavor(mech, flavor->gss.service);
+                        gss_mech_put(mech);
+                        break;
+                }
+        }
+
+        return pseudoflavor;
+}
+
+static int nfs_negotiate_security(const struct dentry *parent,
+                                  const struct dentry *dentry,
+                                  rpc_authflavor_t *flavor)
+{
+        struct page *page;
+        struct nfs4_secinfo_flavors *flavors;
+        int (*secinfo)(struct inode *, const struct qstr *, struct nfs4_secinfo_flavors *);
+        int ret = -EPERM;
+
+        secinfo = NFS_PROTO(parent->d_inode)->secinfo;
+        if (secinfo != NULL) {
+                page = alloc_page(GFP_KERNEL);
+                if (!page) {
+                        ret = -ENOMEM;
+                        goto out;
+                }
+                flavors = page_address(page);
+                ret = secinfo(parent->d_inode, &dentry->d_name, flavors);
+                *flavor = nfs_find_best_sec(flavors);
+                put_page(page);
+        }
+
+out:
+        return ret;
+}
+
+static int nfs_lookup_with_sec(struct nfs_server *server, struct dentry *parent,
+                               struct dentry *dentry, struct path *path,
+                               struct nfs_fh *fh, struct nfs_fattr *fattr,
+                               rpc_authflavor_t *flavor)
+{
+        struct rpc_clnt *clone;
+        struct rpc_auth *auth;
+        int err;
+
+        err = nfs_negotiate_security(parent, path->dentry, flavor);
+        if (err < 0)
+                goto out;
+        clone  = rpc_clone_client(server->client);
+        auth   = rpcauth_create(*flavor, clone);
+        if (!auth) {
+                err = -EIO;
+                goto out_shutdown;
+        }
+        err = server->nfs_client->rpc_ops->lookup(clone, parent->d_inode,
+                                                  &path->dentry->d_name,
+                                                  fh, fattr);
+out_shutdown:
+        rpc_shutdown_client(clone);
+out:
+        return err;
+}
+#else /* CONFIG_NFS_V4 */
+static inline int nfs_lookup_with_sec(struct nfs_server *server,
+                                      struct dentry *parent, struct dentry *dentry,
+                                      struct path *path, struct nfs_fh *fh,
+                                      struct nfs_fattr *fattr,
+                                      rpc_authflavor_t *flavor)
+{
+        return -EPERM;
+}
+#endif /* CONFIG_NFS_V4 */
+
 /*
  * nfs_d_automount - Handle crossing a mountpoint on the server
  * @path - The mountpoint
@@ -116,6 +231,7 @@ struct vfsmount *nfs_d_automount(struct path *path)
         struct nfs_fh *fh = NULL;
         struct nfs_fattr *fattr = NULL;
         int err;
+        rpc_authflavor_t flavor = RPC_AUTH_UNIX;
 
         dprintk("--> nfs_d_automount()\n");
 
@@ -133,9 +249,11 @@ struct vfsmount *nfs_d_automount(struct path *path)
 
         /* Look it up again to get its attributes */
         parent = dget_parent(path->dentry);
-        err = server->nfs_client->rpc_ops->lookup(parent->d_inode,
+        err = server->nfs_client->rpc_ops->lookup(server->client, parent->d_inode,
                                                   &path->dentry->d_name,
                                                   fh, fattr);
+        if (err == -EPERM && NFS_PROTO(parent->d_inode)->secinfo != NULL)
+                err = nfs_lookup_with_sec(server, parent, path->dentry, path, fh, fattr, &flavor);
         dput(parent);
         if (err != 0) {
                 mnt = ERR_PTR(err);
@@ -143,9 +261,9 @@ struct vfsmount *nfs_d_automount(struct path *path)
         }
 
         if (fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL)
-                mnt = nfs_do_refmount(path->mnt, path->dentry);
+                mnt = nfs_do_refmount(path->dentry);
         else
-                mnt = nfs_do_submount(path->mnt, path->dentry, fh, fattr);
+                mnt = nfs_do_submount(path->dentry, fh, fattr, flavor);
         if (IS_ERR(mnt))
                 goto out;
 
@@ -209,22 +327,23 @@ static struct vfsmount *nfs_do_clone_mount(struct nfs_server *server,
 
 /**
  * nfs_do_submount - set up mountpoint when crossing a filesystem boundary
- * @mnt_parent - mountpoint of parent directory
  * @dentry - parent directory
  * @fh - filehandle for new root dentry
  * @fattr - attributes for new root inode
+ * @authflavor - security flavor to use when performing the mount
  *
  */
-static struct vfsmount *nfs_do_submount(const struct vfsmount *mnt_parent,
-                                        const struct dentry *dentry,
+static struct vfsmount *nfs_do_submount(struct dentry *dentry,
                                         struct nfs_fh *fh,
-                                        struct nfs_fattr *fattr)
+                                        struct nfs_fattr *fattr,
+                                        rpc_authflavor_t authflavor)
 {
         struct nfs_clone_mount mountdata = {
-                .sb = mnt_parent->mnt_sb,
+                .sb = dentry->d_sb,
                 .dentry = dentry,
                 .fh = fh,
                 .fattr = fattr,
+                .authflavor = authflavor,
         };
         struct vfsmount *mnt = ERR_PTR(-ENOMEM);
         char *page = (char *) __get_free_page(GFP_USER);
@@ -237,11 +356,11 @@ static struct vfsmount *nfs_do_submount(const struct vfsmount *mnt_parent,
                         dentry->d_name.name);
         if (page == NULL)
                 goto out;
-        devname = nfs_devname(mnt_parent, dentry, page, PAGE_SIZE);
+        devname = nfs_devname(dentry, page, PAGE_SIZE);
         mnt = (struct vfsmount *)devname;
         if (IS_ERR(devname))
                 goto free_page;
-        mnt = nfs_do_clone_mount(NFS_SB(mnt_parent->mnt_sb), devname, &mountdata);
+        mnt = nfs_do_clone_mount(NFS_SB(dentry->d_sb), devname, &mountdata);
 free_page:
         free_page((unsigned long)page);
 out:
diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c
index ce939c062a52..38053d823eb0 100644
--- a/fs/nfs/nfs3proc.c
+++ b/fs/nfs/nfs3proc.c
@@ -141,7 +141,7 @@ nfs3_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr,
 }
 
 static int
-nfs3_proc_lookup(struct inode *dir, struct qstr *name,
+nfs3_proc_lookup(struct rpc_clnt *clnt, struct inode *dir, struct qstr *name,
                  struct nfs_fh *fhandle, struct nfs_fattr *fattr)
 {
         struct nfs3_diropargs   arg = {
@@ -885,4 +885,5 @@ const struct nfs_rpc_ops nfs_v3_clientops = {
         .lock           = nfs3_proc_lock,
         .clear_acl_cache = nfs3_forget_cached_acls,
         .close_context  = nfs_close_context,
+        .init_client    = nfs_init_client,
 };
diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index 7a7474073148..c4a69833dd0d 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -47,6 +47,7 @@ enum nfs4_client_state {
         NFS4CLNT_LAYOUTRECALL,
         NFS4CLNT_SESSION_RESET,
         NFS4CLNT_RECALL_SLOT,
+        NFS4CLNT_LEASE_CONFIRM,
 };
 
 enum nfs4_session_state {
@@ -57,7 +58,8 @@ enum nfs4_session_state {
 struct nfs4_minor_version_ops {
         u32     minor_version;
 
-        int     (*call_sync)(struct nfs_server *server,
+        int     (*call_sync)(struct rpc_clnt *clnt,
+                        struct nfs_server *server,
                         struct rpc_message *msg,
                         struct nfs4_sequence_args *args,
                         struct nfs4_sequence_res *res,
@@ -252,6 +254,9 @@ static inline struct nfs4_session *nfs4_get_session(const struct nfs_server *ser
 extern int nfs4_setup_sequence(const struct nfs_server *server,
                 struct nfs4_sequence_args *args, struct nfs4_sequence_res *res,
                 int cache_reply, struct rpc_task *task);
+extern int nfs41_setup_sequence(struct nfs4_session *session,
+                struct nfs4_sequence_args *args, struct nfs4_sequence_res *res,
+                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 *);
@@ -259,6 +264,21 @@ 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);
+extern int nfs4_proc_layoutcommit(struct nfs4_layoutcommit_data *data,
+                                  bool sync);
+
+static inline bool
+is_ds_only_client(struct nfs_client *clp)
+{
+        return (clp->cl_exchange_flags & EXCHGID4_FLAG_MASK_PNFS) ==
+                EXCHGID4_FLAG_USE_PNFS_DS;
+}
+
+static inline bool
+is_ds_client(struct nfs_client *clp)
+{
+        return clp->cl_exchange_flags & EXCHGID4_FLAG_USE_PNFS_DS;
+}
 #else /* CONFIG_NFS_v4_1 */
 static inline struct nfs4_session *nfs4_get_session(const struct nfs_server *server)
 {
@@ -276,6 +296,18 @@ static inline int nfs4_init_session(struct nfs_server *server)
 {
         return 0;
 }
+
+static inline bool
+is_ds_only_client(struct nfs_client *clp)
+{
+        return false;
+}
+
+static inline bool
+is_ds_client(struct nfs_client *clp)
+{
+        return false;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 extern const struct nfs4_minor_version_ops *nfs_v4_minor_ops[];
@@ -298,6 +330,11 @@ struct rpc_cred *nfs4_get_renew_cred_locked(struct nfs_client *clp);
 #if defined(CONFIG_NFS_V4_1)
 struct rpc_cred *nfs4_get_machine_cred_locked(struct nfs_client *clp);
 struct rpc_cred *nfs4_get_exchange_id_cred(struct nfs_client *clp);
+extern void nfs4_schedule_session_recovery(struct nfs4_session *);
+#else
+static inline void nfs4_schedule_session_recovery(struct nfs4_session *session)
+{
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 extern struct nfs4_state_owner * nfs4_get_state_owner(struct nfs_server *, struct rpc_cred *);
@@ -307,10 +344,9 @@ extern void nfs4_put_open_state(struct nfs4_state *);
 extern void nfs4_close_state(struct path *, struct nfs4_state *, fmode_t);
 extern void nfs4_close_sync(struct path *, struct nfs4_state *, fmode_t);
 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_lease_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 int nfs4_state_mark_reclaim_reboot(struct nfs_client *clp, struct nfs4_state *state);
+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 nfs4_put_lock_state(struct nfs4_lock_state *lsp);
diff --git a/fs/nfs/nfs4filelayout.c b/fs/nfs/nfs4filelayout.c
index 23f930caf1e2..6f8192f4cfc7 100644
--- a/fs/nfs/nfs4filelayout.c
+++ b/fs/nfs/nfs4filelayout.c
@@ -40,32 +40,370 @@ MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Dean Hildebrand <dhildebz@umich.edu>");
 MODULE_DESCRIPTION("The NFSv4 file layout driver");
 
-static int
-filelayout_set_layoutdriver(struct nfs_server *nfss)
-{
-        int status = pnfs_alloc_init_deviceid_cache(nfss->nfs_client,
-                                                nfs4_fl_free_deviceid_callback);
-        if (status) {
-                printk(KERN_WARNING "%s: deviceid cache could not be "
-                        "initialized\n", __func__);
-                return status;
+#define FILELAYOUT_POLL_RETRY_MAX     (15*HZ)
+
+static loff_t
+filelayout_get_dense_offset(struct nfs4_filelayout_segment *flseg,
+                            loff_t offset)
+{
+        u32 stripe_width = flseg->stripe_unit * flseg->dsaddr->stripe_count;
+        u64 tmp;
+
+        offset -= flseg->pattern_offset;
+        tmp = offset;
+        do_div(tmp, stripe_width);
+
+        return tmp * flseg->stripe_unit + do_div(offset, flseg->stripe_unit);
+}
+
+/* This function is used by the layout driver to calculate the
+ * offset of the file on the dserver based on whether the
+ * layout type is STRIPE_DENSE or STRIPE_SPARSE
+ */
+static loff_t
+filelayout_get_dserver_offset(struct pnfs_layout_segment *lseg, loff_t offset)
+{
+        struct nfs4_filelayout_segment *flseg = FILELAYOUT_LSEG(lseg);
+
+        switch (flseg->stripe_type) {
+        case STRIPE_SPARSE:
+                return offset;
+
+        case STRIPE_DENSE:
+                return filelayout_get_dense_offset(flseg, offset);
+        }
+
+        BUG();
+}
+
+/* For data server errors we don't recover from */
+static void
+filelayout_set_lo_fail(struct pnfs_layout_segment *lseg)
+{
+        if (lseg->pls_range.iomode == IOMODE_RW) {
+                dprintk("%s Setting layout IOMODE_RW fail bit\n", __func__);
+                set_bit(lo_fail_bit(IOMODE_RW), &lseg->pls_layout->plh_flags);
+        } else {
+                dprintk("%s Setting layout IOMODE_READ fail bit\n", __func__);
+                set_bit(lo_fail_bit(IOMODE_READ), &lseg->pls_layout->plh_flags);
+        }
+}
+
+static int filelayout_async_handle_error(struct rpc_task *task,
+                                         struct nfs4_state *state,
+                                         struct nfs_client *clp,
+                                         int *reset)
+{
+        if (task->tk_status >= 0)
+                return 0;
+
+        *reset = 0;
+
+        switch (task->tk_status) {
+        case -NFS4ERR_BADSESSION:
+        case -NFS4ERR_BADSLOT:
+        case -NFS4ERR_BAD_HIGH_SLOT:
+        case -NFS4ERR_DEADSESSION:
+        case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
+        case -NFS4ERR_SEQ_FALSE_RETRY:
+        case -NFS4ERR_SEQ_MISORDERED:
+                dprintk("%s ERROR %d, Reset session. Exchangeid "
+                        "flags 0x%x\n", __func__, task->tk_status,
+                        clp->cl_exchange_flags);
+                nfs4_schedule_session_recovery(clp->cl_session);
+                break;
+        case -NFS4ERR_DELAY:
+        case -NFS4ERR_GRACE:
+        case -EKEYEXPIRED:
+                rpc_delay(task, FILELAYOUT_POLL_RETRY_MAX);
+                break;
+        default:
+                dprintk("%s DS error. Retry through MDS %d\n", __func__,
+                        task->tk_status);
+                *reset = 1;
+                break;
+        }
+        task->tk_status = 0;
+        return -EAGAIN;
+}
+
+/* NFS_PROTO call done callback routines */
+
+static int filelayout_read_done_cb(struct rpc_task *task,
+                                struct nfs_read_data *data)
+{
+        struct nfs_client *clp = data->ds_clp;
+        int reset = 0;
+
+        dprintk("%s DS read\n", __func__);
+
+        if (filelayout_async_handle_error(task, data->args.context->state,
+                                          data->ds_clp, &reset) == -EAGAIN) {
+                dprintk("%s calling restart ds_clp %p ds_clp->cl_session %p\n",
+                        __func__, data->ds_clp, data->ds_clp->cl_session);
+                if (reset) {
+                        filelayout_set_lo_fail(data->lseg);
+                        nfs4_reset_read(task, data);
+                        clp = NFS_SERVER(data->inode)->nfs_client;
+                }
+                nfs_restart_rpc(task, clp);
+                return -EAGAIN;
         }
-        dprintk("%s: deviceid cache has been initialized successfully\n",
-                __func__);
+
         return 0;
 }
 
-/* Clear out the layout by destroying its device list */
-static int
-filelayout_clear_layoutdriver(struct nfs_server *nfss)
+/*
+ * We reference the rpc_cred of the first WRITE that triggers the need for
+ * a LAYOUTCOMMIT, and use it to send the layoutcommit compound.
+ * rfc5661 is not clear about which credential should be used.
+ */
+static void
+filelayout_set_layoutcommit(struct nfs_write_data *wdata)
 {
-        dprintk("--> %s\n", __func__);
+        if (FILELAYOUT_LSEG(wdata->lseg)->commit_through_mds ||
+            wdata->res.verf->committed == NFS_FILE_SYNC)
+                return;
+
+        pnfs_set_layoutcommit(wdata);
+        dprintk("%s ionde %lu pls_end_pos %lu\n", __func__, wdata->inode->i_ino,
+                (unsigned long) wdata->lseg->pls_end_pos);
+}
+
+/*
+ * Call ops for the async read/write cases
+ * In the case of dense layouts, the offset needs to be reset to its
+ * original value.
+ */
+static void filelayout_read_prepare(struct rpc_task *task, void *data)
+{
+        struct nfs_read_data *rdata = (struct nfs_read_data *)data;
+
+        rdata->read_done_cb = filelayout_read_done_cb;
+
+        if (nfs41_setup_sequence(rdata->ds_clp->cl_session,
+                                &rdata->args.seq_args, &rdata->res.seq_res,
+                                0, task))
+                return;
+
+        rpc_call_start(task);
+}
+
+static void filelayout_read_call_done(struct rpc_task *task, void *data)
+{
+        struct nfs_read_data *rdata = (struct nfs_read_data *)data;
+
+        dprintk("--> %s task->tk_status %d\n", __func__, task->tk_status);
+
+        /* Note this may cause RPC to be resent */
+        rdata->mds_ops->rpc_call_done(task, data);
+}
+
+static void filelayout_read_release(void *data)
+{
+        struct nfs_read_data *rdata = (struct nfs_read_data *)data;
+
+        rdata->mds_ops->rpc_release(data);
+}
+
+static int filelayout_write_done_cb(struct rpc_task *task,
+                                struct nfs_write_data *data)
+{
+        int reset = 0;
+
+        if (filelayout_async_handle_error(task, data->args.context->state,
+                                          data->ds_clp, &reset) == -EAGAIN) {
+                struct nfs_client *clp;
+
+                dprintk("%s calling restart ds_clp %p ds_clp->cl_session %p\n",
+                        __func__, data->ds_clp, data->ds_clp->cl_session);
+                if (reset) {
+                        filelayout_set_lo_fail(data->lseg);
+                        nfs4_reset_write(task, data);
+                        clp = NFS_SERVER(data->inode)->nfs_client;
+                } else
+                        clp = data->ds_clp;
+                nfs_restart_rpc(task, clp);
+                return -EAGAIN;
+        }
 
-        if (nfss->nfs_client->cl_devid_cache)
-                pnfs_put_deviceid_cache(nfss->nfs_client);
+        filelayout_set_layoutcommit(data);
         return 0;
 }
 
+/* Fake up some data that will cause nfs_commit_release to retry the writes. */
+static void prepare_to_resend_writes(struct nfs_write_data *data)
+{
+        struct nfs_page *first = nfs_list_entry(data->pages.next);
+
+        data->task.tk_status = 0;
+        memcpy(data->verf.verifier, first->wb_verf.verifier,
+               sizeof(first->wb_verf.verifier));
+        data->verf.verifier[0]++; /* ensure verifier mismatch */
+}
+
+static int filelayout_commit_done_cb(struct rpc_task *task,
+                                     struct nfs_write_data *data)
+{
+        int reset = 0;
+
+        if (filelayout_async_handle_error(task, data->args.context->state,
+                                          data->ds_clp, &reset) == -EAGAIN) {
+                dprintk("%s calling restart ds_clp %p ds_clp->cl_session %p\n",
+                        __func__, data->ds_clp, data->ds_clp->cl_session);
+                if (reset) {
+                        prepare_to_resend_writes(data);
+                        filelayout_set_lo_fail(data->lseg);
+                } else
+                        nfs_restart_rpc(task, data->ds_clp);
+                return -EAGAIN;
+        }
+
+        return 0;
+}
+
+static void filelayout_write_prepare(struct rpc_task *task, void *data)
+{
+        struct nfs_write_data *wdata = (struct nfs_write_data *)data;
+
+        if (nfs41_setup_sequence(wdata->ds_clp->cl_session,
+                                &wdata->args.seq_args, &wdata->res.seq_res,
+                                0, task))
+                return;
+
+        rpc_call_start(task);
+}
+
+static void filelayout_write_call_done(struct rpc_task *task, void *data)
+{
+        struct nfs_write_data *wdata = (struct nfs_write_data *)data;
+
+        /* Note this may cause RPC to be resent */
+        wdata->mds_ops->rpc_call_done(task, data);
+}
+
+static void filelayout_write_release(void *data)
+{
+        struct nfs_write_data *wdata = (struct nfs_write_data *)data;
+
+        wdata->mds_ops->rpc_release(data);
+}
+
+static void filelayout_commit_release(void *data)
+{
+        struct nfs_write_data *wdata = (struct nfs_write_data *)data;
+
+        nfs_commit_release_pages(wdata);
+        if (atomic_dec_and_test(&NFS_I(wdata->inode)->commits_outstanding))
+                nfs_commit_clear_lock(NFS_I(wdata->inode));
+        nfs_commitdata_release(wdata);
+}
+
+struct rpc_call_ops filelayout_read_call_ops = {
+        .rpc_call_prepare = filelayout_read_prepare,
+        .rpc_call_done = filelayout_read_call_done,
+        .rpc_release = filelayout_read_release,
+};
+
+struct rpc_call_ops filelayout_write_call_ops = {
+        .rpc_call_prepare = filelayout_write_prepare,
+        .rpc_call_done = filelayout_write_call_done,
+        .rpc_release = filelayout_write_release,
+};
+
+struct rpc_call_ops filelayout_commit_call_ops = {
+        .rpc_call_prepare = filelayout_write_prepare,
+        .rpc_call_done = filelayout_write_call_done,
+        .rpc_release = filelayout_commit_release,
+};
+
+static enum pnfs_try_status
+filelayout_read_pagelist(struct nfs_read_data *data)
+{
+        struct pnfs_layout_segment *lseg = data->lseg;
+        struct nfs4_pnfs_ds *ds;
+        loff_t offset = data->args.offset;
+        u32 j, idx;
+        struct nfs_fh *fh;
+        int status;
+
+        dprintk("--> %s ino %lu pgbase %u req %Zu@%llu\n",
+                __func__, data->inode->i_ino,
+                data->args.pgbase, (size_t)data->args.count, offset);
+
+        /* 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);
+        ds = nfs4_fl_prepare_ds(lseg, idx);
+        if (!ds) {
+                /* Either layout fh index faulty, or ds connect failed */
+                set_bit(lo_fail_bit(IOMODE_RW), &lseg->pls_layout->plh_flags);
+                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));
+
+        /* No multipath support. Use first DS */
+        data->ds_clp = ds->ds_clp;
+        fh = nfs4_fl_select_ds_fh(lseg, j);
+        if (fh)
+                data->args.fh = fh;
+
+        data->args.offset = filelayout_get_dserver_offset(lseg, offset);
+        data->mds_offset = offset;
+
+        /* Perform an asynchronous read to ds */
+        status = nfs_initiate_read(data, ds->ds_clp->cl_rpcclient,
+                                   &filelayout_read_call_ops);
+        BUG_ON(status != 0);
+        return PNFS_ATTEMPTED;
+}
+
+/* Perform async writes. */
+static enum pnfs_try_status
+filelayout_write_pagelist(struct nfs_write_data *data, int sync)
+{
+        struct pnfs_layout_segment *lseg = data->lseg;
+        struct nfs4_pnfs_ds *ds;
+        loff_t offset = data->args.offset;
+        u32 j, idx;
+        struct nfs_fh *fh;
+        int status;
+
+        /* 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);
+        ds = nfs4_fl_prepare_ds(lseg, idx);
+        if (!ds) {
+                printk(KERN_ERR "%s: prepare_ds failed, use MDS\n", __func__);
+                set_bit(lo_fail_bit(IOMODE_RW), &lseg->pls_layout->plh_flags);
+                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__,
+                data->inode->i_ino, sync, (size_t) data->args.count, offset,
+                ntohl(ds->ds_ip_addr), ntohs(ds->ds_port));
+
+        data->write_done_cb = filelayout_write_done_cb;
+        data->ds_clp = ds->ds_clp;
+        fh = nfs4_fl_select_ds_fh(lseg, j);
+        if (fh)
+                data->args.fh = fh;
+        /*
+         * Get the file offset on the dserver. Set the write offset to
+         * this offset and save the original offset.
+         */
+        data->args.offset = filelayout_get_dserver_offset(lseg, offset);
+        data->mds_offset = offset;
+
+        /* Perform an asynchronous write */
+        status = nfs_initiate_write(data, ds->ds_clp->cl_rpcclient,
+                                    &filelayout_write_call_ops, sync);
+        BUG_ON(status != 0);
+        return PNFS_ATTEMPTED;
+}
+
 /*
  * filelayout_check_layout()
  *
@@ -92,14 +430,14 @@ filelayout_check_layout(struct pnfs_layout_hdr *lo,
                 goto out;
         }
 
-        if (fl->stripe_unit % PAGE_SIZE) {
-                dprintk("%s Stripe unit (%u) not page aligned\n",
+        if (!fl->stripe_unit || fl->stripe_unit % PAGE_SIZE) {
+                dprintk("%s Invalid stripe unit (%u)\n",
                         __func__, fl->stripe_unit);
                 goto out;
         }
 
         /* find and reference the deviceid */
-        dsaddr = nfs4_fl_find_get_deviceid(nfss->nfs_client, id);
+        dsaddr = nfs4_fl_find_get_deviceid(id);
         if (dsaddr == NULL) {
                 dsaddr = get_device_info(lo->plh_inode, id);
                 if (dsaddr == NULL)
@@ -134,7 +472,7 @@ out:
         dprintk("--> %s returns %d\n", __func__, status);
         return status;
 out_put:
-        pnfs_put_deviceid(nfss->nfs_client->cl_devid_cache, &dsaddr->deviceid);
+        nfs4_fl_put_deviceid(dsaddr);
         goto out;
 }
 
@@ -164,12 +502,33 @@ filelayout_decode_layout(struct pnfs_layout_hdr *flo,
                          struct nfs4_layoutget_res *lgr,
                          struct nfs4_deviceid *id)
 {
-        uint32_t *p = (uint32_t *)lgr->layout.buf;
+        struct xdr_stream stream;
+        struct xdr_buf buf = {
+                .pages =  lgr->layoutp->pages,
+                .page_len =  lgr->layoutp->len,
+                .buflen =  lgr->layoutp->len,
+                .len = lgr->layoutp->len,
+        };
+        struct page *scratch;
+        __be32 *p;
         uint32_t nfl_util;
         int i;
 
         dprintk("%s: set_layout_map Begin\n", __func__);
 
+        scratch = alloc_page(GFP_KERNEL);
+        if (!scratch)
+                return -ENOMEM;
+
+        xdr_init_decode(&stream, &buf, NULL);
+        xdr_set_scratch_buffer(&stream, page_address(scratch), PAGE_SIZE);
+
+        /* 20 = ufl_util (4), first_stripe_index (4), pattern_offset (8),
+         * num_fh (4) */
+        p = xdr_inline_decode(&stream, NFS4_DEVICEID4_SIZE + 20);
+        if (unlikely(!p))
+                goto out_err;
+
         memcpy(id, p, sizeof(*id));
         p += XDR_QUADLEN(NFS4_DEVICEID4_SIZE);
         print_deviceid(id);
@@ -191,32 +550,57 @@ filelayout_decode_layout(struct pnfs_layout_hdr *flo,
                 __func__, nfl_util, fl->num_fh, fl->first_stripe_index,
                 fl->pattern_offset);
 
+        if (!fl->num_fh)
+                goto out_err;
+
         fl->fh_array = kzalloc(fl->num_fh * sizeof(struct nfs_fh *),
                                GFP_KERNEL);
         if (!fl->fh_array)
-                return -ENOMEM;
+                goto out_err;
 
         for (i = 0; i < fl->num_fh; i++) {
                 /* Do we want to use a mempool here? */
                 fl->fh_array[i] = kmalloc(sizeof(struct nfs_fh), GFP_KERNEL);
-                if (!fl->fh_array[i]) {
-                        filelayout_free_fh_array(fl);
-                        return -ENOMEM;
-                }
+                if (!fl->fh_array[i])
+                        goto out_err_free;
+
+                p = xdr_inline_decode(&stream, 4);
+                if (unlikely(!p))
+                        goto out_err_free;
                 fl->fh_array[i]->size = be32_to_cpup(p++);
                 if (sizeof(struct nfs_fh) < fl->fh_array[i]->size) {
                         printk(KERN_ERR "Too big fh %d received %d\n",
                                i, fl->fh_array[i]->size);
-                        filelayout_free_fh_array(fl);
-                        return -EIO;
+                        goto out_err_free;
                 }
+
+                p = xdr_inline_decode(&stream, fl->fh_array[i]->size);
+                if (unlikely(!p))
+                        goto out_err_free;
                 memcpy(fl->fh_array[i]->data, p, fl->fh_array[i]->size);
-                p += XDR_QUADLEN(fl->fh_array[i]->size);
                 dprintk("DEBUG: %s: fh len %d\n", __func__,
                         fl->fh_array[i]->size);
         }
 
+        __free_page(scratch);
         return 0;
+
+out_err_free:
+        filelayout_free_fh_array(fl);
+out_err:
+        __free_page(scratch);
+        return -EIO;
+}
+
+static void
+filelayout_free_lseg(struct pnfs_layout_segment *lseg)
+{
+        struct nfs4_filelayout_segment *fl = FILELAYOUT_LSEG(lseg);
+
+        dprintk("--> %s\n", __func__);
+        nfs4_fl_put_deviceid(fl->dsaddr);
+        kfree(fl->commit_buckets);
+        _filelayout_free_lseg(fl);
 }
 
 static struct pnfs_layout_segment *
@@ -237,29 +621,252 @@ filelayout_alloc_lseg(struct pnfs_layout_hdr *layoutid,
                 _filelayout_free_lseg(fl);
                 return NULL;
         }
+
+        /* This assumes there is only one IOMODE_RW lseg.  What
+         * we really want to do is have a layout_hdr level
+         * dictionary of <multipath_list4, fh> keys, each
+         * associated with a struct list_head, populated by calls
+         * to filelayout_write_pagelist().
+         * */
+        if ((!fl->commit_through_mds) && (lgr->range.iomode == IOMODE_RW)) {
+                int i;
+                int size = (fl->stripe_type == STRIPE_SPARSE) ?
+                        fl->dsaddr->ds_num : fl->dsaddr->stripe_count;
+
+                fl->commit_buckets = kcalloc(size, sizeof(struct list_head), GFP_KERNEL);
+                if (!fl->commit_buckets) {
+                        filelayout_free_lseg(&fl->generic_hdr);
+                        return NULL;
+                }
+                fl->number_of_buckets = size;
+                for (i = 0; i < size; i++)
+                        INIT_LIST_HEAD(&fl->commit_buckets[i]);
+        }
         return &fl->generic_hdr;
 }
 
-static void
-filelayout_free_lseg(struct pnfs_layout_segment *lseg)
+/*
+ * filelayout_pg_test(). Called by nfs_can_coalesce_requests()
+ *
+ * return 1 :  coalesce page
+ * return 0 :  don't coalesce page
+ */
+int
+filelayout_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev,
+                   struct nfs_page *req)
+{
+        u64 p_stripe, r_stripe;
+        u32 stripe_unit;
+
+        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;
+
+        do_div(p_stripe, stripe_unit);
+        do_div(r_stripe, stripe_unit);
+
+        return (p_stripe == r_stripe);
+}
+
+static bool filelayout_mark_pnfs_commit(struct pnfs_layout_segment *lseg)
+{
+        return !FILELAYOUT_LSEG(lseg)->commit_through_mds;
+}
+
+static u32 select_bucket_index(struct nfs4_filelayout_segment *fl, u32 j)
+{
+        if (fl->stripe_type == STRIPE_SPARSE)
+                return nfs4_fl_calc_ds_index(&fl->generic_hdr, j);
+        else
+                return j;
+}
+
+struct list_head *filelayout_choose_commit_list(struct nfs_page *req)
 {
-        struct nfs_server *nfss = NFS_SERVER(lseg->pls_layout->plh_inode);
+        struct pnfs_layout_segment *lseg = req->wb_commit_lseg;
         struct nfs4_filelayout_segment *fl = FILELAYOUT_LSEG(lseg);
+        u32 i, j;
+        struct list_head *list;
+
+        /* Note that we are calling nfs4_fl_calc_j_index on each page
+         * that ends up being committed to a data server.  An attractive
+         * alternative is to add a field to nfs_write_data and nfs_page
+         * to store the value calculated in filelayout_write_pagelist
+         * and just use that here.
+         */
+        j = nfs4_fl_calc_j_index(lseg,
+                                 (loff_t)req->wb_index << PAGE_CACHE_SHIFT);
+        i = select_bucket_index(fl, j);
+        list = &fl->commit_buckets[i];
+        if (list_empty(list)) {
+                /* Non-empty buckets hold a reference on the lseg */
+                get_lseg(lseg);
+        }
+        return list;
+}
 
-        dprintk("--> %s\n", __func__);
-        pnfs_put_deviceid(nfss->nfs_client->cl_devid_cache,
-                          &fl->dsaddr->deviceid);
-        _filelayout_free_lseg(fl);
+static u32 calc_ds_index_from_commit(struct pnfs_layout_segment *lseg, u32 i)
+{
+        struct nfs4_filelayout_segment *flseg = FILELAYOUT_LSEG(lseg);
+
+        if (flseg->stripe_type == STRIPE_SPARSE)
+                return i;
+        else
+                return nfs4_fl_calc_ds_index(lseg, i);
+}
+
+static struct nfs_fh *
+select_ds_fh_from_commit(struct pnfs_layout_segment *lseg, u32 i)
+{
+        struct nfs4_filelayout_segment *flseg = FILELAYOUT_LSEG(lseg);
+
+        if (flseg->stripe_type == STRIPE_SPARSE) {
+                if (flseg->num_fh == 1)
+                        i = 0;
+                else if (flseg->num_fh == 0)
+                        /* Use the MDS OPEN fh set in nfs_read_rpcsetup */
+                        return NULL;
+        }
+        return flseg->fh_array[i];
+}
+
+static int filelayout_initiate_commit(struct nfs_write_data *data, int how)
+{
+        struct pnfs_layout_segment *lseg = data->lseg;
+        struct nfs4_pnfs_ds *ds;
+        u32 idx;
+        struct nfs_fh *fh;
+
+        idx = calc_ds_index_from_commit(lseg, data->ds_commit_index);
+        ds = nfs4_fl_prepare_ds(lseg, idx);
+        if (!ds) {
+                printk(KERN_ERR "%s: prepare_ds failed, use MDS\n", __func__);
+                set_bit(lo_fail_bit(IOMODE_RW), &lseg->pls_layout->plh_flags);
+                set_bit(lo_fail_bit(IOMODE_READ), &lseg->pls_layout->plh_flags);
+                prepare_to_resend_writes(data);
+                data->mds_ops->rpc_release(data);
+                return -EAGAIN;
+        }
+        dprintk("%s ino %lu, how %d\n", __func__, data->inode->i_ino, how);
+        data->write_done_cb = filelayout_commit_done_cb;
+        data->ds_clp = ds->ds_clp;
+        fh = select_ds_fh_from_commit(lseg, data->ds_commit_index);
+        if (fh)
+                data->args.fh = fh;
+        return nfs_initiate_commit(data, ds->ds_clp->cl_rpcclient,
+                                   &filelayout_commit_call_ops, how);
+}
+
+/*
+ * This is only useful while we are using whole file layouts.
+ */
+static struct pnfs_layout_segment *find_only_write_lseg(struct inode *inode)
+{
+        struct pnfs_layout_segment *lseg, *rv = NULL;
+
+        spin_lock(&inode->i_lock);
+        list_for_each_entry(lseg, &NFS_I(inode)->layout->plh_segs, pls_list)
+                if (lseg->pls_range.iomode == IOMODE_RW)
+                        rv = get_lseg(lseg);
+        spin_unlock(&inode->i_lock);
+        return rv;
+}
+
+static int alloc_ds_commits(struct inode *inode, struct list_head *list)
+{
+        struct pnfs_layout_segment *lseg;
+        struct nfs4_filelayout_segment *fl;
+        struct nfs_write_data *data;
+        int i, j;
+
+        /* Won't need this when non-whole file layout segments are supported
+         * instead we will use a pnfs_layout_hdr structure */
+        lseg = find_only_write_lseg(inode);
+        if (!lseg)
+                return 0;
+        fl = FILELAYOUT_LSEG(lseg);
+        for (i = 0; i < fl->number_of_buckets; i++) {
+                if (list_empty(&fl->commit_buckets[i]))
+                        continue;
+                data = nfs_commitdata_alloc();
+                if (!data)
+                        goto out_bad;
+                data->ds_commit_index = i;
+                data->lseg = lseg;
+                list_add(&data->pages, list);
+        }
+        put_lseg(lseg);
+        return 0;
+
+out_bad:
+        for (j = i; j < fl->number_of_buckets; j++) {
+                if (list_empty(&fl->commit_buckets[i]))
+                        continue;
+                nfs_retry_commit(&fl->commit_buckets[i], lseg);
+                put_lseg(lseg);  /* associated with emptying bucket */
+        }
+        put_lseg(lseg);
+        /* Caller will clean up entries put on list */
+        return -ENOMEM;
+}
+
+/* This follows nfs_commit_list pretty closely */
+static int
+filelayout_commit_pagelist(struct inode *inode, struct list_head *mds_pages,
+                           int how)
+{
+        struct nfs_write_data   *data, *tmp;
+        LIST_HEAD(list);
+
+        if (!list_empty(mds_pages)) {
+                data = nfs_commitdata_alloc();
+                if (!data)
+                        goto out_bad;
+                data->lseg = NULL;
+                list_add(&data->pages, &list);
+        }
+
+        if (alloc_ds_commits(inode, &list))
+                goto out_bad;
+
+        list_for_each_entry_safe(data, tmp, &list, pages) {
+                list_del_init(&data->pages);
+                atomic_inc(&NFS_I(inode)->commits_outstanding);
+                if (!data->lseg) {
+                        nfs_init_commit(data, mds_pages, NULL);
+                        nfs_initiate_commit(data, NFS_CLIENT(inode),
+                                            data->mds_ops, how);
+                } else {
+                        nfs_init_commit(data, &FILELAYOUT_LSEG(data->lseg)->commit_buckets[data->ds_commit_index], data->lseg);
+                        filelayout_initiate_commit(data, how);
+                }
+        }
+        return 0;
+ out_bad:
+        list_for_each_entry_safe(data, tmp, &list, pages) {
+                nfs_retry_commit(&data->pages, data->lseg);
+                list_del_init(&data->pages);
+                nfs_commit_free(data);
+        }
+        nfs_retry_commit(mds_pages, NULL);
+        nfs_commit_clear_lock(NFS_I(inode));
+        return -ENOMEM;
 }
 
 static struct pnfs_layoutdriver_type filelayout_type = {
-        .id = LAYOUT_NFSV4_1_FILES,
-        .name = "LAYOUT_NFSV4_1_FILES",
-        .owner = THIS_MODULE,
-        .set_layoutdriver = filelayout_set_layoutdriver,
-        .clear_layoutdriver = filelayout_clear_layoutdriver,
-        .alloc_lseg              = filelayout_alloc_lseg,
-        .free_lseg               = filelayout_free_lseg,
+        .id                     = LAYOUT_NFSV4_1_FILES,
+        .name                   = "LAYOUT_NFSV4_1_FILES",
+        .owner                  = THIS_MODULE,
+        .alloc_lseg             = filelayout_alloc_lseg,
+        .free_lseg              = filelayout_free_lseg,
+        .pg_test                = filelayout_pg_test,
+        .mark_pnfs_commit       = filelayout_mark_pnfs_commit,
+        .choose_commit_list     = filelayout_choose_commit_list,
+        .commit_pagelist        = filelayout_commit_pagelist,
+        .read_pagelist          = filelayout_read_pagelist,
+        .write_pagelist         = filelayout_write_pagelist,
 };
 
 static int __init nfs4filelayout_init(void)
diff --git a/fs/nfs/nfs4filelayout.h b/fs/nfs/nfs4filelayout.h
index bbf60dd2ab9d..7c44579f5832 100644
--- a/fs/nfs/nfs4filelayout.h
+++ b/fs/nfs/nfs4filelayout.h
@@ -33,7 +33,7 @@
 #include "pnfs.h"
 
 /*
- * Field testing shows we need to support upto 4096 stripe indices.
+ * Field testing shows we need to support up to 4096 stripe indices.
  * We store each index as a u8 (u32 on the wire) to keep the memory footprint
  * reasonable. This in turn means we support a maximum of 256
  * RFC 5661 multipath_list4 structures.
@@ -55,8 +55,14 @@ struct nfs4_pnfs_ds {
         atomic_t                ds_count;
 };
 
+/* nfs4_file_layout_dsaddr flags */
+#define NFS4_DEVICE_ID_NEG_ENTRY        0x00000001
+
 struct nfs4_file_layout_dsaddr {
-        struct pnfs_deviceid_node       deviceid;
+        struct hlist_node               node;
+        struct nfs4_deviceid            deviceid;
+        atomic_t                        ref;
+        unsigned long                   flags;
         u32                             stripe_count;
         u8                              *stripe_indices;
         u32                             ds_num;
@@ -73,6 +79,8 @@ struct nfs4_filelayout_segment {
         struct nfs4_file_layout_dsaddr *dsaddr; /* Point to GETDEVINFO data */
         unsigned int num_fh;
         struct nfs_fh **fh_array;
+        struct list_head *commit_buckets; /* Sort commits to ds */
+        int number_of_buckets;
 };
 
 static inline struct nfs4_filelayout_segment *
@@ -83,11 +91,18 @@ FILELAYOUT_LSEG(struct pnfs_layout_segment *lseg)
                             generic_hdr);
 }
 
-extern void nfs4_fl_free_deviceid_callback(struct pnfs_deviceid_node *);
+extern struct nfs_fh *
+nfs4_fl_select_ds_fh(struct pnfs_layout_segment *lseg, u32 j);
+
 extern void print_ds(struct nfs4_pnfs_ds *ds);
 extern void print_deviceid(struct nfs4_deviceid *dev_id);
+u32 nfs4_fl_calc_j_index(struct pnfs_layout_segment *lseg, loff_t offset);
+u32 nfs4_fl_calc_ds_index(struct pnfs_layout_segment *lseg, u32 j);
+struct nfs4_pnfs_ds *nfs4_fl_prepare_ds(struct pnfs_layout_segment *lseg,
+                                        u32 ds_idx);
 extern struct nfs4_file_layout_dsaddr *
-nfs4_fl_find_get_deviceid(struct nfs_client *, struct nfs4_deviceid *dev_id);
+nfs4_fl_find_get_deviceid(struct nfs4_deviceid *dev_id);
+extern void nfs4_fl_put_deviceid(struct nfs4_file_layout_dsaddr *dsaddr);
 struct nfs4_file_layout_dsaddr *
 get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id);
 
diff --git a/fs/nfs/nfs4filelayoutdev.c b/fs/nfs/nfs4filelayoutdev.c
index f5c9b125e8cc..de5350f2b249 100644
--- a/fs/nfs/nfs4filelayoutdev.c
+++ b/fs/nfs/nfs4filelayoutdev.c
@@ -37,6 +37,30 @@
 #define NFSDBG_FACILITY         NFSDBG_PNFS_LD
 
 /*
+ * Device ID RCU cache. A device ID is unique per client ID and layout type.
+ */
+#define NFS4_FL_DEVICE_ID_HASH_BITS     5
+#define NFS4_FL_DEVICE_ID_HASH_SIZE     (1 << NFS4_FL_DEVICE_ID_HASH_BITS)
+#define NFS4_FL_DEVICE_ID_HASH_MASK     (NFS4_FL_DEVICE_ID_HASH_SIZE - 1)
+
+static inline u32
+nfs4_fl_deviceid_hash(struct nfs4_deviceid *id)
+{
+        unsigned char *cptr = (unsigned char *)id->data;
+        unsigned int nbytes = NFS4_DEVICEID4_SIZE;
+        u32 x = 0;
+
+        while (nbytes--) {
+                x *= 37;
+                x += *cptr++;
+        }
+        return x & NFS4_FL_DEVICE_ID_HASH_MASK;
+}
+
+static struct hlist_head filelayout_deviceid_cache[NFS4_FL_DEVICE_ID_HASH_SIZE];
+static DEFINE_SPINLOCK(filelayout_deviceid_lock);
+
+/*
  * Data server cache
  *
  * Data servers can be mapped to different device ids.
@@ -104,6 +128,67 @@ _data_server_lookup_locked(u32 ip_addr, u32 port)
         return NULL;
 }
 
+/*
+ * Create an rpc connection to the nfs4_pnfs_ds data server
+ * Currently only support IPv4
+ */
+static int
+nfs4_ds_connect(struct nfs_server *mds_srv, struct nfs4_pnfs_ds *ds)
+{
+        struct nfs_client *clp;
+        struct sockaddr_in sin;
+        int status = 0;
+
+        dprintk("--> %s ip:port %x:%hu au_flavor %d\n", __func__,
+                ntohl(ds->ds_ip_addr), ntohs(ds->ds_port),
+                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;
+
+        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;
+        }
+
+        if ((clp->cl_exchange_flags & EXCHGID4_FLAG_MASK_PNFS) != 0) {
+                if (!is_ds_client(clp)) {
+                        status = -ENODEV;
+                        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));
+                goto out;
+        }
+
+        /*
+         * Do not set NFS_CS_CHECK_LEASE_TIME instead set the DS lease to
+         * be equal to the MDS lease. Renewal is scheduled in create_session.
+         */
+        spin_lock(&mds_srv->nfs_client->cl_lock);
+        clp->cl_lease_time = mds_srv->nfs_client->cl_lease_time;
+        spin_unlock(&mds_srv->nfs_client->cl_lock);
+        clp->cl_last_renewal = jiffies;
+
+        /* New nfs_client */
+        status = nfs4_init_ds_session(clp);
+        if (status)
+                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));
+out:
+        return status;
+out_put:
+        nfs_put_client(clp);
+        goto out;
+}
+
 static void
 destroy_ds(struct nfs4_pnfs_ds *ds)
 {
@@ -122,7 +207,7 @@ nfs4_fl_free_deviceid(struct nfs4_file_layout_dsaddr *dsaddr)
         struct nfs4_pnfs_ds *ds;
         int i;
 
-        print_deviceid(&dsaddr->deviceid.de_id);
+        print_deviceid(&dsaddr->deviceid);
 
         for (i = 0; i < dsaddr->ds_num; i++) {
                 ds = dsaddr->ds_list[i];
@@ -139,15 +224,6 @@ nfs4_fl_free_deviceid(struct nfs4_file_layout_dsaddr *dsaddr)
         kfree(dsaddr);
 }
 
-void
-nfs4_fl_free_deviceid_callback(struct pnfs_deviceid_node *device)
-{
-        struct nfs4_file_layout_dsaddr *dsaddr =
-                container_of(device, struct nfs4_file_layout_dsaddr, deviceid);
-
-        nfs4_fl_free_deviceid(dsaddr);
-}
-
 static struct nfs4_pnfs_ds *
 nfs4_pnfs_ds_add(struct inode *inode, u32 ip_addr, u32 port)
 {
@@ -185,7 +261,7 @@ out:
  * Currently only support ipv4, and one multi-path address.
  */
 static struct nfs4_pnfs_ds *
-decode_and_add_ds(__be32 **pp, struct inode *inode)
+decode_and_add_ds(struct xdr_stream *streamp, struct inode *inode)
 {
         struct nfs4_pnfs_ds *ds = NULL;
         char *buf;
@@ -193,25 +269,34 @@ decode_and_add_ds(__be32 **pp, struct inode *inode)
         u32 ip_addr, port;
         int nlen, rlen, i;
         int tmp[2];
-        __be32 *r_netid, *r_addr, *p = *pp;
+        __be32 *p;
 
         /* r_netid */
+        p = xdr_inline_decode(streamp, 4);
+        if (unlikely(!p))
+                goto out_err;
         nlen = be32_to_cpup(p++);
-        r_netid = p;
-        p += XDR_QUADLEN(nlen);
 
-        /* r_addr */
-        rlen = be32_to_cpup(p++);
-        r_addr = p;
-        p += XDR_QUADLEN(rlen);
-        *pp = p;
+        p = xdr_inline_decode(streamp, nlen);
+        if (unlikely(!p))
+                goto out_err;
 
         /* Check that netid is "tcp" */
-        if (nlen != 3 ||  memcmp((char *)r_netid, "tcp", 3)) {
+        if (nlen != 3 ||  memcmp((char *)p, "tcp", 3)) {
                 dprintk("%s: ERROR: non ipv4 TCP r_netid\n", __func__);
                 goto out_err;
         }
 
+        /* r_addr */
+        p = xdr_inline_decode(streamp, 4);
+        if (unlikely(!p))
+                goto out_err;
+        rlen = be32_to_cpup(p);
+
+        p = xdr_inline_decode(streamp, rlen);
+        if (unlikely(!p))
+                goto out_err;
+
         /* ipv6 length plus port is legal */
         if (rlen > INET6_ADDRSTRLEN + 8) {
                 dprintk("%s: Invalid address, length %d\n", __func__,
@@ -219,8 +304,12 @@ decode_and_add_ds(__be32 **pp, struct inode *inode)
                 goto out_err;
         }
         buf = kmalloc(rlen + 1, GFP_KERNEL);
+        if (!buf) {
+                dprintk("%s: Not enough memory\n", __func__);
+                goto out_err;
+        }
         buf[rlen] = '\0';
-        memcpy(buf, r_addr, rlen);
+        memcpy(buf, p, rlen);
 
         /* replace the port dots with dashes for the in4_pton() delimiter*/
         for (i = 0; i < 2; i++) {
@@ -256,118 +345,191 @@ out_err:
 static struct nfs4_file_layout_dsaddr*
 decode_device(struct inode *ino, struct pnfs_device *pdev)
 {
-        int i, dummy;
+        int i;
         u32 cnt, num;
         u8 *indexp;
-        __be32 *p = (__be32 *)pdev->area, *indicesp;
-        struct nfs4_file_layout_dsaddr *dsaddr;
+        __be32 *p;
+        u8 *stripe_indices;
+        u8 max_stripe_index;
+        struct nfs4_file_layout_dsaddr *dsaddr = NULL;
+        struct xdr_stream stream;
+        struct xdr_buf buf = {
+                .pages = pdev->pages,
+                .page_len = pdev->pglen,
+                .buflen = pdev->pglen,
+                .len = pdev->pglen,
+        };
+        struct page *scratch;
+
+        /* set up xdr stream */
+        scratch = alloc_page(GFP_KERNEL);
+        if (!scratch)
+                goto out_err;
+
+        xdr_init_decode(&stream, &buf, NULL);
+        xdr_set_scratch_buffer(&stream, page_address(scratch), PAGE_SIZE);
 
         /* Get the stripe count (number of stripe index) */
-        cnt = be32_to_cpup(p++);
+        p = xdr_inline_decode(&stream, 4);
+        if (unlikely(!p))
+                goto out_err_free_scratch;
+
+        cnt = be32_to_cpup(p);
         dprintk("%s stripe count  %d\n", __func__, cnt);
         if (cnt > NFS4_PNFS_MAX_STRIPE_CNT) {
                 printk(KERN_WARNING "%s: stripe count %d greater than "
                        "supported maximum %d\n", __func__,
                         cnt, NFS4_PNFS_MAX_STRIPE_CNT);
-                goto out_err;
+                goto out_err_free_scratch;
+        }
+
+        /* read stripe indices */
+        stripe_indices = kcalloc(cnt, sizeof(u8), GFP_KERNEL);
+        if (!stripe_indices)
+                goto out_err_free_scratch;
+
+        p = xdr_inline_decode(&stream, cnt << 2);
+        if (unlikely(!p))
+                goto out_err_free_stripe_indices;
+
+        indexp = &stripe_indices[0];
+        max_stripe_index = 0;
+        for (i = 0; i < cnt; i++) {
+                *indexp = be32_to_cpup(p++);
+                max_stripe_index = max(max_stripe_index, *indexp);
+                indexp++;
         }
 
         /* Check the multipath list count */
-        indicesp = p;
-        p += XDR_QUADLEN(cnt << 2);
-        num = be32_to_cpup(p++);
+        p = xdr_inline_decode(&stream, 4);
+        if (unlikely(!p))
+                goto out_err_free_stripe_indices;
+
+        num = be32_to_cpup(p);
         dprintk("%s ds_num %u\n", __func__, num);
         if (num > NFS4_PNFS_MAX_MULTI_CNT) {
                 printk(KERN_WARNING "%s: multipath count %d greater than "
                         "supported maximum %d\n", __func__,
                         num, NFS4_PNFS_MAX_MULTI_CNT);
-                goto out_err;
+                goto out_err_free_stripe_indices;
+        }
+
+        /* validate stripe indices are all < num */
+        if (max_stripe_index >= num) {
+                printk(KERN_WARNING "%s: stripe index %u >= num ds %u\n",
+                        __func__, max_stripe_index, num);
+                goto out_err_free_stripe_indices;
         }
+
         dsaddr = kzalloc(sizeof(*dsaddr) +
                         (sizeof(struct nfs4_pnfs_ds *) * (num - 1)),
                         GFP_KERNEL);
         if (!dsaddr)
-                goto out_err;
-
-        dsaddr->stripe_indices = kzalloc(sizeof(u8) * cnt, GFP_KERNEL);
-        if (!dsaddr->stripe_indices)
-                goto out_err_free;
+                goto out_err_free_stripe_indices;
 
         dsaddr->stripe_count = cnt;
+        dsaddr->stripe_indices = stripe_indices;
+        stripe_indices = NULL;
         dsaddr->ds_num = num;
 
-        memcpy(&dsaddr->deviceid.de_id, &pdev->dev_id, sizeof(pdev->dev_id));
-
-        /* Go back an read stripe indices */
-        p = indicesp;
-        indexp = &dsaddr->stripe_indices[0];
-        for (i = 0; i < dsaddr->stripe_count; i++) {
-                *indexp = be32_to_cpup(p++);
-                if (*indexp >= num)
-                        goto out_err_free;
-                indexp++;
-        }
-        /* Skip already read multipath list count */
-        p++;
+        memcpy(&dsaddr->deviceid, &pdev->dev_id, sizeof(pdev->dev_id));
 
         for (i = 0; i < dsaddr->ds_num; i++) {
                 int j;
+                u32 mp_count;
 
-                dummy = be32_to_cpup(p++); /* multipath count */
-                if (dummy > 1) {
+                p = xdr_inline_decode(&stream, 4);
+                if (unlikely(!p))
+                        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__,
-                                dummy);
+                                mp_count);
                 }
-                for (j = 0; j < dummy; j++) {
+                for (j = 0; j < mp_count; j++) {
                         if (j == 0) {
-                                dsaddr->ds_list[i] = decode_and_add_ds(&p, ino);
+                                dsaddr->ds_list[i] = decode_and_add_ds(&stream,
+                                        ino);
                                 if (dsaddr->ds_list[i] == NULL)
-                                        goto out_err_free;
+                                        goto out_err_free_deviceid;
                         } else {
                                 u32 len;
                                 /* skip extra multipath */
-                                len = be32_to_cpup(p++);
-                                p += XDR_QUADLEN(len);
-                                len = be32_to_cpup(p++);
-                                p += XDR_QUADLEN(len);
-                                continue;
+
+                                /* 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;
                         }
                 }
         }
+
+        __free_page(scratch);
         return dsaddr;
 
-out_err_free:
+out_err_free_deviceid:
         nfs4_fl_free_deviceid(dsaddr);
+        /* stripe_indicies was part of dsaddr */
+        goto out_err_free_scratch;
+out_err_free_stripe_indices:
+        kfree(stripe_indices);
+out_err_free_scratch:
+        __free_page(scratch);
 out_err:
         dprintk("%s ERROR: returning NULL\n", __func__);
         return NULL;
 }
 
 /*
- * Decode the opaque device specified in 'dev'
- * and add it to the list of available devices.
- * If the deviceid is already cached, nfs4_add_deviceid will return
- * a pointer to the cached struct and throw away the new.
+ * Decode the opaque device specified in 'dev' and add it to the cache of
+ * available devices.
  */
-static struct nfs4_file_layout_dsaddr*
+static struct nfs4_file_layout_dsaddr *
 decode_and_add_device(struct inode *inode, struct pnfs_device *dev)
 {
-        struct nfs4_file_layout_dsaddr *dsaddr;
-        struct pnfs_deviceid_node *d;
+        struct nfs4_file_layout_dsaddr *d, *new;
+        long hash;
 
-        dsaddr = decode_device(inode, dev);
-        if (!dsaddr) {
+        new = decode_device(inode, dev);
+        if (!new) {
                 printk(KERN_WARNING "%s: Could not decode or add device\n",
                         __func__);
                 return NULL;
         }
 
-        d = pnfs_add_deviceid(NFS_SERVER(inode)->nfs_client->cl_devid_cache,
-                              &dsaddr->deviceid);
+        spin_lock(&filelayout_deviceid_lock);
+        d = nfs4_fl_find_get_deviceid(&new->deviceid);
+        if (d) {
+                spin_unlock(&filelayout_deviceid_lock);
+                nfs4_fl_free_deviceid(new);
+                return d;
+        }
 
-        return container_of(d, struct nfs4_file_layout_dsaddr, deviceid);
+        INIT_HLIST_NODE(&new->node);
+        atomic_set(&new->ref, 1);
+        hash = nfs4_fl_deviceid_hash(&new->deviceid);
+        hlist_add_head_rcu(&new->node, &filelayout_deviceid_cache[hash]);
+        spin_unlock(&filelayout_deviceid_lock);
+
+        return new;
 }
 
 /*
@@ -409,11 +571,6 @@ get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id)
                         goto out_free;
         }
 
-        /* set pdev->area */
-        pdev->area = vmap(pages, max_pages, VM_MAP, PAGE_KERNEL);
-        if (!pdev->area)
-                goto out_free;
-
         memcpy(&pdev->dev_id, dev_id, sizeof(*dev_id));
         pdev->layout_type = LAYOUT_NFSV4_1_FILES;
         pdev->pages = pages;
@@ -432,8 +589,6 @@ get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id)
          */
         dsaddr = decode_and_add_device(inode, pdev);
 out_free:
-        if (pdev->area != NULL)
-                vunmap(pdev->area);
         for (i = 0; i < max_pages; i++)
                 __free_page(pages[i]);
         kfree(pages);
@@ -442,12 +597,123 @@ out_free:
         return dsaddr;
 }
 
+void
+nfs4_fl_put_deviceid(struct nfs4_file_layout_dsaddr *dsaddr)
+{
+        if (atomic_dec_and_lock(&dsaddr->ref, &filelayout_deviceid_lock)) {
+                hlist_del_rcu(&dsaddr->node);
+                spin_unlock(&filelayout_deviceid_lock);
+
+                synchronize_rcu();
+                nfs4_fl_free_deviceid(dsaddr);
+        }
+}
+
 struct nfs4_file_layout_dsaddr *
-nfs4_fl_find_get_deviceid(struct nfs_client *clp, struct nfs4_deviceid *id)
+nfs4_fl_find_get_deviceid(struct nfs4_deviceid *id)
 {
-        struct pnfs_deviceid_node *d;
+        struct nfs4_file_layout_dsaddr *d;
+        struct hlist_node *n;
+        long hash = nfs4_fl_deviceid_hash(id);
+
+
+        rcu_read_lock();
+        hlist_for_each_entry_rcu(d, n, &filelayout_deviceid_cache[hash], node) {
+                if (!memcmp(&d->deviceid, id, sizeof(*id))) {
+                        if (!atomic_inc_not_zero(&d->ref))
+                                goto fail;
+                        rcu_read_unlock();
+                        return d;
+                }
+        }
+fail:
+        rcu_read_unlock();
+        return NULL;
+}
 
-        d = pnfs_find_get_deviceid(clp->cl_devid_cache, id);
-        return (d == NULL) ? NULL :
-                container_of(d, struct nfs4_file_layout_dsaddr, deviceid);
+/*
+ * Want res = (offset - layout->pattern_offset)/ layout->stripe_unit
+ * Then: ((res + fsi) % dsaddr->stripe_count)
+ */
+u32
+nfs4_fl_calc_j_index(struct pnfs_layout_segment *lseg, loff_t offset)
+{
+        struct nfs4_filelayout_segment *flseg = FILELAYOUT_LSEG(lseg);
+        u64 tmp;
+
+        tmp = offset - flseg->pattern_offset;
+        do_div(tmp, flseg->stripe_unit);
+        tmp += flseg->first_stripe_index;
+        return do_div(tmp, flseg->dsaddr->stripe_count);
+}
+
+u32
+nfs4_fl_calc_ds_index(struct pnfs_layout_segment *lseg, u32 j)
+{
+        return FILELAYOUT_LSEG(lseg)->dsaddr->stripe_indices[j];
+}
+
+struct nfs_fh *
+nfs4_fl_select_ds_fh(struct pnfs_layout_segment *lseg, u32 j)
+{
+        struct nfs4_filelayout_segment *flseg = FILELAYOUT_LSEG(lseg);
+        u32 i;
+
+        if (flseg->stripe_type == STRIPE_SPARSE) {
+                if (flseg->num_fh == 1)
+                        i = 0;
+                else if (flseg->num_fh == 0)
+                        /* Use the MDS OPEN fh set in nfs_read_rpcsetup */
+                        return NULL;
+                else
+                        i = nfs4_fl_calc_ds_index(lseg, j);
+        } else
+                i = j;
+        return flseg->fh_array[i];
+}
+
+static void
+filelayout_mark_devid_negative(struct nfs4_file_layout_dsaddr *dsaddr,
+                               int err, u32 ds_addr)
+{
+        u32 *p = (u32 *)&dsaddr->deviceid;
+
+        printk(KERN_ERR "NFS: data server %x connection error %d."
+                " Deviceid [%x%x%x%x] marked out of use.\n",
+                ds_addr, err, p[0], p[1], p[2], p[3]);
+
+        spin_lock(&filelayout_deviceid_lock);
+        dsaddr->flags |= NFS4_DEVICE_ID_NEG_ENTRY;
+        spin_unlock(&filelayout_deviceid_lock);
+}
+
+struct nfs4_pnfs_ds *
+nfs4_fl_prepare_ds(struct pnfs_layout_segment *lseg, u32 ds_idx)
+{
+        struct nfs4_file_layout_dsaddr *dsaddr = FILELAYOUT_LSEG(lseg)->dsaddr;
+        struct nfs4_pnfs_ds *ds = dsaddr->ds_list[ds_idx];
+
+        if (ds == NULL) {
+                printk(KERN_ERR "%s: No data server for offset index %d\n",
+                        __func__, ds_idx);
+                return NULL;
+        }
+
+        if (!ds->ds_clp) {
+                struct nfs_server *s = NFS_SERVER(lseg->pls_layout->plh_inode);
+                int err;
+
+                if (dsaddr->flags & NFS4_DEVICE_ID_NEG_ENTRY) {
+                        /* Already tried to connect, don't try again */
+                        dprintk("%s Deviceid marked out of use\n", __func__);
+                        return NULL;
+                }
+                err = nfs4_ds_connect(s, ds);
+                if (err) {
+                        filelayout_mark_devid_negative(dsaddr, err,
+                                                       ntohl(ds->ds_ip_addr));
+                        return NULL;
+                }
+        }
+        return ds;
 }
diff --git a/fs/nfs/nfs4namespace.c b/fs/nfs/nfs4namespace.c
index 3c2a1724fbd2..bb80c49b6533 100644
--- a/fs/nfs/nfs4namespace.c
+++ b/fs/nfs/nfs4namespace.c
@@ -54,33 +54,29 @@ Elong:
 /*
  * Determine the mount path as a string
  */
-static char *nfs4_path(const struct vfsmount *mnt_parent,
-                       const struct dentry *dentry,
-                       char *buffer, ssize_t buflen)
+static char *nfs4_path(struct dentry *dentry, char *buffer, ssize_t buflen)
 {
-        const char *srvpath;
-
-        srvpath = strchr(mnt_parent->mnt_devname, ':');
-        if (srvpath)
-                srvpath++;
-        else
-                srvpath = mnt_parent->mnt_devname;
-
-        return nfs_path(srvpath, mnt_parent->mnt_root, dentry, buffer, buflen);
+        char *limit;
+        char *path = nfs_path(&limit, dentry, buffer, buflen);
+        if (!IS_ERR(path)) {
+                char *colon = strchr(path, ':');
+                if (colon && colon < limit)
+                        path = colon + 1;
+        }
+        return path;
 }
 
 /*
  * Check that fs_locations::fs_root [RFC3530 6.3] is a prefix for what we
  * believe to be the server path to this dentry
  */
-static int nfs4_validate_fspath(const struct vfsmount *mnt_parent,
-                                const struct dentry *dentry,
+static int nfs4_validate_fspath(struct dentry *dentry,
                                 const struct nfs4_fs_locations *locations,
                                 char *page, char *page2)
 {
         const char *path, *fs_path;
 
-        path = nfs4_path(mnt_parent, dentry, page, PAGE_SIZE);
+        path = nfs4_path(dentry, page, PAGE_SIZE);
         if (IS_ERR(path))
                 return PTR_ERR(path);
 
@@ -165,20 +161,18 @@ static struct vfsmount *try_location(struct nfs_clone_mount *mountdata,
 
 /**
  * nfs_follow_referral - set up mountpoint when hitting a referral on moved error
- * @mnt_parent - mountpoint of parent directory
  * @dentry - parent directory
  * @locations - array of NFSv4 server location information
  *
  */
-static struct vfsmount *nfs_follow_referral(const struct vfsmount *mnt_parent,
-                                            const struct dentry *dentry,
+static struct vfsmount *nfs_follow_referral(struct dentry *dentry,
                                             const struct nfs4_fs_locations *locations)
 {
         struct vfsmount *mnt = ERR_PTR(-ENOENT);
         struct nfs_clone_mount mountdata = {
-                .sb = mnt_parent->mnt_sb,
+                .sb = dentry->d_sb,
                 .dentry = dentry,
-                .authflavor = NFS_SB(mnt_parent->mnt_sb)->client->cl_auth->au_flavor,
+                .authflavor = NFS_SB(dentry->d_sb)->client->cl_auth->au_flavor,
         };
         char *page = NULL, *page2 = NULL;
         int loc, error;
@@ -198,7 +192,7 @@ static struct vfsmount *nfs_follow_referral(const struct vfsmount *mnt_parent,
                 goto out;
 
         /* Ensure fs path is a prefix of current dentry path */
-        error = nfs4_validate_fspath(mnt_parent, dentry, locations, page, page2);
+        error = nfs4_validate_fspath(dentry, locations, page, page2);
         if (error < 0) {
                 mnt = ERR_PTR(error);
                 goto out;
@@ -225,11 +219,10 @@ out:
 
 /*
  * nfs_do_refmount - handle crossing a referral on server
- * @mnt_parent - mountpoint of referral
  * @dentry - dentry of referral
  *
  */
-struct vfsmount *nfs_do_refmount(const struct vfsmount *mnt_parent, struct dentry *dentry)
+struct vfsmount *nfs_do_refmount(struct dentry *dentry)
 {
         struct vfsmount *mnt = ERR_PTR(-ENOMEM);
         struct dentry *parent;
@@ -262,7 +255,7 @@ struct vfsmount *nfs_do_refmount(const struct vfsmount *mnt_parent, struct dentr
             fs_locations->fs_path.ncomponents <= 0)
                 goto out_free;
 
-        mnt = nfs_follow_referral(mnt_parent, dentry, fs_locations);
+        mnt = nfs_follow_referral(dentry, fs_locations);
 out_free:
         __free_page(page);
         kfree(fs_locations);
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 78936a8f40ab..69c0f3c5ee7a 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -41,10 +41,12 @@
 #include <linux/string.h>
 #include <linux/slab.h>
 #include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/gss_api.h>
 #include <linux/nfs.h>
 #include <linux/nfs4.h>
 #include <linux/nfs_fs.h>
 #include <linux/nfs_page.h>
+#include <linux/nfs_mount.h>
 #include <linux/namei.h>
 #include <linux/mount.h>
 #include <linux/module.h>
@@ -71,7 +73,9 @@ static int _nfs4_proc_open(struct nfs4_opendata *data);
 static int _nfs4_recover_proc_open(struct nfs4_opendata *data);
 static int nfs4_do_fsinfo(struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *);
 static int nfs4_async_handle_error(struct rpc_task *, const struct nfs_server *, struct nfs4_state *);
-static int _nfs4_proc_lookup(struct inode *dir, const struct qstr *name, struct nfs_fh *fhandle, struct nfs_fattr *fattr);
+static int _nfs4_proc_lookup(struct rpc_clnt *client, struct inode *dir,
+                             const struct qstr *name, struct nfs_fh *fhandle,
+                             struct nfs_fattr *fattr);
 static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fattr *fattr);
 static int nfs4_do_setattr(struct inode *inode, struct rpc_cred *cred,
                             struct nfs_fattr *fattr, struct iattr *sattr,
@@ -85,6 +89,11 @@ static int nfs4_map_errors(int err)
         switch (err) {
         case -NFS4ERR_RESOURCE:
                 return -EREMOTEIO;
+        case -NFS4ERR_WRONGSEC:
+                return -EPERM;
+        case -NFS4ERR_BADOWNER:
+        case -NFS4ERR_BADNAME:
+                return -EINVAL;
         default:
                 dprintk("%s could not handle NFSv4 error %d\n",
                                 __func__, -err);
@@ -241,7 +250,7 @@ static int nfs4_delay(struct rpc_clnt *clnt, long *timeout)
 /* This is the error handling routine for processes that are allowed
  * to sleep.
  */
-static int nfs4_handle_exception(const struct nfs_server *server, int errorcode, struct nfs4_exception *exception)
+static int nfs4_handle_exception(struct nfs_server *server, int errorcode, struct nfs4_exception *exception)
 {
         struct nfs_client *clp = server->nfs_client;
         struct nfs4_state *state = exception->state;
@@ -256,12 +265,13 @@ static int nfs4_handle_exception(const struct nfs_server *server, int errorcode,
                 case -NFS4ERR_OPENMODE:
                         if (state == NULL)
                                 break;
-                        nfs4_state_mark_reclaim_nograce(clp, state);
-                        goto do_state_recovery;
+                        nfs4_schedule_stateid_recovery(server, state);
+                        goto wait_on_recovery;
                 case -NFS4ERR_STALE_STATEID:
                 case -NFS4ERR_STALE_CLIENTID:
                 case -NFS4ERR_EXPIRED:
-                        goto do_state_recovery;
+                        nfs4_schedule_lease_recovery(clp);
+                        goto wait_on_recovery;
 #if defined(CONFIG_NFS_V4_1)
                 case -NFS4ERR_BADSESSION:
                 case -NFS4ERR_BADSLOT:
@@ -272,7 +282,7 @@ 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);
-                        nfs4_schedule_state_recovery(clp);
+                        nfs4_schedule_session_recovery(clp->cl_session);
                         exception->retry = 1;
                         break;
 #endif /* defined(CONFIG_NFS_V4_1) */
@@ -292,11 +302,23 @@ static int nfs4_handle_exception(const struct nfs_server *server, int errorcode,
                                 break;
                 case -NFS4ERR_OLD_STATEID:
                         exception->retry = 1;
+                        break;
+                case -NFS4ERR_BADOWNER:
+                        /* The following works around a Linux server bug! */
+                case -NFS4ERR_BADNAME:
+                        if (server->caps & NFS_CAP_UIDGID_NOMAP) {
+                                server->caps &= ~NFS_CAP_UIDGID_NOMAP;
+                                exception->retry = 1;
+                                printk(KERN_WARNING "NFS: v4 server %s "
+                                                "does not accept raw "
+                                                "uid/gids. "
+                                                "Reenabling the idmapper.\n",
+                                                server->nfs_client->cl_hostname);
+                        }
         }
         /* We failed to handle the error */
         return nfs4_map_errors(ret);
-do_state_recovery:
-        nfs4_schedule_state_recovery(clp);
+wait_on_recovery:
         ret = nfs4_wait_clnt_recover(clp);
         if (ret == 0)
                 exception->retry = 1;
@@ -422,8 +444,8 @@ static int nfs41_sequence_done(struct rpc_task *task, struct nfs4_sequence_res *
         if (res->sr_status == 1)
                 res->sr_status = NFS_OK;
 
-        /* -ERESTARTSYS can result in skipping nfs41_sequence_setup */
-        if (!res->sr_slot)
+        /* don't increment the sequence number if the task wasn't sent */
+        if (!RPC_WAS_SENT(task))
                 goto out;
 
         /* Check the SEQUENCE operation status */
@@ -435,8 +457,8 @@ static int nfs41_sequence_done(struct rpc_task *task, struct nfs4_sequence_res *
                 clp = res->sr_session->clp;
                 do_renew_lease(clp, timestamp);
                 /* Check sequence flags */
-                if (atomic_read(&clp->cl_count) > 1)
-                        nfs41_handle_sequence_flag_errors(clp, res->sr_status_flags);
+                if (res->sr_status_flags != 0)
+                        nfs4_schedule_lease_recovery(clp);
                 break;
         case -NFS4ERR_DELAY:
                 /* The server detected a resend of the RPC call and
@@ -505,7 +527,7 @@ out:
         return ret_id;
 }
 
-static int nfs41_setup_sequence(struct nfs4_session *session,
+int nfs41_setup_sequence(struct nfs4_session *session,
                                 struct nfs4_sequence_args *args,
                                 struct nfs4_sequence_res *res,
                                 int cache_reply,
@@ -571,6 +593,7 @@ static int nfs41_setup_sequence(struct nfs4_session *session,
         res->sr_status = 1;
         return 0;
 }
+EXPORT_SYMBOL_GPL(nfs41_setup_sequence);
 
 int nfs4_setup_sequence(const struct nfs_server *server,
                         struct nfs4_sequence_args *args,
@@ -640,7 +663,8 @@ struct rpc_call_ops nfs41_call_priv_sync_ops = {
         .rpc_call_done = nfs41_call_sync_done,
 };
 
-static int nfs4_call_sync_sequence(struct nfs_server *server,
+static int nfs4_call_sync_sequence(struct rpc_clnt *clnt,
+                                   struct nfs_server *server,
                                    struct rpc_message *msg,
                                    struct nfs4_sequence_args *args,
                                    struct nfs4_sequence_res *res,
@@ -656,7 +680,7 @@ static int nfs4_call_sync_sequence(struct nfs_server *server,
                 .cache_reply = cache_reply,
         };
         struct rpc_task_setup task_setup = {
-                .rpc_client = server->client,
+                .rpc_client = clnt,
                 .rpc_message = msg,
                 .callback_ops = &nfs41_call_sync_ops,
                 .callback_data = &data
@@ -675,13 +699,14 @@ static int nfs4_call_sync_sequence(struct nfs_server *server,
         return ret;
 }
 
-int _nfs4_call_sync_session(struct nfs_server *server,
+int _nfs4_call_sync_session(struct rpc_clnt *clnt,
+                            struct nfs_server *server,
                             struct rpc_message *msg,
                             struct nfs4_sequence_args *args,
                             struct nfs4_sequence_res *res,
                             int cache_reply)
 {
-        return nfs4_call_sync_sequence(server, msg, args, res, cache_reply, 0);
+        return nfs4_call_sync_sequence(clnt, server, msg, args, res, cache_reply, 0);
 }
 
 #else
@@ -692,19 +717,28 @@ static int nfs4_sequence_done(struct rpc_task *task,
 }
 #endif /* CONFIG_NFS_V4_1 */
 
-int _nfs4_call_sync(struct nfs_server *server,
+int _nfs4_call_sync(struct rpc_clnt *clnt,
+                    struct nfs_server *server,
                     struct rpc_message *msg,
                     struct nfs4_sequence_args *args,
                     struct nfs4_sequence_res *res,
                     int cache_reply)
 {
         args->sa_session = res->sr_session = NULL;
-        return rpc_call_sync(server->client, msg, 0);
+        return rpc_call_sync(clnt, msg, 0);
 }
 
-#define nfs4_call_sync(server, msg, args, res, cache_reply) \
-        (server)->nfs_client->cl_mvops->call_sync((server), (msg), &(args)->seq_args, \
-                        &(res)->seq_res, (cache_reply))
+static inline
+int nfs4_call_sync(struct rpc_clnt *clnt,
+                   struct nfs_server *server,
+                   struct rpc_message *msg,
+                   struct nfs4_sequence_args *args,
+                   struct nfs4_sequence_res *res,
+                   int cache_reply)
+{
+        return server->nfs_client->cl_mvops->call_sync(clnt, server, msg,
+                                                args, res, cache_reply);
+}
 
 static void update_changeattr(struct inode *dir, struct nfs4_change_info *cinfo)
 {
@@ -1255,14 +1289,13 @@ int nfs4_open_delegation_recall(struct nfs_open_context *ctx, struct nfs4_state
                         case -NFS4ERR_BAD_HIGH_SLOT:
                         case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
                         case -NFS4ERR_DEADSESSION:
-                                nfs4_schedule_state_recovery(
-                                        server->nfs_client);
+                                nfs4_schedule_session_recovery(server->nfs_client->cl_session);
                                 goto out;
                         case -NFS4ERR_STALE_CLIENTID:
                         case -NFS4ERR_STALE_STATEID:
                         case -NFS4ERR_EXPIRED:
                                 /* Don't recall a delegation if it was lost */
-                                nfs4_schedule_state_recovery(server->nfs_client);
+                                nfs4_schedule_lease_recovery(server->nfs_client);
                                 goto out;
                         case -ERESTARTSYS:
                                 /*
@@ -1271,7 +1304,7 @@ int nfs4_open_delegation_recall(struct nfs_open_context *ctx, struct nfs4_state
                                  */
                         case -NFS4ERR_ADMIN_REVOKED:
                         case -NFS4ERR_BAD_STATEID:
-                                nfs4_state_mark_reclaim_nograce(server->nfs_client, state);
+                                nfs4_schedule_stateid_recovery(server, state);
                         case -EKEYEXPIRED:
                                 /*
                                  * User RPCSEC_GSS context has expired.
@@ -1574,9 +1607,8 @@ static int _nfs4_proc_open(struct nfs4_opendata *data)
         return 0;
 }
 
-static int nfs4_recover_expired_lease(struct nfs_server *server)
+static int nfs4_client_recover_expired_lease(struct nfs_client *clp)
 {
-        struct nfs_client *clp = server->nfs_client;
         unsigned int loop;
         int ret;
 
@@ -1587,12 +1619,17 @@ static int nfs4_recover_expired_lease(struct nfs_server *server)
                 if (!test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state) &&
                     !test_bit(NFS4CLNT_CHECK_LEASE,&clp->cl_state))
                         break;
-                nfs4_schedule_state_recovery(clp);
+                nfs4_schedule_state_manager(clp);
                 ret = -EIO;
         }
         return ret;
 }
 
+static int nfs4_recover_expired_lease(struct nfs_server *server)
+{
+        return nfs4_client_recover_expired_lease(server->nfs_client);
+}
+
 /*
  * OPEN_EXPIRED:
  *      reclaim state on the server after a network partition.
@@ -1811,7 +1848,7 @@ static int _nfs4_do_setattr(struct inode *inode, struct rpc_cred *cred,
         } else
                 memcpy(&arg.stateid, &zero_stateid, sizeof(arg.stateid));
 
-        status = nfs4_call_sync(server, &msg, &arg, &res, 1);
+        status = nfs4_call_sync(server->client, server, &msg, &arg.seq_args, &res.seq_res, 1);
         if (status == 0 && state != NULL)
                 renew_lease(server, timestamp);
         return status;
@@ -2070,7 +2107,7 @@ static int _nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *f
         };
         int status;
 
-        status = nfs4_call_sync(server, &msg, &args, &res, 0);
+        status = nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0);
         if (status == 0) {
                 memcpy(server->attr_bitmask, res.attr_bitmask, sizeof(server->attr_bitmask));
                 server->caps &= ~(NFS_CAP_ACLS|NFS_CAP_HARDLINKS|
@@ -2140,7 +2177,7 @@ static int _nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle,
         };
 
         nfs_fattr_init(info->fattr);
-        return nfs4_call_sync(server, &msg, &args, &res, 0);
+        return nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0);
 }
 
 static int nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle,
@@ -2149,22 +2186,75 @@ static int nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle,
         struct nfs4_exception exception = { };
         int err;
         do {
-                err = nfs4_handle_exception(server,
-                                _nfs4_lookup_root(server, fhandle, info),
-                                &exception);
+                err = _nfs4_lookup_root(server, fhandle, info);
+                switch (err) {
+                case 0:
+                case -NFS4ERR_WRONGSEC:
+                        break;
+                default:
+                        err = nfs4_handle_exception(server, err, &exception);
+                }
         } while (exception.retry);
         return err;
 }
 
+static int nfs4_lookup_root_sec(struct nfs_server *server, struct nfs_fh *fhandle,
+                                struct nfs_fsinfo *info, rpc_authflavor_t flavor)
+{
+        struct rpc_auth *auth;
+        int ret;
+
+        auth = rpcauth_create(flavor, server->client);
+        if (!auth) {
+                ret = -EIO;
+                goto out;
+        }
+        ret = nfs4_lookup_root(server, fhandle, info);
+out:
+        return ret;
+}
+
+static int nfs4_find_root_sec(struct nfs_server *server, struct nfs_fh *fhandle,
+                              struct nfs_fsinfo *info)
+{
+        int i, len, status = 0;
+        rpc_authflavor_t flav_array[NFS_MAX_SECFLAVORS];
+
+        len = gss_mech_list_pseudoflavors(&flav_array[0]);
+        flav_array[len] = RPC_AUTH_NULL;
+        len += 1;
+
+        for (i = 0; i < len; i++) {
+                status = nfs4_lookup_root_sec(server, fhandle, info, flav_array[i]);
+                if (status == -NFS4ERR_WRONGSEC || status == -EACCES)
+                        continue;
+                break;
+        }
+        /*
+         * -EACCESS could mean that the user doesn't have correct permissions
+         * to access the mount.  It could also mean that we tried to mount
+         * with a gss auth flavor, but rpc.gssd isn't running.  Either way,
+         * existing mount programs don't handle -EACCES very well so it should
+         * be mapped to -EPERM instead.
+         */
+        if (status == -EACCES)
+                status = -EPERM;
+        return status;
+}
+
 /*
  * get the file handle for the "/" directory on the server
  */
 static int nfs4_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle,
                               struct nfs_fsinfo *info)
 {
-        int status;
-
-        status = nfs4_lookup_root(server, fhandle, info);
+        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);
         if (status == 0)
                 status = nfs4_server_capabilities(server, fhandle);
         if (status == 0)
@@ -2229,7 +2319,7 @@ static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle,
         };
         
         nfs_fattr_init(fattr);
-        return nfs4_call_sync(server, &msg, &args, &res, 0);
+        return nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0);
 }
 
 static int nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fattr *fattr)
@@ -2289,9 +2379,9 @@ nfs4_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr,
         return status;
 }
 
-static int _nfs4_proc_lookupfh(struct nfs_server *server, const struct nfs_fh *dirfh,
-                const struct qstr *name, struct nfs_fh *fhandle,
-                struct nfs_fattr *fattr)
+static int _nfs4_proc_lookupfh(struct rpc_clnt *clnt, struct nfs_server *server,
+                const struct nfs_fh *dirfh, const struct qstr *name,
+                struct nfs_fh *fhandle, struct nfs_fattr *fattr)
 {
         int                    status;
         struct nfs4_lookup_arg args = {
@@ -2313,7 +2403,7 @@ static int _nfs4_proc_lookupfh(struct nfs_server *server, const struct nfs_fh *d
         nfs_fattr_init(fattr);
 
         dprintk("NFS call  lookupfh %s\n", name->name);
-        status = nfs4_call_sync(server, &msg, &args, &res, 0);
+        status = nfs4_call_sync(clnt, server, &msg, &args.seq_args, &res.seq_res, 0);
         dprintk("NFS reply lookupfh: %d\n", status);
         return status;
 }
@@ -2325,7 +2415,7 @@ static int nfs4_proc_lookupfh(struct nfs_server *server, struct nfs_fh *dirfh,
         struct nfs4_exception exception = { };
         int err;
         do {
-                err = _nfs4_proc_lookupfh(server, dirfh, name, fhandle, fattr);
+                err = _nfs4_proc_lookupfh(server->client, server, dirfh, name, fhandle, fattr);
                 /* FIXME: !!!! */
                 if (err == -NFS4ERR_MOVED) {
                         err = -EREMOTE;
@@ -2336,27 +2426,41 @@ static int nfs4_proc_lookupfh(struct nfs_server *server, struct nfs_fh *dirfh,
         return err;
 }
 
-static int _nfs4_proc_lookup(struct inode *dir, const struct qstr *name,
-                struct nfs_fh *fhandle, struct nfs_fattr *fattr)
+static int _nfs4_proc_lookup(struct rpc_clnt *clnt, struct inode *dir,
+                const struct qstr *name, struct nfs_fh *fhandle,
+                struct nfs_fattr *fattr)
 {
         int status;
         
         dprintk("NFS call  lookup %s\n", name->name);
-        status = _nfs4_proc_lookupfh(NFS_SERVER(dir), NFS_FH(dir), name, fhandle, fattr);
+        status = _nfs4_proc_lookupfh(clnt, NFS_SERVER(dir), NFS_FH(dir), name, fhandle, fattr);
         if (status == -NFS4ERR_MOVED)
                 status = nfs4_get_referral(dir, name, fattr, fhandle);
         dprintk("NFS reply lookup: %d\n", status);
         return status;
 }
 
-static int nfs4_proc_lookup(struct inode *dir, struct qstr *name, struct nfs_fh *fhandle, struct nfs_fattr *fattr)
+void nfs_fixup_secinfo_attributes(struct nfs_fattr *fattr, struct nfs_fh *fh)
+{
+        memset(fh, 0, sizeof(struct nfs_fh));
+        fattr->fsid.major = 1;
+        fattr->valid |= NFS_ATTR_FATTR_TYPE | NFS_ATTR_FATTR_MODE |
+                NFS_ATTR_FATTR_NLINK | NFS_ATTR_FATTR_FSID | NFS_ATTR_FATTR_MOUNTPOINT;
+        fattr->mode = S_IFDIR | S_IRUGO | S_IXUGO;
+        fattr->nlink = 2;
+}
+
+static int nfs4_proc_lookup(struct rpc_clnt *clnt, struct inode *dir, struct qstr *name,
+                            struct nfs_fh *fhandle, struct nfs_fattr *fattr)
 {
         struct nfs4_exception exception = { };
         int err;
         do {
                 err = nfs4_handle_exception(NFS_SERVER(dir),
-                                _nfs4_proc_lookup(dir, name, fhandle, fattr),
+                                _nfs4_proc_lookup(clnt, dir, name, fhandle, fattr),
                                 &exception);
+                if (err == -EPERM)
+                        nfs_fixup_secinfo_attributes(fattr, fhandle);
         } while (exception.retry);
         return err;
 }
@@ -2401,7 +2505,7 @@ static int _nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry
         if (res.fattr == NULL)
                 return -ENOMEM;
 
-        status = nfs4_call_sync(server, &msg, &args, &res, 0);
+        status = nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0);
         if (!status) {
                 entry->mask = 0;
                 if (res.access & NFS4_ACCESS_READ)
@@ -2468,7 +2572,7 @@ static int _nfs4_proc_readlink(struct inode *inode, struct page *page,
                 .rpc_resp = &res,
         };
 
-        return nfs4_call_sync(NFS_SERVER(inode), &msg, &args, &res, 0);
+        return nfs4_call_sync(NFS_SERVER(inode)->client, NFS_SERVER(inode), &msg, &args.seq_args, &res.seq_res, 0);
 }
 
 static int nfs4_proc_readlink(struct inode *inode, struct page *page,
@@ -2557,7 +2661,7 @@ static int _nfs4_proc_remove(struct inode *dir, struct qstr *name)
         if (res.dir_attr == NULL)
                 goto out;
 
-        status = nfs4_call_sync(server, &msg, &args, &res, 1);
+        status = nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 1);
         if (status == 0) {
                 update_changeattr(dir, &res.cinfo);
                 nfs_post_op_update_inode(dir, res.dir_attr);
@@ -2658,7 +2762,7 @@ static int _nfs4_proc_rename(struct inode *old_dir, struct qstr *old_name,
         if (res.old_fattr == NULL || res.new_fattr == NULL)
                 goto out;
 
-        status = nfs4_call_sync(server, &msg, &arg, &res, 1);
+        status = nfs4_call_sync(server->client, server, &msg, &arg.seq_args, &res.seq_res, 1);
         if (!status) {
                 update_changeattr(old_dir, &res.old_cinfo);
                 nfs_post_op_update_inode(old_dir, res.old_fattr);
@@ -2709,7 +2813,7 @@ static int _nfs4_proc_link(struct inode *inode, struct inode *dir, struct qstr *
         if (res.fattr == NULL || res.dir_attr == NULL)
                 goto out;
 
-        status = nfs4_call_sync(server, &msg, &arg, &res, 1);
+        status = nfs4_call_sync(server->client, server, &msg, &arg.seq_args, &res.seq_res, 1);
         if (!status) {
                 update_changeattr(dir, &res.cinfo);
                 nfs_post_op_update_inode(dir, res.dir_attr);
@@ -2772,8 +2876,8 @@ static struct nfs4_createdata *nfs4_alloc_createdata(struct inode *dir,
 
 static int nfs4_do_create(struct inode *dir, struct dentry *dentry, struct nfs4_createdata *data)
 {
-        int status = nfs4_call_sync(NFS_SERVER(dir), &data->msg,
-                                    &data->arg, &data->res, 1);
+        int status = nfs4_call_sync(NFS_SERVER(dir)->client, NFS_SERVER(dir), &data->msg,
+                                    &data->arg.seq_args, &data->res.seq_res, 1);
         if (status == 0) {
                 update_changeattr(dir, &data->res.dir_cinfo);
                 nfs_post_op_update_inode(dir, data->res.dir_fattr);
@@ -2885,7 +2989,7 @@ static int _nfs4_proc_readdir(struct dentry *dentry, struct rpc_cred *cred,
                         (unsigned long long)cookie);
         nfs4_setup_readdir(cookie, NFS_COOKIEVERF(dir), dentry, &args);
         res.pgbase = args.pgbase;
-        status = nfs4_call_sync(NFS_SERVER(dir), &msg, &args, &res, 0);
+        status = nfs4_call_sync(NFS_SERVER(dir)->client, NFS_SERVER(dir), &msg, &args.seq_args, &res.seq_res, 0);
         if (status >= 0) {
                 memcpy(NFS_COOKIEVERF(dir), res.verifier.data, NFS4_VERIFIER_SIZE);
                 status += args.pgbase;
@@ -2977,7 +3081,7 @@ static int _nfs4_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle,
         };
 
         nfs_fattr_init(fsstat->fattr);
-        return  nfs4_call_sync(server, &msg, &args, &res, 0);
+        return  nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0);
 }
 
 static int nfs4_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsstat *fsstat)
@@ -3008,7 +3112,7 @@ static int _nfs4_do_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle,
                 .rpc_resp = &res,
         };
 
-        return nfs4_call_sync(server, &msg, &args, &res, 0);
+        return nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0);
 }
 
 static int nfs4_do_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsinfo *fsinfo)
@@ -3053,7 +3157,7 @@ static int _nfs4_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle
         }
 
         nfs_fattr_init(pathconf->fattr);
-        return nfs4_call_sync(server, &msg, &args, &res, 0);
+        return nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0);
 }
 
 static int nfs4_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle,
@@ -3070,15 +3174,10 @@ static int nfs4_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle,
         return err;
 }
 
-static int nfs4_read_done(struct rpc_task *task, struct nfs_read_data *data)
+static int nfs4_read_done_cb(struct rpc_task *task, struct nfs_read_data *data)
 {
         struct nfs_server *server = NFS_SERVER(data->inode);
 
-        dprintk("--> %s\n", __func__);
-
-        if (!nfs4_sequence_done(task, &data->res.seq_res))
-                return -EAGAIN;
-
         if (nfs4_async_handle_error(task, server, data->args.context->state) == -EAGAIN) {
                 nfs_restart_rpc(task, server->nfs_client);
                 return -EAGAIN;
@@ -3090,19 +3189,44 @@ static int nfs4_read_done(struct rpc_task *task, struct nfs_read_data *data)
         return 0;
 }
 
+static int nfs4_read_done(struct rpc_task *task, struct nfs_read_data *data)
+{
+
+        dprintk("--> %s\n", __func__);
+
+        if (!nfs4_sequence_done(task, &data->res.seq_res))
+                return -EAGAIN;
+
+        return data->read_done_cb(task, data);
+}
+
 static void nfs4_proc_read_setup(struct nfs_read_data *data, struct rpc_message *msg)
 {
         data->timestamp   = jiffies;
+        data->read_done_cb = nfs4_read_done_cb;
         msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READ];
 }
 
-static int nfs4_write_done(struct rpc_task *task, struct nfs_write_data *data)
+/* Reset the the nfs_read_data to send the read to the MDS. */
+void nfs4_reset_read(struct rpc_task *task, struct nfs_read_data *data)
+{
+        dprintk("%s Reset task for i/o through\n", __func__);
+        put_lseg(data->lseg);
+        data->lseg = NULL;
+        /* offsets will differ in the dense stripe case */
+        data->args.offset = data->mds_offset;
+        data->ds_clp = NULL;
+        data->args.fh     = NFS_FH(data->inode);
+        data->read_done_cb = nfs4_read_done_cb;
+        task->tk_ops = data->mds_ops;
+        rpc_task_reset_client(task, NFS_CLIENT(data->inode));
+}
+EXPORT_SYMBOL_GPL(nfs4_reset_read);
+
+static int nfs4_write_done_cb(struct rpc_task *task, struct nfs_write_data *data)
 {
         struct inode *inode = data->inode;
         
-        if (!nfs4_sequence_done(task, &data->res.seq_res))
-                return -EAGAIN;
-
         if (nfs4_async_handle_error(task, NFS_SERVER(inode), data->args.context->state) == -EAGAIN) {
                 nfs_restart_rpc(task, NFS_SERVER(inode)->nfs_client);
                 return -EAGAIN;
@@ -3114,23 +3238,50 @@ static int nfs4_write_done(struct rpc_task *task, struct nfs_write_data *data)
         return 0;
 }
 
+static int nfs4_write_done(struct rpc_task *task, struct nfs_write_data *data)
+{
+        if (!nfs4_sequence_done(task, &data->res.seq_res))
+                return -EAGAIN;
+        return data->write_done_cb(task, data);
+}
+
+/* Reset the the nfs_write_data to send the write to the MDS. */
+void nfs4_reset_write(struct rpc_task *task, struct nfs_write_data *data)
+{
+        dprintk("%s Reset task for i/o through\n", __func__);
+        put_lseg(data->lseg);
+        data->lseg          = NULL;
+        data->ds_clp        = NULL;
+        data->write_done_cb = nfs4_write_done_cb;
+        data->args.fh       = NFS_FH(data->inode);
+        data->args.bitmask  = data->res.server->cache_consistency_bitmask;
+        data->args.offset   = data->mds_offset;
+        data->res.fattr     = &data->fattr;
+        task->tk_ops        = data->mds_ops;
+        rpc_task_reset_client(task, NFS_CLIENT(data->inode));
+}
+EXPORT_SYMBOL_GPL(nfs4_reset_write);
+
 static void nfs4_proc_write_setup(struct nfs_write_data *data, struct rpc_message *msg)
 {
         struct nfs_server *server = NFS_SERVER(data->inode);
 
-        data->args.bitmask = server->cache_consistency_bitmask;
+        if (data->lseg) {
+                data->args.bitmask = NULL;
+                data->res.fattr = NULL;
+        } else
+                data->args.bitmask = server->cache_consistency_bitmask;
+        if (!data->write_done_cb)
+                data->write_done_cb = nfs4_write_done_cb;
         data->res.server = server;
         data->timestamp   = jiffies;
 
         msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_WRITE];
 }
 
-static int nfs4_commit_done(struct rpc_task *task, struct nfs_write_data *data)
+static int nfs4_commit_done_cb(struct rpc_task *task, struct nfs_write_data *data)
 {
         struct inode *inode = data->inode;
-        
-        if (!nfs4_sequence_done(task, &data->res.seq_res))
-                return -EAGAIN;
 
         if (nfs4_async_handle_error(task, NFS_SERVER(inode), NULL) == -EAGAIN) {
                 nfs_restart_rpc(task, NFS_SERVER(inode)->nfs_client);
@@ -3140,11 +3291,24 @@ static int nfs4_commit_done(struct rpc_task *task, struct nfs_write_data *data)
         return 0;
 }
 
+static int nfs4_commit_done(struct rpc_task *task, struct nfs_write_data *data)
+{
+        if (!nfs4_sequence_done(task, &data->res.seq_res))
+                return -EAGAIN;
+        return data->write_done_cb(task, data);
+}
+
 static void nfs4_proc_commit_setup(struct nfs_write_data *data, struct rpc_message *msg)
 {
         struct nfs_server *server = NFS_SERVER(data->inode);
-        
-        data->args.bitmask = server->cache_consistency_bitmask;
+
+        if (data->lseg) {
+                data->args.bitmask = NULL;
+                data->res.fattr = NULL;
+        } else
+                data->args.bitmask = server->cache_consistency_bitmask;
+        if (!data->write_done_cb)
+                data->write_done_cb = nfs4_commit_done_cb;
         data->res.server = server;
         msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_COMMIT];
 }
@@ -3178,7 +3342,7 @@ static void nfs4_renew_done(struct rpc_task *task, void *calldata)
         if (task->tk_status < 0) {
                 /* Unless we're shutting down, schedule state recovery! */
                 if (test_bit(NFS_CS_RENEWD, &clp->cl_res_state) != 0)
-                        nfs4_schedule_state_recovery(clp);
+                        nfs4_schedule_lease_recovery(clp);
                 return;
         }
         do_renew_lease(clp, timestamp);
@@ -3252,6 +3416,35 @@ static void buf_to_pages(const void *buf, size_t buflen,
         }
 }
 
+static int buf_to_pages_noslab(const void *buf, size_t buflen,
+                struct page **pages, unsigned int *pgbase)
+{
+        struct page *newpage, **spages;
+        int rc = 0;
+        size_t len;
+        spages = pages;
+
+        do {
+                len = min_t(size_t, PAGE_CACHE_SIZE, buflen);
+                newpage = alloc_page(GFP_KERNEL);
+
+                if (newpage == NULL)
+                        goto unwind;
+                memcpy(page_address(newpage), buf, len);
+                buf += len;
+                buflen -= len;
+                *pages++ = newpage;
+                rc++;
+        } while (buflen != 0);
+
+        return rc;
+
+unwind:
+        for(; rc > 0; rc--)
+                __free_page(spages[rc-1]);
+        return -ENOMEM;
+}
+
 struct nfs4_cached_acl {
         int cached;
         size_t len;
@@ -3353,7 +3546,7 @@ static ssize_t __nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t bu
                 resp_buf = buf;
                 buf_to_pages(buf, buflen, args.acl_pages, &args.acl_pgbase);
         }
-        ret = nfs4_call_sync(NFS_SERVER(inode), &msg, &args, &res, 0);
+        ret = nfs4_call_sync(NFS_SERVER(inode)->client, NFS_SERVER(inode), &msg, &args.seq_args, &res.seq_res, 0);
         if (ret)
                 goto out_free;
         if (res.acl_len > args.acl_len)
@@ -3420,13 +3613,23 @@ static int __nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t bufl
                 .rpc_argp       = &arg,
                 .rpc_resp       = &res,
         };
-        int ret;
+        int ret, i;
 
         if (!nfs4_server_supports_acls(server))
                 return -EOPNOTSUPP;
+        i = buf_to_pages_noslab(buf, buflen, arg.acl_pages, &arg.acl_pgbase);
+        if (i < 0)
+                return i;
         nfs_inode_return_delegation(inode);
-        buf_to_pages(buf, buflen, arg.acl_pages, &arg.acl_pgbase);
-        ret = nfs4_call_sync(server, &msg, &arg, &res, 1);
+        ret = nfs4_call_sync(server->client, server, &msg, &arg.seq_args, &res.seq_res, 1);
+
+        /*
+         * Free each page after tx, so the only ref left is
+         * held by the network stack
+         */
+        for (; i > 0; i--)
+                put_page(pages[i-1]);
+
         /*
          * Acl update can result in inode attribute update.
          * so mark the attribute cache invalid.
@@ -3464,12 +3667,13 @@ nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server,
                 case -NFS4ERR_OPENMODE:
                         if (state == NULL)
                                 break;
-                        nfs4_state_mark_reclaim_nograce(clp, state);
-                        goto do_state_recovery;
+                        nfs4_schedule_stateid_recovery(server, state);
+                        goto wait_on_recovery;
                 case -NFS4ERR_STALE_STATEID:
                 case -NFS4ERR_STALE_CLIENTID:
                 case -NFS4ERR_EXPIRED:
-                        goto do_state_recovery;
+                        nfs4_schedule_lease_recovery(clp);
+                        goto wait_on_recovery;
 #if defined(CONFIG_NFS_V4_1)
                 case -NFS4ERR_BADSESSION:
                 case -NFS4ERR_BADSLOT:
@@ -3480,7 +3684,7 @@ 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);
-                        nfs4_schedule_state_recovery(clp);
+                        nfs4_schedule_session_recovery(clp->cl_session);
                         task->tk_status = 0;
                         return -EAGAIN;
 #endif /* CONFIG_NFS_V4_1 */
@@ -3497,9 +3701,8 @@ nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server,
         }
         task->tk_status = nfs4_map_errors(task->tk_status);
         return 0;
-do_state_recovery:
+wait_on_recovery:
         rpc_sleep_on(&clp->cl_rpcwaitq, task, NULL);
-        nfs4_schedule_state_recovery(clp);
         if (test_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state) == 0)
                 rpc_wake_up_queued_task(&clp->cl_rpcwaitq, task);
         task->tk_status = 0;
@@ -3548,21 +3751,20 @@ int nfs4_proc_setclientid(struct nfs_client *clp, u32 program,
                                 sizeof(setclientid.sc_uaddr), "%s.%u.%u",
                                 clp->cl_ipaddr, port >> 8, port & 255);
 
-                status = rpc_call_sync(clp->cl_rpcclient, &msg, 0);
+                status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT);
                 if (status != -NFS4ERR_CLID_INUSE)
                         break;
-                if (signalled())
+                if (loop != 0) {
+                        ++clp->cl_id_uniquifier;
                         break;
-                if (loop++ & 1)
-                        ssleep(clp->cl_lease_time / HZ + 1);
-                else
-                        if (++clp->cl_id_uniquifier == 0)
-                                break;
+                }
+                ++loop;
+                ssleep(clp->cl_lease_time / HZ + 1);
         }
         return status;
 }
 
-static int _nfs4_proc_setclientid_confirm(struct nfs_client *clp,
+int nfs4_proc_setclientid_confirm(struct nfs_client *clp,
                 struct nfs4_setclientid_res *arg,
                 struct rpc_cred *cred)
 {
@@ -3577,7 +3779,7 @@ static int _nfs4_proc_setclientid_confirm(struct nfs_client *clp,
         int status;
 
         now = jiffies;
-        status = rpc_call_sync(clp->cl_rpcclient, &msg, 0);
+        status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT);
         if (status == 0) {
                 spin_lock(&clp->cl_lock);
                 clp->cl_lease_time = fsinfo.lease_time * HZ;
@@ -3587,26 +3789,6 @@ static int _nfs4_proc_setclientid_confirm(struct nfs_client *clp,
         return status;
 }
 
-int nfs4_proc_setclientid_confirm(struct nfs_client *clp,
-                struct nfs4_setclientid_res *arg,
-                struct rpc_cred *cred)
-{
-        long timeout = 0;
-        int err;
-        do {
-                err = _nfs4_proc_setclientid_confirm(clp, arg, cred);
-                switch (err) {
-                        case 0:
-                                return err;
-                        case -NFS4ERR_RESOURCE:
-                                /* The IBM lawyers misread another document! */
-                        case -NFS4ERR_DELAY:
-                                err = nfs4_delay(clp->cl_rpcclient, &timeout);
-                }
-        } while (err == 0);
-        return err;
-}
-
 struct nfs4_delegreturndata {
         struct nfs4_delegreturnargs args;
         struct nfs4_delegreturnres res;
@@ -3781,7 +3963,7 @@ static int _nfs4_proc_getlk(struct nfs4_state *state, int cmd, struct file_lock
         lsp = request->fl_u.nfs4_fl.owner;
         arg.lock_owner.id = lsp->ls_id.id;
         arg.lock_owner.s_dev = server->s_dev;
-        status = nfs4_call_sync(server, &msg, &arg, &res, 1);
+        status = nfs4_call_sync(server->client, server, &msg, &arg.seq_args, &res.seq_res, 1);
         switch (status) {
                 case 0:
                         request->fl_type = F_UNLCK;
@@ -4110,7 +4292,7 @@ static void nfs4_lock_release(void *calldata)
                 task = nfs4_do_unlck(&data->fl, data->ctx, data->lsp,
                                 data->arg.lock_seqid);
                 if (!IS_ERR(task))
-                        rpc_put_task(task);
+                        rpc_put_task_async(task);
                 dprintk("%s: cancelling lock!\n", __func__);
         } else
                 nfs_free_seqid(data->arg.lock_seqid);
@@ -4134,23 +4316,18 @@ static const struct rpc_call_ops nfs4_recover_lock_ops = {
 
 static void nfs4_handle_setlk_error(struct nfs_server *server, struct nfs4_lock_state *lsp, int new_lock_owner, int error)
 {
-        struct nfs_client *clp = server->nfs_client;
-        struct nfs4_state *state = lsp->ls_state;
-
         switch (error) {
         case -NFS4ERR_ADMIN_REVOKED:
         case -NFS4ERR_BAD_STATEID:
-        case -NFS4ERR_EXPIRED:
+                lsp->ls_seqid.flags &= ~NFS_SEQID_CONFIRMED;
                 if (new_lock_owner != 0 ||
                    (lsp->ls_flags & NFS_LOCK_INITIALIZED) != 0)
-                        nfs4_state_mark_reclaim_nograce(clp, state);
-                lsp->ls_seqid.flags &= ~NFS_SEQID_CONFIRMED;
+                        nfs4_schedule_stateid_recovery(server, lsp->ls_state);
                 break;
         case -NFS4ERR_STALE_STATEID:
-                if (new_lock_owner != 0 ||
-                    (lsp->ls_flags & NFS_LOCK_INITIALIZED) != 0)
-                        nfs4_state_mark_reclaim_reboot(clp, state);
                 lsp->ls_seqid.flags &= ~NFS_SEQID_CONFIRMED;
+        case -NFS4ERR_EXPIRED:
+                nfs4_schedule_lease_recovery(server->nfs_client);
         };
 }
 
@@ -4366,12 +4543,14 @@ int nfs4_lock_delegation_recall(struct nfs4_state *state, struct file_lock *fl)
                         case -NFS4ERR_EXPIRED:
                         case -NFS4ERR_STALE_CLIENTID:
                         case -NFS4ERR_STALE_STATEID:
+                                nfs4_schedule_lease_recovery(server->nfs_client);
+                                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);
+                                nfs4_schedule_session_recovery(server->nfs_client->cl_session);
                                 goto out;
                         case -ERESTARTSYS:
                                 /*
@@ -4381,7 +4560,7 @@ int nfs4_lock_delegation_recall(struct nfs4_state *state, struct file_lock *fl)
                         case -NFS4ERR_ADMIN_REVOKED:
                         case -NFS4ERR_BAD_STATEID:
                         case -NFS4ERR_OPENMODE:
-                                nfs4_state_mark_reclaim_nograce(server->nfs_client, state);
+                                nfs4_schedule_stateid_recovery(server, state);
                                 err = 0;
                                 goto out;
                         case -EKEYEXPIRED:
@@ -4512,12 +4691,46 @@ int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name,
         nfs_fattr_init(&fs_locations->fattr);
         fs_locations->server = server;
         fs_locations->nlocations = 0;
-        status = nfs4_call_sync(server, &msg, &args, &res, 0);
+        status = nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0);
         nfs_fixup_referral_attributes(&fs_locations->fattr);
         dprintk("%s: returned status = %d\n", __func__, status);
         return status;
 }
 
+static int _nfs4_proc_secinfo(struct inode *dir, const struct qstr *name, struct nfs4_secinfo_flavors *flavors)
+{
+        int status;
+        struct nfs4_secinfo_arg args = {
+                .dir_fh = NFS_FH(dir),
+                .name   = name,
+        };
+        struct nfs4_secinfo_res res = {
+                .flavors     = flavors,
+        };
+        struct rpc_message msg = {
+                .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SECINFO],
+                .rpc_argp = &args,
+                .rpc_resp = &res,
+        };
+
+        dprintk("NFS call  secinfo %s\n", name->name);
+        status = nfs4_call_sync(NFS_SERVER(dir)->client, NFS_SERVER(dir), &msg, &args.seq_args, &res.seq_res, 0);
+        dprintk("NFS reply  secinfo: %d\n", status);
+        return status;
+}
+
+int nfs4_proc_secinfo(struct inode *dir, const struct qstr *name, struct nfs4_secinfo_flavors *flavors)
+{
+        struct nfs4_exception exception = { };
+        int err;
+        do {
+                err = nfs4_handle_exception(NFS_SERVER(dir),
+                                _nfs4_proc_secinfo(dir, name, flavors),
+                                &exception);
+        } while (exception.retry);
+        return err;
+}
+
 #ifdef CONFIG_NFS_V4_1
 /*
  * Check the exchange flags returned by the server for invalid flags, having
@@ -4580,7 +4793,7 @@ int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred)
                                 init_utsname()->domainname,
                                 clp->cl_rpcclient->cl_auth->au_flavor);
 
-        status = rpc_call_sync(clp->cl_rpcclient, &msg, 0);
+        status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT);
         if (!status)
                 status = nfs4_check_cl_exchange_flags(clp->cl_exchange_flags);
         dprintk("<-- %s status= %d\n", __func__, status);
@@ -4663,7 +4876,8 @@ int nfs4_proc_get_lease_time(struct nfs_client *clp, struct nfs_fsinfo *fsinfo)
                 .rpc_client = clp->cl_rpcclient,
                 .rpc_message = &msg,
                 .callback_ops = &nfs4_get_lease_time_ops,
-                .callback_data = &data
+                .callback_data = &data,
+                .flags = RPC_TASK_TIMEOUT,
         };
         int status;
 
@@ -4965,7 +5179,7 @@ static int _nfs4_proc_create_session(struct nfs_client *clp)
         nfs4_init_channel_attrs(&args);
         args.flags = (SESSION4_PERSIST | SESSION4_BACK_CHAN);
 
-        status = rpc_call_sync(session->clp->cl_rpcclient, &msg, 0);
+        status = rpc_call_sync(session->clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT);
 
         if (!status)
                 /* Verify the session's negotiated channel_attrs values */
@@ -5032,7 +5246,7 @@ int nfs4_proc_destroy_session(struct nfs4_session *session)
         msg.rpc_argp = session;
         msg.rpc_resp = NULL;
         msg.rpc_cred = NULL;
-        status = rpc_call_sync(session->clp->cl_rpcclient, &msg, 0);
+        status = rpc_call_sync(session->clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT);
 
         if (status)
                 printk(KERN_WARNING
@@ -5073,6 +5287,27 @@ int nfs4_init_session(struct nfs_server *server)
         return ret;
 }
 
+int nfs4_init_ds_session(struct nfs_client *clp)
+{
+        struct nfs4_session *session = clp->cl_session;
+        int ret;
+
+        if (!test_and_clear_bit(NFS4_SESSION_INITING, &session->session_state))
+                return 0;
+
+        ret = nfs4_client_recover_expired_lease(clp);
+        if (!ret)
+                /* Test for the DS role */
+                if (!is_ds_client(clp))
+                        ret = -ENODEV;
+        if (!ret)
+                ret = nfs4_check_client_ready(clp);
+        return ret;
+
+}
+EXPORT_SYMBOL_GPL(nfs4_init_ds_session);
+
+
 /*
  * Renew the cl_session lease.
  */
@@ -5100,7 +5335,7 @@ static int nfs41_sequence_handle_errors(struct rpc_task *task, struct nfs_client
                 rpc_delay(task, NFS4_POLL_RETRY_MAX);
                 return -EAGAIN;
         default:
-                nfs4_schedule_state_recovery(clp);
+                nfs4_schedule_lease_recovery(clp);
         }
         return 0;
 }
@@ -5187,7 +5422,7 @@ static int nfs41_proc_async_sequence(struct nfs_client *clp, struct rpc_cred *cr
         if (IS_ERR(task))
                 ret = PTR_ERR(task);
         else
-                rpc_put_task(task);
+                rpc_put_task_async(task);
         dprintk("<-- %s status=%d\n", __func__, ret);
         return ret;
 }
@@ -5203,8 +5438,13 @@ static int nfs4_proc_sequence(struct nfs_client *clp, struct rpc_cred *cred)
                 goto out;
         }
         ret = rpc_wait_for_completion_task(task);
-        if (!ret)
+        if (!ret) {
+                struct nfs4_sequence_res *res = task->tk_msg.rpc_resp;
+
+                if (task->tk_status == 0)
+                        nfs41_handle_sequence_flag_errors(clp, res->sr_status_flags);
                 ret = task->tk_status;
+        }
         rpc_put_task(task);
 out:
         dprintk("<-- %s status=%d\n", __func__, ret);
@@ -5241,7 +5481,7 @@ static int nfs41_reclaim_complete_handle_errors(struct rpc_task *task, struct nf
                 rpc_delay(task, NFS4_POLL_RETRY_MAX);
                 return -EAGAIN;
         default:
-                nfs4_schedule_state_recovery(clp);
+                nfs4_schedule_lease_recovery(clp);
         }
         return 0;
 }
@@ -5309,6 +5549,9 @@ static int nfs41_proc_reclaim_complete(struct nfs_client *clp)
                 status = PTR_ERR(task);
                 goto out;
         }
+        status = nfs4_wait_for_completion_rpc_task(task);
+        if (status == 0)
+                status = task->tk_status;
         rpc_put_task(task);
         return 0;
 out:
@@ -5371,8 +5614,6 @@ static void nfs4_layoutget_release(void *calldata)
         struct nfs4_layoutget *lgp = calldata;
 
         dprintk("--> %s\n", __func__);
-        if (lgp->res.layout.buf != NULL)
-                free_page((unsigned long) lgp->res.layout.buf);
         put_nfs_open_context(lgp->args.ctx);
         kfree(calldata);
         dprintk("<-- %s\n", __func__);
@@ -5404,12 +5645,7 @@ int nfs4_proc_layoutget(struct nfs4_layoutget *lgp)
 
         dprintk("--> %s\n", __func__);
 
-        lgp->res.layout.buf = (void *)__get_free_page(GFP_NOFS);
-        if (lgp->res.layout.buf == NULL) {
-                nfs4_layoutget_release(lgp);
-                return -ENOMEM;
-        }
-
+        lgp->res.layoutp = &lgp->args.layout;
         lgp->res.seq_res.sr_slot = NULL;
         task = rpc_run_task(&task_setup_data);
         if (IS_ERR(task))
@@ -5441,7 +5677,7 @@ _nfs4_proc_getdeviceinfo(struct nfs_server *server, struct pnfs_device *pdev)
         int status;
 
         dprintk("--> %s\n", __func__);
-        status = nfs4_call_sync(server, &msg, &args, &res, 0);
+        status = nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0);
         dprintk("<-- %s status=%d\n", __func__, status);
 
         return status;
@@ -5461,6 +5697,100 @@ int nfs4_proc_getdeviceinfo(struct nfs_server *server, struct pnfs_device *pdev)
 }
 EXPORT_SYMBOL_GPL(nfs4_proc_getdeviceinfo);
 
+static void nfs4_layoutcommit_prepare(struct rpc_task *task, void *calldata)
+{
+        struct nfs4_layoutcommit_data *data = calldata;
+        struct nfs_server *server = NFS_SERVER(data->args.inode);
+
+        if (nfs4_setup_sequence(server, &data->args.seq_args,
+                                &data->res.seq_res, 1, task))
+                return;
+        rpc_call_start(task);
+}
+
+static void
+nfs4_layoutcommit_done(struct rpc_task *task, void *calldata)
+{
+        struct nfs4_layoutcommit_data *data = calldata;
+        struct nfs_server *server = NFS_SERVER(data->args.inode);
+
+        if (!nfs4_sequence_done(task, &data->res.seq_res))
+                return;
+
+        switch (task->tk_status) { /* Just ignore these failures */
+        case NFS4ERR_DELEG_REVOKED: /* layout was recalled */
+        case NFS4ERR_BADIOMODE:     /* no IOMODE_RW layout for range */
+        case NFS4ERR_BADLAYOUT:     /* no layout */
+        case NFS4ERR_GRACE:         /* loca_recalim always false */
+                task->tk_status = 0;
+        }
+
+        if (nfs4_async_handle_error(task, server, NULL) == -EAGAIN) {
+                nfs_restart_rpc(task, server->nfs_client);
+                return;
+        }
+
+        if (task->tk_status == 0)
+                nfs_post_op_update_inode_force_wcc(data->args.inode,
+                                                   data->res.fattr);
+}
+
+static void nfs4_layoutcommit_release(void *calldata)
+{
+        struct nfs4_layoutcommit_data *data = calldata;
+
+        /* Matched by references in pnfs_set_layoutcommit */
+        put_lseg(data->lseg);
+        put_rpccred(data->cred);
+        kfree(data);
+}
+
+static const struct rpc_call_ops nfs4_layoutcommit_ops = {
+        .rpc_call_prepare = nfs4_layoutcommit_prepare,
+        .rpc_call_done = nfs4_layoutcommit_done,
+        .rpc_release = nfs4_layoutcommit_release,
+};
+
+int
+nfs4_proc_layoutcommit(struct nfs4_layoutcommit_data *data, bool sync)
+{
+        struct rpc_message msg = {
+                .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LAYOUTCOMMIT],
+                .rpc_argp = &data->args,
+                .rpc_resp = &data->res,
+                .rpc_cred = data->cred,
+        };
+        struct rpc_task_setup task_setup_data = {
+                .task = &data->task,
+                .rpc_client = NFS_CLIENT(data->args.inode),
+                .rpc_message = &msg,
+                .callback_ops = &nfs4_layoutcommit_ops,
+                .callback_data = data,
+                .flags = RPC_TASK_ASYNC,
+        };
+        struct rpc_task *task;
+        int status = 0;
+
+        dprintk("NFS: %4d initiating layoutcommit call. sync %d "
+                "lbw: %llu inode %lu\n",
+                data->task.tk_pid, sync,
+                data->args.lastbytewritten,
+                data->args.inode->i_ino);
+
+        task = rpc_run_task(&task_setup_data);
+        if (IS_ERR(task))
+                return PTR_ERR(task);
+        if (sync == false)
+                goto out;
+        status = nfs4_wait_for_completion_rpc_task(task);
+        if (status != 0)
+                goto out;
+        status = task->tk_status;
+out:
+        dprintk("%s: status %d\n", __func__, status);
+        rpc_put_task(task);
+        return status;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 struct nfs4_state_recovery_ops nfs40_reboot_recovery_ops = {
@@ -5595,6 +5925,8 @@ const struct nfs_rpc_ops nfs_v4_clientops = {
         .clear_acl_cache = nfs4_zap_acl_attr,
         .close_context  = nfs4_close_context,
         .open_context   = nfs4_atomic_open,
+        .init_client    = nfs4_init_client,
+        .secinfo        = nfs4_proc_secinfo,
 };
 
 static const struct xattr_handler nfs4_xattr_nfs4_acl_handler = {
diff --git a/fs/nfs/nfs4renewd.c b/fs/nfs/nfs4renewd.c
index 402143d75fc5..df8e7f3ca56d 100644
--- a/fs/nfs/nfs4renewd.c
+++ b/fs/nfs/nfs4renewd.c
@@ -64,12 +64,8 @@ nfs4_renew_state(struct work_struct *work)
         ops = clp->cl_mvops->state_renewal_ops;
         dprintk("%s: start\n", __func__);
 
-        rcu_read_lock();
-        if (list_empty(&clp->cl_superblocks)) {
-                rcu_read_unlock();
+        if (test_bit(NFS_CS_STOP_RENEW, &clp->cl_res_state))
                 goto out;
-        }
-        rcu_read_unlock();
 
         spin_lock(&clp->cl_lock);
         lease = clp->cl_lease_time;
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index e6742b57a04c..036f5adc9e1f 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -64,10 +64,15 @@ static LIST_HEAD(nfs4_clientid_list);
 
 int nfs4_init_clientid(struct nfs_client *clp, struct rpc_cred *cred)
 {
-        struct nfs4_setclientid_res clid;
+        struct nfs4_setclientid_res clid = {
+                .clientid = clp->cl_clientid,
+                .confirm = clp->cl_confirm,
+        };
         unsigned short port;
         int status;
 
+        if (test_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state))
+                goto do_confirm;
         port = nfs_callback_tcpport;
         if (clp->cl_addr.ss_family == AF_INET6)
                 port = nfs_callback_tcpport6;
@@ -75,10 +80,14 @@ int nfs4_init_clientid(struct nfs_client *clp, struct rpc_cred *cred)
         status = nfs4_proc_setclientid(clp, NFS4_CALLBACK, port, cred, &clid);
         if (status != 0)
                 goto out;
+        clp->cl_clientid = clid.clientid;
+        clp->cl_confirm = clid.confirm;
+        set_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state);
+do_confirm:
         status = nfs4_proc_setclientid_confirm(clp, &clid, cred);
         if (status != 0)
                 goto out;
-        clp->cl_clientid = clid.clientid;
+        clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state);
         nfs4_schedule_state_renewal(clp);
 out:
         return status;
@@ -153,6 +162,11 @@ static int nfs41_setup_state_renewal(struct nfs_client *clp)
         int status;
         struct nfs_fsinfo fsinfo;
 
+        if (!test_bit(NFS_CS_CHECK_LEASE_TIME, &clp->cl_res_state)) {
+                nfs4_schedule_state_renewal(clp);
+                return 0;
+        }
+
         status = nfs4_proc_get_lease_time(clp, &fsinfo);
         if (status == 0) {
                 /* Update lease time and schedule renewal */
@@ -225,13 +239,18 @@ int nfs41_init_clientid(struct nfs_client *clp, struct rpc_cred *cred)
 {
         int status;
 
+        if (test_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state))
+                goto do_confirm;
         nfs4_begin_drain_session(clp);
         status = nfs4_proc_exchange_id(clp, cred);
         if (status != 0)
                 goto out;
+        set_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state);
+do_confirm:
         status = nfs4_proc_create_session(clp);
         if (status != 0)
                 goto out;
+        clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state);
         nfs41_setup_state_renewal(clp);
         nfs_mark_client_ready(clp, NFS_CS_READY);
 out:
@@ -585,7 +604,8 @@ nfs4_get_open_state(struct inode *inode, struct nfs4_state_owner *owner)
                 state->owner = owner;
                 atomic_inc(&owner->so_count);
                 list_add(&state->inode_states, &nfsi->open_states);
-                state->inode = igrab(inode);
+                ihold(inode);
+                state->inode = inode;
                 spin_unlock(&inode->i_lock);
                 /* Note: The reclaim code dictates that we add stateless
                  * and read-only stateids to the end of the list */
@@ -1007,9 +1027,9 @@ void nfs4_schedule_state_manager(struct nfs_client *clp)
 }
 
 /*
- * Schedule a state recovery attempt
+ * Schedule a lease recovery attempt
  */
-void nfs4_schedule_state_recovery(struct nfs_client *clp)
+void nfs4_schedule_lease_recovery(struct nfs_client *clp)
 {
         if (!clp)
                 return;
@@ -1018,7 +1038,7 @@ void nfs4_schedule_state_recovery(struct nfs_client *clp)
         nfs4_schedule_state_manager(clp);
 }
 
-int nfs4_state_mark_reclaim_reboot(struct nfs_client *clp, struct nfs4_state *state)
+static int nfs4_state_mark_reclaim_reboot(struct nfs_client *clp, struct nfs4_state *state)
 {
 
         set_bit(NFS_STATE_RECLAIM_REBOOT, &state->flags);
@@ -1032,7 +1052,7 @@ int nfs4_state_mark_reclaim_reboot(struct nfs_client *clp, struct nfs4_state *st
         return 1;
 }
 
-int nfs4_state_mark_reclaim_nograce(struct nfs_client *clp, struct nfs4_state *state)
+static int nfs4_state_mark_reclaim_nograce(struct nfs_client *clp, struct nfs4_state *state)
 {
         set_bit(NFS_STATE_RECLAIM_NOGRACE, &state->flags);
         clear_bit(NFS_STATE_RECLAIM_REBOOT, &state->flags);
@@ -1041,6 +1061,14 @@ int nfs4_state_mark_reclaim_nograce(struct nfs_client *clp, struct nfs4_state *s
         return 1;
 }
 
+void nfs4_schedule_stateid_recovery(const struct nfs_server *server, struct nfs4_state *state)
+{
+        struct nfs_client *clp = server->nfs_client;
+
+        nfs4_state_mark_reclaim_nograce(clp, state);
+        nfs4_schedule_state_manager(clp);
+}
+
 static int nfs4_reclaim_locks(struct nfs4_state *state, const struct nfs4_state_recovery_ops *ops)
 {
         struct inode *inode = state->inode;
@@ -1436,10 +1464,16 @@ static int nfs4_reclaim_lease(struct nfs_client *clp)
 }
 
 #ifdef CONFIG_NFS_V4_1
+void nfs4_schedule_session_recovery(struct nfs4_session *session)
+{
+        nfs4_schedule_lease_recovery(session->clp);
+}
+EXPORT_SYMBOL_GPL(nfs4_schedule_session_recovery);
+
 void nfs41_handle_recall_slot(struct nfs_client *clp)
 {
         set_bit(NFS4CLNT_RECALL_SLOT, &clp->cl_state);
-        nfs4_schedule_state_recovery(clp);
+        nfs4_schedule_state_manager(clp);
 }
 
 static void nfs4_reset_all_state(struct nfs_client *clp)
@@ -1447,7 +1481,7 @@ static void nfs4_reset_all_state(struct nfs_client *clp)
         if (test_and_set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state) == 0) {
                 clp->cl_boot_time = CURRENT_TIME;
                 nfs4_state_start_reclaim_nograce(clp);
-                nfs4_schedule_state_recovery(clp);
+                nfs4_schedule_state_manager(clp);
         }
 }
 
@@ -1455,7 +1489,7 @@ static void nfs41_handle_server_reboot(struct nfs_client *clp)
 {
         if (test_and_set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state) == 0) {
                 nfs4_state_start_reclaim_reboot(clp);
-                nfs4_schedule_state_recovery(clp);
+                nfs4_schedule_state_manager(clp);
         }
 }
 
@@ -1475,7 +1509,7 @@ static void nfs41_handle_cb_path_down(struct nfs_client *clp)
 {
         nfs_expire_all_delegations(clp);
         if (test_and_set_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state) == 0)
-                nfs4_schedule_state_recovery(clp);
+                nfs4_schedule_state_manager(clp);
 }
 
 void nfs41_handle_sequence_flag_errors(struct nfs_client *clp, u32 flags)
@@ -1564,20 +1598,23 @@ static int nfs4_recall_slot(struct nfs_client *clp) { return 0; }
  */
 static void nfs4_set_lease_expired(struct nfs_client *clp, int status)
 {
-        if (nfs4_has_session(clp)) {
-                switch (status) {
-                case -NFS4ERR_DELAY:
-                case -NFS4ERR_CLID_INUSE:
-                case -EAGAIN:
-                        break;
+        switch (status) {
+        case -NFS4ERR_CLID_INUSE:
+        case -NFS4ERR_STALE_CLIENTID:
+                clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state);
+                break;
+        case -NFS4ERR_DELAY:
+        case -ETIMEDOUT:
+        case -EAGAIN:
+                ssleep(1);
+                break;
 
-                case -EKEYEXPIRED:
-                        nfs4_warn_keyexpired(clp->cl_hostname);
-                case -NFS4ERR_NOT_SAME: /* FixMe: implement recovery
-                                         * in nfs4_exchange_id */
-                default:
-                        return;
-                }
+        case -EKEYEXPIRED:
+                nfs4_warn_keyexpired(clp->cl_hostname);
+        case -NFS4ERR_NOT_SAME: /* FixMe: implement recovery
+                                 * in nfs4_exchange_id */
+        default:
+                return;
         }
         set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
 }
@@ -1587,7 +1624,7 @@ static void nfs4_state_manager(struct nfs_client *clp)
         int status = 0;
 
         /* Ensure exclusive access to NFSv4 state */
-        for(;;) {
+        do {
                 if (test_and_clear_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state)) {
                         /* We're going to have to re-establish a clientid */
                         status = nfs4_reclaim_lease(clp);
@@ -1671,7 +1708,7 @@ static void nfs4_state_manager(struct nfs_client *clp)
                         break;
                 if (test_and_set_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state) != 0)
                         break;
-        }
+        } while (atomic_read(&clp->cl_count) > 1);
         return;
 out_error:
         printk(KERN_WARNING "Error: state manager failed on NFSv4 server %s"
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 4e2c168b6ee9..c3ccd2c46834 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -46,6 +46,7 @@
 #include <linux/kdev_t.h>
 #include <linux/sunrpc/clnt.h>
 #include <linux/sunrpc/msg_prot.h>
+#include <linux/sunrpc/gss_api.h>
 #include <linux/nfs.h>
 #include <linux/nfs4.h>
 #include <linux/nfs_fs.h>
@@ -112,7 +113,7 @@ static int nfs4_stat_to_errno(int);
 #define encode_restorefh_maxsz  (op_encode_hdr_maxsz)
 #define decode_restorefh_maxsz  (op_decode_hdr_maxsz)
 #define encode_fsinfo_maxsz     (encode_getattr_maxsz)
-#define decode_fsinfo_maxsz     (op_decode_hdr_maxsz + 11)
+#define decode_fsinfo_maxsz     (op_decode_hdr_maxsz + 15)
 #define encode_renew_maxsz      (op_encode_hdr_maxsz + 3)
 #define decode_renew_maxsz      (op_decode_hdr_maxsz)
 #define encode_setclientid_maxsz \
@@ -253,6 +254,8 @@ static int nfs4_stat_to_errno(int);
                                 (encode_getattr_maxsz)
 #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)))
 
 #if defined(CONFIG_NFS_V4_1)
 #define NFS4_MAX_MACHINE_NAME_LEN (64)
@@ -324,6 +327,18 @@ static int nfs4_stat_to_errno(int);
 #define decode_layoutget_maxsz  (op_decode_hdr_maxsz + 8 + \
                                 decode_stateid_maxsz + \
                                 XDR_QUADLEN(PNFS_LAYOUT_MAXSIZE))
+#define encode_layoutcommit_maxsz (op_encode_hdr_maxsz +          \
+                                2 /* offset */ + \
+                                2 /* length */ + \
+                                1 /* reclaim */ + \
+                                encode_stateid_maxsz + \
+                                1 /* new offset (true) */ + \
+                                2 /* last byte written */ + \
+                                1 /* nt_timechanged (false) */ + \
+                                1 /* layoutupdate4 layout type */ + \
+                                1 /* NULL filelayout layoutupdate4 payload */)
+#define decode_layoutcommit_maxsz (op_decode_hdr_maxsz + 3)
+
 #else /* CONFIG_NFS_V4_1 */
 #define encode_sequence_maxsz   0
 #define decode_sequence_maxsz   0
@@ -676,6 +691,14 @@ static int nfs4_stat_to_errno(int);
                                  decode_putfh_maxsz + \
                                  decode_lookup_maxsz + \
                                  decode_fs_locations_maxsz)
+#define NFS4_enc_secinfo_sz     (compound_encode_hdr_maxsz + \
+                                encode_sequence_maxsz + \
+                                encode_putfh_maxsz + \
+                                encode_secinfo_maxsz)
+#define NFS4_dec_secinfo_sz     (compound_decode_hdr_maxsz + \
+                                decode_sequence_maxsz + \
+                                decode_putfh_maxsz + \
+                                decode_secinfo_maxsz)
 #if defined(CONFIG_NFS_V4_1)
 #define NFS4_enc_exchange_id_sz \
                                 (compound_encode_hdr_maxsz + \
@@ -727,6 +750,17 @@ static int nfs4_stat_to_errno(int);
                                 decode_sequence_maxsz + \
                                 decode_putfh_maxsz +        \
                                 decode_layoutget_maxsz)
+#define NFS4_enc_layoutcommit_sz (compound_encode_hdr_maxsz + \
+                                encode_sequence_maxsz +\
+                                encode_putfh_maxsz + \
+                                encode_layoutcommit_maxsz + \
+                                encode_getattr_maxsz)
+#define NFS4_dec_layoutcommit_sz (compound_decode_hdr_maxsz + \
+                                decode_sequence_maxsz + \
+                                decode_putfh_maxsz + \
+                                decode_layoutcommit_maxsz + \
+                                decode_getattr_maxsz)
+
 
 const u32 nfs41_maxwrite_overhead = ((RPC_MAX_HEADER_WITH_AUTH +
                                       compound_encode_hdr_maxsz +
@@ -844,7 +878,7 @@ static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, const
         if (iap->ia_valid & ATTR_MODE)
                 len += 4;
         if (iap->ia_valid & ATTR_UID) {
-                owner_namelen = nfs_map_uid_to_name(server->nfs_client, iap->ia_uid, owner_name, IDMAP_NAMESZ);
+                owner_namelen = nfs_map_uid_to_name(server, iap->ia_uid, owner_name, IDMAP_NAMESZ);
                 if (owner_namelen < 0) {
                         dprintk("nfs: couldn't resolve uid %d to string\n",
                                         iap->ia_uid);
@@ -856,7 +890,7 @@ static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, const
                 len += 4 + (XDR_QUADLEN(owner_namelen) << 2);
         }
         if (iap->ia_valid & ATTR_GID) {
-                owner_grouplen = nfs_map_gid_to_group(server->nfs_client, iap->ia_gid, owner_group, IDMAP_NAMESZ);
+                owner_grouplen = nfs_map_gid_to_group(server, iap->ia_gid, owner_group, IDMAP_NAMESZ);
                 if (owner_grouplen < 0) {
                         dprintk("nfs: couldn't resolve gid %d to string\n",
                                         iap->ia_gid);
@@ -1384,7 +1418,7 @@ static void encode_putrootfh(struct xdr_stream *xdr, struct compound_hdr *hdr)
         hdr->replen += decode_putrootfh_maxsz;
 }
 
-static void encode_stateid(struct xdr_stream *xdr, const struct nfs_open_context *ctx, const struct nfs_lock_context *l_ctx)
+static void encode_stateid(struct xdr_stream *xdr, const struct nfs_open_context *ctx, const struct nfs_lock_context *l_ctx, int zero_seqid)
 {
         nfs4_stateid stateid;
         __be32 *p;
@@ -1392,6 +1426,8 @@ static void encode_stateid(struct xdr_stream *xdr, const struct nfs_open_context
         p = reserve_space(xdr, NFS4_STATEID_SIZE);
         if (ctx->state != NULL) {
                 nfs4_copy_stateid(&stateid, ctx->state, l_ctx->lockowner, l_ctx->pid);
+                if (zero_seqid)
+                        stateid.stateid.seqid = 0;
                 xdr_encode_opaque_fixed(p, stateid.data, NFS4_STATEID_SIZE);
         } else
                 xdr_encode_opaque_fixed(p, zero_stateid.data, NFS4_STATEID_SIZE);
@@ -1404,7 +1440,8 @@ static void encode_read(struct xdr_stream *xdr, const struct nfs_readargs *args,
         p = reserve_space(xdr, 4);
         *p = cpu_to_be32(OP_READ);
 
-        encode_stateid(xdr, args->context, args->lock_context);
+        encode_stateid(xdr, args->context, args->lock_context,
+                       hdr->minorversion);
 
         p = reserve_space(xdr, 12);
         p = xdr_encode_hyper(p, args->offset);
@@ -1415,26 +1452,25 @@ static void encode_read(struct xdr_stream *xdr, const struct nfs_readargs *args,
 
 static void encode_readdir(struct xdr_stream *xdr, const struct nfs4_readdir_arg *readdir, struct rpc_rqst *req, struct compound_hdr *hdr)
 {
-        uint32_t attrs[2] = {0, 0};
+        uint32_t attrs[2] = {
+                FATTR4_WORD0_RDATTR_ERROR,
+                FATTR4_WORD1_MOUNTED_ON_FILEID,
+        };
         uint32_t dircount = readdir->count >> 1;
         __be32 *p;
 
         if (readdir->plus) {
                 attrs[0] |= FATTR4_WORD0_TYPE|FATTR4_WORD0_CHANGE|FATTR4_WORD0_SIZE|
-                        FATTR4_WORD0_FSID|FATTR4_WORD0_FILEHANDLE;
+                        FATTR4_WORD0_FSID|FATTR4_WORD0_FILEHANDLE|FATTR4_WORD0_FILEID;
                 attrs[1] |= FATTR4_WORD1_MODE|FATTR4_WORD1_NUMLINKS|FATTR4_WORD1_OWNER|
                         FATTR4_WORD1_OWNER_GROUP|FATTR4_WORD1_RAWDEV|
                         FATTR4_WORD1_SPACE_USED|FATTR4_WORD1_TIME_ACCESS|
                         FATTR4_WORD1_TIME_METADATA|FATTR4_WORD1_TIME_MODIFY;
                 dircount >>= 1;
         }
-        attrs[0] |= FATTR4_WORD0_RDATTR_ERROR|FATTR4_WORD0_FILEID;
-        attrs[1] |= FATTR4_WORD1_MOUNTED_ON_FILEID;
-        /* Switch to mounted_on_fileid if the server supports it */
-        if (readdir->bitmask[1] & FATTR4_WORD1_MOUNTED_ON_FILEID)
-                attrs[0] &= ~FATTR4_WORD0_FILEID;
-        else
-                attrs[1] &= ~FATTR4_WORD1_MOUNTED_ON_FILEID;
+        /* Use mounted_on_fileid only if the server supports it */
+        if (!(readdir->bitmask[1] & FATTR4_WORD1_MOUNTED_ON_FILEID))
+                attrs[0] |= FATTR4_WORD0_FILEID;
 
         p = reserve_space(xdr, 12+NFS4_VERIFIER_SIZE+20);
         *p++ = cpu_to_be32(OP_READDIR);
@@ -1592,7 +1628,8 @@ static void encode_write(struct xdr_stream *xdr, const struct nfs_writeargs *arg
         p = reserve_space(xdr, 4);
         *p = cpu_to_be32(OP_WRITE);
 
-        encode_stateid(xdr, args->context, args->lock_context);
+        encode_stateid(xdr, args->context, args->lock_context,
+                       hdr->minorversion);
 
         p = reserve_space(xdr, 16);
         p = xdr_encode_hyper(p, args->offset);
@@ -1616,6 +1653,18 @@ static void encode_delegreturn(struct xdr_stream *xdr, const nfs4_stateid *state
         hdr->replen += decode_delegreturn_maxsz;
 }
 
+static void encode_secinfo(struct xdr_stream *xdr, const struct qstr *name, struct compound_hdr *hdr)
+{
+        int len = name->len;
+        __be32 *p;
+
+        p = reserve_space(xdr, 8 + len);
+        *p++ = cpu_to_be32(OP_SECINFO);
+        xdr_encode_opaque(p, name->name, len);
+        hdr->nops++;
+        hdr->replen += decode_secinfo_maxsz;
+}
+
 #if defined(CONFIG_NFS_V4_1)
 /* NFSv4.1 operations */
 static void encode_exchange_id(struct xdr_stream *xdr,
@@ -1660,7 +1709,7 @@ static void encode_create_session(struct xdr_stream *xdr,
 
         p = reserve_space(xdr, 20 + 2*28 + 20 + len + 12);
         *p++ = cpu_to_be32(OP_CREATE_SESSION);
-        p = xdr_encode_hyper(p, clp->cl_ex_clid);
+        p = xdr_encode_hyper(p, clp->cl_clientid);
         *p++ = cpu_to_be32(clp->cl_seqid);                      /*Sequence id */
         *p++ = cpu_to_be32(args->flags);                        /*flags */
 
@@ -1812,6 +1861,34 @@ encode_layoutget(struct xdr_stream *xdr,
         hdr->nops++;
         hdr->replen += decode_layoutget_maxsz;
 }
+
+static int
+encode_layoutcommit(struct xdr_stream *xdr,
+                    const struct nfs4_layoutcommit_args *args,
+                    struct compound_hdr *hdr)
+{
+        __be32 *p;
+
+        dprintk("%s: lbw: %llu type: %d\n", __func__, args->lastbytewritten,
+                NFS_SERVER(args->inode)->pnfs_curr_ld->id);
+
+        p = reserve_space(xdr, 48 + NFS4_STATEID_SIZE);
+        *p++ = cpu_to_be32(OP_LAYOUTCOMMIT);
+        /* Only whole file layouts */
+        p = xdr_encode_hyper(p, 0); /* offset */
+        p = xdr_encode_hyper(p, NFS4_MAX_UINT64); /* length */
+        *p++ = cpu_to_be32(0); /* reclaim */
+        p = xdr_encode_opaque_fixed(p, args->stateid.data, NFS4_STATEID_SIZE);
+        *p++ = cpu_to_be32(1); /* newoffset = TRUE */
+        p = xdr_encode_hyper(p, args->lastbytewritten);
+        *p++ = cpu_to_be32(0); /* Never send time_modify_changed */
+        *p++ = cpu_to_be32(NFS_SERVER(args->inode)->pnfs_curr_ld->id);/* type */
+        *p++ = cpu_to_be32(0); /* no file layout payload */
+
+        hdr->nops++;
+        hdr->replen += decode_layoutcommit_maxsz;
+        return 0;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 /*
@@ -2271,7 +2348,8 @@ static void nfs4_xdr_enc_write(struct rpc_rqst *req, struct xdr_stream *xdr,
         encode_putfh(xdr, args->fh, &hdr);
         encode_write(xdr, args, &hdr);
         req->rq_snd_buf.flags |= XDRBUF_WRITE;
-        encode_getfattr(xdr, args->bitmask, &hdr);
+        if (args->bitmask)
+                encode_getfattr(xdr, args->bitmask, &hdr);
         encode_nops(&hdr);
 }
 
@@ -2289,7 +2367,8 @@ static void nfs4_xdr_enc_commit(struct rpc_rqst *req, struct xdr_stream *xdr,
         encode_sequence(xdr, &args->seq_args, &hdr);
         encode_putfh(xdr, args->fh, &hdr);
         encode_commit(xdr, args, &hdr);
-        encode_getfattr(xdr, args->bitmask, &hdr);
+        if (args->bitmask)
+                encode_getfattr(xdr, args->bitmask, &hdr);
         encode_nops(&hdr);
 }
 
@@ -2460,6 +2539,24 @@ static void nfs4_xdr_enc_fs_locations(struct rpc_rqst *req,
         encode_nops(&hdr);
 }
 
+/*
+ * Encode SECINFO request
+ */
+static void nfs4_xdr_enc_secinfo(struct rpc_rqst *req,
+                                struct xdr_stream *xdr,
+                                struct nfs4_secinfo_arg *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_putfh(xdr, args->dir_fh, &hdr);
+        encode_secinfo(xdr, args->name, &hdr);
+        encode_nops(&hdr);
+}
+
 #if defined(CONFIG_NFS_V4_1)
 /*
  * EXCHANGE_ID request
@@ -2599,7 +2696,31 @@ static void nfs4_xdr_enc_layoutget(struct rpc_rqst *req,
         encode_sequence(xdr, &args->seq_args, &hdr);
         encode_putfh(xdr, NFS_FH(args->inode), &hdr);
         encode_layoutget(xdr, args, &hdr);
+
+        xdr_inline_pages(&req->rq_rcv_buf, hdr.replen << 2,
+            args->layout.pages, 0, args->layout.pglen);
+
+        encode_nops(&hdr);
+}
+
+/*
+ *  Encode LAYOUTCOMMIT request
+ */
+static int nfs4_xdr_enc_layoutcommit(struct rpc_rqst *req,
+                                     struct xdr_stream *xdr,
+                                     struct nfs4_layoutcommit_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_putfh(xdr, NFS_FH(args->inode), &hdr);
+        encode_layoutcommit(xdr, args, &hdr);
+        encode_getfattr(xdr, args->bitmask, &hdr);
         encode_nops(&hdr);
+        return 0;
 }
 #endif /* CONFIG_NFS_V4_1 */
 
@@ -2920,6 +3041,7 @@ static int decode_attr_error(struct xdr_stream *xdr, uint32_t *bitmap)
                 if (unlikely(!p))
                         goto out_overflow;
                 bitmap[0] &= ~FATTR4_WORD0_RDATTR_ERROR;
+                return -be32_to_cpup(p);
         }
         return 0;
 out_overflow:
@@ -3017,7 +3139,7 @@ static int decode_attr_mounted_on_fileid(struct xdr_stream *xdr, uint32_t *bitma
                         goto out_overflow;
                 xdr_decode_hyper(p, fileid);
                 bitmap[1] &= ~FATTR4_WORD1_MOUNTED_ON_FILEID;
-                ret = NFS_ATTR_FATTR_FILEID;
+                ret = NFS_ATTR_FATTR_MOUNTED_ON_FILEID;
         }
         dprintk("%s: fileid=%Lu\n", __func__, (unsigned long long)*fileid);
         return ret;
@@ -3382,7 +3504,7 @@ out_overflow:
 }
 
 static int decode_attr_owner(struct xdr_stream *xdr, uint32_t *bitmap,
-                struct nfs_client *clp, uint32_t *uid, int may_sleep)
+                const struct nfs_server *server, uint32_t *uid, int may_sleep)
 {
         uint32_t len;
         __be32 *p;
@@ -3402,7 +3524,7 @@ static int decode_attr_owner(struct xdr_stream *xdr, uint32_t *bitmap,
                 if (!may_sleep) {
                         /* do nothing */
                 } else if (len < XDR_MAX_NETOBJ) {
-                        if (nfs_map_name_to_uid(clp, (char *)p, len, uid) == 0)
+                        if (nfs_map_name_to_uid(server, (char *)p, len, uid) == 0)
                                 ret = NFS_ATTR_FATTR_OWNER;
                         else
                                 dprintk("%s: nfs_map_name_to_uid failed!\n",
@@ -3420,7 +3542,7 @@ out_overflow:
 }
 
 static int decode_attr_group(struct xdr_stream *xdr, uint32_t *bitmap,
-                struct nfs_client *clp, uint32_t *gid, int may_sleep)
+                const struct nfs_server *server, uint32_t *gid, int may_sleep)
 {
         uint32_t len;
         __be32 *p;
@@ -3440,7 +3562,7 @@ static int decode_attr_group(struct xdr_stream *xdr, uint32_t *bitmap,
                 if (!may_sleep) {
                         /* do nothing */
                 } else if (len < XDR_MAX_NETOBJ) {
-                        if (nfs_map_group_to_gid(clp, (char *)p, len, gid) == 0)
+                        if (nfs_map_group_to_gid(server, (char *)p, len, gid) == 0)
                                 ret = NFS_ATTR_FATTR_GROUP;
                         else
                                 dprintk("%s: nfs_map_group_to_gid failed!\n",
@@ -3879,7 +4001,6 @@ static int decode_getfattr_attrs(struct xdr_stream *xdr, uint32_t *bitmap,
 {
         int status;
         umode_t fmode = 0;
-        uint64_t fileid;
         uint32_t type;
 
         status = decode_attr_type(xdr, bitmap, &type);
@@ -3907,6 +4028,10 @@ static int decode_getfattr_attrs(struct xdr_stream *xdr, uint32_t *bitmap,
         fattr->valid |= status;
 
         status = decode_attr_error(xdr, bitmap);
+        if (status == -NFS4ERR_WRONGSEC) {
+                nfs_fixup_secinfo_attributes(fattr, fh);
+                status = 0;
+        }
         if (status < 0)
                 goto xdr_error;
 
@@ -3939,14 +4064,12 @@ static int decode_getfattr_attrs(struct xdr_stream *xdr, uint32_t *bitmap,
                 goto xdr_error;
         fattr->valid |= status;
 
-        status = decode_attr_owner(xdr, bitmap, server->nfs_client,
-                        &fattr->uid, may_sleep);
+        status = decode_attr_owner(xdr, bitmap, server, &fattr->uid, may_sleep);
         if (status < 0)
                 goto xdr_error;
         fattr->valid |= status;
 
-        status = decode_attr_group(xdr, bitmap, server->nfs_client,
-                        &fattr->gid, may_sleep);
+        status = decode_attr_group(xdr, bitmap, server, &fattr->gid, may_sleep);
         if (status < 0)
                 goto xdr_error;
         fattr->valid |= status;
@@ -3976,13 +4099,10 @@ static int decode_getfattr_attrs(struct xdr_stream *xdr, uint32_t *bitmap,
                 goto xdr_error;
         fattr->valid |= status;
 
-        status = decode_attr_mounted_on_fileid(xdr, bitmap, &fileid);
+        status = decode_attr_mounted_on_fileid(xdr, bitmap, &fattr->mounted_on_fileid);
         if (status < 0)
                 goto xdr_error;
-        if (status != 0 && !(fattr->valid & status)) {
-                fattr->fileid = fileid;
-                fattr->valid |= status;
-        }
+        fattr->valid |= status;
 
 xdr_error:
         dprintk("%s: xdr returned %d\n", __func__, -status);
@@ -4677,6 +4797,79 @@ static int decode_delegreturn(struct xdr_stream *xdr)
         return decode_op_hdr(xdr, OP_DELEGRETURN);
 }
 
+static int decode_secinfo_gss(struct xdr_stream *xdr, struct nfs4_secinfo_flavor *flavor)
+{
+        __be32 *p;
+
+        p = xdr_inline_decode(xdr, 4);
+        if (unlikely(!p))
+                goto out_overflow;
+        flavor->gss.sec_oid4.len = be32_to_cpup(p);
+        if (flavor->gss.sec_oid4.len > GSS_OID_MAX_LEN)
+                goto out_err;
+
+        p = xdr_inline_decode(xdr, flavor->gss.sec_oid4.len);
+        if (unlikely(!p))
+                goto out_overflow;
+        memcpy(flavor->gss.sec_oid4.data, p, flavor->gss.sec_oid4.len);
+
+        p = xdr_inline_decode(xdr, 8);
+        if (unlikely(!p))
+                goto out_overflow;
+        flavor->gss.qop4 = be32_to_cpup(p++);
+        flavor->gss.service = be32_to_cpup(p);
+
+        return 0;
+
+out_overflow:
+        print_overflow_msg(__func__, xdr);
+        return -EIO;
+out_err:
+        return -EINVAL;
+}
+
+static int decode_secinfo(struct xdr_stream *xdr, struct nfs4_secinfo_res *res)
+{
+        struct nfs4_secinfo_flavor *sec_flavor;
+        int status;
+        __be32 *p;
+        int i, num_flavors;
+
+        status = decode_op_hdr(xdr, OP_SECINFO);
+        if (status)
+                goto out;
+        p = xdr_inline_decode(xdr, 4);
+        if (unlikely(!p))
+                goto out_overflow;
+
+        res->flavors->num_flavors = 0;
+        num_flavors = be32_to_cpup(p);
+
+        for (i = 0; i < num_flavors; i++) {
+                sec_flavor = &res->flavors->flavors[i];
+                if ((char *)&sec_flavor[1] - (char *)res->flavors > PAGE_SIZE)
+                        break;
+
+                p = xdr_inline_decode(xdr, 4);
+                if (unlikely(!p))
+                        goto out_overflow;
+                sec_flavor->flavor = be32_to_cpup(p);
+
+                if (sec_flavor->flavor == RPC_AUTH_GSS) {
+                        status = decode_secinfo_gss(xdr, sec_flavor);
+                        if (status)
+                                goto out;
+                }
+                res->flavors->num_flavors++;
+        }
+
+out:
+        return status;
+out_overflow:
+        print_overflow_msg(__func__, xdr);
+        return -EIO;
+}
+
 #if defined(CONFIG_NFS_V4_1)
 static int decode_exchange_id(struct xdr_stream *xdr,
                               struct nfs41_exchange_id_res *res)
@@ -4694,7 +4887,7 @@ static int decode_exchange_id(struct xdr_stream *xdr,
         p = xdr_inline_decode(xdr, 8);
         if (unlikely(!p))
                 goto out_overflow;
-        xdr_decode_hyper(p, &clp->cl_ex_clid);
+        xdr_decode_hyper(p, &clp->cl_clientid);
         p = xdr_inline_decode(xdr, 12);
         if (unlikely(!p))
                 goto out_overflow;
@@ -4947,6 +5140,9 @@ static int decode_layoutget(struct xdr_stream *xdr, struct rpc_rqst *req,
         __be32 *p;
         int status;
         u32 layout_count;
+        struct xdr_buf *rcvbuf = &req->rq_rcv_buf;
+        struct kvec *iov = rcvbuf->head;
+        u32 hdrlen, recvd;
 
         status = decode_op_hdr(xdr, OP_LAYOUTGET);
         if (status)
@@ -4963,17 +5159,14 @@ static int decode_layoutget(struct xdr_stream *xdr, struct rpc_rqst *req,
                 return -EINVAL;
         }
 
-        p = xdr_inline_decode(xdr, 24);
+        p = xdr_inline_decode(xdr, 28);
         if (unlikely(!p))
                 goto out_overflow;
         p = xdr_decode_hyper(p, &res->range.offset);
         p = xdr_decode_hyper(p, &res->range.length);
         res->range.iomode = be32_to_cpup(p++);
         res->type = be32_to_cpup(p++);
-
-        status = decode_opaque_inline(xdr, &res->layout.len, (char **)&p);
-        if (unlikely(status))
-                return status;
+        res->layoutp->len = be32_to_cpup(p);
 
         dprintk("%s roff:%lu rlen:%lu riomode:%d, lo_type:0x%x, lo.len:%d\n",
                 __func__,
@@ -4981,12 +5174,18 @@ static int decode_layoutget(struct xdr_stream *xdr, struct rpc_rqst *req,
                 (unsigned long)res->range.length,
                 res->range.iomode,
                 res->type,
-                res->layout.len);
+                res->layoutp->len);
 
-        /* nfs4_proc_layoutget allocated a single page */
-        if (res->layout.len > PAGE_SIZE)
-                return -ENOMEM;
-        memcpy(res->layout.buf, p, res->layout.len);
+        hdrlen = (u8 *) xdr->p - (u8 *) iov->iov_base;
+        recvd = req->rq_rcv_buf.len - hdrlen;
+        if (res->layoutp->len > recvd) {
+                dprintk("NFS: server cheating in layoutget reply: "
+                                "layout len %u > recvd %u\n",
+                                res->layoutp->len, recvd);
+                return -EINVAL;
+        }
+
+        xdr_read_pages(xdr, res->layoutp->len);
 
         if (layout_count > 1) {
                 /* We only handle a length one array at the moment.  Any
@@ -5003,6 +5202,35 @@ out_overflow:
         print_overflow_msg(__func__, xdr);
         return -EIO;
 }
+
+static int decode_layoutcommit(struct xdr_stream *xdr,
+                               struct rpc_rqst *req,
+                               struct nfs4_layoutcommit_res *res)
+{
+        __be32 *p;
+        __u32 sizechanged;
+        int status;
+
+        status = decode_op_hdr(xdr, OP_LAYOUTCOMMIT);
+        if (status)
+                return status;
+
+        p = xdr_inline_decode(xdr, 4);
+        if (unlikely(!p))
+                goto out_overflow;
+        sizechanged = be32_to_cpup(p);
+
+        if (sizechanged) {
+                /* throw away new size */
+                p = xdr_inline_decode(xdr, 8);
+                if (unlikely(!p))
+                        goto out_overflow;
+        }
+        return 0;
+out_overflow:
+        print_overflow_msg(__func__, xdr);
+        return -EIO;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 /*
@@ -5690,8 +5918,9 @@ static int nfs4_xdr_dec_write(struct rpc_rqst *rqstp, struct xdr_stream *xdr,
         status = decode_write(xdr, res);
         if (status)
                 goto out;
-        decode_getfattr(xdr, res->fattr, res->server,
-                        !RPC_IS_ASYNC(rqstp->rq_task));
+        if (res->fattr)
+                decode_getfattr(xdr, res->fattr, res->server,
+                                !RPC_IS_ASYNC(rqstp->rq_task));
         if (!status)
                 status = res->count;
 out:
@@ -5719,8 +5948,9 @@ static int nfs4_xdr_dec_commit(struct rpc_rqst *rqstp, struct xdr_stream *xdr,
         status = decode_commit(xdr, res);
         if (status)
                 goto out;
-        decode_getfattr(xdr, res->fattr, res->server,
-                        !RPC_IS_ASYNC(rqstp->rq_task));
+        if (res->fattr)
+                decode_getfattr(xdr, res->fattr, res->server,
+                                !RPC_IS_ASYNC(rqstp->rq_task));
 out:
         return status;
 }
@@ -5915,6 +6145,32 @@ out:
         return status;
 }
 
+/*
+ * Decode SECINFO response
+ */
+static int nfs4_xdr_dec_secinfo(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_putfh(xdr);
+        if (status)
+                goto out;
+        status = decode_secinfo(xdr, res);
+        if (status)
+                goto out;
+out:
+        return status;
+}
+
 #if defined(CONFIG_NFS_V4_1)
 /*
  * Decode EXCHANGE_ID response
@@ -6062,6 +6318,34 @@ static int nfs4_xdr_dec_layoutget(struct rpc_rqst *rqstp,
 out:
         return status;
 }
+
+/*
+ * Decode LAYOUTCOMMIT response
+ */
+static int nfs4_xdr_dec_layoutcommit(struct rpc_rqst *rqstp,
+                                     struct xdr_stream *xdr,
+                                     struct nfs4_layoutcommit_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_putfh(xdr);
+        if (status)
+                goto out;
+        status = decode_layoutcommit(xdr, rqstp, res);
+        if (status)
+                goto out;
+        decode_getfattr(xdr, res->fattr, res->server,
+                        !RPC_IS_ASYNC(rqstp->rq_task));
+out:
+        return status;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 /**
@@ -6125,7 +6409,9 @@ int nfs4_decode_dirent(struct xdr_stream *xdr, struct nfs_entry *entry,
         if (decode_getfattr_attrs(xdr, bitmap, entry->fattr, entry->fh,
                                         entry->server, 1) < 0)
                 goto out_overflow;
-        if (entry->fattr->valid & NFS_ATTR_FATTR_FILEID)
+        if (entry->fattr->valid & NFS_ATTR_FATTR_MOUNTED_ON_FILEID)
+                entry->ino = entry->fattr->mounted_on_fileid;
+        else if (entry->fattr->valid & NFS_ATTR_FATTR_FILEID)
                 entry->ino = entry->fattr->fileid;
 
         entry->d_type = DT_UNKNOWN;
@@ -6167,8 +6453,6 @@ static struct {
         { NFS4ERR_DQUOT,        -EDQUOT         },
         { NFS4ERR_STALE,        -ESTALE         },
         { NFS4ERR_BADHANDLE,    -EBADHANDLE     },
-        { NFS4ERR_BADOWNER,     -EINVAL         },
-        { NFS4ERR_BADNAME,      -EINVAL         },
         { NFS4ERR_BAD_COOKIE,   -EBADCOOKIE     },
         { NFS4ERR_NOTSUPP,      -ENOTSUPP       },
         { NFS4ERR_TOOSMALL,     -ETOOSMALL      },
@@ -6178,10 +6462,6 @@ static struct {
         { NFS4ERR_SYMLINK,      -ELOOP          },
         { NFS4ERR_OP_ILLEGAL,   -EOPNOTSUPP     },
         { NFS4ERR_DEADLOCK,     -EDEADLK        },
-        { NFS4ERR_WRONGSEC,     -EPERM          }, /* FIXME: this needs
-                                                    * to be handled by a
-                                                    * middle-layer.
-                                                    */
         { -1,                   -EIO            }
 };
 
@@ -6256,6 +6536,7 @@ struct rpc_procinfo	nfs4_procedures[] = {
         PROC(SETACL,            enc_setacl,             dec_setacl),
         PROC(FS_LOCATIONS,      enc_fs_locations,       dec_fs_locations),
         PROC(RELEASE_LOCKOWNER, enc_release_lockowner,  dec_release_lockowner),
+        PROC(SECINFO,           enc_secinfo,            dec_secinfo),
 #if defined(CONFIG_NFS_V4_1)
         PROC(EXCHANGE_ID,       enc_exchange_id,        dec_exchange_id),
         PROC(CREATE_SESSION,    enc_create_session,     dec_create_session),
@@ -6265,6 +6546,7 @@ struct rpc_procinfo	nfs4_procedures[] = {
         PROC(RECLAIM_COMPLETE,  enc_reclaim_complete,   dec_reclaim_complete),
         PROC(GETDEVICEINFO,     enc_getdeviceinfo,      dec_getdeviceinfo),
         PROC(LAYOUTGET,         enc_layoutget,          dec_layoutget),
+        PROC(LAYOUTCOMMIT,      enc_layoutcommit,       dec_layoutcommit),
 #endif /* CONFIG_NFS_V4_1 */
 };
 
diff --git a/fs/nfs/nfsroot.c b/fs/nfs/nfsroot.c
index 903908a20023..c541093a5bf2 100644
--- a/fs/nfs/nfsroot.c
+++ b/fs/nfs/nfsroot.c
@@ -86,11 +86,14 @@
 /* Default path we try to mount. "%s" gets replaced by our IP address */
 #define NFS_ROOT                "/tftpboot/%s"
 
+/* Default NFSROOT mount options. */
+#define NFS_DEF_OPTIONS         "udp"
+
 /* Parameters passed from the kernel command line */
 static char nfs_root_parms[256] __initdata = "";
 
 /* Text-based mount options passed to super.c */
-static char nfs_root_options[256] __initdata = "";
+static char nfs_root_options[256] __initdata = NFS_DEF_OPTIONS;
 
 /* Address of NFS server */
 static __be32 servaddr __initdata = htonl(INADDR_NONE);
@@ -160,8 +163,14 @@ static int __init root_nfs_copy(char *dest, const char *src,
 }
 
 static int __init root_nfs_cat(char *dest, const char *src,
-                                  const size_t destlen)
+                               const size_t destlen)
 {
+        size_t len = strlen(dest);
+
+        if (len && dest[len - 1] != ',')
+                if (strlcat(dest, ",", destlen) > destlen)
+                        return -1;
+
         if (strlcat(dest, src, destlen) > destlen)
                 return -1;
         return 0;
@@ -194,16 +203,6 @@ static int __init root_nfs_parse_options(char *incoming, char *exppath,
                 if (root_nfs_cat(nfs_root_options, incoming,
                                                 sizeof(nfs_root_options)))
                         return -1;
-
-        /*
-         * Possibly prepare for more options to be appended
-         */
-        if (nfs_root_options[0] != '\0' &&
-            nfs_root_options[strlen(nfs_root_options)] != ',')
-                if (root_nfs_cat(nfs_root_options, ",",
-                                                sizeof(nfs_root_options)))
-                        return -1;
-
         return 0;
 }
 
@@ -217,7 +216,7 @@ static int __init root_nfs_parse_options(char *incoming, char *exppath,
  */
 static int __init root_nfs_data(char *cmdline)
 {
-        char addr_option[sizeof("nolock,addr=") + INET_ADDRSTRLEN + 1];
+        char mand_options[sizeof("nolock,addr=") + INET_ADDRSTRLEN + 1];
         int len, retval = -1;
         char *tmp = NULL;
         const size_t tmplen = sizeof(nfs_export_path);
@@ -244,9 +243,9 @@ static int __init root_nfs_data(char *cmdline)
          * Append mandatory options for nfsroot so they override
          * what has come before
          */
-        snprintf(addr_option, sizeof(addr_option), "nolock,addr=%pI4",
+        snprintf(mand_options, sizeof(mand_options), "nolock,addr=%pI4",
                         &servaddr);
-        if (root_nfs_cat(nfs_root_options, addr_option,
+        if (root_nfs_cat(nfs_root_options, mand_options,
                                                 sizeof(nfs_root_options)))
                 goto out_optionstoolong;
 
diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c
index e1164e3f9e69..c80add6e2213 100644
--- a/fs/nfs/pagelist.c
+++ b/fs/nfs/pagelist.c
@@ -20,6 +20,7 @@
 #include <linux/nfs_mount.h>
 
 #include "internal.h"
+#include "pnfs.h"
 
 static struct kmem_cache *nfs_page_cachep;
 
@@ -134,14 +135,14 @@ void nfs_clear_page_tag_locked(struct nfs_page *req)
                 nfs_unlock_request(req);
 }
 
-/**
+/*
  * nfs_clear_request - Free up all resources allocated to the request
  * @req:
  *
  * Release page and open context resources associated with a read/write
  * request after it has completed.
  */
-void nfs_clear_request(struct nfs_page *req)
+static void nfs_clear_request(struct nfs_page *req)
 {
         struct page *page = req->wb_page;
         struct nfs_open_context *ctx = req->wb_context;
@@ -213,7 +214,7 @@ nfs_wait_on_request(struct nfs_page *req)
  */
 void nfs_pageio_init(struct nfs_pageio_descriptor *desc,
                      struct inode *inode,
-                     int (*doio)(struct inode *, struct list_head *, unsigned int, size_t, int),
+                     int (*doio)(struct nfs_pageio_descriptor *),
                      size_t bsize,
                      int io_flags)
 {
@@ -222,10 +223,12 @@ void nfs_pageio_init(struct nfs_pageio_descriptor *desc,
         desc->pg_count = 0;
         desc->pg_bsize = bsize;
         desc->pg_base = 0;
+        desc->pg_moreio = 0;
         desc->pg_inode = inode;
         desc->pg_doio = doio;
         desc->pg_ioflags = io_flags;
         desc->pg_error = 0;
+        desc->pg_lseg = NULL;
 }
 
 /**
@@ -240,7 +243,8 @@ void nfs_pageio_init(struct nfs_pageio_descriptor *desc,
  * Return 'true' if this is the case, else return 'false'.
  */
 static int nfs_can_coalesce_requests(struct nfs_page *prev,
-                                     struct nfs_page *req)
+                                     struct nfs_page *req,
+                                     struct nfs_pageio_descriptor *pgio)
 {
         if (req->wb_context->cred != prev->wb_context->cred)
                 return 0;
@@ -254,6 +258,12 @@ static int nfs_can_coalesce_requests(struct nfs_page *prev,
                 return 0;
         if (prev->wb_pgbase + prev->wb_bytes != PAGE_CACHE_SIZE)
                 return 0;
+        /*
+         * Non-whole file layouts need to check that req is inside of
+         * pgio->pg_lseg.
+         */
+        if (pgio->pg_test && !pgio->pg_test(pgio, prev, req))
+                return 0;
         return 1;
 }
 
@@ -286,7 +296,7 @@ static int nfs_pageio_do_add_request(struct nfs_pageio_descriptor *desc,
                 if (newlen > desc->pg_bsize)
                         return 0;
                 prev = nfs_list_entry(desc->pg_list.prev);
-                if (!nfs_can_coalesce_requests(prev, req))
+                if (!nfs_can_coalesce_requests(prev, req, desc))
                         return 0;
         } else
                 desc->pg_base = req->wb_pgbase;
@@ -302,12 +312,7 @@ 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->pg_inode,
-                                          &desc->pg_list,
-                                          nfs_page_array_len(desc->pg_base,
-                                                             desc->pg_count),
-                                          desc->pg_count,
-                                          desc->pg_ioflags);
+                int error = desc->pg_doio(desc);
                 if (error < 0)
                         desc->pg_error = error;
                 else
@@ -331,9 +336,11 @@ int nfs_pageio_add_request(struct nfs_pageio_descriptor *desc,
                            struct nfs_page *req)
 {
         while (!nfs_pageio_do_add_request(desc, req)) {
+                desc->pg_moreio = 1;
                 nfs_pageio_doio(desc);
                 if (desc->pg_error < 0)
                         return 0;
+                desc->pg_moreio = 0;
         }
         return 1;
 }
@@ -391,6 +398,7 @@ int nfs_scan_list(struct nfs_inode *nfsi,
         pgoff_t idx_end;
         int found, i;
         int res;
+        struct list_head *list;
 
         res = 0;
         if (npages == 0)
@@ -411,10 +419,10 @@ int nfs_scan_list(struct nfs_inode *nfsi,
                         idx_start = req->wb_index + 1;
                         if (nfs_set_page_tag_locked(req)) {
                                 kref_get(&req->wb_kref);
-                                nfs_list_remove_request(req);
                                 radix_tree_tag_clear(&nfsi->nfs_page_tree,
                                                 req->wb_index, tag);
-                                nfs_list_add_request(req, dst);
+                                list = pnfs_choose_commit_list(req, dst);
+                                nfs_list_add_request(req, list);
                                 res++;
                                 if (res == INT_MAX)
                                         goto out;
diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c
index 1b1bc1a0fb0a..ff681ab65d31 100644
--- a/fs/nfs/pnfs.c
+++ b/fs/nfs/pnfs.c
@@ -30,6 +30,7 @@
 #include <linux/nfs_fs.h>
 #include "internal.h"
 #include "pnfs.h"
+#include "iostat.h"
 
 #define NFSDBG_FACILITY         NFSDBG_PNFS
 
@@ -74,10 +75,8 @@ find_pnfs_driver(u32 id)
 void
 unset_pnfs_layoutdriver(struct nfs_server *nfss)
 {
-        if (nfss->pnfs_curr_ld) {
-                nfss->pnfs_curr_ld->clear_layoutdriver(nfss);
+        if (nfss->pnfs_curr_ld)
                 module_put(nfss->pnfs_curr_ld->owner);
-        }
         nfss->pnfs_curr_ld = NULL;
 }
 
@@ -115,13 +114,7 @@ set_pnfs_layoutdriver(struct nfs_server *server, u32 id)
                 goto out_no_driver;
         }
         server->pnfs_curr_ld = ld_type;
-        if (ld_type->set_layoutdriver(server)) {
-                printk(KERN_ERR
-                       "%s: Error initializing mount point for layout driver %u.\n",
-                       __func__, id);
-                module_put(ld_type->owner);
-                goto out_no_driver;
-        }
+
         dprintk("%s: pNFS module for %u set\n", __func__, id);
         return;
 
@@ -230,38 +223,43 @@ static void free_lseg(struct pnfs_layout_segment *lseg)
         put_layout_hdr(NFS_I(ino)->layout);
 }
 
-/* The use of tmp_list is necessary because pnfs_curr_ld->free_lseg
- * could sleep, so must be called outside of the lock.
- * Returns 1 if object was removed, otherwise return 0.
- */
-static int
-put_lseg_locked(struct pnfs_layout_segment *lseg,
-                struct list_head *tmp_list)
+static void
+put_lseg_common(struct pnfs_layout_segment *lseg)
 {
+        struct inode *inode = lseg->pls_layout->plh_inode;
+
+        BUG_ON(test_bit(NFS_LSEG_VALID, &lseg->pls_flags));
+        list_del_init(&lseg->pls_list);
+        if (list_empty(&lseg->pls_layout->plh_segs)) {
+                set_bit(NFS_LAYOUT_DESTROYED, &lseg->pls_layout->plh_flags);
+                /* Matched by initial refcount set in alloc_init_layout_hdr */
+                put_layout_hdr_locked(lseg->pls_layout);
+        }
+        rpc_wake_up(&NFS_SERVER(inode)->roc_rpcwaitq);
+}
+
+void
+put_lseg(struct pnfs_layout_segment *lseg)
+{
+        struct inode *inode;
+
+        if (!lseg)
+                return;
+
         dprintk("%s: lseg %p ref %d valid %d\n", __func__, lseg,
                 atomic_read(&lseg->pls_refcount),
                 test_bit(NFS_LSEG_VALID, &lseg->pls_flags));
-        if (atomic_dec_and_test(&lseg->pls_refcount)) {
-                struct inode *ino = lseg->pls_layout->plh_inode;
+        inode = lseg->pls_layout->plh_inode;
+        if (atomic_dec_and_lock(&lseg->pls_refcount, &inode->i_lock)) {
+                LIST_HEAD(free_me);
 
-                BUG_ON(test_bit(NFS_LSEG_VALID, &lseg->pls_flags));
-                list_del(&lseg->pls_list);
-                if (list_empty(&lseg->pls_layout->plh_segs)) {
-                        struct nfs_client *clp;
-
-                        clp = NFS_SERVER(ino)->nfs_client;
-                        spin_lock(&clp->cl_lock);
-                        /* List does not take a reference, so no need for put here */
-                        list_del_init(&lseg->pls_layout->plh_layouts);
-                        spin_unlock(&clp->cl_lock);
-                        clear_bit(NFS_LAYOUT_BULK_RECALL, &lseg->pls_layout->plh_flags);
-                }
-                rpc_wake_up(&NFS_SERVER(ino)->roc_rpcwaitq);
-                list_add(&lseg->pls_list, tmp_list);
-                return 1;
+                put_lseg_common(lseg);
+                list_add(&lseg->pls_list, &free_me);
+                spin_unlock(&inode->i_lock);
+                pnfs_free_lseg_list(&free_me);
         }
-        return 0;
 }
+EXPORT_SYMBOL_GPL(put_lseg);
 
 static bool
 should_free_lseg(u32 lseg_iomode, u32 recall_iomode)
@@ -281,7 +279,13 @@ static int mark_lseg_invalid(struct pnfs_layout_segment *lseg,
                  * list.  It will now be removed when all
                  * outstanding io is finished.
                  */
-                rv = put_lseg_locked(lseg, tmp_list);
+                dprintk("%s: lseg %p ref %d\n", __func__, lseg,
+                        atomic_read(&lseg->pls_refcount));
+                if (atomic_dec_and_test(&lseg->pls_refcount)) {
+                        put_lseg_common(lseg);
+                        list_add(&lseg->pls_list, tmp_list);
+                        rv = 1;
+                }
         }
         return rv;
 }
@@ -299,6 +303,11 @@ mark_matching_lsegs_invalid(struct pnfs_layout_hdr *lo,
 
         dprintk("%s:Begin lo %p\n", __func__, lo);
 
+        if (list_empty(&lo->plh_segs)) {
+                if (!test_and_set_bit(NFS_LAYOUT_DESTROYED, &lo->plh_flags))
+                        put_layout_hdr_locked(lo);
+                return 0;
+        }
         list_for_each_entry_safe(lseg, next, &lo->plh_segs, pls_list)
                 if (should_free_lseg(lseg->pls_range.iomode, iomode)) {
                         dprintk("%s: freeing lseg %p iomode %d "
@@ -312,11 +321,27 @@ mark_matching_lsegs_invalid(struct pnfs_layout_hdr *lo,
         return invalid - removed;
 }
 
+/* note free_me must contain lsegs from a single layout_hdr */
 void
 pnfs_free_lseg_list(struct list_head *free_me)
 {
         struct pnfs_layout_segment *lseg, *tmp;
+        struct pnfs_layout_hdr *lo;
 
+        if (list_empty(free_me))
+                return;
+
+        lo = list_first_entry(free_me, struct pnfs_layout_segment,
+                              pls_list)->pls_layout;
+
+        if (test_bit(NFS_LAYOUT_DESTROYED, &lo->plh_flags)) {
+                struct nfs_client *clp;
+
+                clp = NFS_SERVER(lo->plh_inode)->nfs_client;
+                spin_lock(&clp->cl_lock);
+                list_del_init(&lo->plh_layouts);
+                spin_unlock(&clp->cl_lock);
+        }
         list_for_each_entry_safe(lseg, tmp, free_me, pls_list) {
                 list_del(&lseg->pls_list);
                 free_lseg(lseg);
@@ -332,10 +357,8 @@ pnfs_destroy_layout(struct nfs_inode *nfsi)
         spin_lock(&nfsi->vfs_inode.i_lock);
         lo = nfsi->layout;
         if (lo) {
-                set_bit(NFS_LAYOUT_DESTROYED, &nfsi->layout->plh_flags);
+                lo->plh_block_lgets++; /* permanently block new LAYOUTGETs */
                 mark_matching_lsegs_invalid(lo, &tmp_list, IOMODE_ANY);
-                /* Matched by refcount set to 1 in alloc_init_layout_hdr */
-                put_layout_hdr_locked(lo);
         }
         spin_unlock(&nfsi->vfs_inode.i_lock);
         pnfs_free_lseg_list(&tmp_list);
@@ -403,6 +426,7 @@ pnfs_layoutgets_blocked(struct pnfs_layout_hdr *lo, nfs4_stateid *stateid,
             (int)(lo->plh_barrier - be32_to_cpu(stateid->stateid.seqid)) >= 0)
                 return true;
         return lo->plh_block_lgets ||
+                test_bit(NFS_LAYOUT_DESTROYED, &lo->plh_flags) ||
                 test_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags) ||
                 (list_empty(&lo->plh_segs) &&
                  (atomic_read(&lo->plh_outstanding) > lget));
@@ -448,6 +472,9 @@ send_layoutget(struct pnfs_layout_hdr *lo,
         struct nfs_server *server = NFS_SERVER(ino);
         struct nfs4_layoutget *lgp;
         struct pnfs_layout_segment *lseg = NULL;
+        struct page **pages = NULL;
+        int i;
+        u32 max_resp_sz, max_pages;
 
         dprintk("--> %s\n", __func__);
 
@@ -455,6 +482,21 @@ send_layoutget(struct pnfs_layout_hdr *lo,
         lgp = kzalloc(sizeof(*lgp), GFP_KERNEL);
         if (lgp == NULL)
                 return NULL;
+
+        /* allocate pages for xdr post processing */
+        max_resp_sz = server->nfs_client->cl_session->fc_attrs.max_resp_sz;
+        max_pages = max_resp_sz >> PAGE_SHIFT;
+
+        pages = kzalloc(max_pages * sizeof(struct page *), GFP_KERNEL);
+        if (!pages)
+                goto out_err_free;
+
+        for (i = 0; i < max_pages; i++) {
+                pages[i] = alloc_page(GFP_KERNEL);
+                if (!pages[i])
+                        goto out_err_free;
+        }
+
         lgp->args.minlength = NFS4_MAX_UINT64;
         lgp->args.maxcount = PNFS_LAYOUT_MAXSIZE;
         lgp->args.range.iomode = iomode;
@@ -463,6 +505,8 @@ send_layoutget(struct pnfs_layout_hdr *lo,
         lgp->args.type = server->pnfs_curr_ld->id;
         lgp->args.inode = ino;
         lgp->args.ctx = get_nfs_open_context(ctx);
+        lgp->args.layout.pages = pages;
+        lgp->args.layout.pglen = max_pages * PAGE_SIZE;
         lgp->lsegpp = &lseg;
 
         /* Synchronously retrieve layout information from server and
@@ -473,7 +517,26 @@ send_layoutget(struct pnfs_layout_hdr *lo,
                 /* remember that LAYOUTGET failed and suspend trying */
                 set_bit(lo_fail_bit(iomode), &lo->plh_flags);
         }
+
+        /* free xdr pages */
+        for (i = 0; i < max_pages; i++)
+                __free_page(pages[i]);
+        kfree(pages);
+
         return lseg;
+
+out_err_free:
+        /* free any allocated xdr pages, lgp as it's not used */
+        if (pages) {
+                for (i = 0; i < max_pages; i++) {
+                        if (!pages[i])
+                                break;
+                        __free_page(pages[i]);
+                }
+                kfree(pages);
+        }
+        kfree(lgp);
+        return NULL;
 }
 
 bool pnfs_roc(struct inode *ino)
@@ -674,7 +737,7 @@ pnfs_find_lseg(struct pnfs_layout_hdr *lo, u32 iomode)
         list_for_each_entry(lseg, &lo->plh_segs, pls_list) {
                 if (test_bit(NFS_LSEG_VALID, &lseg->pls_flags) &&
                     is_matching_lseg(lseg, iomode)) {
-                        ret = lseg;
+                        ret = get_lseg(lseg);
                         break;
                 }
                 if (cmp_layout(iomode, lseg->pls_range.iomode) > 0)
@@ -699,6 +762,7 @@ pnfs_update_layout(struct inode *ino,
         struct nfs_client *clp = NFS_SERVER(ino)->nfs_client;
         struct pnfs_layout_hdr *lo;
         struct pnfs_layout_segment *lseg = NULL;
+        bool first = false;
 
         if (!pnfs_enabled_sb(NFS_SERVER(ino)))
                 return NULL;
@@ -715,21 +779,25 @@ pnfs_update_layout(struct inode *ino,
                 dprintk("%s matches recall, use MDS\n", __func__);
                 goto out_unlock;
         }
-        /* Check to see if the layout for the given range already exists */
-        lseg = pnfs_find_lseg(lo, iomode);
-        if (lseg)
-                goto out_unlock;
 
         /* if LAYOUTGET already failed once we don't try again */
         if (test_bit(lo_fail_bit(iomode), &nfsi->layout->plh_flags))
                 goto out_unlock;
 
+        /* Check to see if the layout for the given range already exists */
+        lseg = pnfs_find_lseg(lo, iomode);
+        if (lseg)
+                goto out_unlock;
+
         if (pnfs_layoutgets_blocked(lo, NULL, 0))
                 goto out_unlock;
         atomic_inc(&lo->plh_outstanding);
 
         get_layout_hdr(lo);
-        if (list_empty(&lo->plh_segs)) {
+        if (list_empty(&lo->plh_segs))
+                first = true;
+        spin_unlock(&ino->i_lock);
+        if (first) {
                 /* The lo must be on the clp list if there is any
                  * chance of a CB_LAYOUTRECALL(FILE) coming in.
                  */
@@ -738,24 +806,18 @@ pnfs_update_layout(struct inode *ino,
                 list_add_tail(&lo->plh_layouts, &clp->cl_layouts);
                 spin_unlock(&clp->cl_lock);
         }
-        spin_unlock(&ino->i_lock);
 
         lseg = send_layoutget(lo, ctx, iomode);
-        if (!lseg) {
-                spin_lock(&ino->i_lock);
-                if (list_empty(&lo->plh_segs)) {
-                        spin_lock(&clp->cl_lock);
-                        list_del_init(&lo->plh_layouts);
-                        spin_unlock(&clp->cl_lock);
-                        clear_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags);
-                }
-                spin_unlock(&ino->i_lock);
+        if (!lseg && first) {
+                spin_lock(&clp->cl_lock);
+                list_del_init(&lo->plh_layouts);
+                spin_unlock(&clp->cl_lock);
         }
         atomic_dec(&lo->plh_outstanding);
         put_layout_hdr(lo);
 out:
         dprintk("%s end, state 0x%lx lseg %p\n", __func__,
-                nfsi->layout->plh_flags, lseg);
+                nfsi->layout ? nfsi->layout->plh_flags : -1, lseg);
         return lseg;
 out_unlock:
         spin_unlock(&ino->i_lock);
@@ -808,7 +870,7 @@ pnfs_layout_process(struct nfs4_layoutget *lgp)
         }
         init_lseg(lo, lseg);
         lseg->pls_range = res->range;
-        *lgp->lsegpp = lseg;
+        *lgp->lsegpp = get_lseg(lseg);
         pnfs_insert_layout(lo, lseg);
 
         if (res->return_on_close) {
@@ -829,137 +891,205 @@ out_forget_reply:
         goto out;
 }
 
+static int pnfs_read_pg_test(struct nfs_pageio_descriptor *pgio,
+                             struct nfs_page *prev,
+                             struct nfs_page *req)
+{
+        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,
+                                                   IOMODE_READ);
+        }
+        return NFS_SERVER(pgio->pg_inode)->pnfs_curr_ld->pg_test(pgio, prev, req);
+}
+
+void
+pnfs_pageio_init_read(struct nfs_pageio_descriptor *pgio, struct inode *inode)
+{
+        struct pnfs_layoutdriver_type *ld;
+
+        ld = NFS_SERVER(inode)->pnfs_curr_ld;
+        pgio->pg_test = (ld && ld->pg_test) ? pnfs_read_pg_test : NULL;
+}
+
+static int pnfs_write_pg_test(struct nfs_pageio_descriptor *pgio,
+                              struct nfs_page *prev,
+                              struct nfs_page *req)
+{
+        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,
+                                                   IOMODE_RW);
+        }
+        return NFS_SERVER(pgio->pg_inode)->pnfs_curr_ld->pg_test(pgio, prev, req);
+}
+
+void
+pnfs_pageio_init_write(struct nfs_pageio_descriptor *pgio, struct inode *inode)
+{
+        struct pnfs_layoutdriver_type *ld;
+
+        ld = NFS_SERVER(inode)->pnfs_curr_ld;
+        pgio->pg_test = (ld && ld->pg_test) ? pnfs_write_pg_test : NULL;
+}
+
+enum pnfs_try_status
+pnfs_try_to_write_data(struct nfs_write_data *wdata,
+                        const struct rpc_call_ops *call_ops, int how)
+{
+        struct inode *inode = wdata->inode;
+        enum pnfs_try_status trypnfs;
+        struct nfs_server *nfss = NFS_SERVER(inode);
+
+        wdata->mds_ops = call_ops;
+
+        dprintk("%s: Writing ino:%lu %u@%llu (how %d)\n", __func__,
+                inode->i_ino, wdata->args.count, wdata->args.offset, how);
+
+        trypnfs = nfss->pnfs_curr_ld->write_pagelist(wdata, how);
+        if (trypnfs == PNFS_NOT_ATTEMPTED) {
+                put_lseg(wdata->lseg);
+                wdata->lseg = NULL;
+        } else
+                nfs_inc_stats(inode, NFSIOS_PNFS_WRITE);
+
+        dprintk("%s End (trypnfs:%d)\n", __func__, trypnfs);
+        return trypnfs;
+}
+
 /*
- * Device ID cache. Currently supports one layout type per struct nfs_client.
- * Add layout type to the lookup key to expand to support multiple types.
+ * Call the appropriate parallel I/O subsystem read function.
  */
-int
-pnfs_alloc_init_deviceid_cache(struct nfs_client *clp,
-                         void (*free_callback)(struct pnfs_deviceid_node *))
+enum pnfs_try_status
+pnfs_try_to_read_data(struct nfs_read_data *rdata,
+                       const struct rpc_call_ops *call_ops)
 {
-        struct pnfs_deviceid_cache *c;
+        struct inode *inode = rdata->inode;
+        struct nfs_server *nfss = NFS_SERVER(inode);
+        enum pnfs_try_status trypnfs;
 
-        c = kzalloc(sizeof(struct pnfs_deviceid_cache), GFP_KERNEL);
-        if (!c)
-                return -ENOMEM;
-        spin_lock(&clp->cl_lock);
-        if (clp->cl_devid_cache != NULL) {
-                atomic_inc(&clp->cl_devid_cache->dc_ref);
-                dprintk("%s [kref [%d]]\n", __func__,
-                        atomic_read(&clp->cl_devid_cache->dc_ref));
-                kfree(c);
+        rdata->mds_ops = call_ops;
+
+        dprintk("%s: Reading ino:%lu %u@%llu\n",
+                __func__, inode->i_ino, rdata->args.count, rdata->args.offset);
+
+        trypnfs = nfss->pnfs_curr_ld->read_pagelist(rdata);
+        if (trypnfs == PNFS_NOT_ATTEMPTED) {
+                put_lseg(rdata->lseg);
+                rdata->lseg = NULL;
         } else {
-                /* kzalloc initializes hlists */
-                spin_lock_init(&c->dc_lock);
-                atomic_set(&c->dc_ref, 1);
-                c->dc_free_callback = free_callback;
-                clp->cl_devid_cache = c;
-                dprintk("%s [new]\n", __func__);
+                nfs_inc_stats(inode, NFSIOS_PNFS_READ);
         }
-        spin_unlock(&clp->cl_lock);
-        return 0;
+        dprintk("%s End (trypnfs:%d)\n", __func__, trypnfs);
+        return trypnfs;
 }
-EXPORT_SYMBOL_GPL(pnfs_alloc_init_deviceid_cache);
 
 /*
- * Called from pnfs_layoutdriver_type->free_lseg
- * last layout segment reference frees deviceid
+ * Currently there is only one (whole file) write lseg.
  */
-void
-pnfs_put_deviceid(struct pnfs_deviceid_cache *c,
-                  struct pnfs_deviceid_node *devid)
+static struct pnfs_layout_segment *pnfs_list_write_lseg(struct inode *inode)
 {
-        struct nfs4_deviceid *id = &devid->de_id;
-        struct pnfs_deviceid_node *d;
-        struct hlist_node *n;
-        long h = nfs4_deviceid_hash(id);
+        struct pnfs_layout_segment *lseg, *rv = NULL;
 
-        dprintk("%s [%d]\n", __func__, atomic_read(&devid->de_ref));
-        if (!atomic_dec_and_lock(&devid->de_ref, &c->dc_lock))
-                return;
+        list_for_each_entry(lseg, &NFS_I(inode)->layout->plh_segs, pls_list)
+                if (lseg->pls_range.iomode == IOMODE_RW)
+                        rv = lseg;
+        return rv;
+}
 
-        hlist_for_each_entry_rcu(d, n, &c->dc_deviceids[h], de_node)
-                if (!memcmp(&d->de_id, id, sizeof(*id))) {
-                        hlist_del_rcu(&d->de_node);
-                        spin_unlock(&c->dc_lock);
-                        synchronize_rcu();
-                        c->dc_free_callback(devid);
-                        return;
-                }
-        spin_unlock(&c->dc_lock);
-        /* Why wasn't it found in  the list? */
-        BUG();
-}
-EXPORT_SYMBOL_GPL(pnfs_put_deviceid);
-
-/* Find and reference a deviceid */
-struct pnfs_deviceid_node *
-pnfs_find_get_deviceid(struct pnfs_deviceid_cache *c, struct nfs4_deviceid *id)
-{
-        struct pnfs_deviceid_node *d;
-        struct hlist_node *n;
-        long hash = nfs4_deviceid_hash(id);
-
-        dprintk("--> %s hash %ld\n", __func__, hash);
-        rcu_read_lock();
-        hlist_for_each_entry_rcu(d, n, &c->dc_deviceids[hash], de_node) {
-                if (!memcmp(&d->de_id, id, sizeof(*id))) {
-                        if (!atomic_inc_not_zero(&d->de_ref)) {
-                                goto fail;
-                        } else {
-                                rcu_read_unlock();
-                                return d;
-                        }
-                }
+void
+pnfs_set_layoutcommit(struct nfs_write_data *wdata)
+{
+        struct nfs_inode *nfsi = NFS_I(wdata->inode);
+        loff_t end_pos = wdata->args.offset + wdata->res.count;
+        bool mark_as_dirty = false;
+
+        spin_lock(&nfsi->vfs_inode.i_lock);
+        if (!test_and_set_bit(NFS_INO_LAYOUTCOMMIT, &nfsi->flags)) {
+                /* references matched in nfs4_layoutcommit_release */
+                get_lseg(wdata->lseg);
+                wdata->lseg->pls_lc_cred =
+                        get_rpccred(wdata->args.context->state->owner->so_cred);
+                mark_as_dirty = true;
+                dprintk("%s: Set layoutcommit for inode %lu ",
+                        __func__, wdata->inode->i_ino);
         }
-fail:
-        rcu_read_unlock();
-        return NULL;
+        if (end_pos > wdata->lseg->pls_end_pos)
+                wdata->lseg->pls_end_pos = end_pos;
+        spin_unlock(&nfsi->vfs_inode.i_lock);
+
+        /* if pnfs_layoutcommit_inode() runs between inode locks, the next one
+         * will be a noop because NFS_INO_LAYOUTCOMMIT will not be set */
+        if (mark_as_dirty)
+                mark_inode_dirty_sync(wdata->inode);
 }
-EXPORT_SYMBOL_GPL(pnfs_find_get_deviceid);
+EXPORT_SYMBOL_GPL(pnfs_set_layoutcommit);
 
 /*
- * Add a deviceid to the cache.
- * GETDEVICEINFOs for same deviceid can race. If deviceid is found, discard new
+ * For the LAYOUT4_NFSV4_1_FILES layout type, NFS_DATA_SYNC WRITEs and
+ * NFS_UNSTABLE WRITEs with a COMMIT to data servers must store enough
+ * data to disk to allow the server to recover the data if it crashes.
+ * LAYOUTCOMMIT is only needed when the NFL4_UFLG_COMMIT_THRU_MDS flag
+ * is off, and a COMMIT is sent to a data server, or
+ * if WRITEs to a data server return NFS_DATA_SYNC.
  */
-struct pnfs_deviceid_node *
-pnfs_add_deviceid(struct pnfs_deviceid_cache *c, struct pnfs_deviceid_node *new)
-{
-        struct pnfs_deviceid_node *d;
-        long hash = nfs4_deviceid_hash(&new->de_id);
-
-        dprintk("--> %s hash %ld\n", __func__, hash);
-        spin_lock(&c->dc_lock);
-        d = pnfs_find_get_deviceid(c, &new->de_id);
-        if (d) {
-                spin_unlock(&c->dc_lock);
-                dprintk("%s [discard]\n", __func__);
-                c->dc_free_callback(new);
-                return d;
-        }
-        INIT_HLIST_NODE(&new->de_node);
-        atomic_set(&new->de_ref, 1);
-        hlist_add_head_rcu(&new->de_node, &c->dc_deviceids[hash]);
-        spin_unlock(&c->dc_lock);
-        dprintk("%s [new]\n", __func__);
-        return new;
-}
-EXPORT_SYMBOL_GPL(pnfs_add_deviceid);
-
-void
-pnfs_put_deviceid_cache(struct nfs_client *clp)
+int
+pnfs_layoutcommit_inode(struct inode *inode, bool sync)
 {
-        struct pnfs_deviceid_cache *local = clp->cl_devid_cache;
+        struct nfs4_layoutcommit_data *data;
+        struct nfs_inode *nfsi = NFS_I(inode);
+        struct pnfs_layout_segment *lseg;
+        struct rpc_cred *cred;
+        loff_t end_pos;
+        int status = 0;
 
-        dprintk("--> %s ({%d})\n", __func__, atomic_read(&local->dc_ref));
-        if (atomic_dec_and_lock(&local->dc_ref, &clp->cl_lock)) {
-                int i;
-                /* Verify cache is empty */
-                for (i = 0; i < NFS4_DEVICE_ID_HASH_SIZE; i++)
-                        BUG_ON(!hlist_empty(&local->dc_deviceids[i]));
-                clp->cl_devid_cache = NULL;
-                spin_unlock(&clp->cl_lock);
-                kfree(local);
+        dprintk("--> %s inode %lu\n", __func__, inode->i_ino);
+
+        if (!test_bit(NFS_INO_LAYOUTCOMMIT, &nfsi->flags))
+                return 0;
+
+        /* Note kzalloc ensures data->res.seq_res.sr_slot == NULL */
+        data = kzalloc(sizeof(*data), GFP_NOFS);
+        if (!data) {
+                mark_inode_dirty_sync(inode);
+                status = -ENOMEM;
+                goto out;
         }
+
+        spin_lock(&inode->i_lock);
+        if (!test_and_clear_bit(NFS_INO_LAYOUTCOMMIT, &nfsi->flags)) {
+                spin_unlock(&inode->i_lock);
+                kfree(data);
+                goto out;
+        }
+        /*
+         * Currently only one (whole file) write lseg which is referenced
+         * in pnfs_set_layoutcommit and will be found.
+         */
+        lseg = pnfs_list_write_lseg(inode);
+
+        end_pos = lseg->pls_end_pos;
+        cred = lseg->pls_lc_cred;
+        lseg->pls_end_pos = 0;
+        lseg->pls_lc_cred = NULL;
+
+        memcpy(&data->args.stateid.data, nfsi->layout->plh_stateid.data,
+                sizeof(nfsi->layout->plh_stateid.data));
+        spin_unlock(&inode->i_lock);
+
+        data->args.inode = inode;
+        data->lseg = lseg;
+        data->cred = cred;
+        nfs_fattr_init(&data->fattr);
+        data->args.bitmask = NFS_SERVER(inode)->cache_consistency_bitmask;
+        data->res.fattr = &data->fattr;
+        data->args.lastbytewritten = end_pos - 1;
+        data->res.server = NFS_SERVER(inode);
+
+        status = nfs4_proc_layoutcommit(data, sync);
+out:
+        dprintk("<-- %s status %d\n", __func__, status);
+        return status;
 }
-EXPORT_SYMBOL_GPL(pnfs_put_deviceid_cache);
diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h
index e2612ea0cbed..bc4827202e7a 100644
--- a/fs/nfs/pnfs.h
+++ b/fs/nfs/pnfs.h
@@ -30,6 +30,8 @@
 #ifndef FS_NFS_PNFS_H
 #define FS_NFS_PNFS_H
 
+#include <linux/nfs_page.h>
+
 enum {
         NFS_LSEG_VALID = 0,     /* cleared when lseg is recalled/returned */
         NFS_LSEG_ROC,           /* roc bit received from server */
@@ -41,6 +43,13 @@ struct pnfs_layout_segment {
         atomic_t pls_refcount;
         unsigned long pls_flags;
         struct pnfs_layout_hdr *pls_layout;
+        struct rpc_cred *pls_lc_cred; /* LAYOUTCOMMIT credential */
+        loff_t pls_end_pos; /* LAYOUTCOMMIT write end */
+};
+
+enum pnfs_try_status {
+        PNFS_ATTEMPTED     = 0,
+        PNFS_NOT_ATTEMPTED = 1,
 };
 
 #ifdef CONFIG_NFS_V4_1
@@ -61,10 +70,25 @@ struct pnfs_layoutdriver_type {
         const u32 id;
         const char *name;
         struct module *owner;
-        int (*set_layoutdriver) (struct nfs_server *);
-        int (*clear_layoutdriver) (struct nfs_server *);
         struct pnfs_layout_segment * (*alloc_lseg) (struct pnfs_layout_hdr *layoutid, struct nfs4_layoutget_res *lgr);
         void (*free_lseg) (struct pnfs_layout_segment *lseg);
+
+        /* test for nfs page cache coalescing */
+        int (*pg_test)(struct nfs_pageio_descriptor *, struct nfs_page *, struct nfs_page *);
+
+        /* Returns true if layoutdriver wants to divert this request to
+         * driver's commit routine.
+         */
+        bool (*mark_pnfs_commit)(struct pnfs_layout_segment *lseg);
+        struct list_head * (*choose_commit_list) (struct nfs_page *req);
+        int (*commit_pagelist)(struct inode *inode, struct list_head *mds_pages, int how);
+
+        /*
+         * Return PNFS_ATTEMPTED to indicate the layout code has attempted
+         * I/O, else return PNFS_NOT_ATTEMPTED to fall back to normal NFS
+         */
+        enum pnfs_try_status (*read_pagelist) (struct nfs_read_data *nfs_data);
+        enum pnfs_try_status (*write_pagelist) (struct nfs_write_data *nfs_data, int how);
 };
 
 struct pnfs_layout_hdr {
@@ -85,57 +109,10 @@ struct pnfs_device {
         unsigned int  layout_type;
         unsigned int  mincount;
         struct page **pages;
-        void          *area;
         unsigned int  pgbase;
         unsigned int  pglen;
 };
 
-/*
- * Device ID RCU cache. A device ID is unique per client ID and layout type.
- */
-#define NFS4_DEVICE_ID_HASH_BITS        5
-#define NFS4_DEVICE_ID_HASH_SIZE        (1 << NFS4_DEVICE_ID_HASH_BITS)
-#define NFS4_DEVICE_ID_HASH_MASK        (NFS4_DEVICE_ID_HASH_SIZE - 1)
-
-static inline u32
-nfs4_deviceid_hash(struct nfs4_deviceid *id)
-{
-        unsigned char *cptr = (unsigned char *)id->data;
-        unsigned int nbytes = NFS4_DEVICEID4_SIZE;
-        u32 x = 0;
-
-        while (nbytes--) {
-                x *= 37;
-                x += *cptr++;
-        }
-        return x & NFS4_DEVICE_ID_HASH_MASK;
-}
-
-struct pnfs_deviceid_node {
-        struct hlist_node       de_node;
-        struct nfs4_deviceid    de_id;
-        atomic_t                de_ref;
-};
-
-struct pnfs_deviceid_cache {
-        spinlock_t              dc_lock;
-        atomic_t                dc_ref;
-        void                    (*dc_free_callback)(struct pnfs_deviceid_node *);
-        struct hlist_head       dc_deviceids[NFS4_DEVICE_ID_HASH_SIZE];
-};
-
-extern int pnfs_alloc_init_deviceid_cache(struct nfs_client *,
-                        void (*free_callback)(struct pnfs_deviceid_node *));
-extern void pnfs_put_deviceid_cache(struct nfs_client *);
-extern struct pnfs_deviceid_node *pnfs_find_get_deviceid(
-                                struct pnfs_deviceid_cache *,
-                                struct nfs4_deviceid *);
-extern struct pnfs_deviceid_node *pnfs_add_deviceid(
-                                struct pnfs_deviceid_cache *,
-                                struct pnfs_deviceid_node *);
-extern void pnfs_put_deviceid(struct pnfs_deviceid_cache *c,
-                              struct pnfs_deviceid_node *devid);
-
 extern int pnfs_register_layoutdriver(struct pnfs_layoutdriver_type *);
 extern void pnfs_unregister_layoutdriver(struct pnfs_layoutdriver_type *);
 
@@ -146,11 +123,18 @@ extern int nfs4_proc_layoutget(struct nfs4_layoutget *lgp);
 
 /* 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,
                    enum pnfs_iomode access_type);
 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_pageio_init_read(struct nfs_pageio_descriptor *, struct inode *);
+void pnfs_pageio_init_write(struct nfs_pageio_descriptor *, struct inode *);
 int pnfs_layout_process(struct nfs4_layoutget *lgp);
 void pnfs_free_lseg_list(struct list_head *tmp_list);
 void pnfs_destroy_layout(struct nfs_inode *);
@@ -169,7 +153,8 @@ bool pnfs_roc(struct inode *ino);
 void pnfs_roc_release(struct inode *ino);
 void pnfs_roc_set_barrier(struct inode *ino, u32 barrier);
 bool pnfs_roc_drain(struct inode *ino, u32 *barrier);
-
+void pnfs_set_layoutcommit(struct nfs_write_data *wdata);
+int pnfs_layoutcommit_inode(struct inode *inode, bool sync);
 
 static inline int lo_fail_bit(u32 iomode)
 {
@@ -177,12 +162,67 @@ static inline int lo_fail_bit(u32 iomode)
                          NFS_LAYOUT_RW_FAILED : NFS_LAYOUT_RO_FAILED;
 }
 
+static inline struct pnfs_layout_segment *
+get_lseg(struct pnfs_layout_segment *lseg)
+{
+        if (lseg) {
+                atomic_inc(&lseg->pls_refcount);
+                smp_mb__after_atomic_inc();
+        }
+        return lseg;
+}
+
 /* Return true if a layout driver is being used for this mountpoint */
 static inline int pnfs_enabled_sb(struct nfs_server *nfss)
 {
         return nfss->pnfs_curr_ld != NULL;
 }
 
+static inline void
+pnfs_mark_request_commit(struct nfs_page *req, struct pnfs_layout_segment *lseg)
+{
+        if (lseg) {
+                struct pnfs_layoutdriver_type *ld;
+
+                ld = NFS_SERVER(req->wb_page->mapping->host)->pnfs_curr_ld;
+                if (ld->mark_pnfs_commit && ld->mark_pnfs_commit(lseg)) {
+                        set_bit(PG_PNFS_COMMIT, &req->wb_flags);
+                        req->wb_commit_lseg = get_lseg(lseg);
+                }
+        }
+}
+
+static inline int
+pnfs_commit_list(struct inode *inode, struct list_head *mds_pages, int how)
+{
+        if (!test_and_clear_bit(NFS_INO_PNFS_COMMIT, &NFS_I(inode)->flags))
+                return PNFS_NOT_ATTEMPTED;
+        return NFS_SERVER(inode)->pnfs_curr_ld->commit_pagelist(inode, mds_pages, how);
+}
+
+static inline struct list_head *
+pnfs_choose_commit_list(struct nfs_page *req, struct list_head *mds)
+{
+        struct list_head *rv;
+
+        if (test_and_clear_bit(PG_PNFS_COMMIT, &req->wb_flags)) {
+                struct inode *inode = req->wb_commit_lseg->pls_layout->plh_inode;
+
+                set_bit(NFS_INO_PNFS_COMMIT, &NFS_I(inode)->flags);
+                rv = NFS_SERVER(inode)->pnfs_curr_ld->choose_commit_list(req);
+                /* matched by ref taken when PG_PNFS_COMMIT is set */
+                put_lseg(req->wb_commit_lseg);
+        } else
+                rv = mds;
+        return rv;
+}
+
+static inline void pnfs_clear_request_commit(struct nfs_page *req)
+{
+        if (test_and_clear_bit(PG_PNFS_COMMIT, &req->wb_flags))
+                put_lseg(req->wb_commit_lseg);
+}
+
 #else  /* CONFIG_NFS_V4_1 */
 
 static inline void pnfs_destroy_all_layouts(struct nfs_client *clp)
@@ -194,12 +234,36 @@ static inline void pnfs_destroy_layout(struct nfs_inode *nfsi)
 }
 
 static inline struct pnfs_layout_segment *
+get_lseg(struct pnfs_layout_segment *lseg)
+{
+        return NULL;
+}
+
+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,
                    enum pnfs_iomode access_type)
 {
         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)
+{
+        return PNFS_NOT_ATTEMPTED;
+}
+
+static inline enum pnfs_try_status
+pnfs_try_to_write_data(struct nfs_write_data *data,
+                       const struct rpc_call_ops *call_ops, int how)
+{
+        return PNFS_NOT_ATTEMPTED;
+}
+
 static inline bool
 pnfs_roc(struct inode *ino)
 {
@@ -230,6 +294,43 @@ static inline void unset_pnfs_layoutdriver(struct nfs_server *s)
 {
 }
 
+static inline void
+pnfs_pageio_init_read(struct nfs_pageio_descriptor *pgio, struct inode *ino)
+{
+        pgio->pg_test = NULL;
+}
+
+static inline void
+pnfs_pageio_init_write(struct nfs_pageio_descriptor *pgio, struct inode *ino)
+{
+        pgio->pg_test = NULL;
+}
+
+static inline void
+pnfs_mark_request_commit(struct nfs_page *req, struct pnfs_layout_segment *lseg)
+{
+}
+
+static inline int
+pnfs_commit_list(struct inode *inode, struct list_head *mds_pages, int how)
+{
+        return PNFS_NOT_ATTEMPTED;
+}
+
+static inline struct list_head *
+pnfs_choose_commit_list(struct nfs_page *req, struct list_head *mds)
+{
+        return mds;
+}
+
+static inline void pnfs_clear_request_commit(struct nfs_page *req)
+{
+}
+
+static inline int pnfs_layoutcommit_inode(struct inode *inode, bool sync)
+{
+        return 0;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 #endif /* FS_NFS_PNFS_H */
diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c
index 77d5e21c4ad6..ac40b8535d7e 100644
--- a/fs/nfs/proc.c
+++ b/fs/nfs/proc.c
@@ -177,7 +177,7 @@ nfs_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr,
 }
 
 static int
-nfs_proc_lookup(struct inode *dir, struct qstr *name,
+nfs_proc_lookup(struct rpc_clnt *clnt, struct inode *dir, struct qstr *name,
                 struct nfs_fh *fhandle, struct nfs_fattr *fattr)
 {
         struct nfs_diropargs    arg = {
@@ -741,4 +741,5 @@ const struct nfs_rpc_ops nfs_v2_clientops = {
         .lock           = nfs_proc_lock,
         .lock_check_bounds = nfs_lock_check_bounds,
         .close_context  = nfs_close_context,
+        .init_client    = nfs_init_client,
 };
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index aedcaa7f291f..7cded2b12a05 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -18,19 +18,20 @@
 #include <linux/sunrpc/clnt.h>
 #include <linux/nfs_fs.h>
 #include <linux/nfs_page.h>
+#include <linux/module.h>
 
 #include <asm/system.h>
+#include "pnfs.h"
 
 #include "nfs4_fs.h"
 #include "internal.h"
 #include "iostat.h"
 #include "fscache.h"
-#include "pnfs.h"
 
 #define NFSDBG_FACILITY         NFSDBG_PAGECACHE
 
-static int nfs_pagein_multi(struct inode *, struct list_head *, unsigned int, size_t, int);
-static int nfs_pagein_one(struct inode *, struct list_head *, unsigned int, size_t, int);
+static int nfs_pagein_multi(struct nfs_pageio_descriptor *desc);
+static int nfs_pagein_one(struct nfs_pageio_descriptor *desc);
 static const struct rpc_call_ops nfs_read_partial_ops;
 static const struct rpc_call_ops nfs_read_full_ops;
 
@@ -69,6 +70,7 @@ void nfs_readdata_free(struct nfs_read_data *p)
 
 static void nfs_readdata_release(struct nfs_read_data *rdata)
 {
+        put_lseg(rdata->lseg);
         put_nfs_open_context(rdata->args.context);
         nfs_readdata_free(rdata);
 }
@@ -114,14 +116,13 @@ static void nfs_readpage_truncate_uninitialised_page(struct nfs_read_data *data)
 int nfs_readpage_async(struct nfs_open_context *ctx, struct inode *inode,
                        struct page *page)
 {
-        LIST_HEAD(one_request);
         struct nfs_page *new;
         unsigned int len;
+        struct nfs_pageio_descriptor pgio;
 
         len = nfs_page_length(page);
         if (len == 0)
                 return nfs_return_empty_page(page);
-        pnfs_update_layout(inode, ctx, IOMODE_READ);
         new = nfs_create_request(ctx, inode, page, 0, len);
         if (IS_ERR(new)) {
                 unlock_page(page);
@@ -130,11 +131,14 @@ 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_list_add_request(new, &one_request);
+        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(inode, &one_request, 1, len, 0);
+                nfs_pagein_multi(&pgio);
         else
-                nfs_pagein_one(inode, &one_request, 1, len, 0);
+                nfs_pagein_one(&pgio);
         return 0;
 }
 
@@ -155,24 +159,20 @@ static void nfs_readpage_release(struct nfs_page *req)
         nfs_release_request(req);
 }
 
-/*
- * Set up the NFS read request struct
- */
-static int nfs_read_rpcsetup(struct nfs_page *req, struct nfs_read_data *data,
-                const struct rpc_call_ops *call_ops,
-                unsigned int count, unsigned int offset)
+int nfs_initiate_read(struct nfs_read_data *data, struct rpc_clnt *clnt,
+                      const struct rpc_call_ops *call_ops)
 {
-        struct inode *inode = req->wb_context->path.dentry->d_inode;
+        struct inode *inode = data->inode;
         int swap_flags = IS_SWAPFILE(inode) ? NFS_RPC_SWAPFLAGS : 0;
         struct rpc_task *task;
         struct rpc_message msg = {
                 .rpc_argp = &data->args,
                 .rpc_resp = &data->res,
-                .rpc_cred = req->wb_context->cred,
+                .rpc_cred = data->cred,
         };
         struct rpc_task_setup task_setup_data = {
                 .task = &data->task,
-                .rpc_client = NFS_CLIENT(inode),
+                .rpc_client = clnt,
                 .rpc_message = &msg,
                 .callback_ops = call_ops,
                 .callback_data = data,
@@ -180,9 +180,39 @@ static int nfs_read_rpcsetup(struct nfs_page *req, struct nfs_read_data *data,
                 .flags = RPC_TASK_ASYNC | swap_flags,
         };
 
+        /* Set up the initial task struct. */
+        NFS_PROTO(inode)->read_setup(data, &msg);
+
+        dprintk("NFS: %5u initiated read call (req %s/%lld, %u bytes @ "
+                        "offset %llu)\n",
+                        data->task.tk_pid,
+                        inode->i_sb->s_id,
+                        (long long)NFS_FILEID(inode),
+                        data->args.count,
+                        (unsigned long long)data->args.offset);
+
+        task = rpc_run_task(&task_setup_data);
+        if (IS_ERR(task))
+                return PTR_ERR(task);
+        rpc_put_task(task);
+        return 0;
+}
+EXPORT_SYMBOL_GPL(nfs_initiate_read);
+
+/*
+ * Set up the NFS read request struct
+ */
+static int nfs_read_rpcsetup(struct nfs_page *req, struct nfs_read_data *data,
+                const struct rpc_call_ops *call_ops,
+                unsigned int count, unsigned int offset,
+                struct pnfs_layout_segment *lseg)
+{
+        struct inode *inode = req->wb_context->path.dentry->d_inode;
+
         data->req         = req;
         data->inode       = inode;
-        data->cred        = msg.rpc_cred;
+        data->cred        = req->wb_context->cred;
+        data->lseg        = get_lseg(lseg);
 
         data->args.fh     = NFS_FH(inode);
         data->args.offset = req_offset(req) + offset;
@@ -197,21 +227,11 @@ static int nfs_read_rpcsetup(struct nfs_page *req, struct nfs_read_data *data,
         data->res.eof     = 0;
         nfs_fattr_init(&data->fattr);
 
-        /* Set up the initial task struct. */
-        NFS_PROTO(inode)->read_setup(data, &msg);
-
-        dprintk("NFS: %5u initiated read call (req %s/%Ld, %u bytes @ offset %Lu)\n",
-                        data->task.tk_pid,
-                        inode->i_sb->s_id,
-                        (long long)NFS_FILEID(inode),
-                        count,
-                        (unsigned long long)data->args.offset);
+        if (data->lseg &&
+            (pnfs_try_to_read_data(data, call_ops) == PNFS_ATTEMPTED))
+                return 0;
 
-        task = rpc_run_task(&task_setup_data);
-        if (IS_ERR(task))
-                return PTR_ERR(task);
-        rpc_put_task(task);
-        return 0;
+        return nfs_initiate_read(data, NFS_CLIENT(inode), call_ops);
 }
 
 static void
@@ -240,20 +260,21 @@ nfs_async_read_error(struct list_head *head)
  * won't see the new data until our attribute cache is updated.  This is more
  * or less conventional NFS client behavior.
  */
-static int nfs_pagein_multi(struct inode *inode, struct list_head *head, unsigned int npages, size_t count, int flags)
+static int nfs_pagein_multi(struct nfs_pageio_descriptor *desc)
 {
-        struct nfs_page *req = nfs_list_entry(head->next);
+        struct nfs_page *req = nfs_list_entry(desc->pg_list.next);
         struct page *page = req->wb_page;
         struct nfs_read_data *data;
-        size_t rsize = NFS_SERVER(inode)->rsize, nbytes;
+        size_t rsize = NFS_SERVER(desc->pg_inode)->rsize, nbytes;
         unsigned int offset;
         int requests = 0;
         int ret = 0;
+        struct pnfs_layout_segment *lseg;
         LIST_HEAD(list);
 
         nfs_list_remove_request(req);
 
-        nbytes = count;
+        nbytes = desc->pg_count;
         do {
                 size_t len = min(nbytes,rsize);
 
@@ -266,9 +287,11 @@ static int nfs_pagein_multi(struct inode *inode, struct list_head *head, unsigne
         } 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, IOMODE_READ);
         ClearPageError(page);
         offset = 0;
-        nbytes = count;
+        nbytes = desc->pg_count;
         do {
                 int ret2;
 
@@ -280,12 +303,14 @@ static int nfs_pagein_multi(struct inode *inode, struct list_head *head, unsigne
                 if (nbytes < rsize)
                         rsize = nbytes;
                 ret2 = nfs_read_rpcsetup(req, data, &nfs_read_partial_ops,
-                                  rsize, offset);
+                                         rsize, offset, lseg);
                 if (ret == 0)
                         ret = ret2;
                 offset += rsize;
                 nbytes -= rsize;
         } while (nbytes != 0);
+        put_lseg(lseg);
+        desc->pg_lseg = NULL;
 
         return ret;
 
@@ -300,16 +325,21 @@ out_bad:
         return -ENOMEM;
 }
 
-static int nfs_pagein_one(struct inode *inode, struct list_head *head, unsigned int npages, size_t count, int flags)
+static int nfs_pagein_one(struct nfs_pageio_descriptor *desc)
 {
         struct nfs_page         *req;
         struct page             **pages;
         struct nfs_read_data    *data;
+        struct list_head *head = &desc->pg_list;
+        struct pnfs_layout_segment *lseg = desc->pg_lseg;
         int ret = -ENOMEM;
 
-        data = nfs_readdata_alloc(npages);
-        if (!data)
-                goto out_bad;
+        data = nfs_readdata_alloc(nfs_page_array_len(desc->pg_base,
+                                                     desc->pg_count));
+        if (!data) {
+                nfs_async_read_error(head);
+                goto out;
+        }
 
         pages = data->pagevec;
         while (!list_empty(head)) {
@@ -320,10 +350,14 @@ static int nfs_pagein_one(struct inode *inode, struct list_head *head, unsigned
                 *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, IOMODE_READ);
 
-        return nfs_read_rpcsetup(req, data, &nfs_read_full_ops, count, 0);
-out_bad:
-        nfs_async_read_error(head);
+        ret = nfs_read_rpcsetup(req, data, &nfs_read_full_ops, desc->pg_count,
+                                0, lseg);
+out:
+        put_lseg(lseg);
+        desc->pg_lseg = NULL;
         return ret;
 }
 
@@ -366,6 +400,7 @@ static void nfs_readpage_retry(struct rpc_task *task, struct nfs_read_data *data
                 return;
 
         /* Yes, so retry the read at the end of the data */
+        data->mds_offset += resp->count;
         argp->offset += resp->count;
         argp->pgbase += resp->count;
         argp->count -= resp->count;
@@ -625,7 +660,7 @@ int nfs_readpages(struct file *filp, struct address_space *mapping,
         if (ret == 0)
                 goto read_complete; /* all pages were read */
 
-        pnfs_update_layout(inode, desc.ctx, IOMODE_READ);
+        pnfs_pageio_init_read(&pgio, inode);
         if (rsize < PAGE_CACHE_SIZE)
                 nfs_pageio_init(&pgio, inode, nfs_pagein_multi, rsize, 0);
         else
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index b68c8607770f..e288f06d3fa7 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -263,8 +263,11 @@ static match_table_t nfs_local_lock_tokens = {
 static void nfs_umount_begin(struct super_block *);
 static int  nfs_statfs(struct dentry *, struct kstatfs *);
 static int  nfs_show_options(struct seq_file *, struct vfsmount *);
+static int  nfs_show_devname(struct seq_file *, struct vfsmount *);
+static int  nfs_show_path(struct seq_file *, struct vfsmount *);
 static int  nfs_show_stats(struct seq_file *, struct vfsmount *);
-static int nfs_get_sb(struct file_system_type *, int, const char *, void *, struct vfsmount *);
+static struct dentry *nfs_fs_mount(struct file_system_type *,
+                int, const char *, void *);
 static struct dentry *nfs_xdev_mount(struct file_system_type *fs_type,
                 int flags, const char *dev_name, void *raw_data);
 static void nfs_put_super(struct super_block *);
@@ -274,7 +277,7 @@ static int nfs_remount(struct super_block *sb, int *flags, char *raw_data);
 static struct file_system_type nfs_fs_type = {
         .owner          = THIS_MODULE,
         .name           = "nfs",
-        .get_sb         = nfs_get_sb,
+        .mount          = nfs_fs_mount,
         .kill_sb        = nfs_kill_super,
         .fs_flags       = FS_RENAME_DOES_D_MOVE|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
 };
@@ -296,6 +299,8 @@ static const struct super_operations nfs_sops = {
         .evict_inode    = nfs_evict_inode,
         .umount_begin   = nfs_umount_begin,
         .show_options   = nfs_show_options,
+        .show_devname   = nfs_show_devname,
+        .show_path      = nfs_show_path,
         .show_stats     = nfs_show_stats,
         .remount_fs     = nfs_remount,
 };
@@ -303,16 +308,16 @@ static const struct super_operations nfs_sops = {
 #ifdef CONFIG_NFS_V4
 static int nfs4_validate_text_mount_data(void *options,
         struct nfs_parsed_mount_data *args, const char *dev_name);
-static int nfs4_try_mount(int flags, const char *dev_name,
-        struct nfs_parsed_mount_data *data, struct vfsmount *mnt);
-static int nfs4_get_sb(struct file_system_type *fs_type,
-        int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt);
+static struct dentry *nfs4_try_mount(int flags, const char *dev_name,
+        struct nfs_parsed_mount_data *data);
+static struct dentry *nfs4_mount(struct file_system_type *fs_type,
+        int flags, const char *dev_name, void *raw_data);
 static struct dentry *nfs4_remote_mount(struct file_system_type *fs_type,
         int flags, const char *dev_name, void *raw_data);
 static struct dentry *nfs4_xdev_mount(struct file_system_type *fs_type,
         int flags, const char *dev_name, void *raw_data);
-static int nfs4_referral_get_sb(struct file_system_type *fs_type,
-        int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt);
+static struct dentry *nfs4_referral_mount(struct file_system_type *fs_type,
+        int flags, const char *dev_name, void *raw_data);
 static struct dentry *nfs4_remote_referral_mount(struct file_system_type *fs_type,
         int flags, const char *dev_name, void *raw_data);
 static void nfs4_kill_super(struct super_block *sb);
@@ -320,7 +325,7 @@ static void nfs4_kill_super(struct super_block *sb);
 static struct file_system_type nfs4_fs_type = {
         .owner          = THIS_MODULE,
         .name           = "nfs4",
-        .get_sb         = nfs4_get_sb,
+        .mount          = nfs4_mount,
         .kill_sb        = nfs4_kill_super,
         .fs_flags       = FS_RENAME_DOES_D_MOVE|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
 };
@@ -352,7 +357,7 @@ static struct file_system_type nfs4_remote_referral_fs_type = {
 struct file_system_type nfs4_referral_fs_type = {
         .owner          = THIS_MODULE,
         .name           = "nfs4",
-        .get_sb         = nfs4_referral_get_sb,
+        .mount          = nfs4_referral_mount,
         .kill_sb        = nfs4_kill_super,
         .fs_flags       = FS_RENAME_DOES_D_MOVE|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
 };
@@ -366,6 +371,8 @@ static const struct super_operations nfs4_sops = {
         .evict_inode    = nfs4_evict_inode,
         .umount_begin   = nfs_umount_begin,
         .show_options   = nfs_show_options,
+        .show_devname   = nfs_show_devname,
+        .show_path      = nfs_show_path,
         .show_stats     = nfs_show_stats,
         .remount_fs     = nfs_remount,
 };
@@ -726,6 +733,28 @@ static int nfs_show_options(struct seq_file *m, struct vfsmount *mnt)
         return 0;
 }
 
+static int nfs_show_devname(struct seq_file *m, struct vfsmount *mnt)
+{
+        char *page = (char *) __get_free_page(GFP_KERNEL);
+        char *devname, *dummy;
+        int err = 0;
+        if (!page)
+                return -ENOMEM;
+        devname = nfs_path(&dummy, mnt->mnt_root, page, PAGE_SIZE);
+        if (IS_ERR(devname))
+                err = PTR_ERR(devname);
+        else
+                seq_escape(m, devname, " \t\n\\");
+        free_page((unsigned long)page);
+        return err;
+}
+
+static int nfs_show_path(struct seq_file *m, struct vfsmount *mnt)
+{
+        seq_puts(m, "/");
+        return 0;
+}
+
 /*
  * Present statistical information for this VFS mountpoint
  */
@@ -975,10 +1004,32 @@ static int nfs_parse_security_flavors(char *value,
                 return 0;
         }
 
+        mnt->flags |= NFS_MOUNT_SECFLAVOUR;
         mnt->auth_flavor_len = 1;
         return 1;
 }
 
+static int nfs_get_option_str(substring_t args[], char **option)
+{
+        kfree(*option);
+        *option = match_strdup(args);
+        return !option;
+}
+
+static int nfs_get_option_ul(substring_t args[], unsigned long *option)
+{
+        int rc;
+        char *string;
+
+        string = match_strdup(args);
+        if (string == NULL)
+                return -ENOMEM;
+        rc = strict_strtoul(string, 10, option);
+        kfree(string);
+
+        return rc;
+}
+
 /*
  * Error-check and convert a string of mount options from user space into
  * a data structure.  The whole mount string is processed; bad options are
@@ -1127,155 +1178,82 @@ static int nfs_parse_mount_options(char *raw,
                  * options that take numeric values
                  */
                 case Opt_port:
-                        string = match_strdup(args);
-                        if (string == NULL)
-                                goto out_nomem;
-                        rc = strict_strtoul(string, 10, &option);
-                        kfree(string);
-                        if (rc != 0 || option > USHRT_MAX)
+                        if (nfs_get_option_ul(args, &option) ||
+                            option > USHRT_MAX)
                                 goto out_invalid_value;
                         mnt->nfs_server.port = option;
                         break;
                 case Opt_rsize:
-                        string = match_strdup(args);
-                        if (string == NULL)
-                                goto out_nomem;
-                        rc = strict_strtoul(string, 10, &option);
-                        kfree(string);
-                        if (rc != 0)
+                        if (nfs_get_option_ul(args, &option))
                                 goto out_invalid_value;
                         mnt->rsize = option;
                         break;
                 case Opt_wsize:
-                        string = match_strdup(args);
-                        if (string == NULL)
-                                goto out_nomem;
-                        rc = strict_strtoul(string, 10, &option);
-                        kfree(string);
-                        if (rc != 0)
+                        if (nfs_get_option_ul(args, &option))
                                 goto out_invalid_value;
                         mnt->wsize = option;
                         break;
                 case Opt_bsize:
-                        string = match_strdup(args);
-                        if (string == NULL)
-                                goto out_nomem;
-                        rc = strict_strtoul(string, 10, &option);
-                        kfree(string);
-                        if (rc != 0)
+                        if (nfs_get_option_ul(args, &option))
                                 goto out_invalid_value;
                         mnt->bsize = option;
                         break;
                 case Opt_timeo:
-                        string = match_strdup(args);
-                        if (string == NULL)
-                                goto out_nomem;
-                        rc = strict_strtoul(string, 10, &option);
-                        kfree(string);
-                        if (rc != 0 || option == 0)
+                        if (nfs_get_option_ul(args, &option) || option == 0)
                                 goto out_invalid_value;
                         mnt->timeo = option;
                         break;
                 case Opt_retrans:
-                        string = match_strdup(args);
-                        if (string == NULL)
-                                goto out_nomem;
-                        rc = strict_strtoul(string, 10, &option);
-                        kfree(string);
-                        if (rc != 0 || option == 0)
+                        if (nfs_get_option_ul(args, &option) || option == 0)
                                 goto out_invalid_value;
                         mnt->retrans = option;
                         break;
                 case Opt_acregmin:
-                        string = match_strdup(args);
-                        if (string == NULL)
-                                goto out_nomem;
-                        rc = strict_strtoul(string, 10, &option);
-                        kfree(string);
-                        if (rc != 0)
+                        if (nfs_get_option_ul(args, &option))
                                 goto out_invalid_value;
                         mnt->acregmin = option;
                         break;
                 case Opt_acregmax:
-                        string = match_strdup(args);
-                        if (string == NULL)
-                                goto out_nomem;
-                        rc = strict_strtoul(string, 10, &option);
-                        kfree(string);
-                        if (rc != 0)
+                        if (nfs_get_option_ul(args, &option))
                                 goto out_invalid_value;
                         mnt->acregmax = option;
                         break;
                 case Opt_acdirmin:
-                        string = match_strdup(args);
-                        if (string == NULL)
-                                goto out_nomem;
-                        rc = strict_strtoul(string, 10, &option);
-                        kfree(string);
-                        if (rc != 0)
+                        if (nfs_get_option_ul(args, &option))
                                 goto out_invalid_value;
                         mnt->acdirmin = option;
                         break;
                 case Opt_acdirmax:
-                        string = match_strdup(args);
-                        if (string == NULL)
-                                goto out_nomem;
-                        rc = strict_strtoul(string, 10, &option);
-                        kfree(string);
-                        if (rc != 0)
+                        if (nfs_get_option_ul(args, &option))
                                 goto out_invalid_value;
                         mnt->acdirmax = option;
                         break;
                 case Opt_actimeo:
-                        string = match_strdup(args);
-                        if (string == NULL)
-                                goto out_nomem;
-                        rc = strict_strtoul(string, 10, &option);
-                        kfree(string);
-                        if (rc != 0)
+                        if (nfs_get_option_ul(args, &option))
                                 goto out_invalid_value;
                         mnt->acregmin = mnt->acregmax =
                         mnt->acdirmin = mnt->acdirmax = option;
                         break;
                 case Opt_namelen:
-                        string = match_strdup(args);
-                        if (string == NULL)
-                                goto out_nomem;
-                        rc = strict_strtoul(string, 10, &option);
-                        kfree(string);
-                        if (rc != 0)
+                        if (nfs_get_option_ul(args, &option))
                                 goto out_invalid_value;
                         mnt->namlen = option;
                         break;
                 case Opt_mountport:
-                        string = match_strdup(args);
-                        if (string == NULL)
-                                goto out_nomem;
-                        rc = strict_strtoul(string, 10, &option);
-                        kfree(string);
-                        if (rc != 0 || option > USHRT_MAX)
+                        if (nfs_get_option_ul(args, &option) ||
+                            option > USHRT_MAX)
                                 goto out_invalid_value;
                         mnt->mount_server.port = option;
                         break;
                 case Opt_mountvers:
-                        string = match_strdup(args);
-                        if (string == NULL)
-                                goto out_nomem;
-                        rc = strict_strtoul(string, 10, &option);
-                        kfree(string);
-                        if (rc != 0 ||
+                        if (nfs_get_option_ul(args, &option) ||
                             option < NFS_MNT_VERSION ||
                             option > NFS_MNT3_VERSION)
                                 goto out_invalid_value;
                         mnt->mount_server.version = option;
                         break;
                 case Opt_nfsvers:
-                        string = match_strdup(args);
-                        if (string == NULL)
-                                goto out_nomem;
-                        rc = strict_strtoul(string, 10, &option);
-                        kfree(string);
-                        if (rc != 0)
+                        if (nfs_get_option_ul(args, &option))
                                 goto out_invalid_value;
                         switch (option) {
                         case NFS2_VERSION:
@@ -1295,12 +1273,7 @@ static int nfs_parse_mount_options(char *raw,
                         }
                         break;
                 case Opt_minorversion:
-                        string = match_strdup(args);
-                        if (string == NULL)
-                                goto out_nomem;
-                        rc = strict_strtoul(string, 10, &option);
-                        kfree(string);
-                        if (rc != 0)
+                        if (nfs_get_option_ul(args, &option))
                                 goto out_invalid_value;
                         if (option > NFS4_MAX_MINOR_VERSION)
                                 goto out_invalid_value;
@@ -1336,21 +1309,18 @@ static int nfs_parse_mount_options(char *raw,
                         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;
-                                kfree(string);
                                 break;
                         case Opt_xprt_rdma:
                                 /* vector side protocols to TCP */
                                 mnt->flags |= NFS_MOUNT_TCP;
                                 mnt->nfs_server.protocol = XPRT_TRANSPORT_RDMA;
                                 xprt_load_transport(string);
-                                kfree(string);
                                 break;
                         default:
                                 dfprintk(MOUNT, "NFS:   unrecognized "
@@ -1358,6 +1328,7 @@ static int nfs_parse_mount_options(char *raw,
                                 kfree(string);
                                 return 0;
                         }
+                        kfree(string);
                         break;
                 case Opt_mountproto:
                         string = match_strdup(args);
@@ -1400,18 +1371,13 @@ static int nfs_parse_mount_options(char *raw,
                                 goto out_invalid_address;
                         break;
                 case Opt_clientaddr:
-                        string = match_strdup(args);
-                        if (string == NULL)
+                        if (nfs_get_option_str(args, &mnt->client_address))
                                 goto out_nomem;
-                        kfree(mnt->client_address);
-                        mnt->client_address = string;
                         break;
                 case Opt_mounthost:
-                        string = match_strdup(args);
-                        if (string == NULL)
+                        if (nfs_get_option_str(args,
+                                               &mnt->mount_server.hostname))
                                 goto out_nomem;
-                        kfree(mnt->mount_server.hostname);
-                        mnt->mount_server.hostname = string;
                         break;
                 case Opt_mountaddr:
                         string = match_strdup(args);
@@ -1451,11 +1417,8 @@ static int nfs_parse_mount_options(char *raw,
                         };
                         break;
                 case Opt_fscache_uniq:
-                        string = match_strdup(args);
-                        if (string == NULL)
+                        if (nfs_get_option_str(args, &mnt->fscache_uniq))
                                 goto out_nomem;
-                        kfree(mnt->fscache_uniq);
-                        mnt->fscache_uniq = string;
                         mnt->options |= NFS_OPTION_FSCACHE;
                         break;
                 case Opt_local_lock:
@@ -1665,99 +1628,59 @@ static int nfs_try_mount(struct nfs_parsed_mount_data *args,
         return nfs_walk_authlist(args, &request);
 }
 
-static int nfs_parse_simple_hostname(const char *dev_name,
-                                     char **hostname, size_t maxnamlen,
-                                     char **export_path, size_t maxpathlen)
+/*
+ * Split "dev_name" into "hostname:export_path".
+ *
+ * The leftmost colon demarks the split between the server's hostname
+ * and the export path.  If the hostname starts with a left square
+ * bracket, then it may contain colons.
+ *
+ * Note: caller frees hostname and export path, even on error.
+ */
+static int nfs_parse_devname(const char *dev_name,
+                             char **hostname, size_t maxnamlen,
+                             char **export_path, size_t maxpathlen)
 {
         size_t len;
-        char *colon, *comma;
-
-        colon = strchr(dev_name, ':');
-        if (colon == NULL)
-                goto out_bad_devname;
-
-        len = colon - dev_name;
-        if (len > maxnamlen)
-                goto out_hostname;
-
-        /* N.B. caller will free nfs_server.hostname in all cases */
-        *hostname = kstrndup(dev_name, len, GFP_KERNEL);
-        if (!*hostname)
-                goto out_nomem;
+        char *end;
 
-        /* kill possible hostname list: not supported */
-        comma = strchr(*hostname, ',');
-        if (comma != NULL) {
-                if (comma == *hostname)
+        /* Is the host name protected with square brakcets? */
+        if (*dev_name == '[') {
+                end = strchr(++dev_name, ']');
+                if (end == NULL || end[1] != ':')
                         goto out_bad_devname;
-                *comma = '\0';
-        }
-
-        colon++;
-        len = strlen(colon);
-        if (len > maxpathlen)
-                goto out_path;
-        *export_path = kstrndup(colon, len, GFP_KERNEL);
-        if (!*export_path)
-                goto out_nomem;
-
-        dfprintk(MOUNT, "NFS: MNTPATH: '%s'\n", *export_path);
-        return 0;
-
-out_bad_devname:
-        dfprintk(MOUNT, "NFS: device name not in host:path format\n");
-        return -EINVAL;
 
-out_nomem:
-        dfprintk(MOUNT, "NFS: not enough memory to parse device name\n");
-        return -ENOMEM;
-
-out_hostname:
-        dfprintk(MOUNT, "NFS: server hostname too long\n");
-        return -ENAMETOOLONG;
-
-out_path:
-        dfprintk(MOUNT, "NFS: export pathname too long\n");
-        return -ENAMETOOLONG;
-}
-
-/*
- * Hostname has square brackets around it because it contains one or
- * more colons.  We look for the first closing square bracket, and a
- * colon must follow it.
- */
-static int nfs_parse_protected_hostname(const char *dev_name,
-                                        char **hostname, size_t maxnamlen,
-                                        char **export_path, size_t maxpathlen)
-{
-        size_t len;
-        char *start, *end;
+                len = end - dev_name;
+                end++;
+        } else {
+                char *comma;
 
-        start = (char *)(dev_name + 1);
+                end = strchr(dev_name, ':');
+                if (end == NULL)
+                        goto out_bad_devname;
+                len = end - dev_name;
 
-        end = strchr(start, ']');
-        if (end == NULL)
-                goto out_bad_devname;
-        if (*(end + 1) != ':')
-                goto out_bad_devname;
+                /* kill possible hostname list: not supported */
+                comma = strchr(dev_name, ',');
+                if (comma != NULL && comma < end)
+                        *comma = 0;
+        }
 
-        len = end - start;
         if (len > maxnamlen)
                 goto out_hostname;
 
         /* N.B. caller will free nfs_server.hostname in all cases */
-        *hostname = kstrndup(start, len, GFP_KERNEL);
+        *hostname = kstrndup(dev_name, len, GFP_KERNEL);
         if (*hostname == NULL)
                 goto out_nomem;
-
-        end += 2;
-        len = strlen(end);
+        len = strlen(++end);
         if (len > maxpathlen)
                 goto out_path;
         *export_path = kstrndup(end, len, GFP_KERNEL);
         if (!*export_path)
                 goto out_nomem;
 
+        dfprintk(MOUNT, "NFS: MNTPATH: '%s'\n", *export_path);
         return 0;
 
 out_bad_devname:
@@ -1778,29 +1701,6 @@ out_path:
 }
 
 /*
- * Split "dev_name" into "hostname:export_path".
- *
- * The leftmost colon demarks the split between the server's hostname
- * and the export path.  If the hostname starts with a left square
- * bracket, then it may contain colons.
- *
- * Note: caller frees hostname and export path, even on error.
- */
-static int nfs_parse_devname(const char *dev_name,
-                             char **hostname, size_t maxnamlen,
-                             char **export_path, size_t maxpathlen)
-{
-        if (*dev_name == '[')
-                return nfs_parse_protected_hostname(dev_name,
-                                                    hostname, maxnamlen,
-                                                    export_path, maxpathlen);
-
-        return nfs_parse_simple_hostname(dev_name,
-                                         hostname, maxnamlen,
-                                         export_path, maxpathlen);
-}
-
-/*
  * Validate the NFS2/NFS3 mount data
  * - fills in the mount root filehandle
  *
@@ -2077,6 +1977,15 @@ nfs_remount(struct super_block *sb, int *flags, char *raw_data)
         if (error < 0)
                 goto out;
 
+        /*
+         * noac is a special case. It implies -o sync, but that's not
+         * necessarily reflected in the mtab options. do_remount_sb
+         * will clear MS_SYNCHRONOUS if -o sync wasn't specified in the
+         * remount options, so we have to explicitly reset it.
+         */
+        if (data->flags & NFS_MOUNT_NOAC)
+                *flags |= MS_SYNCHRONOUS;
+
         /* compare new mount options with old ones */
         error = nfs_compare_remount_data(nfss, data);
 out:
@@ -2267,19 +2176,19 @@ static int nfs_bdi_register(struct nfs_server *server)
         return bdi_register_dev(&server->backing_dev_info, server->s_dev);
 }
 
-static int nfs_get_sb(struct file_system_type *fs_type,
-        int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt)
+static struct dentry *nfs_fs_mount(struct file_system_type *fs_type,
+        int flags, const char *dev_name, void *raw_data)
 {
         struct nfs_server *server = NULL;
         struct super_block *s;
         struct nfs_parsed_mount_data *data;
         struct nfs_fh *mntfh;
-        struct dentry *mntroot;
+        struct dentry *mntroot = ERR_PTR(-ENOMEM);
         int (*compare_super)(struct super_block *, void *) = nfs_compare_super;
         struct nfs_sb_mountdata sb_mntdata = {
                 .mntflags = flags,
         };
-        int error = -ENOMEM;
+        int error;
 
         data = nfs_alloc_parsed_mount_data(NFS_DEFAULT_VERSION);
         mntfh = nfs_alloc_fhandle();
@@ -2290,12 +2199,14 @@ static int nfs_get_sb(struct file_system_type *fs_type,
 
         /* Validate the mount data */
         error = nfs_validate_mount_data(raw_data, data, mntfh, dev_name);
-        if (error < 0)
+        if (error < 0) {
+                mntroot = ERR_PTR(error);
                 goto out;
+        }
 
 #ifdef CONFIG_NFS_V4
         if (data->version == 4) {
-                error = nfs4_try_mount(flags, dev_name, data, mnt);
+                mntroot = nfs4_try_mount(flags, dev_name, data);
                 kfree(data->client_address);
                 kfree(data->nfs_server.export_path);
                 goto out;
@@ -2305,7 +2216,7 @@ static int nfs_get_sb(struct file_system_type *fs_type,
         /* Get a volume representation */
         server = nfs_create_server(data, mntfh);
         if (IS_ERR(server)) {
-                error = PTR_ERR(server);
+                mntroot = ERR_CAST(server);
                 goto out;
         }
         sb_mntdata.server = server;
@@ -2316,7 +2227,7 @@ static int nfs_get_sb(struct file_system_type *fs_type,
         /* Get a superblock - note that we may end up sharing one that already exists */
         s = sget(fs_type, compare_super, nfs_set_super, &sb_mntdata);
         if (IS_ERR(s)) {
-                error = PTR_ERR(s);
+                mntroot = ERR_CAST(s);
                 goto out_err_nosb;
         }
 
@@ -2325,31 +2236,27 @@ static int nfs_get_sb(struct file_system_type *fs_type,
                 server = NULL;
         } else {
                 error = nfs_bdi_register(server);
-                if (error)
+                if (error) {
+                        mntroot = ERR_PTR(error);
                         goto error_splat_bdi;
+                }
         }
 
         if (!s->s_root) {
                 /* initial superblock/root creation */
                 nfs_fill_super(s, data);
-                nfs_fscache_get_super_cookie(
-                        s, data ? data->fscache_uniq : NULL, NULL);
+                nfs_fscache_get_super_cookie(s, data->fscache_uniq, NULL);
         }
 
-        mntroot = nfs_get_root(s, mntfh);
-        if (IS_ERR(mntroot)) {
-                error = PTR_ERR(mntroot);
+        mntroot = nfs_get_root(s, mntfh, dev_name);
+        if (IS_ERR(mntroot))
                 goto error_splat_super;
-        }
 
         error = security_sb_set_mnt_opts(s, &data->lsm_opts);
         if (error)
                 goto error_splat_root;
 
         s->s_flags |= MS_ACTIVE;
-        mnt->mnt_sb = s;
-        mnt->mnt_root = mntroot;
-        error = 0;
 
 out:
         kfree(data->nfs_server.hostname);
@@ -2359,7 +2266,7 @@ out:
 out_free_fh:
         nfs_free_fhandle(mntfh);
         kfree(data);
-        return error;
+        return mntroot;
 
 out_err_nosb:
         nfs_free_server(server);
@@ -2367,6 +2274,7 @@ out_err_nosb:
 
 error_splat_root:
         dput(mntroot);
+        mntroot = ERR_PTR(error);
 error_splat_super:
         if (server && !s->s_root)
                 bdi_unregister(&server->backing_dev_info);
@@ -2450,7 +2358,7 @@ nfs_xdev_mount(struct file_system_type *fs_type, int flags,
                 nfs_fscache_get_super_cookie(s, NULL, data);
         }
 
-        mntroot = nfs_get_root(s, data->fh);
+        mntroot = nfs_get_root(s, data->fh, dev_name);
         if (IS_ERR(mntroot)) {
                 error = PTR_ERR(mntroot);
                 goto error_splat_super;
@@ -2718,7 +2626,7 @@ nfs4_remote_mount(struct file_system_type *fs_type, int flags,
                         s, data ? data->fscache_uniq : NULL, NULL);
         }
 
-        mntroot = nfs4_get_root(s, mntfh);
+        mntroot = nfs4_get_root(s, mntfh, dev_name);
         if (IS_ERR(mntroot)) {
                 error = PTR_ERR(mntroot);
                 goto error_splat_super;
@@ -2771,27 +2679,6 @@ static struct vfsmount *nfs_do_root_mount(struct file_system_type *fs_type,
         return root_mnt;
 }
 
-static void nfs_fix_devname(const struct path *path, struct vfsmount *mnt)
-{
-        char *page = (char *) __get_free_page(GFP_KERNEL);
-        char *devname, *tmp;
-
-        if (page == NULL)
-                return;
-        devname = nfs_path(path->mnt->mnt_devname,
-                        path->mnt->mnt_root, path->dentry,
-                        page, PAGE_SIZE);
-        if (IS_ERR(devname))
-                goto out_freepage;
-        tmp = kstrdup(devname, GFP_KERNEL);
-        if (tmp == NULL)
-                goto out_freepage;
-        kfree(mnt->mnt_devname);
-        mnt->mnt_devname = tmp;
-out_freepage:
-        free_page((unsigned long)page);
-}
-
 struct nfs_referral_count {
         struct list_head list;
         const struct task_struct *task;
@@ -2858,17 +2745,18 @@ static void nfs_referral_loop_unprotect(void)
         kfree(p);
 }
 
-static int nfs_follow_remote_path(struct vfsmount *root_mnt,
-                const char *export_path, struct vfsmount *mnt_target)
+static struct dentry *nfs_follow_remote_path(struct vfsmount *root_mnt,
+                const char *export_path)
 {
         struct nameidata *nd = NULL;
         struct mnt_namespace *ns_private;
         struct super_block *s;
+        struct dentry *dentry;
         int ret;
 
         nd = kmalloc(sizeof(*nd), GFP_KERNEL);
         if (nd == NULL)
-                return -ENOMEM;
+                return ERR_PTR(-ENOMEM);
 
         ns_private = create_mnt_ns(root_mnt);
         ret = PTR_ERR(ns_private);
@@ -2890,32 +2778,27 @@ static int nfs_follow_remote_path(struct vfsmount *root_mnt,
 
         s = nd->path.mnt->mnt_sb;
         atomic_inc(&s->s_active);
-        mnt_target->mnt_sb = s;
-        mnt_target->mnt_root = dget(nd->path.dentry);
-
-        /* Correct the device pathname */
-        nfs_fix_devname(&nd->path, mnt_target);
+        dentry = dget(nd->path.dentry);
 
         path_put(&nd->path);
         kfree(nd);
         down_write(&s->s_umount);
-        return 0;
+        return dentry;
 out_put_mnt_ns:
         put_mnt_ns(ns_private);
 out_mntput:
         mntput(root_mnt);
 out_err:
         kfree(nd);
-        return ret;
+        return ERR_PTR(ret);
 }
 
-static int nfs4_try_mount(int flags, const char *dev_name,
-                         struct nfs_parsed_mount_data *data,
-                         struct vfsmount *mnt)
+static struct dentry *nfs4_try_mount(int flags, const char *dev_name,
+                         struct nfs_parsed_mount_data *data)
 {
         char *export_path;
         struct vfsmount *root_mnt;
-        int error;
+        struct dentry *res;
 
         dfprintk(MOUNT, "--> nfs4_try_mount()\n");
 
@@ -2925,26 +2808,25 @@ static int nfs4_try_mount(int flags, const char *dev_name,
                         data->nfs_server.hostname);
         data->nfs_server.export_path = export_path;
 
-        error = PTR_ERR(root_mnt);
-        if (IS_ERR(root_mnt))
-                goto out;
-
-        error = nfs_follow_remote_path(root_mnt, export_path, mnt);
+        res = ERR_CAST(root_mnt);
+        if (!IS_ERR(root_mnt))
+                res = nfs_follow_remote_path(root_mnt, export_path);
 
-out:
-        dfprintk(MOUNT, "<-- nfs4_try_mount() = %d%s\n", error,
-                        error != 0 ? " [error]" : "");
-        return error;
+        dfprintk(MOUNT, "<-- nfs4_try_mount() = %ld%s\n",
+                        IS_ERR(res) ? PTR_ERR(res) : 0,
+                        IS_ERR(res) ? " [error]" : "");
+        return res;
 }
 
 /*
  * Get the superblock for an NFS4 mountpoint
  */
-static int nfs4_get_sb(struct file_system_type *fs_type,
-        int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt)
+static struct dentry *nfs4_mount(struct file_system_type *fs_type,
+        int flags, const char *dev_name, void *raw_data)
 {
         struct nfs_parsed_mount_data *data;
         int error = -ENOMEM;
+        struct dentry *res = ERR_PTR(-ENOMEM);
 
         data = nfs_alloc_parsed_mount_data(4);
         if (data == NULL)
@@ -2952,10 +2834,14 @@ static int nfs4_get_sb(struct file_system_type *fs_type,
 
         /* Validate the mount data */
         error = nfs4_validate_mount_data(raw_data, data, dev_name);
-        if (error < 0)
+        if (error < 0) {
+                res = ERR_PTR(error);
                 goto out;
+        }
 
-        error = nfs4_try_mount(flags, dev_name, data, mnt);
+        res = nfs4_try_mount(flags, dev_name, data);
+        if (IS_ERR(res))
+                error = PTR_ERR(res);
 
 out:
         kfree(data->client_address);
@@ -2964,9 +2850,9 @@ out:
         kfree(data->fscache_uniq);
 out_free_data:
         kfree(data);
-        dprintk("<-- nfs4_get_sb() = %d%s\n", error,
+        dprintk("<-- nfs4_mount() = %d%s\n", error,
                         error != 0 ? " [error]" : "");
-        return error;
+        return res;
 }
 
 static void nfs4_kill_super(struct super_block *sb)
@@ -3033,7 +2919,7 @@ nfs4_xdev_mount(struct file_system_type *fs_type, int flags,
                 nfs_fscache_get_super_cookie(s, NULL, data);
         }
 
-        mntroot = nfs4_get_root(s, data->fh);
+        mntroot = nfs4_get_root(s, data->fh, dev_name);
         if (IS_ERR(mntroot)) {
                 error = PTR_ERR(mntroot);
                 goto error_splat_super;
@@ -3120,7 +3006,7 @@ nfs4_remote_referral_mount(struct file_system_type *fs_type, int flags,
                 nfs_fscache_get_super_cookie(s, NULL, data);
         }
 
-        mntroot = nfs4_get_root(s, mntfh);
+        mntroot = nfs4_get_root(s, mntfh, dev_name);
         if (IS_ERR(mntroot)) {
                 error = PTR_ERR(mntroot);
                 goto error_splat_super;
@@ -3160,16 +3046,15 @@ error_splat_bdi:
 /*
  * Create an NFS4 server record on referral traversal
  */
-static int nfs4_referral_get_sb(struct file_system_type *fs_type,
-                int flags, const char *dev_name, void *raw_data,
-                struct vfsmount *mnt)
+static struct dentry *nfs4_referral_mount(struct file_system_type *fs_type,
+                int flags, const char *dev_name, void *raw_data)
 {
         struct nfs_clone_mount *data = raw_data;
         char *export_path;
         struct vfsmount *root_mnt;
-        int error;
+        struct dentry *res;
 
-        dprintk("--> nfs4_referral_get_sb()\n");
+        dprintk("--> nfs4_referral_mount()\n");
 
         export_path = data->mnt_path;
         data->mnt_path = "/";
@@ -3178,15 +3063,13 @@ static int nfs4_referral_get_sb(struct file_system_type *fs_type,
                         flags, data, data->hostname);
         data->mnt_path = export_path;
 
-        error = PTR_ERR(root_mnt);
-        if (IS_ERR(root_mnt))
-                goto out;
-
-        error = nfs_follow_remote_path(root_mnt, export_path, mnt);
-out:
-        dprintk("<-- nfs4_referral_get_sb() = %d%s\n", error,
-                        error != 0 ? " [error]" : "");
-        return error;
+        res = ERR_CAST(root_mnt);
+        if (!IS_ERR(root_mnt))
+                res = nfs_follow_remote_path(root_mnt, export_path);
+        dprintk("<-- nfs4_referral_mount() = %ld%s\n",
+                        IS_ERR(res) ? PTR_ERR(res) : 0,
+                        IS_ERR(res) ? " [error]" : "");
+        return res;
 }
 
 #endif /* CONFIG_NFS_V4 */
diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c
index e313a51acdd1..8d6864c2a5fa 100644
--- a/fs/nfs/unlink.c
+++ b/fs/nfs/unlink.c
@@ -148,6 +148,7 @@ static int nfs_do_call_unlink(struct dentry *parent, struct inode *dir, struct n
         alias = d_lookup(parent, &data->args.name);
         if (alias != NULL) {
                 int ret = 0;
+                void *devname_garbage = NULL;
 
                 /*
                  * Hey, we raced with lookup... See if we need to transfer
@@ -157,6 +158,7 @@ static int nfs_do_call_unlink(struct dentry *parent, struct inode *dir, struct n
                 spin_lock(&alias->d_lock);
                 if (alias->d_inode != NULL &&
                     !(alias->d_flags & DCACHE_NFSFS_RENAMED)) {
+                        devname_garbage = alias->d_fsdata;
                         alias->d_fsdata = data;
                         alias->d_flags |= DCACHE_NFSFS_RENAMED;
                         ret = 1;
@@ -164,6 +166,13 @@ static int nfs_do_call_unlink(struct dentry *parent, struct inode *dir, struct n
                 spin_unlock(&alias->d_lock);
                 nfs_dec_sillycount(dir);
                 dput(alias);
+                /*
+                 * If we'd displaced old cached devname, free it.  At that
+                 * point dentry is definitely not a root, so we won't need
+                 * that anymore.
+                 */
+                if (devname_garbage)
+                        kfree(devname_garbage);
                 return ret;
         }
         data->dir = igrab(dir);
@@ -180,7 +189,7 @@ static int nfs_do_call_unlink(struct dentry *parent, struct inode *dir, struct n
         task_setup_data.rpc_client = NFS_CLIENT(dir);
         task = rpc_run_task(&task_setup_data);
         if (!IS_ERR(task))
-                rpc_put_task(task);
+                rpc_put_task_async(task);
         return 1;
 }
 
@@ -252,6 +261,7 @@ nfs_async_unlink(struct inode *dir, struct dentry *dentry)
 {
         struct nfs_unlinkdata *data;
         int status = -ENOMEM;
+        void *devname_garbage = NULL;
 
         data = kzalloc(sizeof(*data), GFP_KERNEL);
         if (data == NULL)
@@ -269,8 +279,16 @@ nfs_async_unlink(struct inode *dir, struct dentry *dentry)
         if (dentry->d_flags & DCACHE_NFSFS_RENAMED)
                 goto out_unlock;
         dentry->d_flags |= DCACHE_NFSFS_RENAMED;
+        devname_garbage = dentry->d_fsdata;
         dentry->d_fsdata = data;
         spin_unlock(&dentry->d_lock);
+        /*
+         * If we'd displaced old cached devname, free it.  At that
+         * point dentry is definitely not a root, so we won't need
+         * that anymore.
+         */
+        if (devname_garbage)
+                kfree(devname_garbage);
         return 0;
 out_unlock:
         spin_unlock(&dentry->d_lock);
@@ -299,6 +317,7 @@ nfs_complete_unlink(struct dentry *dentry, struct inode *inode)
         if (dentry->d_flags & DCACHE_NFSFS_RENAMED) {
                 dentry->d_flags &= ~DCACHE_NFSFS_RENAMED;
                 data = dentry->d_fsdata;
+                dentry->d_fsdata = NULL;
         }
         spin_unlock(&dentry->d_lock);
 
@@ -315,6 +334,7 @@ nfs_cancel_async_unlink(struct dentry *dentry)
                 struct nfs_unlinkdata *data = dentry->d_fsdata;
 
                 dentry->d_flags &= ~DCACHE_NFSFS_RENAMED;
+                dentry->d_fsdata = NULL;
                 spin_unlock(&dentry->d_lock);
                 nfs_free_unlinkdata(data);
                 return;
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index c8278f4046cb..3bd5d7e80f6c 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -28,6 +28,7 @@
 #include "iostat.h"
 #include "nfs4_fs.h"
 #include "fscache.h"
+#include "pnfs.h"
 
 #define NFSDBG_FACILITY         NFSDBG_PAGECACHE
 
@@ -58,6 +59,7 @@ struct nfs_write_data *nfs_commitdata_alloc(void)
         }
         return p;
 }
+EXPORT_SYMBOL_GPL(nfs_commitdata_alloc);
 
 void nfs_commit_free(struct nfs_write_data *p)
 {
@@ -65,6 +67,7 @@ void nfs_commit_free(struct nfs_write_data *p)
                 kfree(p->pagevec);
         mempool_free(p, nfs_commit_mempool);
 }
+EXPORT_SYMBOL_GPL(nfs_commit_free);
 
 struct nfs_write_data *nfs_writedata_alloc(unsigned int pagecount)
 {
@@ -96,6 +99,7 @@ void nfs_writedata_free(struct nfs_write_data *p)
 
 static void nfs_writedata_release(struct nfs_write_data *wdata)
 {
+        put_lseg(wdata->lseg);
         put_nfs_open_context(wdata->args.context);
         nfs_writedata_free(wdata);
 }
@@ -177,8 +181,8 @@ static int wb_priority(struct writeback_control *wbc)
         if (wbc->for_reclaim)
                 return FLUSH_HIGHPRI | FLUSH_STABLE;
         if (wbc->for_kupdate || wbc->for_background)
-                return FLUSH_LOWPRI;
-        return 0;
+                return FLUSH_LOWPRI | FLUSH_COND_STABLE;
+        return FLUSH_COND_STABLE;
 }
 
 /*
@@ -385,11 +389,8 @@ static int nfs_inode_add_request(struct inode *inode, struct nfs_page *req)
         spin_lock(&inode->i_lock);
         error = radix_tree_insert(&nfsi->nfs_page_tree, req->wb_index, req);
         BUG_ON(error);
-        if (!nfsi->npages) {
-                igrab(inode);
-                if (nfs_have_delegation(inode, FMODE_WRITE))
-                        nfsi->change_attr++;
-        }
+        if (!nfsi->npages && nfs_have_delegation(inode, FMODE_WRITE))
+                nfsi->change_attr++;
         set_bit(PG_MAPPED, &req->wb_flags);
         SetPagePrivate(req->wb_page);
         set_page_private(req->wb_page, (unsigned long)req);
@@ -419,11 +420,7 @@ static void nfs_inode_remove_request(struct nfs_page *req)
         clear_bit(PG_MAPPED, &req->wb_flags);
         radix_tree_delete(&nfsi->nfs_page_tree, req->wb_index);
         nfsi->npages--;
-        if (!nfsi->npages) {
-                spin_unlock(&inode->i_lock);
-                iput(inode);
-        } else
-                spin_unlock(&inode->i_lock);
+        spin_unlock(&inode->i_lock);
         nfs_release_request(req);
 }
 
@@ -439,7 +436,7 @@ nfs_mark_request_dirty(struct nfs_page *req)
  * Add a request to the inode's commit list.
  */
 static void
-nfs_mark_request_commit(struct nfs_page *req)
+nfs_mark_request_commit(struct nfs_page *req, struct pnfs_layout_segment *lseg)
 {
         struct inode *inode = req->wb_context->path.dentry->d_inode;
         struct nfs_inode *nfsi = NFS_I(inode);
@@ -451,6 +448,7 @@ nfs_mark_request_commit(struct nfs_page *req)
                         NFS_PAGE_TAG_COMMIT);
         nfsi->ncommit++;
         spin_unlock(&inode->i_lock);
+        pnfs_mark_request_commit(req, lseg);
         inc_zone_page_state(req->wb_page, NR_UNSTABLE_NFS);
         inc_bdi_stat(req->wb_page->mapping->backing_dev_info, BDI_RECLAIMABLE);
         __mark_inode_dirty(inode, I_DIRTY_DATASYNC);
@@ -472,14 +470,18 @@ nfs_clear_request_commit(struct nfs_page *req)
 static inline
 int nfs_write_need_commit(struct nfs_write_data *data)
 {
-        return data->verf.committed != NFS_FILE_SYNC;
+        if (data->verf.committed == NFS_DATA_SYNC)
+                return data->lseg == NULL;
+        else
+                return data->verf.committed != NFS_FILE_SYNC;
 }
 
 static inline
-int nfs_reschedule_unstable_write(struct nfs_page *req)
+int nfs_reschedule_unstable_write(struct nfs_page *req,
+                                  struct nfs_write_data *data)
 {
         if (test_and_clear_bit(PG_NEED_COMMIT, &req->wb_flags)) {
-                nfs_mark_request_commit(req);
+                nfs_mark_request_commit(req, data->lseg);
                 return 1;
         }
         if (test_and_clear_bit(PG_NEED_RESCHED, &req->wb_flags)) {
@@ -490,7 +492,7 @@ int nfs_reschedule_unstable_write(struct nfs_page *req)
 }
 #else
 static inline void
-nfs_mark_request_commit(struct nfs_page *req)
+nfs_mark_request_commit(struct nfs_page *req, struct pnfs_layout_segment *lseg)
 {
 }
 
@@ -507,7 +509,8 @@ int nfs_write_need_commit(struct nfs_write_data *data)
 }
 
 static inline
-int nfs_reschedule_unstable_write(struct nfs_page *req)
+int nfs_reschedule_unstable_write(struct nfs_page *req,
+                                  struct nfs_write_data *data)
 {
         return 0;
 }
@@ -539,11 +542,15 @@ nfs_scan_commit(struct inode *inode, struct list_head *dst, pgoff_t idx_start, u
         if (!nfs_need_commit(nfsi))
                 return 0;
 
+        spin_lock(&inode->i_lock);
         ret = nfs_scan_list(nfsi, dst, idx_start, npages, NFS_PAGE_TAG_COMMIT);
         if (ret > 0)
                 nfsi->ncommit -= ret;
+        spin_unlock(&inode->i_lock);
+
         if (nfs_need_commit(NFS_I(inode)))
                 __mark_inode_dirty(inode, I_DIRTY_DATASYNC);
+
         return ret;
 }
 #else
@@ -610,9 +617,11 @@ static struct nfs_page *nfs_try_to_update_request(struct inode *inode,
         }
 
         if (nfs_clear_request_commit(req) &&
-                        radix_tree_tag_clear(&NFS_I(inode)->nfs_page_tree,
-                                req->wb_index, NFS_PAGE_TAG_COMMIT) != NULL)
+            radix_tree_tag_clear(&NFS_I(inode)->nfs_page_tree,
+                                 req->wb_index, NFS_PAGE_TAG_COMMIT) != NULL) {
                 NFS_I(inode)->ncommit--;
+                pnfs_clear_request_commit(req);
+        }
 
         /* Okay, the request matches. Update the region */
         if (offset < req->wb_offset) {
@@ -671,7 +680,6 @@ static int nfs_writepage_setup(struct nfs_open_context *ctx, struct page *page,
         req = nfs_setup_write_request(ctx, page, offset, count);
         if (IS_ERR(req))
                 return PTR_ERR(req);
-        nfs_mark_request_dirty(req);
         /* Update file length */
         nfs_grow_file(page, offset, count);
         nfs_mark_uptodate(page, req->wb_pgbase, req->wb_bytes);
@@ -760,11 +768,12 @@ int nfs_updatepage(struct file *file, struct page *page,
         return status;
 }
 
-static void nfs_writepage_release(struct nfs_page *req)
+static void nfs_writepage_release(struct nfs_page *req,
+                                  struct nfs_write_data *data)
 {
         struct page *page = req->wb_page;
 
-        if (PageError(req->wb_page) || !nfs_reschedule_unstable_write(req))
+        if (PageError(req->wb_page) || !nfs_reschedule_unstable_write(req, data))
                 nfs_inode_remove_request(req);
         nfs_clear_page_tag_locked(req);
         nfs_end_page_writeback(page);
@@ -781,25 +790,21 @@ static int flush_task_priority(int how)
         return RPC_PRIORITY_NORMAL;
 }
 
-/*
- * Set up the argument/result storage required for the RPC call.
- */
-static int nfs_write_rpcsetup(struct nfs_page *req,
-                struct nfs_write_data *data,
-                const struct rpc_call_ops *call_ops,
-                unsigned int count, unsigned int offset,
-                int how)
+int nfs_initiate_write(struct nfs_write_data *data,
+                       struct rpc_clnt *clnt,
+                       const struct rpc_call_ops *call_ops,
+                       int how)
 {
-        struct inode *inode = req->wb_context->path.dentry->d_inode;
+        struct inode *inode = data->inode;
         int priority = flush_task_priority(how);
         struct rpc_task *task;
         struct rpc_message msg = {
                 .rpc_argp = &data->args,
                 .rpc_resp = &data->res,
-                .rpc_cred = req->wb_context->cred,
+                .rpc_cred = data->cred,
         };
         struct rpc_task_setup task_setup_data = {
-                .rpc_client = NFS_CLIENT(inode),
+                .rpc_client = clnt,
                 .task = &data->task,
                 .rpc_message = &msg,
                 .callback_ops = call_ops,
@@ -810,12 +815,52 @@ static int nfs_write_rpcsetup(struct nfs_page *req,
         };
         int ret = 0;
 
+        /* Set up the initial task struct.  */
+        NFS_PROTO(inode)->write_setup(data, &msg);
+
+        dprintk("NFS: %5u initiated write call "
+                "(req %s/%lld, %u bytes @ offset %llu)\n",
+                data->task.tk_pid,
+                inode->i_sb->s_id,
+                (long long)NFS_FILEID(inode),
+                data->args.count,
+                (unsigned long long)data->args.offset);
+
+        task = rpc_run_task(&task_setup_data);
+        if (IS_ERR(task)) {
+                ret = PTR_ERR(task);
+                goto out;
+        }
+        if (how & FLUSH_SYNC) {
+                ret = rpc_wait_for_completion_task(task);
+                if (ret == 0)
+                        ret = task->tk_status;
+        }
+        rpc_put_task(task);
+out:
+        return ret;
+}
+EXPORT_SYMBOL_GPL(nfs_initiate_write);
+
+/*
+ * Set up the argument/result storage required for the RPC call.
+ */
+static int nfs_write_rpcsetup(struct nfs_page *req,
+                struct nfs_write_data *data,
+                const struct rpc_call_ops *call_ops,
+                unsigned int count, unsigned int offset,
+                struct pnfs_layout_segment *lseg,
+                int how)
+{
+        struct inode *inode = req->wb_context->path.dentry->d_inode;
+
         /* Set up the RPC argument and reply structs
          * NB: take care not to mess about with data->commit et al. */
 
         data->req = req;
         data->inode = inode = req->wb_context->path.dentry->d_inode;
-        data->cred = msg.rpc_cred;
+        data->cred = req->wb_context->cred;
+        data->lseg = get_lseg(lseg);
 
         data->args.fh     = NFS_FH(inode);
         data->args.offset = req_offset(req) + offset;
@@ -825,7 +870,7 @@ static int nfs_write_rpcsetup(struct nfs_page *req,
         data->args.context = get_nfs_open_context(req->wb_context);
         data->args.lock_context = req->wb_lock_context;
         data->args.stable  = NFS_UNSTABLE;
-        if (how & FLUSH_STABLE) {
+        if (how & (FLUSH_STABLE | FLUSH_COND_STABLE)) {
                 data->args.stable = NFS_DATA_SYNC;
                 if (!nfs_need_commit(NFS_I(inode)))
                         data->args.stable = NFS_FILE_SYNC;
@@ -836,30 +881,11 @@ static int nfs_write_rpcsetup(struct nfs_page *req,
         data->res.verf    = &data->verf;
         nfs_fattr_init(&data->fattr);
 
-        /* Set up the initial task struct.  */
-        NFS_PROTO(inode)->write_setup(data, &msg);
-
-        dprintk("NFS: %5u initiated write call "
-                "(req %s/%lld, %u bytes @ offset %llu)\n",
-                data->task.tk_pid,
-                inode->i_sb->s_id,
-                (long long)NFS_FILEID(inode),
-                count,
-                (unsigned long long)data->args.offset);
+        if (data->lseg &&
+            (pnfs_try_to_write_data(data, call_ops, how) == PNFS_ATTEMPTED))
+                return 0;
 
-        task = rpc_run_task(&task_setup_data);
-        if (IS_ERR(task)) {
-                ret = PTR_ERR(task);
-                goto out;
-        }
-        if (how & FLUSH_SYNC) {
-                ret = rpc_wait_for_completion_task(task);
-                if (ret == 0)
-                        ret = task->tk_status;
-        }
-        rpc_put_task(task);
-out:
-        return ret;
+        return nfs_initiate_write(data, NFS_CLIENT(inode), call_ops, how);
 }
 
 /* If a nfs_flush_* function fails, it should remove reqs from @head and
@@ -879,20 +905,27 @@ static void nfs_redirty_request(struct nfs_page *req)
  * Generate multiple small requests to write out a single
  * contiguous dirty area on one page.
  */
-static int nfs_flush_multi(struct inode *inode, struct list_head *head, unsigned int npages, size_t count, int how)
+static int nfs_flush_multi(struct nfs_pageio_descriptor *desc)
 {
-        struct nfs_page *req = nfs_list_entry(head->next);
+        struct nfs_page *req = nfs_list_entry(desc->pg_list.next);
         struct page *page = req->wb_page;
         struct nfs_write_data *data;
-        size_t wsize = NFS_SERVER(inode)->wsize, nbytes;
+        size_t wsize = NFS_SERVER(desc->pg_inode)->wsize, nbytes;
         unsigned int offset;
         int requests = 0;
         int ret = 0;
+        struct pnfs_layout_segment *lseg;
         LIST_HEAD(list);
 
         nfs_list_remove_request(req);
 
-        nbytes = count;
+        if ((desc->pg_ioflags & FLUSH_COND_STABLE) &&
+            (desc->pg_moreio || NFS_I(desc->pg_inode)->ncommit ||
+             desc->pg_count > wsize))
+                desc->pg_ioflags &= ~FLUSH_COND_STABLE;
+
+
+        nbytes = desc->pg_count;
         do {
                 size_t len = min(nbytes, wsize);
 
@@ -905,9 +938,11 @@ static int nfs_flush_multi(struct inode *inode, struct list_head *head, unsigned
         } while (nbytes != 0);
         atomic_set(&req->wb_complete, requests);
 
+        BUG_ON(desc->pg_lseg);
+        lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_RW);
         ClearPageError(page);
         offset = 0;
-        nbytes = count;
+        nbytes = desc->pg_count;
         do {
                 int ret2;
 
@@ -919,13 +954,15 @@ static int nfs_flush_multi(struct inode *inode, struct list_head *head, unsigned
                 if (nbytes < wsize)
                         wsize = nbytes;
                 ret2 = nfs_write_rpcsetup(req, data, &nfs_write_partial_ops,
-                                   wsize, offset, how);
+                                          wsize, offset, lseg, desc->pg_ioflags);
                 if (ret == 0)
                         ret = ret2;
                 offset += wsize;
                 nbytes -= wsize;
         } while (nbytes != 0);
 
+        put_lseg(lseg);
+        desc->pg_lseg = NULL;
         return ret;
 
 out_bad:
@@ -946,16 +983,26 @@ out_bad:
  * This is the case if nfs_updatepage detects a conflicting request
  * that has been written but not committed.
  */
-static int nfs_flush_one(struct inode *inode, struct list_head *head, unsigned int npages, size_t count, int how)
+static int nfs_flush_one(struct nfs_pageio_descriptor *desc)
 {
         struct nfs_page         *req;
         struct page             **pages;
         struct nfs_write_data   *data;
+        struct list_head *head = &desc->pg_list;
+        struct pnfs_layout_segment *lseg = desc->pg_lseg;
+        int ret;
 
-        data = nfs_writedata_alloc(npages);
-        if (!data)
-                goto out_bad;
-
+        data = nfs_writedata_alloc(nfs_page_array_len(desc->pg_base,
+                                                      desc->pg_count));
+        if (!data) {
+                while (!list_empty(head)) {
+                        req = nfs_list_entry(head->next);
+                        nfs_list_remove_request(req);
+                        nfs_redirty_request(req);
+                }
+                ret = -ENOMEM;
+                goto out;
+        }
         pages = data->pagevec;
         while (!list_empty(head)) {
                 req = nfs_list_entry(head->next);
@@ -965,16 +1012,19 @@ static int nfs_flush_one(struct inode *inode, struct list_head *head, unsigned i
                 *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, IOMODE_RW);
+
+        if ((desc->pg_ioflags & FLUSH_COND_STABLE) &&
+            (desc->pg_moreio || NFS_I(desc->pg_inode)->ncommit))
+                desc->pg_ioflags &= ~FLUSH_COND_STABLE;
 
         /* Set up the argument struct */
-        return nfs_write_rpcsetup(req, data, &nfs_write_full_ops, count, 0, how);
- out_bad:
-        while (!list_empty(head)) {
-                req = nfs_list_entry(head->next);
-                nfs_list_remove_request(req);
-                nfs_redirty_request(req);
-        }
-        return -ENOMEM;
+        ret = nfs_write_rpcsetup(req, data, &nfs_write_full_ops, desc->pg_count, 0, lseg, desc->pg_ioflags);
+out:
+        put_lseg(lseg); /* Cleans any gotten in ->pg_test */
+        desc->pg_lseg = NULL;
+        return ret;
 }
 
 static void nfs_pageio_init_write(struct nfs_pageio_descriptor *pgio,
@@ -982,6 +1032,8 @@ static void nfs_pageio_init_write(struct nfs_pageio_descriptor *pgio,
 {
         size_t wsize = NFS_SERVER(inode)->wsize;
 
+        pnfs_pageio_init_write(pgio, inode);
+
         if (wsize < PAGE_CACHE_SIZE)
                 nfs_pageio_init(pgio, inode, nfs_flush_multi, wsize, ioflags);
         else
@@ -1039,7 +1091,7 @@ static void nfs_writeback_release_partial(void *calldata)
 
 out:
         if (atomic_dec_and_test(&req->wb_complete))
-                nfs_writepage_release(req);
+                nfs_writepage_release(req, data);
         nfs_writedata_release(calldata);
 }
 
@@ -1106,7 +1158,7 @@ static void nfs_writeback_release_full(void *calldata)
 
                 if (nfs_write_need_commit(data)) {
                         memcpy(&req->wb_verf, &data->verf, sizeof(req->wb_verf));
-                        nfs_mark_request_commit(req);
+                        nfs_mark_request_commit(req, data->lseg);
                         dprintk(" marked for commit\n");
                         goto next;
                 }
@@ -1132,7 +1184,7 @@ static const struct rpc_call_ops nfs_write_full_ops = {
 /*
  * This function is called when the WRITE call is complete.
  */
-int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
+void nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
 {
         struct nfs_writeargs    *argp = &data->args;
         struct nfs_writeres     *resp = &data->res;
@@ -1151,7 +1203,7 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
          */
         status = NFS_PROTO(data->inode)->write_done(task, data);
         if (status != 0)
-                return status;
+                return;
         nfs_add_stats(data->inode, NFSIOS_SERVERWRITTENBYTES, resp->count);
 
 #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
@@ -1166,6 +1218,7 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
                  */
                 static unsigned long    complain;
 
+                /* Note this will print the MDS for a DS write */
                 if (time_before(complain, jiffies)) {
                         dprintk("NFS:       faulty NFS server %s:"
                                 " (committed = %d) != (stable = %d)\n",
@@ -1186,6 +1239,7 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
                         /* Was this an NFSv2 write or an NFSv3 stable write? */
                         if (resp->verf->committed != NFS_UNSTABLE) {
                                 /* Resend from where the server left off */
+                                data->mds_offset += resp->count;
                                 argp->offset += resp->count;
                                 argp->pgbase += resp->count;
                                 argp->count -= resp->count;
@@ -1196,7 +1250,7 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
                                 argp->stable = NFS_FILE_SYNC;
                         }
                         nfs_restart_rpc(task, server->nfs_client);
-                        return -EAGAIN;
+                        return;
                 }
                 if (time_before(complain, jiffies)) {
                         printk(KERN_WARNING
@@ -1207,64 +1261,89 @@ 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;
         }
-        return 0;
+        return;
 }
 
 
 #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
 static int nfs_commit_set_lock(struct nfs_inode *nfsi, int may_wait)
 {
+        int ret;
+
         if (!test_and_set_bit(NFS_INO_COMMIT, &nfsi->flags))
                 return 1;
-        if (may_wait && !out_of_line_wait_on_bit_lock(&nfsi->flags,
-                                NFS_INO_COMMIT, nfs_wait_bit_killable,
-                                TASK_KILLABLE))
-                return 1;
-        return 0;
+        if (!may_wait)
+                return 0;
+        ret = out_of_line_wait_on_bit_lock(&nfsi->flags,
+                                NFS_INO_COMMIT,
+                                nfs_wait_bit_killable,
+                                TASK_KILLABLE);
+        return (ret < 0) ? ret : 1;
 }
 
-static void nfs_commit_clear_lock(struct nfs_inode *nfsi)
+void nfs_commit_clear_lock(struct nfs_inode *nfsi)
 {
         clear_bit(NFS_INO_COMMIT, &nfsi->flags);
         smp_mb__after_clear_bit();
         wake_up_bit(&nfsi->flags, NFS_INO_COMMIT);
 }
+EXPORT_SYMBOL_GPL(nfs_commit_clear_lock);
 
-
-static void nfs_commitdata_release(void *data)
+void nfs_commitdata_release(void *data)
 {
         struct nfs_write_data *wdata = data;
 
+        put_lseg(wdata->lseg);
         put_nfs_open_context(wdata->args.context);
         nfs_commit_free(wdata);
 }
+EXPORT_SYMBOL_GPL(nfs_commitdata_release);
 
-/*
- * Set up the argument/result storage required for the RPC call.
- */
-static int nfs_commit_rpcsetup(struct list_head *head,
-                struct nfs_write_data *data,
-                int how)
+int nfs_initiate_commit(struct nfs_write_data *data, struct rpc_clnt *clnt,
+                        const struct rpc_call_ops *call_ops,
+                        int how)
 {
-        struct nfs_page *first = nfs_list_entry(head->next);
-        struct inode *inode = first->wb_context->path.dentry->d_inode;
-        int priority = flush_task_priority(how);
         struct rpc_task *task;
+        int priority = flush_task_priority(how);
         struct rpc_message msg = {
                 .rpc_argp = &data->args,
                 .rpc_resp = &data->res,
-                .rpc_cred = first->wb_context->cred,
+                .rpc_cred = data->cred,
         };
         struct rpc_task_setup task_setup_data = {
                 .task = &data->task,
-                .rpc_client = NFS_CLIENT(inode),
+                .rpc_client = clnt,
                 .rpc_message = &msg,
-                .callback_ops = &nfs_commit_ops,
+                .callback_ops = call_ops,
                 .callback_data = data,
                 .workqueue = nfsiod_workqueue,
                 .flags = RPC_TASK_ASYNC,
                 .priority = priority,
         };
+        /* Set up the initial task struct.  */
+        NFS_PROTO(data->inode)->commit_setup(data, &msg);
+
+        dprintk("NFS: %5u initiated commit call\n", data->task.tk_pid);
+
+        task = rpc_run_task(&task_setup_data);
+        if (IS_ERR(task))
+                return PTR_ERR(task);
+        if (how & FLUSH_SYNC)
+                rpc_wait_for_completion_task(task);
+        rpc_put_task(task);
+        return 0;
+}
+EXPORT_SYMBOL_GPL(nfs_initiate_commit);
+
+/*
+ * Set up the argument/result storage required for the RPC call.
+ */
+void nfs_init_commit(struct nfs_write_data *data,
+                            struct list_head *head,
+                            struct pnfs_layout_segment *lseg)
+{
+        struct nfs_page *first = nfs_list_entry(head->next);
+        struct inode *inode = first->wb_context->path.dentry->d_inode;
 
         /* Set up the RPC argument and reply structs
          * NB: take care not to mess about with data->commit et al. */
@@ -1272,7 +1351,9 @@ static int nfs_commit_rpcsetup(struct list_head *head,
         list_splice_init(head, &data->pages);
 
         data->inode       = inode;
-        data->cred        = msg.rpc_cred;
+        data->cred        = first->wb_context->cred;
+        data->lseg        = lseg; /* reference transferred */
+        data->mds_ops     = &nfs_commit_ops;
 
         data->args.fh     = NFS_FH(data->inode);
         /* Note: we always request a commit of the entire inode */
@@ -1283,18 +1364,25 @@ static int nfs_commit_rpcsetup(struct list_head *head,
         data->res.fattr   = &data->fattr;
         data->res.verf    = &data->verf;
         nfs_fattr_init(&data->fattr);
+}
+EXPORT_SYMBOL_GPL(nfs_init_commit);
 
-        /* Set up the initial task struct.  */
-        NFS_PROTO(inode)->commit_setup(data, &msg);
-
-        dprintk("NFS: %5u initiated commit call\n", data->task.tk_pid);
+void nfs_retry_commit(struct list_head *page_list,
+                      struct pnfs_layout_segment *lseg)
+{
+        struct nfs_page *req;
 
-        task = rpc_run_task(&task_setup_data);
-        if (IS_ERR(task))
-                return PTR_ERR(task);
-        rpc_put_task(task);
-        return 0;
+        while (!list_empty(page_list)) {
+                req = nfs_list_entry(page_list->next);
+                nfs_list_remove_request(req);
+                nfs_mark_request_commit(req, lseg);
+                dec_zone_page_state(req->wb_page, NR_UNSTABLE_NFS);
+                dec_bdi_stat(req->wb_page->mapping->backing_dev_info,
+                             BDI_RECLAIMABLE);
+                nfs_clear_page_tag_locked(req);
+        }
 }
+EXPORT_SYMBOL_GPL(nfs_retry_commit);
 
 /*
  * Commit dirty pages
@@ -1303,7 +1391,6 @@ static int
 nfs_commit_list(struct inode *inode, struct list_head *head, int how)
 {
         struct nfs_write_data   *data;
-        struct nfs_page         *req;
 
         data = nfs_commitdata_alloc();
 
@@ -1311,17 +1398,10 @@ nfs_commit_list(struct inode *inode, struct list_head *head, int how)
                 goto out_bad;
 
         /* Set up the argument struct */
-        return nfs_commit_rpcsetup(head, data, how);
+        nfs_init_commit(data, head, NULL);
+        return nfs_initiate_commit(data, NFS_CLIENT(inode), data->mds_ops, how);
  out_bad:
-        while (!list_empty(head)) {
-                req = nfs_list_entry(head->next);
-                nfs_list_remove_request(req);
-                nfs_mark_request_commit(req);
-                dec_zone_page_state(req->wb_page, NR_UNSTABLE_NFS);
-                dec_bdi_stat(req->wb_page->mapping->backing_dev_info,
-                                BDI_RECLAIMABLE);
-                nfs_clear_page_tag_locked(req);
-        }
+        nfs_retry_commit(head, NULL);
         nfs_commit_clear_lock(NFS_I(inode));
         return -ENOMEM;
 }
@@ -1337,14 +1417,12 @@ static void nfs_commit_done(struct rpc_task *task, void *calldata)
                                 task->tk_pid, task->tk_status);
 
         /* Call the NFS version-specific code */
-        if (NFS_PROTO(data->inode)->commit_done(task, data) != 0)
-                return;
+        NFS_PROTO(data->inode)->commit_done(task, data);
 }
 
-static void nfs_commit_release(void *calldata)
+void nfs_commit_release_pages(struct nfs_write_data *data)
 {
-        struct nfs_write_data   *data = calldata;
-        struct nfs_page         *req;
+        struct nfs_page *req;
         int status = data->task.tk_status;
 
         while (!list_empty(&data->pages)) {
@@ -1378,6 +1456,14 @@ static void nfs_commit_release(void *calldata)
         next:
                 nfs_clear_page_tag_locked(req);
         }
+}
+EXPORT_SYMBOL_GPL(nfs_commit_release_pages);
+
+static void nfs_commit_release(void *calldata)
+{
+        struct nfs_write_data *data = calldata;
+
+        nfs_commit_release_pages(data);
         nfs_commit_clear_lock(NFS_I(data->inode));
         nfs_commitdata_release(calldata);
 }
@@ -1394,23 +1480,28 @@ int nfs_commit_inode(struct inode *inode, int how)
 {
         LIST_HEAD(head);
         int may_wait = how & FLUSH_SYNC;
-        int res = 0;
+        int res;
 
-        if (!nfs_commit_set_lock(NFS_I(inode), may_wait))
+        res = nfs_commit_set_lock(NFS_I(inode), may_wait);
+        if (res <= 0)
                 goto out_mark_dirty;
-        spin_lock(&inode->i_lock);
         res = nfs_scan_commit(inode, &head, 0, 0);
-        spin_unlock(&inode->i_lock);
         if (res) {
-                int error = nfs_commit_list(inode, &head, how);
+                int error;
+
+                error = pnfs_commit_list(inode, &head, how);
+                if (error == PNFS_NOT_ATTEMPTED)
+                        error = nfs_commit_list(inode, &head, how);
                 if (error < 0)
                         return error;
-                if (may_wait)
-                        wait_on_bit(&NFS_I(inode)->flags, NFS_INO_COMMIT,
-                                        nfs_wait_bit_killable,
-                                        TASK_KILLABLE);
-                else
+                if (!may_wait)
                         goto out_mark_dirty;
+                error = wait_on_bit(&NFS_I(inode)->flags,
+                                NFS_INO_COMMIT,
+                                nfs_wait_bit_killable,
+                                TASK_KILLABLE);
+                if (error < 0)
+                        return error;
         } else
                 nfs_commit_clear_lock(NFS_I(inode));
         return res;
@@ -1464,7 +1555,22 @@ static int nfs_commit_unstable_pages(struct inode *inode, struct writeback_contr
 
 int nfs_write_inode(struct inode *inode, struct writeback_control *wbc)
 {
-        return nfs_commit_unstable_pages(inode, wbc);
+        int ret;
+
+        ret = nfs_commit_unstable_pages(inode, wbc);
+        if (ret >= 0 && test_bit(NFS_INO_LAYOUTCOMMIT, &NFS_I(inode)->flags)) {
+                int status;
+                bool sync = true;
+
+                if (wbc->sync_mode == WB_SYNC_NONE || wbc->nonblocking ||
+                    wbc->for_background)
+                        sync = false;
+
+                status = pnfs_layoutcommit_inode(inode, sync);
+                if (status < 0)
+                        return status;
+        }
+        return ret;
 }
 
 /*