Merge branch 'nfs-for-2.6.39' of git://git.linux-nfs.org/projects/trondmy/nfs-2.6

* 'nfs-for-2.6.39' of git://git.linux-nfs.org/projects/trondmy/nfs-2.6: (28 commits)
  Cleanup XDR parsing for LAYOUTGET, GETDEVICEINFO
  NFSv4.1 convert layoutcommit sync to boolean
  NFSv4.1 pnfs_layoutcommit_inode fixes
  NFS: Determine initial mount security
  NFS: use secinfo when crossing mountpoints
  NFS: Add secinfo procedure
  NFS: lookup supports alternate client
  NFS: convert call_sync() to a function
  NFSv4.1 remove temp code that prevented ds commits
  NFSv4.1: layoutcommit
  NFSv4.1: filelayout driver specific code for COMMIT
  NFSv4.1: remove GETATTR from ds commits
  NFSv4.1: add generic layer hooks for pnfs COMMIT
  NFSv4.1: alloc and free commit_buckets
  NFSv4.1: shift filelayout_free_lseg
  NFSv4.1: pull out code from nfs_commit_release
  NFSv4.1: pull error handling out of nfs_commit_list
  NFSv4.1: add callback to nfs4_commit_done
  NFSv4.1: rearrange nfs_commit_rpcsetup
  NFSv4.1: don't send COMMIT to ds for data sync writes
  ...

26 files changed, 1728 insertions, 256 deletions

diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index abdf38d5971..7237672216c 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))
@@ -1224,7 +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) {
@@ -1562,7 +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/file.c b/fs/nfs/file.c
index d85a534b15c..3ac5bd695e5 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -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;
 }
 
diff --git a/fs/nfs/getroot.c b/fs/nfs/getroot.c
index 1084792bc0f..dcb61548887 100644
--- a/fs/nfs/getroot.c
+++ b/fs/nfs/getroot.c
@@ -222,6 +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");
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 01768e5e2c9..57bb31ad7a5 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -254,7 +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;
@@ -298,8 +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
@@ -639,7 +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);
@@ -1471,6 +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
 }
 
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index 72e0bddf7a2..ce118ce885d 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;
@@ -214,6 +220,7 @@ 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);
@@ -276,11 +283,25 @@ extern int nfs_initiate_read(struct nfs_read_data *data, struct rpc_clnt *clnt,
 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 *);
@@ -296,12 +317,14 @@ extern int nfs4_init_client(struct nfs_client *clp,
 			    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 nfs_server *server,
+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,
diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c
index bf1c68009ff..ad92bf731ff 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
@@ -27,7 +28,8 @@ int nfs_mountpoint_expiry_timeout = 500 * HZ;
 
 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
@@ -116,6 +118,100 @@ Elong:
 	return ERR_PTR(-ENAMETOOLONG);
 }
 
+#ifdef CONFIG_NFS_V4
+static rpc_authflavor_t nfs_find_best_sec(struct nfs4_secinfo_flavors *flavors, struct inode *inode)
+{
+	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 rpc_authflavor_t nfs_negotiate_security(const struct dentry *parent, const struct dentry *dentry)
+{
+	int status = 0;
+	struct page *page;
+	struct nfs4_secinfo_flavors *flavors;
+	int (*secinfo)(struct inode *, const struct qstr *, struct nfs4_secinfo_flavors *);
+	rpc_authflavor_t flavor = RPC_AUTH_UNIX;
+
+	secinfo = NFS_PROTO(parent->d_inode)->secinfo;
+	if (secinfo != NULL) {
+		page = alloc_page(GFP_KERNEL);
+		if (!page) {
+			status = -ENOMEM;
+			goto out;
+		}
+		flavors = page_address(page);
+		status = secinfo(parent->d_inode, &dentry->d_name, flavors);
+		flavor = nfs_find_best_sec(flavors, dentry->d_inode);
+		put_page(page);
+	}
+
+	return flavor;
+
+out:
+	status = -ENOMEM;
+	return status;
+}
+
+static rpc_authflavor_t 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;
+
+	flavor = nfs_negotiate_security(parent, path->dentry);
+	if (flavor < 0)
+		goto out;
+	clone  = rpc_clone_client(server->client);
+	auth   = rpcauth_create(flavor, clone);
+	if (!auth) {
+		flavor = -EIO;
+		goto out;
+	}
+	err = server->nfs_client->rpc_ops->lookup(clone, parent->d_inode,
+						  &path->dentry->d_name,
+						  fh, fattr);
+	if (err < 0)
+		flavor = err;
+out:
+	return flavor;
+}
+#else /* CONFIG_NFS_V4 */
+static inline rpc_authflavor_t 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)
+{
+	return -EPERM;
+}
+#endif /* CONFIG_NFS_V4 */
+
 /*
  * nfs_d_automount - Handle crossing a mountpoint on the server
  * @path - The mountpoint
@@ -136,6 +232,7 @@ struct vfsmount *nfs_d_automount(struct path *path)
 	struct nfs_fh *fh = NULL;
 	struct nfs_fattr *fattr = NULL;
 	int err;
+	rpc_authflavor_t flavor = 1;
 
 	dprintk("--> nfs_d_automount()\n");
 
@@ -153,9 +250,16 @@ 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) {
+		flavor = nfs_lookup_with_sec(server, parent, path->dentry, path, fh, fattr);
+		if (flavor < 0)
+			err = flavor;
+		else
+			err = 0;
+	}
 	dput(parent);
 	if (err != 0) {
 		mnt = ERR_PTR(err);
@@ -165,7 +269,7 @@ struct vfsmount *nfs_d_automount(struct path *path)
 	if (fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL)
 		mnt = nfs_do_refmount(path->dentry);
 	else
-		mnt = nfs_do_submount(path->dentry, fh, fattr);
+		mnt = nfs_do_submount(path->dentry, fh, fattr, flavor);
 	if (IS_ERR(mnt))
 		goto out;
 
@@ -232,17 +336,20 @@ static struct vfsmount *nfs_do_clone_mount(struct nfs_server *server,
  * @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(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 = 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);
diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c
index d0c80d8b3f9..38053d823eb 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 = {
diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index c64be1cff08..e1c261ddd65 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -57,7 +57,8 @@ 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,
@@ -262,6 +263,8 @@ 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)
diff --git a/fs/nfs/nfs4filelayout.c b/fs/nfs/nfs4filelayout.c
index 42855846481..6f8192f4cfc 100644
--- a/fs/nfs/nfs4filelayout.c
+++ b/fs/nfs/nfs4filelayout.c
@@ -154,6 +154,23 @@ static int filelayout_read_done_cb(struct rpc_task *task,
 }
 
 /*
+ * 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)
+{
+	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.
@@ -210,6 +227,38 @@ static int filelayout_write_done_cb(struct rpc_task *task,
 		return -EAGAIN;
 	}
 
+	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;
 }
 
@@ -240,6 +289,16 @@ static void filelayout_write_release(void *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,
@@ -252,6 +311,12 @@ struct rpc_call_ops filelayout_write_call_ops = {
 	.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)
 {
@@ -320,10 +385,6 @@ filelayout_write_pagelist(struct nfs_write_data *data, int sync)
 		data->inode->i_ino, sync, (size_t) data->args.count, offset,
 		ntohl(ds->ds_ip_addr), ntohs(ds->ds_port));
 
-	/* We can't handle commit to ds yet */
-	if (!FILELAYOUT_LSEG(lseg)->commit_through_mds)
-		data->args.stable = NFS_FILE_SYNC;
-
 	data->write_done_cb = filelayout_write_done_cb;
 	data->ds_clp = ds->ds_clp;
 	fh = nfs4_fl_select_ds_fh(lseg, j);
@@ -441,12 +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);
@@ -468,32 +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 *
@@ -514,17 +621,28 @@ filelayout_alloc_lseg(struct pnfs_layout_hdr *layoutid,
 		_filelayout_free_lseg(fl);
 		return NULL;
 	}
-	return &fl->generic_hdr;
-}
 
-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);
-	_filelayout_free_lseg(fl);
+	/* 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;
 }
 
 /*
@@ -552,6 +670,191 @@ filelayout_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev,
 	return (p_stripe == r_stripe);
 }
 
+static bool filelayout_mark_pnfs_commit(struct pnfs_layout_segment *lseg)
+{
+	return !FILELAYOUT_LS