Merge branch 'for-3.4' of git://linux-nfs.org/~bfields/linux

Pull nfsd changes from Bruce Fields:

Highlights:
 - Benny Halevy and Tigran Mkrtchyan implemented some more 4.1 features,
   moving us closer to a complete 4.1 implementation.
 - Bernd Schubert fixed a long-standing problem with readdir cookies on
   ext2/3/4.
 - Jeff Layton performed a long-overdue overhaul of the server reboot
   recovery code which will allow us to deprecate the current code (a
   rather unusual user of the vfs), and give us some needed flexibility
   for further improvements.
 - Like the client, we now support numeric uid's and gid's in the
   auth_sys case, allowing easier upgrades from NFSv2/v3 to v4.x.

Plus miscellaneous bugfixes and cleanup.

Thanks to everyone!

There are also some delegation fixes waiting on vfs review that I
suppose will have to wait for 3.5.  With that done I think we'll finally
turn off the "EXPERIMENTAL" dependency for v4 (though that's mostly
symbolic as it's been on by default in distro's for a while).

And the list of 4.1 todo's should be achievable for 3.5 as well:

   http://wiki.linux-nfs.org/wiki/index.php/Server_4.0_and_4.1_issues

though we may still want a bit more experience with it before turning it
on by default.

* 'for-3.4' of git://linux-nfs.org/~bfields/linux: (55 commits)
  nfsd: only register cld pipe notifier when CONFIG_NFSD_V4 is enabled
  nfsd4: use auth_unix unconditionally on backchannel
  nfsd: fix NULL pointer dereference in cld_pipe_downcall
  nfsd4: memory corruption in numeric_name_to_id()
  sunrpc: skip portmap calls on sessions backchannel
  nfsd4: allow numeric idmapping
  nfsd: don't allow legacy client tracker init for anything but init_net
  nfsd: add notifier to handle mount/unmount of rpc_pipefs sb
  nfsd: add the infrastructure to handle the cld upcall
  nfsd: add a header describing upcall to nfsdcld
  nfsd: add a per-net-namespace struct for nfsd
  sunrpc: create nfsd dir in rpc_pipefs
  nfsd: add nfsd4_client_tracking_ops struct and a way to set it
  nfsd: convert nfs4_client->cl_cb_flags to a generic flags field
  NFSD: Fix nfs4_verifier memory alignment
  NFSD: Fix warnings when NFSD_DEBUG is not defined
  nfsd: vfs_llseek() with 32 or 64 bit offsets (hashes)
  nfsd: rename 'int access' to 'int may_flags' in nfsd_open()
  ext4: return 32/64-bit dir name hash according to usage type
  fs: add new FMODE flags: FMODE_32bithash and FMODE_64bithash
  ...

Documentation/kernel-parameters.txt

@@ -1699,6 +1699,12 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
 			The default is to send the implementation identification
 			information.
 
+	nfsd.nfs4_disable_idmapping=
+			[NFSv4] When set to the default of '1', the NFSv4
+			server will return only numeric uids and gids to
+			clients using auth_sys, and will accept numeric uids
+			and gids from such clients.  This is intended to ease
+			migration from NFSv2/v3.
 
 	objlayoutdriver.osd_login_prog=
 			[NFS] [OBJLAYOUT] sets the pathname to the program which

fs/ext4/dir.c

@@ -32,24 +32,8 @@ static unsigned char ext4_filetype_table[] = {
 	DT_UNKNOWN, DT_REG, DT_DIR, DT_CHR, DT_BLK, DT_FIFO, DT_SOCK, DT_LNK
 };
 
-static int ext4_readdir(struct file *, void *, filldir_t);
 static int ext4_dx_readdir(struct file *filp,
 			   void *dirent, filldir_t filldir);
-static int ext4_release_dir(struct inode *inode,
-				struct file *filp);
-
-const struct file_operations ext4_dir_operations = {
-	.llseek		= ext4_llseek,
-	.read		= generic_read_dir,
-	.readdir	= ext4_readdir,		/* we take BKL. needed?*/
-	.unlocked_ioctl = ext4_ioctl,
-#ifdef CONFIG_COMPAT
-	.compat_ioctl	= ext4_compat_ioctl,
-#endif
-	.fsync		= ext4_sync_file,
-	.release	= ext4_release_dir,
-};
-
 
 static unsigned char get_dtype(struct super_block *sb, int filetype)
 {
@@ -60,6 +44,26 @@ static unsigned char get_dtype(struct super_block *sb, int filetype)
 	return (ext4_filetype_table[filetype]);
 }
 
+/**
+ * Check if the given dir-inode refers to an htree-indexed directory
+ * (or a directory which chould potentially get coverted to use htree
+ * indexing).
+ *
+ * Return 1 if it is a dx dir, 0 if not
+ */
+static int is_dx_dir(struct inode *inode)
+{
+	struct super_block *sb = inode->i_sb;
+
+	if (EXT4_HAS_COMPAT_FEATURE(inode->i_sb,
+		     EXT4_FEATURE_COMPAT_DIR_INDEX) &&
+	    ((ext4_test_inode_flag(inode, EXT4_INODE_INDEX)) ||
+	     ((inode->i_size >> sb->s_blocksize_bits) == 1)))
+		return 1;
+
+	return 0;
+}
+
 /*
  * Return 0 if the directory entry is OK, and 1 if there is a problem
  *
@@ -115,18 +119,13 @@ static int ext4_readdir(struct file *filp,
 	unsigned int offset;
 	int i, stored;
 	struct ext4_dir_entry_2 *de;
-	struct super_block *sb;
 	int err;
 	struct inode *inode = filp->f_path.dentry->d_inode;
+	struct super_block *sb = inode->i_sb;
 	int ret = 0;
 	int dir_has_error = 0;
 
-	sb = inode->i_sb;
-
-	if (EXT4_HAS_COMPAT_FEATURE(inode->i_sb,
-				    EXT4_FEATURE_COMPAT_DIR_INDEX) &&
-	    ((ext4_test_inode_flag(inode, EXT4_INODE_INDEX)) ||
-	     ((inode->i_size >> sb->s_blocksize_bits) == 1))) {
+	if (is_dx_dir(inode)) {
 		err = ext4_dx_readdir(filp, dirent, filldir);
 		if (err != ERR_BAD_DX_DIR) {
 			ret = err;
@@ -254,22 +253,134 @@ static int ext4_readdir(struct file *filp,
 	return ret;
 }
 
+static inline int is_32bit_api(void)
+{
+#ifdef CONFIG_COMPAT
+	return is_compat_task();
+#else
+	return (BITS_PER_LONG == 32);
+#endif
+}
+
 /*
  * These functions convert from the major/minor hash to an f_pos
- * value.
+ * value for dx directories
+ *
+ * Upper layer (for example NFS) should specify FMODE_32BITHASH or
+ * FMODE_64BITHASH explicitly. On the other hand, we allow ext4 to be mounted
+ * directly on both 32-bit and 64-bit nodes, under such case, neither
+ * FMODE_32BITHASH nor FMODE_64BITHASH is specified.
+ */
+static inline loff_t hash2pos(struct file *filp, __u32 major, __u32 minor)
+{
+	if ((filp->f_mode & FMODE_32BITHASH) ||
+	    (!(filp->f_mode & FMODE_64BITHASH) && is_32bit_api()))
+		return major >> 1;
+	else
+		return ((__u64)(major >> 1) << 32) | (__u64)minor;
+}
+
+static inline __u32 pos2maj_hash(struct file *filp, loff_t pos)
+{
+	if ((filp->f_mode & FMODE_32BITHASH) ||
+	    (!(filp->f_mode & FMODE_64BITHASH) && is_32bit_api()))
+		return (pos << 1) & 0xffffffff;
+	else
+		return ((pos >> 32) << 1) & 0xffffffff;
+}
+
+static inline __u32 pos2min_hash(struct file *filp, loff_t pos)
+{
+	if ((filp->f_mode & FMODE_32BITHASH) ||
+	    (!(filp->f_mode & FMODE_64BITHASH) && is_32bit_api()))
+		return 0;
+	else
+		return pos & 0xffffffff;
+}
+
+/*
+ * Return 32- or 64-bit end-of-file for dx directories
+ */
+static inline loff_t ext4_get_htree_eof(struct file *filp)
+{
+	if ((filp->f_mode & FMODE_32BITHASH) ||
+	    (!(filp->f_mode & FMODE_64BITHASH) && is_32bit_api()))
+		return EXT4_HTREE_EOF_32BIT;
+	else
+		return EXT4_HTREE_EOF_64BIT;
+}
+
+
+/*
+ * ext4_dir_llseek() based on generic_file_llseek() to handle both
+ * non-htree and htree directories, where the "offset" is in terms
+ * of the filename hash value instead of the byte offset.
  *
- * Currently we only use major hash numer.  This is unfortunate, but
- * on 32-bit machines, the same VFS interface is used for lseek and
- * llseek, so if we use the 64 bit offset, then the 32-bit versions of
- * lseek/telldir/seekdir will blow out spectacularly, and from within
- * the ext2 low-level routine, we don't know if we're being called by
- * a 64-bit version of the system call or the 32-bit version of the
- * system call.  Worse yet, NFSv2 only allows for a 32-bit readdir
- * cookie.  Sigh.
+ * NOTE: offsets obtained *before* ext4_set_inode_flag(dir, EXT4_INODE_INDEX)
+ *       will be invalid once the directory was converted into a dx directory
+ */
+loff_t ext4_dir_llseek(struct file *file, loff_t offset, int origin)
+{
+	struct inode *inode = file->f_mapping->host;
+	loff_t ret = -EINVAL;
+	int dx_dir = is_dx_dir(inode);
+
+	mutex_lock(&inode->i_mutex);
+
+	/* NOTE: relative offsets with dx directories might not work
+	 *       as expected, as it is difficult to figure out the
+	 *       correct offset between dx hashes */
+
+	switch (origin) {
+	case SEEK_END:
+		if (unlikely(offset > 0))
+			goto out_err; /* not supported for directories */
+
+		/* so only negative offsets are left, does that have a
+		 * meaning for directories at all? */
+		if (dx_dir)
+			offset += ext4_get_htree_eof(file);
+		else
+			offset += inode->i_size;
+		break;
+	case SEEK_CUR:
+		/*
+		 * Here we special-case the lseek(fd, 0, SEEK_CUR)
+		 * position-querying operation.  Avoid rewriting the "same"
+		 * f_pos value back to the file because a concurrent read(),
+		 * write() or lseek() might have altered it
 		 */
-#define hash2pos(major, minor)	(major >> 1)
-#define pos2maj_hash(pos)	((pos << 1) & 0xffffffff)
-#define pos2min_hash(pos)	(0)
+		if (offset == 0) {
+			offset = file->f_pos;
+			goto out_ok;
+		}
+
+		offset += file->f_pos;
+		break;
+	}
+
+	if (unlikely(offset < 0))
+		goto out_err;
+
+	if (!dx_dir) {
+		if (offset > inode->i_sb->s_maxbytes)
+			goto out_err;
+	} else if (offset > ext4_get_htree_eof(file))
+		goto out_err;
+
+	/* Special lock needed here? */
+	if (offset != file->f_pos) {
+		file->f_pos = offset;
+		file->f_version = 0;
+	}
+
+out_ok:
+	ret = offset;
+out_err:
+	mutex_unlock(&inode->i_mutex);
+
+	return ret;
+}
 
 /*
  * This structure holds the nodes of the red-black tree used to store
@@ -330,15 +441,16 @@ static void free_rb_tree_fname(struct rb_root *root)
 }
 
 
-static struct dir_private_info *ext4_htree_create_dir_info(loff_t pos)
+static struct dir_private_info *ext4_htree_create_dir_info(struct file *filp,
+							   loff_t pos)
 {
 	struct dir_private_info *p;
 
 	p = kzalloc(sizeof(struct dir_private_info), GFP_KERNEL);
 	if (!p)
 		return NULL;
-	p->curr_hash = pos2maj_hash(pos);
-	p->curr_minor_hash = pos2min_hash(pos);
+	p->curr_hash = pos2maj_hash(filp, pos);
+	p->curr_minor_hash = pos2min_hash(filp, pos);
 	return p;
 }
 
@@ -430,7 +542,7 @@ static int call_filldir(struct file *filp, void *dirent,
 			 inode->i_ino, current->comm);
 		return 0;
 	}
-	curr_pos = hash2pos(fname->hash, fname->minor_hash);
+	curr_pos = hash2pos(filp, fname->hash, fname->minor_hash);
 	while (fname) {
 		error = filldir(dirent, fname->name,
 				fname->name_len, curr_pos,
@@ -455,13 +567,13 @@ static int ext4_dx_readdir(struct file *filp,
 	int	ret;
 
 	if (!info) {
-		info = ext4_htree_create_dir_info(filp->f_pos);
+		info = ext4_htree_create_dir_info(filp, filp->f_pos);
 		if (!info)
 			return -ENOMEM;
 		filp->private_data = info;
 	}
 
-	if (filp->f_pos == EXT4_HTREE_EOF)
+	if (filp->f_pos == ext4_get_htree_eof(filp))
 		return 0;	/* EOF */
 
 	/* Some one has messed with f_pos; reset the world */
@@ -469,8 +581,8 @@ static int ext4_dx_readdir(struct file *filp,
 		free_rb_tree_fname(&info->root);
 		info->curr_node = NULL;
 		info->extra_fname = NULL;
-		info->curr_hash = pos2maj_hash(filp->f_pos);
-		info->curr_minor_hash = pos2min_hash(filp->f_pos);
+		info->curr_hash = pos2maj_hash(filp, filp->f_pos);
+		info->curr_minor_hash = pos2min_hash(filp, filp->f_pos);
 	}
 
 	/*
@@ -502,7 +614,7 @@ static int ext4_dx_readdir(struct file *filp,
 			if (ret < 0)
 				return ret;
 			if (ret == 0) {
-				filp->f_pos = EXT4_HTREE_EOF;
+				filp->f_pos = ext4_get_htree_eof(filp);
 				break;
 			}
 			info->curr_node = rb_first(&info->root);
@@ -522,7 +634,7 @@ static int ext4_dx_readdir(struct file *filp,
 			info->curr_minor_hash = fname->minor_hash;
 		} else {
 			if (info->next_hash == ~0) {
-				filp->f_pos = EXT4_HTREE_EOF;
+				filp->f_pos = ext4_get_htree_eof(filp);
 				break;
 			}
 			info->curr_hash = info->next_hash;
@@ -541,3 +653,15 @@ static int ext4_release_dir(struct inode *inode, struct file *filp)
 
 	return 0;
 }
+
+const struct file_operations ext4_dir_operations = {
+	.llseek		= ext4_dir_llseek,
+	.read		= generic_read_dir,
+	.readdir	= ext4_readdir,
+	.unlocked_ioctl = ext4_ioctl,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl	= ext4_compat_ioctl,
+#endif
+	.fsync		= ext4_sync_file,
+	.release	= ext4_release_dir,
+};

fs/ext4/ext4.h

@@ -1623,7 +1623,11 @@ struct dx_hash_info
 	u32		*seed;
 };
 
-#define EXT4_HTREE_EOF	0x7fffffff
+
+/* 32 and 64 bit signed EOF for dx directories */
+#define EXT4_HTREE_EOF_32BIT   ((1UL  << (32 - 1)) - 1)
+#define EXT4_HTREE_EOF_64BIT   ((1ULL << (64 - 1)) - 1)
+
 
 /*
  * Control parameters used by ext4_htree_next_block

fs/ext4/hash.c

@@ -200,8 +200,8 @@ int ext4fs_dirhash(const char *name, int len, struct dx_hash_info *hinfo)
 		return -1;
 	}
 	hash = hash & ~1;
-	if (hash == (EXT4_HTREE_EOF << 1))
-		hash = (EXT4_HTREE_EOF-1) << 1;
+	if (hash == (EXT4_HTREE_EOF_32BIT << 1))
+		hash = (EXT4_HTREE_EOF_32BIT - 1) << 1;
 	hinfo->hash = hash;
 	hinfo->minor_hash = minor_hash;
 	return 0;

fs/lockd/svc.c

@@ -496,7 +496,7 @@ static int param_set_##name(const char *val, struct kernel_param *kp)	\
 	__typeof__(type) num = which_strtol(val, &endp, 0);		\
 	if (endp == val || *endp || num < (min) || num > (max))		\
 		return -EINVAL;						\
-	*((int *) kp->arg) = num;					\
+	*((type *) kp->arg) = num;					\
 	return 0;							\
 }
 

fs/nfsd/current_stateid.h

+#ifndef _NFSD4_CURRENT_STATE_H
+#define _NFSD4_CURRENT_STATE_H
+
+#include "state.h"
+#include "xdr4.h"
+
+extern void clear_current_stateid(struct nfsd4_compound_state *cstate);
+/*
+ * functions to set current state id
+ */
+extern void nfsd4_set_opendowngradestateid(struct nfsd4_compound_state *cstate, struct nfsd4_open_downgrade *);
+extern void nfsd4_set_openstateid(struct nfsd4_compound_state *, struct nfsd4_open *);
+extern void nfsd4_set_lockstateid(struct nfsd4_compound_state *, struct nfsd4_lock *);
+extern void nfsd4_set_closestateid(struct nfsd4_compound_state *, struct nfsd4_close *);
+
+/*
+ * functions to consume current state id
+ */
+extern void nfsd4_get_opendowngradestateid(struct nfsd4_compound_state *cstate, struct nfsd4_open_downgrade *);
+extern void nfsd4_get_delegreturnstateid(struct nfsd4_compound_state *, struct nfsd4_delegreturn *);
+extern void nfsd4_get_freestateid(struct nfsd4_compound_state *, struct nfsd4_free_stateid *);
+extern void nfsd4_get_setattrstateid(struct nfsd4_compound_state *, struct nfsd4_setattr *);
+extern void nfsd4_get_closestateid(struct nfsd4_compound_state *, struct nfsd4_close *);
+extern void nfsd4_get_lockustateid(struct nfsd4_compound_state *, struct nfsd4_locku *);
+extern void nfsd4_get_readstateid(struct nfsd4_compound_state *, struct nfsd4_read *);
+extern void nfsd4_get_writestateid(struct nfsd4_compound_state *, struct nfsd4_write *);
+
+#endif   /* _NFSD4_CURRENT_STATE_H */

fs/nfsd/export.c

@@ -87,7 +87,7 @@ static int expkey_parse(struct cache_detail *cd, char *mesg, int mlen)
 	struct svc_expkey key;
 	struct svc_expkey *ek = NULL;
 
-	if (mlen < 1 || mesg[mlen-1] != '\n')
+	if (mesg[mlen - 1] != '\n')
 		return -EINVAL;
 	mesg[mlen-1] = 0;
 

fs/nfsd/netns.h

+/*
+ * per net namespace data structures for nfsd
+ *
+ * Copyright (C) 2012, Jeff Layton <jlayton@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc., 51
+ * Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
+ */
+
+#ifndef __NFSD_NETNS_H__
+#define __NFSD_NETNS_H__
+
+#include <net/net_namespace.h>
+#include <net/netns/generic.h>
+
+struct cld_net;
+
+struct nfsd_net {
+	struct cld_net *cld_net;
+};
+
+extern int nfsd_net_id;
+#endif /* __NFSD_NETNS_H__ */

fs/nfsd/nfs4callback.c

@@ -645,7 +645,6 @@ static int setup_callback_client(struct nfs4_client *clp, struct nfs4_cb_conn *c
 		.timeout	= &timeparms,
 		.program	= &cb_program,
 		.version	= 0,
-		.authflavor	= clp->cl_flavor,
 		.flags		= (RPC_CLNT_CREATE_NOPING | RPC_CLNT_CREATE_QUIET),
 	};
 	struct rpc_clnt *client;
@@ -656,6 +655,7 @@ static int setup_callback_client(struct nfs4_client *clp, struct nfs4_cb_conn *c
 		args.client_name = clp->cl_principal;
 		args.prognumber	= conn->cb_prog,
 		args.protocol = XPRT_TRANSPORT_TCP;
+		args.authflavor = clp->cl_flavor;
 		clp->cl_cb_ident = conn->cb_ident;
 	} else {
 		if (!conn->cb_xprt)
@@ -665,6 +665,7 @@ static int setup_callback_client(struct nfs4_client *clp, struct nfs4_cb_conn *c
 		args.bc_xprt = conn->cb_xprt;
 		args.prognumber = clp->cl_cb_session->se_cb_prog;
 		args.protocol = XPRT_TRANSPORT_BC_TCP;
+		args.authflavor = RPC_AUTH_UNIX;
 	}
 	/* Create RPC client */
 	client = rpc_create(&args);
@@ -754,9 +755,9 @@ static void do_probe_callback(struct nfs4_client *clp)
  */
 void nfsd4_probe_callback(struct nfs4_client *clp)
 {
-	/* XXX: atomicity?  Also, should we be using cl_cb_flags? */
+	/* XXX: atomicity?  Also, should we be using cl_flags? */
 	clp->cl_cb_state = NFSD4_CB_UNKNOWN;
-	set_bit(NFSD4_CLIENT_CB_UPDATE, &clp->cl_cb_flags);
+	set_bit(NFSD4_CLIENT_CB_UPDATE, &clp->cl_flags);
 	do_probe_callback(clp);
 }
 
@@ -915,7 +916,7 @@ void nfsd4_destroy_callback_queue(void)
 /* must be called under the state lock */
 void nfsd4_shutdown_callback(struct nfs4_client *clp)
 {
-	set_bit(NFSD4_CLIENT_KILL, &clp->cl_cb_flags);
+	set_bit(NFSD4_CLIENT_CB_KILL, &clp->cl_flags);
 	/*
 	 * Note this won't actually result in a null callback;
 	 * instead, nfsd4_do_callback_rpc() will detect the killed
@@ -966,15 +967,15 @@ static void nfsd4_process_cb_update(struct nfsd4_callback *cb)
 		svc_xprt_put(clp->cl_cb_conn.cb_xprt);
 		clp->cl_cb_conn.cb_xprt = NULL;
 	}
-	if (test_bit(NFSD4_CLIENT_KILL, &clp->cl_cb_flags))
+	if (test_bit(NFSD4_CLIENT_CB_KILL, &clp->cl_flags))
 		return;
 	spin_lock(&clp->cl_lock);
 	/*
 	 * Only serialized callback code is allowed to clear these
 	 * flags; main nfsd code can only set them:
 	 */
-	BUG_ON(!clp->cl_cb_flags);
-	clear_bit(NFSD4_CLIENT_CB_UPDATE, &clp->cl_cb_flags);
+	BUG_ON(!(clp->cl_flags & NFSD4_CLIENT_CB_FLAG_MASK));
+	clear_bit(NFSD4_CLIENT_CB_UPDATE, &clp->cl_flags);
 	memcpy(&conn, &cb->cb_clp->cl_cb_conn, sizeof(struct nfs4_cb_conn));
 	c = __nfsd4_find_backchannel(clp);
 	if (c) {
@@ -986,7 +987,7 @@ static void nfsd4_process_cb_update(struct nfsd4_callback *cb)
 
 	err = setup_callback_client(clp, &conn, ses);
 	if (err) {
-		warn_no_callback_path(clp, err);
+		nfsd4_mark_cb_down(clp, err);
 		return;
 	}
 	/* Yay, the callback channel's back! Restart any callbacks: */
@@ -1000,7 +1001,7 @@ void nfsd4_do_callback_rpc(struct work_struct *w)
 	struct nfs4_client *clp = cb->cb_clp;
 	struct rpc_clnt *clnt;
 
-	if (clp->cl_cb_flags)
+	if (clp->cl_flags & NFSD4_CLIENT_CB_FLAG_MASK)
 		nfsd4_process_cb_update(cb);
 
 	clnt = clp->cl_cb_client;

fs/nfsd/nfs4idmap.c

@@ -40,6 +40,14 @@
 #include "idmap.h"
 #include "nfsd.h"
 
+/*
+ * Turn off idmapping when using AUTH_SYS.
+ */
+static bool nfs4_disable_idmapping = true;
+module_param(nfs4_disable_idmapping, bool, 0644);
+MODULE_PARM_DESC(nfs4_disable_idmapping,
+		"Turn off server's NFSv4 idmapping when using 'sec=sys'");
+
 /*
  * Cache entry
  */
@@ -561,28 +569,65 @@ idmap_id_to_name(struct svc_rqst *rqstp, int type, uid_t id, char *name)
 	return ret;
 }
 
+static bool
+numeric_name_to_id(struct svc_rqst *rqstp, int type, const char *name, u32 namelen, uid_t *id)
+{
+	int ret;
+	char buf[11];
+
+	if (namelen + 1 > sizeof(buf))
+		/* too long to represent a 32-bit id: */
+		return false;
+	/* Just to make sure it's null-terminated: */
+	memcpy(buf, name, namelen);
+	buf[namelen] = '\0';
+	ret = kstrtouint(name, 10, id);
+	return ret == 0;
+}
+
+static __be32
+do_name_to_id(struct svc_rqst *rqstp, int type, const char *name, u32 namelen, uid_t *id)
+{
+	if (nfs4_disable_idmapping && rqstp->rq_flavor < RPC_AUTH_GSS)
+		if (numeric_name_to_id(rqstp, type, name, namelen, id))
+			return 0;
+		/*
+		 * otherwise, fall through and try idmapping, for
+		 * backwards compatibility with clients sending names:
+		 */
+	return idmap_name_to_id(rqstp, type, name, namelen, id);
+}
+
+static int
+do_id_to_name(struct svc_rqst *rqstp, int type, uid_t id, char *name)
+{
+	if (nfs4_disable_idmapping && rqstp->rq_flavor < RPC_AUTH_GSS)
+		return sprintf(name, "%u", id);
+	return idmap_id_to_name(rqstp, type, id, name);
+}
+
 __be32
 nfsd_map_name_to_uid(struct svc_rqst *rqstp, const char *name, size_t namelen,
 		__u32 *id)
 {
-	return idmap_name_to_id(rqstp, IDMAP_TYPE_USER, name, namelen, id);
+	return do_name_to_id(rqstp, IDMAP_TYPE_USER, name, namelen, id);
 }
 
 __be32
 nfsd_map_name_to_gid(struct svc_rqst *rqstp, const char *name, size_t namelen,
 		__u32 *id)
 {
-	return idmap_name_to_id(rqstp, IDMAP_TYPE_GROUP, name, namelen, id);
+	return do_name_to_id(rqstp, IDMAP_TYPE_GROUP, name, namelen, id);
 }
 
 int
 nfsd_map_uid_to_name(struct svc_rqst *rqstp, __u32 id, char *name)
 {
-	return idmap_id_to_name(rqstp, IDMAP_TYPE_USER, id, name);
+	return do_id_to_name(rqstp, IDMAP_TYPE_USER, id, name);
 }
 
 int
 nfsd_map_gid_to_name(struct svc_rqst *rqstp, __u32 id, char *name)
 {
-	return idmap_id_to_name(rqstp, IDMAP_TYPE_GROUP, id, name);
+	return do_id_to_name(rqstp, IDMAP_TYPE_GROUP, id, name);
 }

fs/nfsd/nfs4proc.c

@@ -39,6 +39,7 @@
 #include "cache.h"
 #include "xdr4.h"
 #include "vfs.h"
+#include "current_stateid.h"
 
 #define NFSDDBG_FACILITY		NFSDDBG_PROC
 
@@ -192,10 +193,13 @@ static __be32 nfsd_check_obj_isreg(struct svc_fh *fh)
 static __be32
 do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open)
 {
-	struct svc_fh resfh;
+	struct svc_fh *resfh;
 	__be32 status;
 
-	fh_init(&resfh, NFS4_FHSIZE);
+	resfh = kmalloc(sizeof(struct svc_fh), GFP_KERNEL);
+	if (!resfh)
+		return nfserr_jukebox;
+	fh_init(resfh, NFS4_FHSIZE);
 	open->op_truncate = 0;
 
 	if (open->op_create) {
@@ -220,7 +224,7 @@ do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_o
 		 */
 		status = do_nfsd_create(rqstp, current_fh, open->op_fname.data,
 					open->op_fname.len, &open->op_iattr,
-					&resfh, open->op_createmode,
+					resfh, open->op_createmode,
 					(u32 *)open->op_verf.data,
 					&open->op_truncate, &open->op_created);
 
@@ -234,30 +238,29 @@ do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_o
 						FATTR4_WORD1_TIME_MODIFY);
 	} else {
 		status = nfsd_lookup(rqstp, current_fh,
-				     open->op_fname.data, open->op_fname.len, &resfh);
+				     open->op_fname.data, open->op_fname.len, resfh);
 		fh_unlock(current_fh);
 		if (status)
 			goto out;
-		status = nfsd_check_obj_isreg(&resfh);
+		status = nfsd_check_obj_isreg(resfh);
 	}
 	if (status)
 		goto out;
 
 	if (is_create_with_attrs(open) && open->op_acl != NULL)
-		do_set_nfs4_acl(rqstp, &resfh, open->op_acl, open->op_bmval);
-
-	set_change_info(&open->op_cinfo, current_fh);
-	fh_dup2(current_fh, &resfh);
+		do_set_nfs4_acl(rqstp, resfh, open->op_acl, open->op_bmval);
 
 	/* set reply cache */
 	fh_copy_shallow(&open->op_openowner->oo_owner.so_replay.rp_openfh,
-			&resfh.fh_handle);
+			&resfh->fh_handle);
 	if (!open->op_created)
-		status = do_open_permission(rqstp, current_fh, open,
+		status = do_open_permission(rqstp, resfh, open,
 					    NFSD_MAY_NOP);
-
+	set_change_info(&open->op_cinfo, current_fh);
+	fh_dup2(current_fh, resfh);
 out:
-	fh_put(&resfh);
+	fh_put(resfh);
+	kfree(resfh);
 	return status;
 }
 
@@ -310,16 +313,14 @@ nfsd4_open(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	if (open->op_create && open->op_claim_type != NFS4_OPEN_CLAIM_NULL)
 		return nfserr_inval;
 
-	/* We don't yet support WANT bits: */
-	open->op_share_access &= NFS4_SHARE_ACCESS_MASK;
-
 	open->op_created = 0;
 	/*
 	 * RFC5661 18.51.3
 	 * Before RECLAIM_COMPLETE done, server should deny new lock
 	 */
 	if (nfsd4_has_session(cstate) &&
-	    !cstate->session->se_client->cl_firststate &&
+	    !test_bit(NFSD4_CLIENT_RECLAIM_COMPLETE,
+		      &cstate->session->se_client->cl_flags) &&
 	    open->op_claim_type != NFS4_OPEN_CLAIM_PREVIOUS)
 		return nfserr_grace;
 
@@ -452,6 +453,10 @@ nfsd4_restorefh(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		return nfserr_restorefh;
 
 	fh_dup2(&cstate->current_fh, &cstate->save_fh);
+	if (HAS_STATE_ID(cstate, SAVED_STATE_ID_FLAG)) {
+		memcpy(&cstate->current_stateid, &cstate->save_stateid, sizeof(stateid_t));
+		SET_STATE_ID(cstate, CURRENT_STATE_ID_FLAG);
+	}
 	return nfs_ok;
 }
 
@@ -463,6 +468,10 @@ nfsd4_savefh(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		return nfserr_nofilehandle;
 
 	fh_dup2(&cstate->save_fh, &cstate->current_fh);
+	if (HAS_STATE_ID(cstate, CURRENT_STATE_ID_FLAG)) {
+		memcpy(&cstate->save_stateid, &cstate->current_stateid, sizeof(stateid_t));
+		SET_STATE_ID(cstate, SAVED_STATE_ID_FLAG);
+	}
 	return nfs_ok;
 }
 
@@ -481,14 +490,20 @@ nfsd4_access(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 			   &access->ac_supported);
 }
 
+static void gen_boot_verifier(nfs4_verifier *verifier)
+{
+	__be32 verf[2];
+
+	verf[0] = (__be32)nfssvc_boot.tv_sec;
+	verf[1] = (__be32)nfssvc_boot.tv_usec;
+	memcpy(verifier->data, verf, sizeof(verifier->data));
+}
+
 static __be32
 nfsd4_commit(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	     struct nfsd4_commit *commit)
 {
-	u32 *p = (u32 *)commit->co_verf.data;
-	*p++ = nfssvc_boot.tv_sec;
-	*p++ = nfssvc_boot.tv_usec;
-
+	gen_boot_verifier(&commit->co_verf);
 	return nfsd_commit(rqstp, &cstate->current_fh, commit->co_offset,
 			     commit->co_count);
 }
@@ -865,7 +880,6 @@ nfsd4_write(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 {
 	stateid_t *stateid = &write->wr_stateid;
 	struct file *filp = NULL;
-	u32 *p;
 	__be32 status = nfs_ok;
 	unsigned long cnt;
 
@@ -887,9 +901,7 @@ nfsd4_write(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 
 	cnt = write->wr_buflen;
 	write->wr_how_written = write->wr_stable_how;
-	p = (u32 *)write->wr_verifier.data;
-	*p++ = nfssvc_boot.tv_sec;
-	*p++ = nfssvc_boot.tv_usec;
+	gen_boot_verifier(&write->wr_verifier);
 
 	status =  nfsd_write(rqstp, &cstate->current_fh, filp,
 			     write->wr_offset, rqstp->rq_vec, write->wr_vlen,
@@ -1000,6 +1012,8 @@ static inline void nfsd4_increment_op_stats(u32 opnum)
 typedef __be32(*nfsd4op_func)(struct svc_rqst *, struct nfsd4_compound_state *,
 			      void *);
 typedef u32(*nfsd4op_rsize)(struct svc_rqst *, struct nfsd4_op *op);
+typedef void(*stateid_setter)(struct nfsd4_compound_state *, void *);
+typedef void(*stateid_getter)(struct nfsd4_compound_state *, void *);
 
 enum nfsd4_op_flags {
 	ALLOWED_WITHOUT_FH = 1 << 0,	/* No current filehandle required */
@@ -1025,6 +1039,10 @@ enum nfsd4_op_flags {
 	 * the v4.0 case).
 	 */
 	OP_CACHEME = 1 << 6,
+	/*
+	 * These are ops which clear current state id.
+	 */
+	OP_CLEAR_STATEID = 1 << 7,
 };
 
 struct nfsd4_operation {
@@ -1033,11 +1051,15 @@ struct nfsd4_operation {
 	char *op_name;
 	/* Try to get response size before operation */
 	nfsd4op_rsize op_rsize_bop;
+	stateid_setter op_get_currentstateid;
+	stateid_getter op_set_currentstateid;
 };
 
 static struct nfsd4_operation nfsd4_ops[];
 
+#ifdef NFSD_DEBUG
 static const char *nfsd4_op_name(unsigned opnum);
+#endif
 
 /*
  * Enforce NFSv4.1 COMPOUND ordering rules:
@@ -1215,13 +1237,23 @@ nfsd4_proc_compound(struct svc_rqst *rqstp,
 		if (op->status)
 			goto encode_op;
 
-		if (opdesc->op_func)
+		if (opdesc->op_func) {
+			if (opdesc->op_get_currentstateid)
+				opdesc->op_get_currentstateid(cstate, &op->u);
 			op->status = opdesc->op_func(rqstp, cstate, &op->u);
-		else
+		} else
 			BUG_ON(op->status == nfs_ok);
 
-		if (!op->status && need_wrongsec_check(rqstp))
+		if (!op->status) {
+			if (opdesc->op_set_currentstateid)
+				opdesc->op_set_currentstateid(cstate, &op->u);
+
+			if (opdesc->op_flags & OP_CLEAR_STATEID)
+				clear_current_stateid(cstate);
+
+			if (need_wrongsec_check(rqstp))
 				op->status = check_nfsd_access(cstate->current_fh.fh_export, rqstp);
+		}
 
 encode_op:
 		/* Only from SEQUENCE */
@@ -1413,6 +1445,8 @@ static struct nfsd4_operation nfsd4_ops[] = {
 		.op_flags = OP_MODIFIES_SOMETHING,
 		.op_name = "OP_CLOSE",
 		.op_rsize_bop = (nfsd4op_rsize)nfsd4_status_stateid_rsize,
+		.op_get_currentstateid = (stateid_getter)nfsd4_get_closestateid,
+		.op_set_currentstateid = (stateid_setter)nfsd4_set_closestateid,
 	},
 	[OP_COMMIT] = {
 		.op_func = (nfsd4op_func)nfsd4_commit,
@@ -1422,7 +1456,7 @@ static struct nfsd4_operation nfsd4_ops[] = {
 	},
 	[OP_CREATE] = {
 		.op_func = (nfsd4op_func)nfsd4_create,
-		.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME,
+		.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME | OP_CLEAR_STATEID,
 		.op_name = "OP_CREATE",
 		.op_rsize_bop = (nfsd4op_rsize)nfsd4_create_rsize,
 	},
@@ -1431,6 +1465,7 @@ static struct nfsd4_operation nfsd4_ops[] = {
 		.op_flags = OP_MODIFIES_SOMETHING,
 		.op_name = "OP_DELEGRETURN",
 		.op_rsize_bop = nfsd4_only_status_rsize,
+		.op_get_currentstateid = (stateid_getter)nfsd4_get_delegreturnstateid,
 	},
 	[OP_GETATTR] = {
 		.op_func = (nfsd4op_func)nfsd4_getattr,
@@ -1453,6 +1488,7 @@ static struct nfsd4_operation nfsd4_ops[] = {
 		.op_flags = OP_MODIFIES_SOMETHING,
 		.op_name = "OP_LOCK",
 		.op_rsize_bop = (nfsd4op_rsize)nfsd4_lock_rsize,
+		.op_set_currentstateid = (stateid_setter)nfsd4_set_lockstateid,
 	},
 	[OP_LOCKT] = {
 		.op_func = (nfsd4op_func)nfsd4_lockt,
@@ -1463,15 +1499,16 @@ static struct nfsd4_operation nfsd4_ops[] = {
 		.op_flags = OP_MODIFIES_SOMETHING,
 		.op_name = "OP_LOCKU",
 		.op_rsize_bop = (nfsd4op_rsize)nfsd4_status_stateid_rsize,
+		.op_get_currentstateid = (stateid_getter)nfsd4_get_lockustateid,
 	},
 	[OP_LOOKUP] = {
 		.op_func = (nfsd4op_func)nfsd4_lookup,
-		.op_flags = OP_HANDLES_WRONGSEC,
+		.op_flags = OP_HANDLES_WRONGSEC | OP_CLEAR_STATEID,
 		.op_name = "OP_LOOKUP",
 	},
 	[OP_LOOKUPP] = {
 		.op_func = (nfsd4op_func)nfsd4_lookupp,
-		.op_flags = OP_HANDLES_WRONGSEC,
+		.op_flags = OP_HANDLES_WRONGSEC | OP_CLEAR_STATEID,
 		.op_name = "OP_LOOKUPP",
 	},
 	[OP_NVERIFY] = {
@@ -1483,6 +1520,7 @@ static struct nfsd4_operation nfsd4_ops[] = {
 		.op_flags = OP_HANDLES_WRONGSEC | OP_MODIFIES_SOMETHING,
 		.op_name = "OP_OPEN",
 		.op_rsize_bop = (nfsd4op_rsize)nfsd4_open_rsize,
+		.op_set_currentstateid = (stateid_setter)nfsd4_set_openstateid,
 	},
 	[OP_OPEN_CONFIRM] = {
 		.op_func = (nfsd4op_func)nfsd4_open_confirm,
@@ -1495,25 +1533,30 @@ static struct nfsd4_operation nfsd4_ops[] = {
 		.op_flags = OP_MODIFIES_SOMETHING,
 		.op_name = "OP_OPEN_DOWNGRADE",
 		.op_rsize_bop = (nfsd4op_rsize)nfsd4_status_stateid_rsize,
+		.op_get_currentstateid = (stateid_getter)nfsd4_get_opendowngradestateid,
+		.op_set_currentstateid = (stateid_setter)nfsd4_set_opendowngradestateid,
 	},
 	[OP_PUTFH] = {
 		.op_func = (nfsd4op_func)nfsd4_putfh,
 		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
-				| OP_IS_PUTFH_LIKE | OP_MODIFIES_SOMETHING,
+				| OP_IS_PUTFH_LIKE | OP_MODIFIES_SOMETHING
+				| OP_CLEAR_STATEID,
 		.op_name = "OP_PUTFH",
 		.op_rsize_bop = (nfsd4op_rsize)nfsd4_only_status_rsize,
 	},
 	[OP_PUTPUBFH] = {
 		.op_func = (nfsd4op_func)nfsd4_putrootfh,
 		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
-				| OP_IS_PUTFH_LIKE | OP_MODIFIES_SOMETHING,
+				| OP_IS_PUTFH_LIKE | OP_MODIFIES_SOMETHING
+				| OP_CLEAR_STATEID,
 		.op_name = "OP_PUTPUBFH",
 		.op_rsize_bop = (nfsd4op_rsize)nfsd4_only_status_rsize,
 	},
 	[OP_PUTROOTFH] = {
 		.op_func = (nfsd4op_func)nfsd4_putrootfh,
 		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
-				| OP_IS_PUTFH_LIKE | OP_MODIFIES_SOMETHING,
+				| OP_IS_PUTFH_LIKE | OP_MODIFIES_SOMETHING
+				| OP_CLEAR_STATEID,
 		.op_name = "OP_PUTROOTFH",
 		.op_rsize_bop = (nfsd4op_rsize)nfsd4_only_status_rsize,
 	},
@@ -1522,6 +1565,7 @@ static struct nfsd4_operation nfsd4_ops[] = {
 		.op_flags = OP_MODIFIES_SOMETHING,
 		.op_name = "OP_READ",
 		.op_rsize_bop = (nfsd4op_rsize)nfsd4_read_rsize,
+		.op_get_currentstateid = (stateid_getter)nfsd4_get_readstateid,
 	},
 	[OP_READDIR] = {
 		.op_func = (nfsd4op_func)nfsd4_readdir,
@@ -1576,6 +1620,7 @@ static struct nfsd4_operation nfsd4_ops[] = {
 		.op_name = "OP_SETATTR",
 		.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME,
 		.op_rsize_bop = (nfsd4op_rsize)nfsd4_setattr_rsize,
+		.op_get_currentstateid = (stateid_getter)nfsd4_get_setattrstateid,
 	},
 	[OP_SETCLIENTID] = {
 		.op_func = (nfsd4op_func)nfsd4_setclientid,
@@ -1600,6 +1645,7 @@ static struct nfsd4_operation nfsd4_ops[] = {
 		.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME,
 		.op_name = "OP_WRITE",
 		.op_rsize_bop = (nfsd4op_rsize)nfsd4_write_rsize,
+		.op_get_currentstateid = (stateid_getter)nfsd4_get_writestateid,
 	},
 	[OP_RELEASE_LOCKOWNER] = {
 		.op_func = (nfsd4op_func)nfsd4_release_lockowner,
@@ -1674,12 +1720,14 @@ static struct nfsd4_operation nfsd4_ops[] = {
 	},
 };
 
+#ifdef NFSD_DEBUG
 static const char *nfsd4_op_name(unsigned opnum)
 {
 	if (opnum < ARRAY_SIZE(nfsd4_ops))
 		return nfsd4_ops[opnum].op_name;
 	return "unknown_operation";
 }
+#endif
 
 #define nfsd4_voidres			nfsd4_voidargs
 struct nfsd4_voidargs { int dummy; };

fs/nfsd/nfs4recover.c

 /*
 *  Copyright (c) 2004 The Regents of the University of Michigan.
+*  Copyright (c) 2012 Jeff Layton <jlayton@redhat.com>
 *  All rights reserved.
 *
 *  Andy Adamson <andros@citi.umich.edu>
@@ -36,16 +37,34 @@
 #include <linux/namei.h>
 #include <linux/crypto.h>
 #include <linux/sched.h>
+#include <linux/fs.h>
+#include <linux/module.h>
+#include <net/net_namespace.h>
+#include <linux/sunrpc/rpc_pipe_fs.h>
+#include <linux/sunrpc/clnt.h>
+#include <linux/nfsd/cld.h>
 
 #include "nfsd.h"
 #include "state.h"
 #include "vfs.h"
+#include "netns.h"
 
 #define NFSDDBG_FACILITY                NFSDDBG_PROC
 
+/* Declarations */
+struct nfsd4_client_tracking_ops {
+	int (*init)(struct net *);
+	void (*exit)(struct net *);
+	void (*create)(struct nfs4_client *);
+	void (*remove)(struct nfs4_client *);
+	int (*check)(struct nfs4_client *);
+	void (*grace_done)(struct net *, time_t);
+};
+
 /* Globals */
 static struct file *rec_file;
 static char user_recovery_dirname[PATH_MAX] = "/var/lib/nfs/v4recovery";
+static struct nfsd4_client_tracking_ops *client_tracking_ops;
 
 static int
 nfs4_save_creds(const struct cred **original_creds)
@@ -117,7 +136,8 @@ nfs4_make_rec_clidname(char *dname, struct xdr_netobj *clname)
 	return status;
 }
 
-void nfsd4_create_clid_dir(struct nfs4_client *clp)
+static void
+nfsd4_create_clid_dir(struct nfs4_client *clp)
 {
 	const struct cred *original_cred;
 	char *dname = clp->cl_recdir;
@@ -126,9 +146,8 @@ void nfsd4_create_clid_dir(struct nfs4_client *clp)
 
 	dprintk("NFSD: nfsd4_create_clid_dir for \"%s\"\n", dname);
 
-	if (clp->cl_firststate)
+	if (test_and_set_bit(NFSD4_CLIENT_STABLE, &clp->cl_flags))
 		return;
-	clp->cl_firststate = 1;
 	if (!rec_file)
 		return;
 	status = nfs4_save_creds(&original_cred);
@@ -265,19 +284,19 @@ nfsd4_unlink_clid_dir(char *name, int namlen)
 	return status;
 }
 
-void
+static void
 nfsd4_remove_clid_dir(struct nfs4_client *clp)
 {
 	const struct cred *original_cred;
 	int status;
 
-	if (!rec_file || !clp->cl_firststate)
+	if (!rec_file || !test_bit(NFSD4_CLIENT_STABLE, &clp->cl_flags))
 		return;
 
 	status = mnt_want_write_file(rec_file);
 	if (status)
 		goto out;
-	clp->cl_firststate = 0;
+	clear_bit(NFSD4_CLIENT_STABLE, &clp->cl_flags);
 
 	status = nfs4_save_creds(&original_cred);
 	if (status < 0)
@@ -292,7 +311,6 @@ nfsd4_remove_clid_dir(struct nfs4_client *clp)
 	if (status)
 		printk("NFSD: Failed to remove expired client state directory"
 				" %.*s\n", HEXDIR_LEN, clp->cl_recdir);
-	return;
 }
 
 static int
@@ -311,8 +329,9 @@ purge_old(struct dentry *parent, struct dentry *child)
 	return 0;
 }
 
-void
-nfsd4_recdir_purge_old(void) {
+static void
+nfsd4_recdir_purge_old(struct net *net, time_t boot_time)
+{
 	int status;
 
 	if (!rec_file)
@@ -343,7 +362,7 @@ load_recdir(struct dentry *parent, struct dentry *child)
 	return 0;
 }
 
-int
+static int
 nfsd4_recdir_load(void) {
 	int status;
 
@@ -361,8 +380,8 @@ nfsd4_recdir_load(void) {
  * Hold reference to the recovery directory.
  */
 
-void
-nfsd4_init_recdir()
+static int
+nfsd4_init_recdir(void)
 {
 	const struct cred *original_cred;
 	int status;
@@ -377,20 +396,44 @@ nfsd4_init_recdir()
 		printk("NFSD: Unable to change credentials to find recovery"
 		       " directory: error %d\n",
 		       status);
-		return;
+		return status;
 	}
 
 	rec_file = filp_open(user_recovery_dirname, O_RDONLY | O_DIRECTORY, 0);
 	if (IS_ERR(rec_file)) {
 		printk("NFSD: unable to find recovery directory %s\n",
 				user_recovery_dirname);
+		status = PTR_ERR(rec_file);
 		rec_file = NULL;
 	}
 
 	nfs4_reset_creds(original_cred);
+	return status;
 }
 
-void
+static int
+nfsd4_load_reboot_recovery_data(struct net *net)
+{
+	int status;
+
+	/* XXX: The legacy code won't work in a container */
+	if (net != &init_net) {
+		WARN(1, KERN_ERR "NFSD: attempt to initialize legacy client "
+			"tracking in a container!\n");
+		return -EINVAL;
+	}
+
+	nfs4_lock_state();
+	status = nfsd4_init_recdir();
+	if (!status)
+		status = nfsd4_recdir_load();
+	nfs4_unlock_state();
+	if (status)
+		printk(KERN_ERR "NFSD: Failure reading reboot recovery data\n");
+	return status;
+}
+
+static void
 nfsd4_shutdown_recdir(void)
 {
 	if (!rec_file)
@@ -399,6 +442,13 @@ nfsd4_shutdown_recdir(void)
 	rec_file = NULL;
 }
 
+static void
+nfsd4_legacy_tracking_exit(struct net *net)
+{
+	nfs4_release_reclaim();
+	nfsd4_shutdown_recdir();
+}
+
 /*
  * Change the NFSv4 recovery directory to recdir.
  */
@@ -425,3 +475,572 @@ nfs4_recoverydir(void)
 {
 	return user_recovery_dirname;
 }
+
+static int
+nfsd4_check_legacy_client(struct nfs4_client *clp)
+{
+	/* did we already find that this client is stable? */
+	if (test_bit(NFSD4_CLIENT_STABLE, &clp->cl_flags))
+		return 0;
+
+	/* look for it in the reclaim hashtable otherwise */
+	if (nfsd4_find_reclaim_client(clp)) {
+		set_bit(NFSD4_CLIENT_STABLE, &clp->cl_flags);
+		return 0;
+	}
+
+	return -ENOENT;
+}
+
+static struct nfsd4_client_tracking_ops nfsd4_legacy_tracking_ops = {
+	.init		= nfsd4_load_reboot_recovery_data,
+	.exit		= nfsd4_legacy_tracking_exit,
+	.create		= nfsd4_create_clid_dir,
+	.remove		= nfsd4_remove_clid_dir,
+	.check		= nfsd4_check_legacy_client,
+	.grace_done	= nfsd4_recdir_purge_old,
+};
+
+/* Globals */
+#define NFSD_PIPE_DIR		"nfsd"
+#define NFSD_CLD_PIPE		"cld"
+
+/* per-net-ns structure for holding cld upcall info */
+struct cld_net {
+	struct rpc_pipe		*cn_pipe;
+	spinlock_t		 cn_lock;
+	struct list_head	 cn_list;
+	unsigned int		 cn_xid;
+};
+
+struct cld_upcall {
+	struct list_head	 cu_list;
+	struct cld_net		*cu_net;
+	struct task_struct	*cu_task;
+	struct cld_msg		 cu_msg;
+};
+
+static int
+__cld_pipe_upcall(struct rpc_pipe *pipe, struct cld_msg *cmsg)
+{
+	int ret;
+	struct rpc_pipe_msg msg;
+
+	memset(&msg, 0, sizeof(msg));
+	msg.data = cmsg;
+	msg.len = sizeof(*cmsg);
+
+	/*
+	 * Set task state before we queue the upcall. That prevents
+	 * wake_up_process in the downcall from racing with schedule.
+	 */
+	set_current_state(TASK_UNINTERRUPTIBLE);
+	ret = rpc_queue_upcall(pipe, &msg);
+	if (ret < 0) {
+		set_current_state(TASK_RUNNING);
+		goto out;
+	}
+
+	schedule();
+	set_current_state(TASK_RUNNING);
+
+	if (msg.errno < 0)
+		ret = msg.errno;
+out:
+	return ret;
+}
+
+static int
+cld_pipe_upcall(struct rpc_pipe *pipe, struct cld_msg *cmsg)
+{
+	int ret;
+
+	/*
+	 * -EAGAIN occurs when pipe is closed and reopened while there are
+	 *  upcalls queued.
+	 */
+	do {
+		ret = __cld_pipe_upcall(pipe, cmsg);
+	} while (ret == -EAGAIN);
+
+	return ret;
+}
+
+static ssize_t
+cld_pipe_downcall(struct file *filp, const char __user *src, size_t mlen)
+{
+	struct cld_upcall *tmp, *cup;
+	struct cld_msg *cmsg = (struct cld_msg *)src;
+	uint32_t xid;
+	struct nfsd_net *nn = net_generic(filp->f_dentry->d_sb->s_fs_info,
+						nfsd_net_id);
+	struct cld_net *cn = nn->cld_net;
+
+	if (mlen != sizeof(*cmsg)) {
+		dprintk("%s: got %lu bytes, expected %lu\n", __func__, mlen,
+			sizeof(*cmsg));
+		return -EINVAL;
+	}
+
+	/* copy just the xid so we can try to find that */
+	if (copy_from_user(&xid, &cmsg->cm_xid, sizeof(xid)) != 0) {
+		dprintk("%s: error when copying xid from userspace", __func__);
+		return -EFAULT;
+	}
+
+	/* walk the list and find corresponding xid */
+	cup = NULL;
+	spin_lock(&cn->cn_lock);
+	list_for_each_entry(tmp, &cn->cn_list, cu_list) {
+		if (get_unaligned(&tmp->cu_msg.cm_xid) == xid) {
+			cup = tmp;
+			list_del_init(&cup->cu_list);
+			break;
+		}
+	}
+	spin_unlock(&cn->cn_lock);
+
+	/* couldn't find upcall? */
+	if (!cup) {
+		dprintk("%s: couldn't find upcall -- xid=%u\n", __func__, xid);
+		return -EINVAL;
+	}
+
+	if (copy_from_user(&cup->cu_msg, src, mlen) != 0)
+		return -EFAULT;
+
+	wake_up_process(cup->cu_task);
+	return mlen;
+}
+
+static void
+cld_pipe_destroy_msg(struct rpc_pipe_msg *msg)
+{
+	struct cld_msg *cmsg = msg->data;
+	struct cld_upcall *cup = container_of(cmsg, struct cld_upcall,
+						 cu_msg);
+
+	/* errno >= 0 means we got a downcall */
+	if (msg->errno >= 0)
+		return;
+
+	wake_up_process(cup->cu_task);
+}
+
+static const struct rpc_pipe_ops cld_upcall_ops = {
+	.upcall		= rpc_pipe_generic_upcall,
+	.downcall	= cld_pipe_downcall,
+	.destroy_msg	= cld_pipe_destroy_msg,
+};
+
+static struct dentry *
+nfsd4_cld_register_sb(struct super_block *sb, struct rpc_pipe *pipe)
+{
+	struct dentry *dir, *dentry;
+
+	dir = rpc_d_lookup_sb(sb, NFSD_PIPE_DIR);
+	if (dir == NULL)
+		return ERR_PTR(-ENOENT);
+	dentry = rpc_mkpipe_dentry(dir, NFSD_CLD_PIPE, NULL, pipe);
+	dput(dir);
+	return dentry;
+}
+
+static void
+nfsd4_cld_unregister_sb(struct rpc_pipe *pipe)
+{
+	if (pipe->dentry)
+		rpc_unlink(pipe->dentry);
+}
+
+static struct dentry *
+nfsd4_cld_register_net(struct net *net, struct rpc_pipe *pipe)
+{
+	struct super_block *sb;
+	struct dentry *dentry;
+
+	sb = rpc_get_sb_net(net);
+	if (!sb)
+		return NULL;
+	dentry = nfsd4_cld_register_sb(sb, pipe);
+	rpc_put_sb_net(net);
+	return dentry;
+}
+
+static void
+nfsd4_cld_unregister_net(struct net *net, struct rpc_pipe *pipe)
+{
+	struct super_block *sb;
+
+	sb = rpc_get_sb_net(net);
+	if (sb) {
+		nfsd4_cld_unregister_sb(pipe);
+		rpc_put_sb_net(net);
+	}
+}
+
+/* Initialize rpc_pipefs pipe for communication with client tracking daemon */
+static int
+nfsd4_init_cld_pipe(struct net *net)
+{
+	int ret;
+	struct dentry *dentry;
+	struct nfsd_net *nn = net_generic(net, nfsd_net_id);
+	struct cld_net *cn;
+
+	if (nn->cld_net)
+		return 0;
+
+	cn = kzalloc(sizeof(*cn), GFP_KERNEL);
+	if (!cn) {
+		ret = -ENOMEM;
+		goto err;
+	}
+
+	cn->cn_pipe = rpc_mkpipe_data(&cld_upcall_ops, RPC_PIPE_WAIT_FOR_OPEN);
+	if (IS_ERR(cn->cn_pipe)) {
+		ret = PTR_ERR(cn->cn_pipe);
+		goto err;
+	}
+	spin_lock_init(&cn->cn_lock);
+	INIT_LIST_HEAD(&cn->cn_list);
+
+	dentry = nfsd4_cld_register_net(net, cn->cn_pipe);
+	if (IS_ERR(dentry)) {
+		ret = PTR_ERR(dentry);
+		goto err_destroy_data;
+	}
+
+	cn->cn_pipe->dentry = dentry;
+	nn->cld_net = cn;
+	return 0;
+
+err_destroy_data:
+	rpc_destroy_pipe_data(cn->cn_pipe);
+err:
+	kfree(cn);
+	printk(KERN_ERR "NFSD: unable to create nfsdcld upcall pipe (%d)\n",
+			ret);
+	return ret;
+}
+
+static void
+nfsd4_remove_cld_pipe(struct net *net)
+{
+	struct nfsd_net *nn = net_generic(net, nfsd_net_id);
+	struct cld_net *cn = nn->cld_net;
+
+	nfsd4_cld_unregister_net(net, cn->cn_pipe);
+	rpc_destroy_pipe_data(cn->cn_pipe);
+	kfree(nn->cld_net);
+	nn->cld_net = NULL;
+}
+
+static struct cld_upcall *
+alloc_cld_upcall(struct cld_net *cn)
+{
+	struct cld_upcall *new, *tmp;
+
+	new = kzalloc(sizeof(*new), GFP_KERNEL);
+	if (!new)
+		return new;
+
+	/* FIXME: hard cap on number in flight? */
+restart_search:
+	spin_lock(&cn->cn_lock);
+	list_for_each_entry(tmp, &cn->cn_list, cu_list) {
+		if (tmp->cu_msg.cm_xid == cn->cn_xid) {
+			cn->cn_xid++;
+			spin_unlock(&cn->cn_lock);
+			goto restart_search;
+		}
+	}
+	new->cu_task = current;
+	new->cu_msg.cm_vers = CLD_UPCALL_VERSION;
+	put_unaligned(cn->cn_xid++, &new->cu_msg.cm_xid);
+	new->cu_net = cn;
+	list_add(&new->cu_list, &cn->cn_list);
+	spin_unlock(&cn->cn_lock);
+
+	dprintk("%s: allocated xid %u\n", __func__, new->cu_msg.cm_xid);
+
+	return new;
+}
+
+static void
+free_cld_upcall(struct cld_upcall *victim)
+{
+	struct cld_net *cn = victim->cu_net;
+
+	spin_lock(&cn->cn_lock);
+	list_del(&victim->cu_list);
+	spin_unlock(&cn->cn_lock);
+	kfree(victim);
+}
+
+/* Ask daemon to create a new record */
+static void
+nfsd4_cld_create(struct nfs4_client *clp)
+{
+	int ret;
+	struct cld_upcall *cup;
+	/* FIXME: determine net from clp */
+	struct nfsd_net *nn = net_generic(&init_net, nfsd_net_id);
+	struct cld_net *cn = nn->cld_net;
+
+	/* Don't upcall if it's already stored */
+	if (test_bit(NFSD4_CLIENT_STABLE, &clp->cl_flags))
+		return;
+
+	cup = alloc_cld_upcall(cn);
+	if (!cup) {
+		ret = -ENOMEM;
+		goto out_err;
+	}
+
+	cup->cu_msg.cm_cmd = Cld_Create;
+	cup->cu_msg.cm_u.cm_name.cn_len = clp->cl_name.len;
+	memcpy(cup->cu_msg.cm_u.cm_name.cn_id, clp->cl_name.data,
+			clp->cl_name.len);
+
+	ret = cld_pipe_upcall(cn->cn_pipe, &cup->cu_msg);
+	if (!ret) {
+		ret = cup->cu_msg.cm_status;
+		set_bit(NFSD4_CLIENT_STABLE, &clp->cl_flags);
+	}
+
+	free_cld_upcall(cup);
+out_err:
+	if (ret)
+		printk(KERN_ERR "NFSD: Unable to create client "
+				"record on stable storage: %d\n", ret);
+}
+
+/* Ask daemon to create a new record */
+static void
+nfsd4_cld_remove(struct nfs4_client *clp)
+{
+	int ret;
+	struct cld_upcall *cup;
+	/* FIXME: determine net from clp */
+	struct nfsd_net *nn = net_generic(&init_net, nfsd_net_id);
+	struct cld_net *cn = nn->cld_net;
+
+	/* Don't upcall if it's already removed */
+	if (!test_bit(NFSD4_CLIENT_STABLE, &clp->cl_flags))
+		return;
+
+	cup = alloc_cld_upcall(cn);
+	if (!cup) {
+		ret = -ENOMEM;
+		goto out_err;
+	}
+
+	cup->cu_msg.cm_cmd = Cld_Remove;
+	cup->cu_msg.cm_u.cm_name.cn_len = clp->cl_name.len;
+	memcpy(cup->cu_msg.cm_u.cm_name.cn_id, clp->cl_name.data,
+			clp->cl_name.len);
+
+	ret = cld_pipe_upcall(cn->cn_pipe, &cup->cu_msg);
+	if (!ret) {
+		ret = cup->cu_msg.cm_status;
+		clear_bit(NFSD4_CLIENT_STABLE, &clp->cl_flags);
+	}
+
+	free_cld_upcall(cup);
+out_err:
+	if (ret)
+		printk(KERN_ERR "NFSD: Unable to remove client "
+				"record from stable storage: %d\n", ret);
+}
+
+/* Check for presence of a record, and update its timestamp */
+static int
+nfsd4_cld_check(struct nfs4_client *clp)
+{
+	int ret;
+	struct cld_upcall *cup;
+	/* FIXME: determine net from clp */
+	struct nfsd_net *nn = net_generic(&init_net, nfsd_net_id);
+	struct cld_net *cn = nn->cld_net;
+
+	/* Don't upcall if one was already stored during this grace pd */
+	if (test_bit(NFSD4_CLIENT_STABLE, &clp->cl_flags))
+		return 0;
+
+	cup = alloc_cld_upcall(cn);
+	if (!cup) {
+		printk(KERN_ERR "NFSD: Unable to check client record on "
+				"stable storage: %d\n", -ENOMEM);
+		return -ENOMEM;
+	}
+
+	cup->cu_msg.cm_cmd = Cld_Check;
+	cup->cu_msg.cm_u.cm_name.cn_len = clp->cl_name.len;
+	memcpy(cup->cu_msg.cm_u.cm_name.cn_id, clp->cl_name.data,
+			clp->cl_name.len);
+
+	ret = cld_pipe_upcall(cn->cn_pipe, &cup->cu_msg);
+	if (!ret) {
+		ret = cup->cu_msg.cm_status;
+		set_bit(NFSD4_CLIENT_STABLE, &clp->cl_flags);
+	}
+
+	free_cld_upcall(cup);
+	return ret;
+}
+
+static void
+nfsd4_cld_grace_done(struct net *net, time_t boot_time)
+{
+	int ret;
+	struct cld_upcall *cup;
+	struct nfsd_net *nn = net_generic(net, nfsd_net_id);
+	struct cld_net *cn = nn->cld_net;
+
+	cup = alloc_cld_upcall(cn);
+	if (!cup) {
+		ret = -ENOMEM;
+		goto out_err;
+	}
+
+	cup->cu_msg.cm_cmd = Cld_GraceDone;
+	cup->cu_msg.cm_u.cm_gracetime = (int64_t)boot_time;
+	ret = cld_pipe_upcall(cn->cn_pipe, &cup->cu_msg);
+	if (!ret)
+		ret = cup->cu_msg.cm_status;
+
+	free_cld_upcall(cup);
+out_err:
+	if (ret)
+		printk(KERN_ERR "NFSD: Unable to end grace period: %d\n", ret);
+}
+
+static struct nfsd4_client_tracking_ops nfsd4_cld_tracking_ops = {
+	.init		= nfsd4_init_cld_pipe,
+	.exit		= nfsd4_remove_cld_pipe,
+	.create		= nfsd4_cld_create,
+	.remove		= nfsd4_cld_remove,
+	.check		= nfsd4_cld_check,
+	.grace_done	= nfsd4_cld_grace_done,
+};
+
+int
+nfsd4_client_tracking_init(struct net *net)
+{
+	int status;
+	struct path path;
+
+	if (!client_tracking_ops) {
+		client_tracking_ops = &nfsd4_cld_tracking_ops;
+		status = kern_path(nfs4_recoverydir(), LOOKUP_FOLLOW, &path);
+		if (!status) {
+			if (S_ISDIR(path.dentry->d_inode->i_mode))
+				client_tracking_ops =
+						&nfsd4_legacy_tracking_ops;
+			path_put(&path);
+		}
+	}
+
+	status = client_tracking_ops->init(net);
+	if (status) {
+		printk(KERN_WARNING "NFSD: Unable to initialize client "
+				    "recovery tracking! (%d)\n", status);
+		client_tracking_ops = NULL;
+	}
+	return status;
+}
+
+void
+nfsd4_client_tracking_exit(struct net *net)
+{
+	if (client_tracking_ops) {
+		client_tracking_ops->exit(net);
+		client_tracking_ops = NULL;
+	}
+}
+
+void
+nfsd4_client_record_create(struct nfs4_client *clp)
+{
+	if (client_tracking_ops)
+		client_tracking_ops->create(clp);
+}
+
+void
+nfsd4_client_record_remove(struct nfs4_client *clp)
+{
+	if (client_tracking_ops)
+		client_tracking_ops->remove(clp);
+}
+
+int
+nfsd4_client_record_check(struct nfs4_client *clp)
+{
+	if (client_tracking_ops)
+		return client_tracking_ops->check(clp);
+
+	return -EOPNOTSUPP;
+}
+
+void
+nfsd4_record_grace_done(struct net *net, time_t boot_time)
+{
+	if (client_tracking_ops)
+		client_tracking_ops->grace_done(net, boot_time);
+}
+
+static int
+rpc_pipefs_event(struct notifier_block *nb, unsigned long event, void *ptr)
+{
+	struct super_block *sb = ptr;
+	struct net *net = sb->s_fs_info;
+	struct nfsd_net *nn = net_generic(net, nfsd_net_id);
+	struct cld_net *cn = nn->cld_net;
+	struct dentry *dentry;
+	int ret = 0;
+
+	if (!try_module_get(THIS_MODULE))
+		return 0;
+
+	if (!cn) {
+		module_put(THIS_MODULE);
+		return 0;
+	}
+
+	switch (event) {
+	case RPC_PIPEFS_MOUNT:
+		dentry = nfsd4_cld_register_sb(sb, cn->cn_pipe);
+		if (IS_ERR(dentry)) {
+			ret = PTR_ERR(dentry);
+			break;
+		}
+		cn->cn_pipe->dentry = dentry;
+		break;
+	case RPC_PIPEFS_UMOUNT:
+		if (cn->cn_pipe->dentry)
+			nfsd4_cld_unregister_sb(cn->cn_pipe);
+		break;
+	default:
+		ret = -ENOTSUPP;
+		break;
+	}
+	module_put(THIS_MODULE);
+	return ret;
+}
+
+struct notifier_block nfsd4_cld_block = {
+	.notifier_call = rpc_pipefs_event,
+};
+
+int
+register_cld_notifier(void)
+{
+	return rpc_pipefs_notifier_register(&nfsd4_cld_block);
+}
+
+void
+unregister_cld_notifier(void)
+{
+	rpc_pipefs_notifier_unregister(&nfsd4_cld_block);
+}

fs/nfsd/nfs4state.c

@@ -58,11 +58,15 @@ static const stateid_t one_stateid = {
 static const stateid_t zero_stateid = {
 	/* all fields zero */
 };
+static const stateid_t currentstateid = {
+	.si_generation = 1,
+};
 
 static u64 current_sessionid = 1;
 
 #define ZERO_STATEID(stateid) (!memcmp((stateid), &zero_stateid, sizeof(stateid_t)))
 #define ONE_STATEID(stateid)  (!memcmp((stateid), &one_stateid, sizeof(stateid_t)))
+#define CURRENT_STATEID(stateid) (!memcmp((stateid), &currentstateid, sizeof(stateid_t)))
 
 /* forward declarations */
 static int check_for_locks(struct nfs4_file *filp, struct nfs4_lockowner *lowner);
@@ -91,6 +95,19 @@ nfs4_lock_state(void)
 	mutex_lock(&client_mutex);
 }
 
+static void free_session(struct kref *);
+
+/* Must be called under the client_lock */
+static void nfsd4_put_session_locked(struct nfsd4_session *ses)
+{
+	kref_put(&ses->se_ref, free_session);
+}
+
+static void nfsd4_get_session(struct nfsd4_session *ses)
+{
+	kref_get(&ses->se_ref);
+}
+
 void
 nfs4_unlock_state(void)
 {
@@ -605,12 +622,20 @@ hash_sessionid(struct nfs4_sessionid *sessionid)
 	return sid->sequence % SESSION_HASH_SIZE;
 }
 
+#ifdef NFSD_DEBUG
 static inline void
 dump_sessionid(const char *fn, struct nfs4_sessionid *sessionid)
 {
 	u32 *ptr = (u32 *)(&sessionid->data[0]);
 	dprintk("%s: %u:%u:%u:%u\n", fn, ptr[0], ptr[1], ptr[2], ptr[3]);
 }
+#else
+static inline void
+dump_sessionid(const char *fn, struct nfs4_sessionid *sessionid)
+{
+}
+#endif
+
 
 static void
 gen_sessionid(struct nfsd4_session *ses)
@@ -832,11 +857,12 @@ static void nfsd4_del_conns(struct nfsd4_session *s)
 	spin_unlock(&clp->cl_lock);
 }
 
-void free_session(struct kref *kref)
+static void free_session(struct kref *kref)
 {
 	struct nfsd4_session *ses;
 	int mem;
 
+	BUG_ON(!spin_is_locked(&client_lock));
 	ses = container_of(kref, struct nfsd4_session, se_ref);
 	nfsd4_del_conns(ses);
 	spin_lock(&nfsd_drc_lock);
@@ -847,6 +873,13 @@ void free_session(struct kref *kref)
 	kfree(ses);
 }
 
+void nfsd4_put_session(struct nfsd4_session *ses)
+{
+	spin_lock(&client_lock);
+	nfsd4_put_session_locked(ses);
+	spin_unlock(&client_lock);
+}
+
 static struct nfsd4_session *alloc_init_session(struct svc_rqst *rqstp, struct nfs4_client *clp, struct nfsd4_create_session *cses)
 {
 	struct nfsd4_session *new;
@@ -894,7 +927,9 @@ static struct nfsd4_session *alloc_init_session(struct svc_rqst *rqstp, struct n
 	status = nfsd4_new_conn_from_crses(rqstp, new);
 	/* whoops: benny points out, status is ignored! (err, or bogus) */
 	if (status) {
+		spin_lock(&client_lock);
 		free_session(&new->se_ref);
+		spin_unlock(&client_lock);
 		return NULL;
 	}
 	if (cses->flags & SESSION4_BACK_CHAN) {
@@ -1006,12 +1041,13 @@ static struct nfs4_client *alloc_client(struct xdr_netobj name)
 static inline void
 free_client(struct nfs4_client *clp)
 {
+	BUG_ON(!spin_is_locked(&client_lock));
 	while (!list_empty(&clp->cl_sessions)) {
 		struct nfsd4_session *ses;
 		ses = list_entry(clp->cl_sessions.next, struct nfsd4_session,
 				se_perclnt);
 		list_del(&ses->se_perclnt);
-		nfsd4_put_session(ses);
+		nfsd4_put_session_locked(ses);
 	}
 	if (clp->cl_cred.cr_group_info)
 		put_group_info(clp->cl_cred.cr_group_info);
@@ -1138,12 +1174,12 @@ static void gen_clid(struct nfs4_client *clp)
 
 static void gen_confirm(struct nfs4_client *clp)
 {
+	__be32 verf[2];
 	static u32 i;
-	u32 *p;
 
-	p = (u32 *)clp->cl_confirm.data;
-	*p++ = get_seconds();
-	*p++ = i++;
+	verf[0] = (__be32)get_seconds();
+	verf[1] = (__be32)i++;
+	memcpy(clp->cl_confirm.data, verf, sizeof(clp->cl_confirm.data));
 }
 
 static struct nfs4_stid *find_stateid(struct nfs4_client *cl, stateid_t *t)
@@ -1180,7 +1216,9 @@ static struct nfs4_client *create_client(struct xdr_netobj name, char *recdir,
 	if (princ) {
 		clp->cl_principal = kstrdup(princ, GFP_KERNEL);
 		if (clp->cl_principal == NULL) {
+			spin_lock(&client_lock);
 			free_client(clp);
+			spin_unlock(&client_lock);
 			return NULL;
 		}
 	}
@@ -1347,6 +1385,7 @@ nfsd4_store_cache_entry(struct nfsd4_compoundres *resp)
 	slot->sl_opcnt = resp->opcnt;
 	slot->sl_status = resp->cstate.status;
 
+	slot->sl_flags |= NFSD4_SLOT_INITIALIZED;
 	if (nfsd4_not_cached(resp)) {
 		slot->sl_datalen = 0;
 		return;
@@ -1374,15 +1413,12 @@ nfsd4_enc_sequence_replay(struct nfsd4_compoundargs *args,
 	struct nfsd4_op *op;
 	struct nfsd4_slot *slot = resp->cstate.slot;
 
-	dprintk("--> %s resp->opcnt %d cachethis %u \n", __func__,
-		resp->opcnt, resp->cstate.slot->sl_cachethis);
-
 	/* Encode the replayed sequence operation */
 	op = &args->ops[resp->opcnt - 1];
 	nfsd4_encode_operation(resp, op);
 
 	/* Return nfserr_retry_uncached_rep in next operation. */
-	if (args->opcnt > 1 && slot->sl_cachethis == 0) {
+	if (args->opcnt > 1 && !(slot->sl_flags & NFSD4_SLOT_CACHETHIS)) {
 		op = &args->ops[resp->opcnt++];
 		op->status = nfserr_retry_uncached_rep;
 		nfsd4_encode_operation(resp, op);
@@ -1575,16 +1611,11 @@ check_slot_seqid(u32 seqid, u32 slot_seqid, int slot_inuse)
 		else
 			return nfserr_seq_misordered;
 	}
-	/* Normal */
+	/* Note unsigned 32-bit arithmetic handles wraparound: */
 	if (likely(seqid == slot_seqid + 1))
 		return nfs_ok;
-	/* Replay */
 	if (seqid == slot_seqid)
 		return nfserr_replay_cache;
-	/* Wraparound */
-	if (seqid == 1 && (slot_seqid + 1) == 0)
-		return nfs_ok;
-	/* Misordered replay or misordered new request */
 	return nfserr_seq_misordered;
 }
 
@@ -1815,9 +1846,10 @@ nfsd4_destroy_session(struct svc_rqst *r,
 	nfsd4_probe_callback_sync(ses->se_client);
 	nfs4_unlock_state();
 
+	spin_lock(&client_lock);
 	nfsd4_del_conns(ses);
-
-	nfsd4_put_session(ses);
+	nfsd4_put_session_locked(ses);
+	spin_unlock(&client_lock);
 	status = nfs_ok;
 out:
 	dprintk("%s returns %d\n", __func__, ntohl(status));
@@ -1921,8 +1953,12 @@ nfsd4_sequence(struct svc_rqst *rqstp,
 	 * sr_highest_slotid and the sr_target_slot id to maxslots */
 	seq->maxslots = session->se_fchannel.maxreqs;
 
-	status = check_slot_seqid(seq->seqid, slot->sl_seqid, slot->sl_inuse);
+	status = check_slot_seqid(seq->seqid, slot->sl_seqid,
+					slot->sl_flags & NFSD4_SLOT_INUSE);
 	if (status == nfserr_replay_cache) {
+		status = nfserr_seq_misordered;
+		if (!(slot->sl_flags & NFSD4_SLOT_INITIALIZED))
+			goto out;
 		cstate->slot = slot;
 		cstate->session = session;
 		/* Return the cached reply status and set cstate->status
@@ -1938,9 +1974,12 @@ nfsd4_sequence(struct svc_rqst *rqstp,
 	conn = NULL;
 
 	/* Success! bump slot seqid */
-	slot->sl_inuse = true;
 	slot->sl_seqid = seq->seqid;
-	slot->sl_cachethis = seq->cachethis;
+	slot->sl_flags |= NFSD4_SLOT_INUSE;
+	if (seq->cachethis)
+		slot->sl_flags |= NFSD4_SLOT_CACHETHIS;
+	else
+		slot->sl_flags &= ~NFSD4_SLOT_CACHETHIS;
 
 	cstate->slot = slot;
 	cstate->session = session;
@@ -2030,7 +2069,8 @@ nfsd4_reclaim_complete(struct svc_rqst *rqstp, struct nfsd4_compound_state *csta
 
 	nfs4_lock_state();
 	status = nfserr_complete_already;
-	if (cstate->session->se_client->cl_firststate)
+	if (test_and_set_bit(NFSD4_CLIENT_RECLAIM_COMPLETE,
+			     &cstate->session->se_client->cl_flags))
 		goto out;
 
 	status = nfserr_stale_clientid;
@@ -2045,7 +2085,7 @@ nfsd4_reclaim_complete(struct svc_rqst *rqstp, struct nfsd4_compound_state *csta
 		goto out;
 
 	status = nfs_ok;
-	nfsd4_create_clid_dir(cstate->session->se_client);
+	nfsd4_client_record_create(cstate->session->se_client);
 out:
 	nfs4_unlock_state();
 	return status;
@@ -2240,7 +2280,7 @@ nfsd4_setclientid_confirm(struct svc_rqst *rqstp,
 			conf = find_confirmed_client_by_str(unconf->cl_recdir,
 							    hash);
 			if (conf) {
-				nfsd4_remove_clid_dir(conf);
+				nfsd4_client_record_remove(conf);
 				expire_client(conf);
 			}
 			move_to_confirmed(unconf);
@@ -2633,8 +2673,6 @@ nfs4_check_delegmode(struct nfs4_delegation *dp, int flags)
 
 static int share_access_to_flags(u32 share_access)
 {
-	share_access &= ~NFS4_SHARE_WANT_MASK;
-
 	return share_access == NFS4_SHARE_ACCESS_READ ? RD_STATE : WR_STATE;
 }
 
@@ -2776,10 +2814,9 @@ nfs4_upgrade_open(struct svc_rqst *rqstp, struct nfs4_file *fp, struct svc_fh *c
 
 
 static void
-nfs4_set_claim_prev(struct nfsd4_open *open)
+nfs4_set_claim_prev(struct nfsd4_open *open, bool has_session)
 {
 	open->op_openowner->oo_flags |= NFS4_OO_CONFIRMED;
-	open->op_openowner->oo_owner.so_client->cl_firststate = 1;
 }
 
 /* Should we give out recallable state?: */
@@ -2855,6 +2892,27 @@ static int nfs4_set_delegation(struct nfs4_delegation *dp, int flag)
 	return 0;
 }
 
+static void nfsd4_open_deleg_none_ext(struct nfsd4_open *open, int status)
+{
+	open->op_delegate_type = NFS4_OPEN_DELEGATE_NONE_EXT;
+	if (status == -EAGAIN)
+		open->op_why_no_deleg = WND4_CONTENTION;
+	else {
+		open->op_why_no_deleg = WND4_RESOURCE;
+		switch (open->op_deleg_want) {
+		case NFS4_SHARE_WANT_READ_DELEG:
+		case NFS4_SHARE_WANT_WRITE_DELEG:
+		case NFS4_SHARE_WANT_ANY_DELEG:
+			break;
+		case NFS4_SHARE_WANT_CANCEL:
+			open->op_why_no_deleg = WND4_CANCELLED;
+			break;
+		case NFS4_SHARE_WANT_NO_DELEG:
+			BUG();	/* not supposed to get here */
+		}
+	}
+}
+
 /*
  * Attempt to hand out a delegation.
  */
@@ -2864,7 +2922,7 @@ nfs4_open_delegation(struct svc_fh *fh, struct nfsd4_open *open, struct nfs4_ol_
 	struct nfs4_delegation *dp;
 	struct nfs4_openowner *oo = container_of(stp->st_stateowner, struct nfs4_openowner, oo_owner);
 	int cb_up;
-	int status, flag = 0;
+	int status = 0, flag = 0;
 
 	cb_up = nfsd4_cb_channel_good(oo->oo_owner.so_client);
 	flag = NFS4_OPEN_DELEGATE_NONE;
@@ -2905,11 +2963,16 @@ nfs4_open_delegation(struct svc_fh *fh, struct nfsd4_open *open, struct nfs4_ol_
 	dprintk("NFSD: delegation stateid=" STATEID_FMT "\n",
 		STATEID_VAL(&dp->dl_stid.sc_stateid));
 out:
-	if (open->op_claim_type == NFS4_OPEN_CLAIM_PREVIOUS
-			&& flag == NFS4_OPEN_DELEGATE_NONE
-			&& open->op_delegate_type != NFS4_OPEN_DELEGATE_NONE)
-		dprintk("NFSD: WARNING: refusing delegation reclaim\n");
 	open->op_delegate_type = flag;
+	if (flag == NFS4_OPEN_DELEGATE_NONE) {
+		if (open->op_claim_type == NFS4_OPEN_CLAIM_PREVIOUS &&
+		    open->op_delegate_type != NFS4_OPEN_DELEGATE_NONE)
+			dprintk("NFSD: WARNING: refusing delegation reclaim\n");
+
+		/* 4.1 client asking for a delegation? */
+		if (open->op_deleg_want)
+			nfsd4_open_deleg_none_ext(open, status);
+	}
 	return;
 out_free:
 	nfs4_put_delegation(dp);
@@ -2918,6 +2981,24 @@ nfs4_open_delegation(struct svc_fh *fh, struct nfsd4_open *open, struct nfs4_ol_
 	goto out;
 }
 
+static void nfsd4_deleg_xgrade_none_ext(struct nfsd4_open *open,
+					struct nfs4_delegation *dp)
+{
+	if (open->op_deleg_want == NFS4_SHARE_WANT_READ_DELEG &&
+	    dp->dl_type == NFS4_OPEN_DELEGATE_WRITE) {
+		open->op_delegate_type = NFS4_OPEN_DELEGATE_NONE_EXT;
+		open->op_why_no_deleg = WND4_NOT_SUPP_DOWNGRADE;
+	} else if (open->op_deleg_want == NFS4_SHARE_WANT_WRITE_DELEG &&
+		   dp->dl_type == NFS4_OPEN_DELEGATE_WRITE) {
+		open->op_delegate_type = NFS4_OPEN_DELEGATE_NONE_EXT;
+		open->op_why_no_deleg = WND4_NOT_SUPP_UPGRADE;
+	}
+	/* Otherwise the client must be confused wanting a delegation
+	 * it already has, therefore we don't return
+	 * NFS4_OPEN_DELEGATE_NONE_EXT and reason.
+	 */
+}
+
 /*
  * called with nfs4_lock_state() held.
  */
@@ -2979,24 +3060,36 @@ nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nf
 	update_stateid(&stp->st_stid.sc_stateid);
 	memcpy(&open->op_stateid, &stp->st_stid.sc_stateid, sizeof(stateid_t));
 
-	if (nfsd4_has_session(&resp->cstate))
+	if (nfsd4_has_session(&resp->cstate)) {
 		open->op_openowner->oo_flags |= NFS4_OO_CONFIRMED;
 
+		if (open->op_deleg_want & NFS4_SHARE_WANT_NO_DELEG) {
+			open->op_delegate_type = NFS4_OPEN_DELEGATE_NONE_EXT;
+			open->op_why_no_deleg = WND4_NOT_WANTED;
+			goto nodeleg;
+		}
+	}
+
 	/*
 	* Attempt to hand out a delegation. No error return, because the
 	* OPEN succeeds even if we fail.
 	*/
 	nfs4_open_delegation(current_fh, open, stp);
-
+nodeleg:
 	status = nfs_ok;
 
 	dprintk("%s: stateid=" STATEID_FMT "\n", __func__,
 		STATEID_VAL(&stp->st_stid.sc_stateid));
 out:
+	/* 4.1 client trying to upgrade/downgrade delegation? */
+	if (open->op_delegate_type == NFS4_OPEN_DELEGATE_NONE && dp &&
+	    open->op_deleg_want)
+		nfsd4_deleg_xgrade_none_ext(open, dp);
+
 	if (fp)
 		put_nfs4_file(fp);
 	if (status == 0 && open->op_claim_type == NFS4_OPEN_CLAIM_PREVIOUS)
-		nfs4_set_claim_prev(open);
+		nfs4_set_claim_prev(open, nfsd4_has_session(&resp->cstate));
 	/*
 	* To finish the open response, we just need to set the rflags.
 	*/
@@ -3066,7 +3159,7 @@ static void
 nfsd4_end_grace(void)
 {
 	dprintk("NFSD: end of grace period\n");
-	nfsd4_recdir_purge_old();
+	nfsd4_record_grace_done(&init_net, boot_time);
 	locks_end_grace(&nfsd4_manager);
 	/*
 	 * Now that every NFSv4 client has had the chance to recover and
@@ -3115,7 +3208,7 @@ nfs4_laundromat(void)
 		clp = list_entry(pos, struct nfs4_client, cl_lru);
 		dprintk("NFSD: purging unused client (clientid %08x)\n",
 			clp->cl_clientid.cl_id);
-		nfsd4_remove_clid_dir(clp);
+		nfsd4_client_record_remove(clp);
 		expire_client(clp);
 	}
 	spin_lock(&recall_lock);
@@ -3400,7 +3493,14 @@ __be32
 nfsd4_test_stateid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		   struct nfsd4_test_stateid *test_stateid)
 {
-	/* real work is done during encoding */
+	struct nfsd4_test_stateid_id *stateid;
+	struct nfs4_client *cl = cstate->session->se_client;
+
+	nfs4_lock_state();
+	list_for_each_entry(stateid, &test_stateid->ts_stateid_list, ts_id_list)
+		stateid->ts_id_status = nfs4_validate_stateid(cl, &stateid->ts_id_stateid);
+	nfs4_unlock_state();
+
 	return nfs_ok;
 }
 
@@ -3539,7 +3639,7 @@ nfsd4_open_confirm(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	dprintk("NFSD: %s: success, seqid=%d stateid=" STATEID_FMT "\n",
 		__func__, oc->oc_seqid, STATEID_VAL(&stp->st_stid.sc_stateid));
 
-	nfsd4_create_clid_dir(oo->oo_owner.so_client);
+	nfsd4_client_record_create(oo->oo_owner.so_client);
 	status = nfs_ok;
 out:
 	if (!cstate->replay_owner)
@@ -3596,7 +3696,9 @@ nfsd4_open_downgrade(struct svc_rqst *rqstp,
 			cstate->current_fh.fh_dentry->d_name.name);
 
 	/* We don't yet support WANT bits: */
-	od->od_share_access &= NFS4_SHARE_ACCESS_MASK;
+	if (od->od_deleg_want)
+		dprintk("NFSD: %s: od_deleg_want=0x%x ignored\n", __func__,
+			od->od_deleg_want);
 
 	nfs4_lock_state();
 	status = nfs4_preprocess_confirmed_seqid_op(cstate, od->od_seqid,
@@ -4353,7 +4455,9 @@ nfs4_has_reclaimed_state(const char *name, bool use_exchange_id)
 	struct nfs4_client *clp;
 
 	clp = find_confirmed_client_by_str(name, strhashval);
-	return clp ? 1 : 0;
+	if (!clp)
+		return 0;
+	return test_bit(NFSD4_CLIENT_STABLE, &clp->cl_flags);
 }
 
 /*
@@ -4377,7 +4481,7 @@ nfs4_client_to_reclaim(const char *name)
 	return 1;
 }
 
-static void
+void
 nfs4_release_reclaim(void)
 {
 	struct nfs4_client_reclaim *crp = NULL;
@@ -4397,19 +4501,12 @@ nfs4_release_reclaim(void)
 
 /*
  * called from OPEN, CLAIM_PREVIOUS with a new clientid. */
-static struct nfs4_client_reclaim *
-nfs4_find_reclaim_client(clientid_t *clid)
+struct nfs4_client_reclaim *
+nfsd4_find_reclaim_client(struct nfs4_client *clp)
 {
 	unsigned int strhashval;
-	struct nfs4_client *clp;
 	struct nfs4_client_reclaim *crp = NULL;
 
-
-	/* find clientid in conf_id_hashtbl */
-	clp = find_confirmed_client(clid);
-	if (clp == NULL)
-		return NULL;
-
 	dprintk("NFSD: nfs4_find_reclaim_client for %.*s with recdir %s\n",
 		            clp->cl_name.len, clp->cl_name.data,
 			    clp->cl_recdir);
@@ -4430,7 +4527,14 @@ nfs4_find_reclaim_client(clientid_t *clid)
 __be32
 nfs4_check_open_reclaim(clientid_t *clid)
 {
-	return nfs4_find_reclaim_client(clid) ? nfs_ok : nfserr_reclaim_bad;
+	struct nfs4_client *clp;
+
+	/* find clientid in conf_id_hashtbl */
+	clp = find_confirmed_client(clid);
+	if (clp == NULL)
+		return nfserr_reclaim_bad;
+
+	return nfsd4_client_record_check(clp) ? nfserr_reclaim_bad : nfs_ok;
 }
 
 #ifdef CONFIG_NFSD_FAULT_INJECTION
@@ -4442,7 +4546,7 @@ void nfsd_forget_clients(u64 num)
 
 	nfs4_lock_state();
 	list_for_each_entry_safe(clp, next, &client_lru, cl_lru) {
-		nfsd4_remove_clid_dir(clp);
+		nfsd4_client_record_remove(clp);
 		expire_client(clp);
 		if (++count == num)
 			break;
@@ -4577,19 +4681,6 @@ nfs4_state_init(void)
 	reclaim_str_hashtbl_size = 0;
 }
 
-static void
-nfsd4_load_reboot_recovery_data(void)
-{
-	int status;
-
-	nfs4_lock_state();
-	nfsd4_init_recdir();
-	status = nfsd4_recdir_load();
-	nfs4_unlock_state();
-	if (status)
-		printk("NFSD: Failure reading reboot recovery data\n");
-}
-
 /*
  * Since the lifetime of a delegation isn't limited to that of an open, a
  * client may quite reasonably hang on to a delegation as long as it has
@@ -4613,21 +4704,34 @@ set_max_delegations(void)
 
 /* initialization to perform when the nfsd service is started: */
 
-static int
-__nfs4_state_start(void)
+int
+nfs4_state_start(void)
 {
 	int ret;
 
+	/*
+	 * FIXME: For now, we hang most of the pernet global stuff off of
+	 * init_net until nfsd is fully containerized. Eventually, we'll
+	 * need to pass a net pointer into this function, take a reference
+	 * to that instead and then do most of the rest of this on a per-net
+	 * basis.
+	 */
+	get_net(&init_net);
+	nfsd4_client_tracking_init(&init_net);
 	boot_time = get_seconds();
 	locks_start_grace(&nfsd4_manager);
 	printk(KERN_INFO "NFSD: starting %ld-second grace period\n",
 	       nfsd4_grace);
 	ret = set_callback_cred();
-	if (ret)
-		return -ENOMEM;
+	if (ret) {
+		ret = -ENOMEM;
+		goto out_recovery;
+	}
 	laundry_wq = create_singlethread_workqueue("nfsd4");
-	if (laundry_wq == NULL)
-		return -ENOMEM;
+	if (laundry_wq == NULL) {
+		ret = -ENOMEM;
+		goto out_recovery;
+	}
 	ret = nfsd4_create_callback_queue();
 	if (ret)
 		goto out_free_laundry;
@@ -4636,16 +4740,12 @@ __nfs4_state_start(void)
 	return 0;
 out_free_laundry:
 	destroy_workqueue(laundry_wq);
+out_recovery:
+	nfsd4_client_tracking_exit(&init_net);
+	put_net(&init_net);
 	return ret;
 }
 
-int
-nfs4_state_start(void)
-{
-	nfsd4_load_reboot_recovery_data();
-	return __nfs4_state_start();
-}
-
 static void
 __nfs4_state_shutdown(void)
 {
@@ -4676,7 +4776,8 @@ __nfs4_state_shutdown(void)
 		unhash_delegation(dp);
 	}
 
-	nfsd4_shutdown_recdir();
+	nfsd4_client_tracking_exit(&init_net);
+	put_net(&init_net);
 }
 
 void
@@ -4686,8 +4787,108 @@ nfs4_state_shutdown(void)
 	destroy_workqueue(laundry_wq);
 	locks_end_grace(&nfsd4_manager);
 	nfs4_lock_state();
-	nfs4_release_reclaim();
 	__nfs4_state_shutdown();
 	nfs4_unlock_state();
 	nfsd4_destroy_callback_queue();
 }
+
+static void
+get_stateid(struct nfsd4_compound_state *cstate, stateid_t *stateid)
+{
+	if (HAS_STATE_ID(cstate, CURRENT_STATE_ID_FLAG) && CURRENT_STATEID(stateid))
+		memcpy(stateid, &cstate->current_stateid, sizeof(stateid_t));
+}
+
+static void
+put_stateid(struct nfsd4_compound_state *cstate, stateid_t *stateid)
+{
+	if (cstate->minorversion) {
+		memcpy(&cstate->current_stateid, stateid, sizeof(stateid_t));
+		SET_STATE_ID(cstate, CURRENT_STATE_ID_FLAG);
+	}
+}
+
+void
+clear_current_stateid(struct nfsd4_compound_state *cstate)
+{
+	CLEAR_STATE_ID(cstate, CURRENT_STATE_ID_FLAG);
+}
+
+/*
+ * functions to set current state id
+ */
+void
+nfsd4_set_opendowngradestateid(struct nfsd4_compound_state *cstate, struct nfsd4_open_downgrade *odp)
+{
+	put_stateid(cstate, &odp->od_stateid);
+}
+
+void
+nfsd4_set_openstateid(struct nfsd4_compound_state *cstate, struct nfsd4_open *open)
+{
+	put_stateid(cstate, &open->op_stateid);
+}
+
+void
+nfsd4_set_closestateid(struct nfsd4_compound_state *cstate, struct nfsd4_close *close)
+{
+	put_stateid(cstate, &close->cl_stateid);
+}
+
+void
+nfsd4_set_lockstateid(struct nfsd4_compound_state *cstate, struct nfsd4_lock *lock)
+{
+	put_stateid(cstate, &lock->lk_resp_stateid);
+}
+
+/*
+ * functions to consume current state id
+ */
+
+void
+nfsd4_get_opendowngradestateid(struct nfsd4_compound_state *cstate, struct nfsd4_open_downgrade *odp)
+{
+	get_stateid(cstate, &odp->od_stateid);
+}
+
+void
+nfsd4_get_delegreturnstateid(struct nfsd4_compound_state *cstate, struct nfsd4_delegreturn *drp)
+{
+	get_stateid(cstate, &drp->dr_stateid);
+}
+
+void
+nfsd4_get_freestateid(struct nfsd4_compound_state *cstate, struct nfsd4_free_stateid *fsp)
+{
+	get_stateid(cstate, &fsp->fr_stateid);
+}
+
+void
+nfsd4_get_setattrstateid(struct nfsd4_compound_state *cstate, struct nfsd4_setattr *setattr)
+{
+	get_stateid(cstate, &setattr->sa_stateid);
+}
+
+void
+nfsd4_get_closestateid(struct nfsd4_compound_state *cstate, struct nfsd4_close *close)
+{
+	get_stateid(cstate, &close->cl_stateid);
+}
+
+void
+nfsd4_get_lockustateid(struct nfsd4_compound_state *cstate, struct nfsd4_locku *locku)
+{
+	get_stateid(cstate, &locku->lu_stateid);
+}
+
+void
+nfsd4_get_readstateid(struct nfsd4_compound_state *cstate, struct nfsd4_read *read)
+{
+	get_stateid(cstate, &read->rd_stateid);
+}
+
+void
+nfsd4_get_writestateid(struct nfsd4_compound_state *cstate, struct nfsd4_write *write)
+{
+	get_stateid(cstate, &write->wr_stateid);
+}

fs/nfsd/nfs4xdr.c

@@ -133,22 +133,6 @@ xdr_error:					\
 	}					\
 } while (0)
 
-static void save_buf(struct nfsd4_compoundargs *argp, struct nfsd4_saved_compoundargs *savep)
-{
-	savep->p        = argp->p;
-	savep->end      = argp->end;
-	savep->pagelen  = argp->pagelen;
-	savep->pagelist = argp->pagelist;
-}
-
-static void restore_buf(struct nfsd4_compoundargs *argp, struct nfsd4_saved_compoundargs *savep)
-{
-	argp->p        = savep->p;
-	argp->end      = savep->end;
-	argp->pagelen  = savep->pagelen;
-	argp->pagelist = savep->pagelist;
-}
-
 static __be32 *read_buf(struct nfsd4_compoundargs *argp, u32 nbytes)
 {
 	/* We want more bytes than seem to be available.
@@ -638,14 +622,18 @@ nfsd4_decode_lookup(struct nfsd4_compoundargs *argp, struct nfsd4_lookup *lookup
 	DECODE_TAIL;
 }
 
-static __be32 nfsd4_decode_share_access(struct nfsd4_compoundargs *argp, u32 *x)
+static __be32 nfsd4_decode_share_access(struct nfsd4_compoundargs *argp, u32 *share_access, u32 *deleg_want, u32 *deleg_when)
 {
 	__be32 *p;
 	u32 w;
 
 	READ_BUF(4);
 	READ32(w);
-	*x = w;
+	*share_access = w & NFS4_SHARE_ACCESS_MASK;
+	*deleg_want = w & NFS4_SHARE_WANT_MASK;
+	if (deleg_when)
+		*deleg_when = w & NFS4_SHARE_WHEN_MASK;
+
 	switch (w & NFS4_SHARE_ACCESS_MASK) {
 	case NFS4_SHARE_ACCESS_READ:
 	case NFS4_SHARE_ACCESS_WRITE:
@@ -673,6 +661,9 @@ static __be32 nfsd4_decode_share_access(struct nfsd4_compoundargs *argp, u32 *x)
 	w &= ~NFS4_SHARE_WANT_MASK;
 	if (!w)
 		return nfs_ok;
+
+	if (!deleg_when)	/* open_downgrade */
+		return nfserr_inval;
 	switch (w) {
 	case NFS4_SHARE_SIGNAL_DELEG_WHEN_RESRC_AVAIL:
 	case NFS4_SHARE_PUSH_DELEG_WHEN_UNCONTENDED:
@@ -719,6 +710,7 @@ static __be32
 nfsd4_decode_open(struct nfsd4_compoundargs *argp, struct nfsd4_open *open)
 {
 	DECODE_HEAD;
+	u32 dummy;
 
 	memset(open->op_bmval, 0, sizeof(open->op_bmval));
 	open->op_iattr.ia_valid = 0;
@@ -727,7 +719,9 @@ nfsd4_decode_open(struct nfsd4_compoundargs *argp, struct nfsd4_open *open)
 	/* seqid, share_access, share_deny, clientid, ownerlen */
 	READ_BUF(4);
 	READ32(open->op_seqid);
-	status = nfsd4_decode_share_access(argp, &open->op_share_access);
+	/* decode, yet ignore deleg_when until supported */
+	status = nfsd4_decode_share_access(argp, &open->op_share_access,
+					   &open->op_deleg_want, &dummy);
 	if (status)
 		goto xdr_error;
 	status = nfsd4_decode_share_deny(argp, &open->op_share_deny);
@@ -755,14 +749,14 @@ nfsd4_decode_open(struct nfsd4_compoundargs *argp, struct nfsd4_open *open)
 				goto out;
 			break;
 		case NFS4_CREATE_EXCLUSIVE:
-			READ_BUF(8);
-			COPYMEM(open->op_verf.data, 8);
+			READ_BUF(NFS4_VERIFIER_SIZE);
+			COPYMEM(open->op_verf.data, NFS4_VERIFIER_SIZE);
 			break;
 		case NFS4_CREATE_EXCLUSIVE4_1:
 			if (argp->minorversion < 1)
 				goto xdr_error;
-			READ_BUF(8);
-			COPYMEM(open->op_verf.data, 8);
+			READ_BUF(NFS4_VERIFIER_SIZE);
+			COPYMEM(open->op_verf.data, NFS4_VERIFIER_SIZE);
 			status = nfsd4_decode_fattr(argp, open->op_bmval,
 				&open->op_iattr, &open->op_acl);
 			if (status)
@@ -848,7 +842,8 @@ nfsd4_decode_open_downgrade(struct nfsd4_compoundargs *argp, struct nfsd4_open_d
 		return status;
 	READ_BUF(4);
 	READ32(open_down->od_seqid);
-	status = nfsd4_decode_share_access(argp, &open_down->od_share_access);
+	status = nfsd4_decode_share_access(argp, &open_down->od_share_access,
+					   &open_down->od_deleg_want, NULL);
 	if (status)
 		return status;
 	status = nfsd4_decode_share_deny(argp, &open_down->od_share_deny);
@@ -994,8 +989,8 @@ nfsd4_decode_setclientid(struct nfsd4_compoundargs *argp, struct nfsd4_setclient
 {
 	DECODE_HEAD;
 
-	READ_BUF(8);
-	COPYMEM(setclientid->se_verf.data, 8);
+	READ_BUF(NFS4_VERIFIER_SIZE);
+	COPYMEM(setclientid->se_verf.data, NFS4_VERIFIER_SIZE);
 
 	status = nfsd4_decode_opaque(argp, &setclientid->se_name);
 	if (status)
@@ -1020,9 +1015,9 @@ nfsd4_decode_setclientid_confirm(struct nfsd4_compoundargs *argp, struct nfsd4_s
 {
 	DECODE_HEAD;
 
-	READ_BUF(8 + sizeof(nfs4_verifier));
+	READ_BUF(8 + NFS4_VERIFIER_SIZE);
 	COPYMEM(&scd_c->sc_clientid, 8);
-	COPYMEM(&scd_c->sc_confirm, sizeof(nfs4_verifier));
+	COPYMEM(&scd_c->sc_confirm, NFS4_VERIFIER_SIZE);
 
 	DECODE_TAIL;
 }
@@ -1385,26 +1380,29 @@ nfsd4_decode_sequence(struct nfsd4_compoundargs *argp,
 static __be32
 nfsd4_decode_test_stateid(struct nfsd4_compoundargs *argp, struct nfsd4_test_stateid *test_stateid)
 {
-	unsigned int nbytes;
-	stateid_t si;
 	int i;
-	__be32 *p;
-	__be32 status;
+	__be32 *p, status;
+	struct nfsd4_test_stateid_id *stateid;
 
 	READ_BUF(4);
 	test_stateid->ts_num_ids = ntohl(*p++);
 
-	nbytes = test_stateid->ts_num_ids * sizeof(stateid_t);
-	if (nbytes > (u32)((char *)argp->end - (char *)argp->p))
-		goto xdr_error;
-
-	test_stateid->ts_saved_args = argp;
-	save_buf(argp, &test_stateid->ts_savedp);
+	INIT_LIST_HEAD(&test_stateid->ts_stateid_list);
 
 	for (i = 0; i < test_stateid->ts_num_ids; i++) {
-		status = nfsd4_decode_stateid(argp, &si);
+		stateid = kmalloc(sizeof(struct nfsd4_test_stateid_id), GFP_KERNEL);
+		if (!stateid) {
+			status = PTR_ERR(stateid);
+			goto out;
+		}
+
+		defer_free(argp, kfree, stateid);
+		INIT_LIST_HEAD(&stateid->ts_id_list);
+		list_add_tail(&stateid->ts_id_list, &test_stateid->ts_stateid_list);
+
+		status = nfsd4_decode_stateid(argp, &stateid->ts_id_stateid);
 		if (status)
-			return status;
+			goto out;
 	}
 
 	status = 0;
@@ -2661,8 +2659,8 @@ nfsd4_encode_commit(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_
 	__be32 *p;
 
 	if (!nfserr) {
-		RESERVE_SPACE(8);
-		WRITEMEM(commit->co_verf.data, 8);
+		RESERVE_SPACE(NFS4_VERIFIER_SIZE);
+		WRITEMEM(commit->co_verf.data, NFS4_VERIFIER_SIZE);
 		ADJUST_ARGS();
 	}
 	return nfserr;
@@ -2851,6 +2849,20 @@ nfsd4_encode_open(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_op
 		WRITE32(0);   /* XXX: is NULL principal ok? */
 		ADJUST_ARGS();
 		break;
+	case NFS4_OPEN_DELEGATE_NONE_EXT: /* 4.1 */
+		switch (open->op_why_no_deleg) {
+		case WND4_CONTENTION:
+		case WND4_RESOURCE:
+			RESERVE_SPACE(8);
+			WRITE32(open->op_why_no_deleg);
+			WRITE32(0);	/* deleg signaling not supported yet */
+			break;
+		default:
+			RESERVE_SPACE(4);
+			WRITE32(open->op_why_no_deleg);
+		}
+		ADJUST_ARGS();
+		break;
 	default:
 		BUG();
 	}
@@ -3008,7 +3020,7 @@ nfsd4_encode_readdir(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4
 	if (resp->xbuf->page_len)
 		return nfserr_resource;
 
-	RESERVE_SPACE(8);  /* verifier */
+	RESERVE_SPACE(NFS4_VERIFIER_SIZE);
 	savep = p;
 
 	/* XXX: Following NFSv3, we ignore the READDIR verifier for now. */
@@ -3209,9 +3221,9 @@ nfsd4_encode_setclientid(struct nfsd4_compoundres *resp, __be32 nfserr, struct n
 	__be32 *p;
 
 	if (!nfserr) {
-		RESERVE_SPACE(8 + sizeof(nfs4_verifier));
+		RESERVE_SPACE(8 + NFS4_VERIFIER_SIZE);
 		WRITEMEM(&scd->se_clientid, 8);
-		WRITEMEM(&scd->se_confirm, sizeof(nfs4_verifier));
+		WRITEMEM(&scd->se_confirm, NFS4_VERIFIER_SIZE);
 		ADJUST_ARGS();
 	}
 	else if (nfserr == nfserr_clid_inuse) {
@@ -3232,7 +3244,7 @@ nfsd4_encode_write(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_w
 		RESERVE_SPACE(16);
 		WRITE32(write->wr_bytes_written);
 		WRITE32(write->wr_how_written);
-		WRITEMEM(write->wr_verifier.data, 8);
+		WRITEMEM(write->wr_verifier.data, NFS4_VERIFIER_SIZE);
 		ADJUST_ARGS();
 	}
 	return nfserr;
@@ -3391,30 +3403,17 @@ __be32
 nfsd4_encode_test_stateid(struct nfsd4_compoundres *resp, int nfserr,
 			  struct nfsd4_test_stateid *test_stateid)
 {
-	struct nfsd4_compoundargs *argp;
-	struct nfs4_client *cl = resp->cstate.session->se_client;
-	stateid_t si;
+	struct nfsd4_test_stateid_id *stateid, *next;
 	__be32 *p;
-	int i;
-	int valid;
 
-	restore_buf(test_stateid->ts_saved_args, &test_stateid->ts_savedp);
-	argp = test_stateid->ts_saved_args;
-
-	RESERVE_SPACE(4);
+	RESERVE_SPACE(4 + (4 * test_stateid->ts_num_ids));
 	*p++ = htonl(test_stateid->ts_num_ids);
-	resp->p = p;
 
-	nfs4_lock_state();
-	for (i = 0; i < test_stateid->ts_num_ids; i++) {
-		nfsd4_decode_stateid(argp, &si);
-		valid = nfs4_validate_stateid(cl, &si);
-		RESERVE_SPACE(4);
-		*p++ = htonl(valid);
-		resp->p = p;
+	list_for_each_entry_safe(stateid, next, &test_stateid->ts_stateid_list, ts_id_list) {
+		*p++ = htonl(stateid->ts_id_status);
 	}
-	nfs4_unlock_state();
 
+	ADJUST_ARGS();
 	return nfserr;
 }
 
@@ -3532,7 +3531,7 @@ int nfsd4_check_resp_size(struct nfsd4_compoundres *resp, u32 pad)
 	if (length > session->se_fchannel.maxresp_sz)
 		return nfserr_rep_too_big;
 
-	if (slot->sl_cachethis == 1 &&
+	if ((slot->sl_flags & NFSD4_SLOT_CACHETHIS) &&
 	    length > session->se_fchannel.maxresp_cached)
 		return nfserr_rep_too_big_to_cache;
 
@@ -3656,8 +3655,7 @@ nfs4svc_encode_compoundres(struct svc_rqst *rqstp, __be32 *p, struct nfsd4_compo
 	if (nfsd4_has_session(cs)) {
 		if (cs->status != nfserr_replay_cache) {
 			nfsd4_store_cache_entry(resp);
-			dprintk("%s: SET SLOT STATE TO AVAILABLE\n", __func__);
-			cs->slot->sl_inuse = false;
+			cs->slot->sl_flags &= ~NFSD4_SLOT_INUSE;
 		}
 		/* Renew the clientid on success and on replay */
 		release_session_client(cs->session);

fs/nfsd/nfsctl.c

@@ -13,12 +13,14 @@
 #include <linux/sunrpc/clnt.h>
 #include <linux/sunrpc/gss_api.h>
 #include <linux/sunrpc/gss_krb5_enctypes.h>
+#include <linux/sunrpc/rpc_pipe_fs.h>
 #include <linux/module.h>
 
 #include "idmap.h"
 #include "nfsd.h"
 #include "cache.h"
 #include "fault_inject.h"
+#include "netns.h"
 
 /*
  *	We have a single directory with several nodes in it.
@@ -1124,14 +1126,26 @@ static int create_proc_exports_entry(void)
 }
 #endif
 
+int nfsd_net_id;
+static struct pernet_operations nfsd_net_ops = {
+	.id   = &nfsd_net_id,
+	.size = sizeof(struct nfsd_net),
+};
+
 static int __init init_nfsd(void)
 {
 	int retval;
 	printk(KERN_INFO "Installing knfsd (copyright (C) 1996 okir@monad.swb.de).\n");
 
-	retval = nfsd4_init_slabs();
+	retval = register_cld_notifier();
 	if (retval)
 		return retval;
+	retval = register_pernet_subsys(&nfsd_net_ops);
+	if (retval < 0)
+		goto out_unregister_notifier;
+	retval = nfsd4_init_slabs();
+	if (retval)
+		goto out_unregister_pernet;
 	nfs4_state_init();
 	retval = nfsd_fault_inject_init(); /* nfsd fault injection controls */
 	if (retval)
@@ -1169,6 +1183,10 @@ static int __init init_nfsd(void)
 	nfsd_fault_inject_cleanup();
 out_free_slabs:
 	nfsd4_free_slabs();
+out_unregister_pernet:
+	unregister_pernet_subsys(&nfsd_net_ops);
+out_unregister_notifier:
+	unregister_cld_notifier();
 	return retval;
 }
 
@@ -1184,6 +1202,8 @@ static void __exit exit_nfsd(void)
 	nfsd4_free_slabs();
 	nfsd_fault_inject_cleanup();
 	unregister_filesystem(&nfsd_fs_type);
+	unregister_pernet_subsys(&nfsd_net_ops);
+	unregister_cld_notifier();
 }
 
 MODULE_AUTHOR("Olaf Kirch <okir@monad.swb.de>");

fs/nfsd/nfsd.h

@@ -364,12 +364,17 @@ static inline u32 nfsd_suppattrs2(u32 minorversion)
 	NFSD_WRITEABLE_ATTRS_WORD2
 
 extern int nfsd4_is_junction(struct dentry *dentry);
-#else
+extern int register_cld_notifier(void);
+extern void unregister_cld_notifier(void);
+#else /* CONFIG_NFSD_V4 */
 static inline int nfsd4_is_junction(struct dentry *dentry)
 {
 	return 0;
 }
 
+#define register_cld_notifier() 0
+#define unregister_cld_notifier() do { } while(0)
+
 #endif /* CONFIG_NFSD_V4 */
 
 #endif /* LINUX_NFSD_NFSD_H */

fs/nfsd/nfssvc.c

@@ -307,33 +307,37 @@ static void set_max_drc(void)
 	dprintk("%s nfsd_drc_max_mem %u \n", __func__, nfsd_drc_max_mem);
 }
 
-int nfsd_create_serv(void)
+static int nfsd_get_default_max_blksize(void)
 {
-	int err = 0;
+	struct sysinfo i;
+	unsigned long long target;
+	unsigned long ret;
+
+	si_meminfo(&i);
+	target = (i.totalram - i.totalhigh) << PAGE_SHIFT;
+	/*
+	 * Aim for 1/4096 of memory per thread This gives 1MB on 4Gig
+	 * machines, but only uses 32K on 128M machines.  Bottom out at
+	 * 8K on 32M and smaller.  Of course, this is only a default.
+	 */
+	target >>= 12;
+
+	ret = NFSSVC_MAXBLKSIZE;
+	while (ret > target && ret >= 8*1024*2)
+		ret /= 2;
+	return ret;
+}
 
+int nfsd_create_serv(void)
+{
 	WARN_ON(!mutex_is_locked(&nfsd_mutex));
 	if (nfsd_serv) {
 		svc_get(nfsd_serv);
 		return 0;
 	}
-	if (nfsd_max_blksize == 0) {
-		/* choose a suitable default */
-		struct sysinfo i;
-		si_meminfo(&i);
-		/* Aim for 1/4096 of memory per thread
-		 * This gives 1MB on 4Gig machines
-		 * But only uses 32K on 128M machines.
-		 * Bottom out at 8K on 32M and smaller.
-		 * Of course, this is only a default.
-		 */
-		nfsd_max_blksize = NFSSVC_MAXBLKSIZE;
-		i.totalram <<= PAGE_SHIFT - 12;
-		while (nfsd_max_blksize > i.totalram &&
-		       nfsd_max_blksize >= 8*1024*2)
-			nfsd_max_blksize /= 2;
-	}
+	if (nfsd_max_blksize == 0)
+		nfsd_max_blksize = nfsd_get_default_max_blksize();
 	nfsd_reset_versions();
-
 	nfsd_serv = svc_create_pooled(&nfsd_program, nfsd_max_blksize,
 				      nfsd_last_thread, nfsd, THIS_MODULE);
 	if (nfsd_serv == NULL)
@@ -341,7 +345,7 @@ int nfsd_create_serv(void)
 
 	set_max_drc();
 	do_gettimeofday(&nfssvc_boot);		/* record boot time */
-	return err;
+	return 0;
 }
 
 int nfsd_nrpools(void)

fs/nfsd/state.h

@@ -128,12 +128,14 @@ static inline struct nfs4_delegation *delegstateid(struct nfs4_stid *s)
 		(NFSD_CACHE_SIZE_SLOTS_PER_SESSION * NFSD_SLOT_CACHE_SIZE)
 
 struct nfsd4_slot {
-	bool	sl_inuse;
-	bool	sl_cachethis;
-	u16	sl_opcnt;
 	u32	sl_seqid;
 	__be32	sl_status;
 	u32	sl_datalen;
+	u16	sl_opcnt;
+#define NFSD4_SLOT_INUSE	(1 << 0)
+#define NFSD4_SLOT_CACHETHIS	(1 << 1)
+#define NFSD4_SLOT_INITIALIZED	(1 << 2)
+	u8	sl_flags;
 	char	sl_data[];
 };
 
@@ -196,18 +198,7 @@ struct nfsd4_session {
 	struct nfsd4_slot	*se_slots[];	/* forward channel slots */
 };
 
-static inline void
-nfsd4_put_session(struct nfsd4_session *ses)
-{
-	extern void free_session(struct kref *kref);
-	kref_put(&ses->se_ref, free_session);
-}
-
-static inline void
-nfsd4_get_session(struct nfsd4_session *ses)
-{
-	kref_get(&ses->se_ref);
-}
+extern void nfsd4_put_session(struct nfsd4_session *ses);
 
 /* formatted contents of nfs4_sessionid */
 struct nfsd4_sessionid {
@@ -245,14 +236,17 @@ struct nfs4_client {
 	struct svc_cred		cl_cred; 	/* setclientid principal */
 	clientid_t		cl_clientid;	/* generated by server */
 	nfs4_verifier		cl_confirm;	/* generated by server */
-	u32			cl_firststate;	/* recovery dir creation */
 	u32			cl_minorversion;
 
 	/* for v4.0 and v4.1 callbacks: */
 	struct nfs4_cb_conn	cl_cb_conn;
-#define NFSD4_CLIENT_CB_UPDATE	1
-#define NFSD4_CLIENT_KILL	2
-	unsigned long		cl_cb_flags;
+#define NFSD4_CLIENT_CB_UPDATE		(0)
+#define NFSD4_CLIENT_CB_KILL		(1)
+#define NFSD4_CLIENT_STABLE		(2)	/* client on stable storage */
+#define NFSD4_CLIENT_RECLAIM_COMPLETE	(3)	/* reclaim_complete done */
+#define NFSD4_CLIENT_CB_FLAG_MASK	(1 << NFSD4_CLIENT_CB_UPDATE | \
+					 1 << NFSD4_CLIENT_CB_KILL)
+	unsigned long		cl_flags;
 	struct rpc_clnt		*cl_cb_client;
 	u32			cl_cb_ident;
 #define NFSD4_CB_UP		0
@@ -463,6 +457,8 @@ extern __be32 nfs4_preprocess_stateid_op(struct nfsd4_compound_state *cstate,
 extern void nfs4_lock_state(void);
 extern void nfs4_unlock_state(void);
 extern int nfs4_in_grace(void);
+extern void nfs4_release_reclaim(void);
+extern struct nfs4_client_reclaim *nfsd4_find_reclaim_client(struct nfs4_client *crp);
 extern __be32 nfs4_check_open_reclaim(clientid_t *clid);
 extern void nfs4_free_openowner(struct nfs4_openowner *);
 extern void nfs4_free_lockowner(struct nfs4_lockowner *);
@@ -477,16 +473,17 @@ extern void nfsd4_destroy_callback_queue(void);
 extern void nfsd4_shutdown_callback(struct nfs4_client *);
 extern void nfs4_put_delegation(struct nfs4_delegation *dp);
 extern __be32 nfs4_make_rec_clidname(char *clidname, struct xdr_netobj *clname);
-extern void nfsd4_init_recdir(void);
-extern int nfsd4_recdir_load(void);
-extern void nfsd4_shutdown_recdir(void);
 extern int nfs4_client_to_reclaim(const char *name);
 extern int nfs4_has_reclaimed_state(const char *name, bool use_exchange_id);
-extern void nfsd4_recdir_purge_old(void);
-extern void nfsd4_create_clid_dir(struct nfs4_client *clp);
-extern void nfsd4_remove_clid_dir(struct nfs4_client *clp);
 extern void release_session_client(struct nfsd4_session *);
 extern __be32 nfs4_validate_stateid(struct nfs4_client *, stateid_t *);
 extern void nfsd4_purge_closed_stateid(struct nfs4_stateowner *);
 
+/* nfs4recover operations */
+extern int nfsd4_client_tracking_init(struct net *net);
+extern void nfsd4_client_tracking_exit(struct net *net);
+extern void nfsd4_client_record_create(struct nfs4_client *clp);
+extern void nfsd4_client_record_remove(struct nfs4_client *clp);
+extern int nfsd4_client_record_check(struct nfs4_client *clp);
+extern void nfsd4_record_grace_done(struct net *net, time_t boot_time);
 #endif   /* NFSD4_STATE_H */

fs/nfsd/vfs.c

@@ -737,12 +737,13 @@ static int nfsd_open_break_lease(struct inode *inode, int access)
 
 /*
  * Open an existing file or directory.
- * The access argument indicates the type of open (read/write/lock)
+ * The may_flags argument indicates the type of open (read/write/lock)
+ * and additional flags.
  * N.B. After this call fhp needs an fh_put
  */
 __be32
 nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, umode_t type,
-			int access, struct file **filp)
+			int may_flags, struct file **filp)
 {
 	struct dentry	*dentry;
 	struct inode	*inode;
@@ -757,7 +758,7 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, umode_t type,
 	 * and (hopefully) checked permission - so allow OWNER_OVERRIDE
 	 * in case a chmod has now revoked permission.
 	 */
-	err = fh_verify(rqstp, fhp, type, access | NFSD_MAY_OWNER_OVERRIDE);
+	err = fh_verify(rqstp, fhp, type, may_flags | NFSD_MAY_OWNER_OVERRIDE);
 	if (err)
 		goto out;
 
@@ -768,7 +769,7 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, umode_t type,
 	 * or any access when mandatory locking enabled
 	 */
 	err = nfserr_perm;
-	if (IS_APPEND(inode) && (access & NFSD_MAY_WRITE))
+	if (IS_APPEND(inode) && (may_flags & NFSD_MAY_WRITE))
 		goto out;
 	/*
 	 * We must ignore files (but only files) which might have mandatory
@@ -781,12 +782,12 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, umode_t type,
 	if (!inode->i_fop)
 		goto out;
 
-	host_err = nfsd_open_break_lease(inode, access);
+	host_err = nfsd_open_break_lease(inode, may_flags);
 	if (host_err) /* NOMEM or WOULDBLOCK */
 		goto out_nfserr;
 
-	if (access & NFSD_MAY_WRITE) {
-		if (access & NFSD_MAY_READ)
+	if (may_flags & NFSD_MAY_WRITE) {
+		if (may_flags & NFSD_MAY_READ)
 			flags = O_RDWR|O_LARGEFILE;
 		else
 			flags = O_WRONLY|O_LARGEFILE;
@@ -795,8 +796,15 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, umode_t type,
 			    flags, current_cred());
 	if (IS_ERR(*filp))
 		host_err = PTR_ERR(*filp);
+	else {
+		host_err = ima_file_check(*filp, may_flags);
+
+		if (may_flags & NFSD_MAY_64BIT_COOKIE)
+			(*filp)->f_mode |= FMODE_64BITHASH;
 		else
-		host_err = ima_file_check(*filp, access);
+			(*filp)->f_mode |= FMODE_32BITHASH;
+	}
+
 out_nfserr:
 	err = nfserrno(host_err);
 out:
@@ -2021,8 +2029,13 @@ nfsd_readdir(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t *offsetp,
 	__be32		err;
 	struct file	*file;
 	loff_t		offset = *offsetp;
+	int             may_flags = NFSD_MAY_READ;
+
+	/* NFSv2 only supports 32 bit cookies */
+	if (rqstp->rq_vers > 2)
+		may_flags |= NFSD_MAY_64BIT_COOKIE;
 
-	err = nfsd_open(rqstp, fhp, S_IFDIR, NFSD_MAY_READ, &file);
+	err = nfsd_open(rqstp, fhp, S_IFDIR, may_flags, &file);
 	if (err)
 		goto out;
 

fs/nfsd/vfs.h

@@ -27,6 +27,8 @@
 #define NFSD_MAY_BYPASS_GSS		0x400
 #define NFSD_MAY_READ_IF_EXEC		0x800
 
+#define NFSD_MAY_64BIT_COOKIE		0x1000 /* 64 bit readdir cookies for >= NFSv3 */
+
 #define NFSD_MAY_CREATE		(NFSD_MAY_EXEC|NFSD_MAY_WRITE)
 #define NFSD_MAY_REMOVE		(NFSD_MAY_EXEC|NFSD_MAY_WRITE|NFSD_MAY_TRUNC)
 

fs/nfsd/xdr4.h

@@ -43,6 +43,13 @@
 #define NFSD4_MAX_TAGLEN	128
 #define XDR_LEN(n)                     (((n) + 3) & ~3)
 
+#define CURRENT_STATE_ID_FLAG (1<<0)
+#define SAVED_STATE_ID_FLAG (1<<1)
+
+#define SET_STATE_ID(c, f) ((c)->sid_flags |= (f))
+#define HAS_STATE_ID(c, f) ((c)->sid_flags & (f))
+#define CLEAR_STATE_ID(c, f) ((c)->sid_flags &= ~(f))
+
 struct nfsd4_compound_state {
 	struct svc_fh		current_fh;
 	struct svc_fh		save_fh;
@@ -54,6 +61,10 @@ struct nfsd4_compound_state {
 	size_t			iovlen;
 	u32			minorversion;
 	u32			status;
+	stateid_t	current_stateid;
+	stateid_t	save_stateid;
+	/* to indicate current and saved state id presents */
+	u32		sid_flags;
 };
 
 static inline bool nfsd4_has_session(struct nfsd4_compound_state *cs)
@@ -212,16 +223,19 @@ struct nfsd4_open {
 	struct xdr_netobj op_fname;	    /* request - everything but CLAIM_PREV */
 	u32		op_delegate_type;   /* request - CLAIM_PREV only */
 	stateid_t       op_delegate_stateid; /* request - response */
+	u32		op_why_no_deleg;    /* response - DELEG_NONE_EXT only */
 	u32		op_create;     	    /* request */
 	u32		op_createmode;      /* request */
 	u32		op_bmval[3];        /* request */
 	struct iattr	iattr;              /* UNCHECKED4, GUARDED4, EXCLUSIVE4_1 */
-	nfs4_verifier	verf;               /* EXCLUSIVE4 */
+	nfs4_verifier	op_verf __attribute__((aligned(32)));
+					    /* EXCLUSIVE4 */
 	clientid_t	op_clientid;        /* request */
 	struct xdr_netobj op_owner;           /* request */
 	u32		op_seqid;           /* request */
 	u32		op_share_access;    /* request */
 	u32		op_share_deny;      /* request */
+	u32		op_deleg_want;      /* request */
 	stateid_t	op_stateid;         /* response */
 	u32		op_recall;          /* recall */
 	struct nfsd4_change_info  op_cinfo; /* response */
@@ -234,7 +248,6 @@ struct nfsd4_open {
 	struct nfs4_acl *op_acl;
 };
 #define op_iattr	iattr
-#define op_verf		verf
 
 struct nfsd4_open_confirm {
 	stateid_t	oc_req_stateid		/* request */;
@@ -245,8 +258,9 @@ struct nfsd4_open_confirm {
 struct nfsd4_open_downgrade {
 	stateid_t       od_stateid;
 	u32             od_seqid;
-	u32             od_share_access;
-	u32             od_share_deny;
+	u32             od_share_access;	/* request */
+	u32		od_deleg_want;		/* request */
+	u32             od_share_deny;		/* request */
 };
 
 
@@ -343,10 +357,15 @@ struct nfsd4_saved_compoundargs {
 	struct page **pagelist;
 };
 
+struct nfsd4_test_stateid_id {
+	__be32			ts_id_status;
+	stateid_t		ts_id_stateid;
+	struct list_head	ts_id_list;
+};
+
 struct nfsd4_test_stateid {
 	__be32		ts_num_ids;
-	struct nfsd4_compoundargs *ts_saved_args;
-	struct nfsd4_saved_compoundargs ts_savedp;
+	struct list_head ts_stateid_list;
 };
 
 struct nfsd4_free_stateid {
@@ -503,7 +522,8 @@ static inline bool nfsd4_is_solo_sequence(struct nfsd4_compoundres *resp)
 
 static inline bool nfsd4_not_cached(struct nfsd4_compoundres *resp)
 {
-	return !resp->cstate.slot->sl_cachethis || nfsd4_is_solo_sequence(resp);
+	return !(resp->cstate.slot->sl_flags & NFSD4_SLOT_CACHETHIS)
+		|| nfsd4_is_solo_sequence(resp);
 }
 
 #define NFS4_SVC_XDRSIZE		sizeof(struct nfsd4_compoundargs)

include/linux/fs.h

@@ -92,6 +92,10 @@ struct inodes_stat_t {
 /* File is opened using open(.., 3, ..) and is writeable only for ioctls
    (specialy hack for floppy.c) */
 #define FMODE_WRITE_IOCTL	((__force fmode_t)0x100)
+/* 32bit hashes as llseek() offset (for directories) */
+#define FMODE_32BITHASH         ((__force fmode_t)0x200)
+/* 64bit hashes as llseek() offset (for directories) */
+#define FMODE_64BITHASH         ((__force fmode_t)0x400)
 
 /*
  * Don't update ctime and mtime.

include/linux/nfs4.h

@@ -438,7 +438,20 @@ enum limit_by4 {
 enum open_delegation_type4 {
 	NFS4_OPEN_DELEGATE_NONE = 0,
 	NFS4_OPEN_DELEGATE_READ = 1,
-	NFS4_OPEN_DELEGATE_WRITE = 2
+	NFS4_OPEN_DELEGATE_WRITE = 2,
+	NFS4_OPEN_DELEGATE_NONE_EXT = 3, /* 4.1 */
+};
+
+enum why_no_delegation4 { /* new to v4.1 */
+	WND4_NOT_WANTED = 0,
+	WND4_CONTENTION = 1,
+	WND4_RESOURCE = 2,
+	WND4_NOT_SUPP_FTYPE = 3,
+	WND4_WRITE_DELEG_NOT_SUPP_FTYPE = 4,
+	WND4_NOT_SUPP_UPGRADE = 5,
+	WND4_NOT_SUPP_DOWNGRADE = 6,
+	WND4_CANCELLED = 7,
+	WND4_IS_DIR = 8,
 };
 
 enum lock_type4 {

include/linux/nfsd/cld.h

+/*
+ * Upcall description for nfsdcld communication
+ *
+ * Copyright (c) 2012 Red Hat, Inc.
+ * Author(s): Jeff Layton <jlayton@redhat.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#ifndef _NFSD_CLD_H
+#define _NFSD_CLD_H
+
+/* latest upcall version available */
+#define CLD_UPCALL_VERSION 1
+
+/* defined by RFC3530 */
+#define NFS4_OPAQUE_LIMIT 1024
+
+enum cld_command {
+	Cld_Create,		/* create a record for this cm_id */
+	Cld_Remove,		/* remove record of this cm_id */
+	Cld_Check,		/* is this cm_id allowed? */
+	Cld_GraceDone,		/* grace period is complete */
+};
+
+/* representation of long-form NFSv4 client ID */
+struct cld_name {
+	uint16_t	cn_len;				/* length of cm_id */
+	unsigned char	cn_id[NFS4_OPAQUE_LIMIT];	/* client-provided */
+} __attribute__((packed));
+
+/* message struct for communication with userspace */
+struct cld_msg {
+	uint8_t		cm_vers;		/* upcall version */
+	uint8_t		cm_cmd;			/* upcall command */
+	int16_t		cm_status;		/* return code */
+	uint32_t	cm_xid;			/* transaction id */
+	union {
+		int64_t		cm_gracetime;	/* grace period start time */
+		struct cld_name	cm_name;
+	} __attribute__((packed)) cm_u;
+} __attribute__((packed));
+
+#endif /* !_NFSD_CLD_H */

include/linux/sunrpc/svc_rdma.h

@@ -190,7 +190,7 @@ extern int svc_rdma_xdr_encode_error(struct svcxprt_rdma *,
 extern void svc_rdma_xdr_encode_write_list(struct rpcrdma_msg *, int);
 extern void svc_rdma_xdr_encode_reply_array(struct rpcrdma_write_array *, int);
 extern void svc_rdma_xdr_encode_array_chunk(struct rpcrdma_write_array *, int,
-					    u32, u64, u32);
+					    __be32, __be64, u32);
 extern void svc_rdma_xdr_encode_reply_header(struct svcxprt_rdma *,
 					     struct rpcrdma_msg *,
 					     struct rpcrdma_msg *,
@@ -292,7 +292,7 @@ svc_rdma_get_reply_array(struct rpcrdma_msg *rmsgp)
 	if (wr_ary) {
 		rp_ary = (struct rpcrdma_write_array *)
 			&wr_ary->
-			wc_array[wr_ary->wc_nchunks].wc_target.rs_length;
+			wc_array[ntohl(wr_ary->wc_nchunks)].wc_target.rs_length;
 
 		goto found_it;
 	}

net/sunrpc/cache.c

@@ -830,6 +830,8 @@ static ssize_t cache_do_downcall(char *kaddr, const char __user *buf,
 {
 	ssize_t ret;
 
+	if (count == 0)
+		return -EINVAL;
 	if (copy_from_user(kaddr, buf, count))
 		return -EFAULT;
 	kaddr[count] = '\0';

net/sunrpc/rpc_pipe.c

@@ -1014,6 +1014,7 @@ enum {
 	RPCAUTH_statd,
 	RPCAUTH_nfsd4_cb,
 	RPCAUTH_cache,
+	RPCAUTH_nfsd,
 	RPCAUTH_RootEOF
 };
 
@@ -1046,6 +1047,10 @@ static const struct rpc_filelist files[] = {
 		.name = "cache",
 		.mode = S_IFDIR | S_IRUGO | S_IXUGO,
 	},
+	[RPCAUTH_nfsd] = {
+		.name = "nfsd",
+		.mode = S_IFDIR | S_IRUGO | S_IXUGO,
+	},
 };
 
 /*

net/sunrpc/svcauth_unix.c

@@ -507,7 +507,7 @@ static int unix_gid_parse(struct cache_detail *cd,
 	time_t expiry;
 	struct unix_gid ug, *ugp;
 
-	if (mlen <= 0 || mesg[mlen-1] != '\n')
+	if (mesg[mlen - 1] != '\n')
 		return -EINVAL;
 	mesg[mlen-1] = 0;
 

net/sunrpc/svcsock.c

@@ -1381,8 +1381,6 @@ void svc_sock_update_bufs(struct svc_serv *serv)
 	spin_lock_bh(&serv->sv_lock);
 	list_for_each_entry(svsk, &serv->sv_permsocks, sk_xprt.xpt_list)
 		set_bit(XPT_CHNGBUF, &svsk->sk_xprt.xpt_flags);
-	list_for_each_entry(svsk, &serv->sv_tempsocks, sk_xprt.xpt_list)
-		set_bit(XPT_CHNGBUF, &svsk->sk_xprt.xpt_flags);
 	spin_unlock_bh(&serv->sv_lock);
 }
 EXPORT_SYMBOL_GPL(svc_sock_update_bufs);

net/sunrpc/xprtrdma/svc_rdma.c

@@ -47,6 +47,7 @@
 #include <linux/sunrpc/clnt.h>
 #include <linux/sunrpc/sched.h>
 #include <linux/sunrpc/svc_rdma.h>
+#include "xprt_rdma.h"
 
 #define RPCDBG_FACILITY	RPCDBG_SVCXPRT
 

net/sunrpc/xprtrdma/svc_rdma_marshal.c

@@ -60,21 +60,11 @@ static u32 *decode_read_list(u32 *va, u32 *vaend)
 	struct rpcrdma_read_chunk *ch = (struct rpcrdma_read_chunk *)va;
 
 	while (ch->rc_discrim != xdr_zero) {
-		u64 ch_offset;
-
 		if (((unsigned long)ch + sizeof(struct rpcrdma_read_chunk)) >
 		    (unsigned long)vaend) {
 			dprintk("svcrdma: vaend=%p, ch=%p\n", vaend, ch);
 			return NULL;
 		}
-
-		ch->rc_discrim = ntohl(ch->rc_discrim);
-		ch->rc_position = ntohl(ch->rc_position);
-		ch->rc_target.rs_handle = ntohl(ch->rc_target.rs_handle);
-		ch->rc_target.rs_length = ntohl(ch->rc_target.rs_length);
-		va = (u32 *)&ch->rc_target.rs_offset;
-		xdr_decode_hyper(va, &ch_offset);
-		put_unaligned(ch_offset, (u64 *)va);
 		ch++;
 	}
 	return (u32 *)&ch->rc_position;
@@ -91,7 +81,7 @@ void svc_rdma_rcl_chunk_counts(struct rpcrdma_read_chunk *ch,
 	*byte_count = 0;
 	*ch_count = 0;
 	for (; ch->rc_discrim != 0; ch++) {
-		*byte_count = *byte_count + ch->rc_target.rs_length;
+		*byte_count = *byte_count + ntohl(ch->rc_target.rs_length);
 		*ch_count = *ch_count + 1;
 	}
 }
@@ -108,7 +98,8 @@ void svc_rdma_rcl_chunk_counts(struct rpcrdma_read_chunk *ch,
  */
 static u32 *decode_write_list(u32 *va, u32 *vaend)
 {
-	int ch_no;
+	int nchunks;
+
 	struct rpcrdma_write_array *ary =
 		(struct rpcrdma_write_array *)va;
 
@@ -121,37 +112,24 @@ static u32 *decode_write_list(u32 *va, u32 *vaend)
 		dprintk("svcrdma: ary=%p, vaend=%p\n", ary, vaend);
 		return NULL;
 	}
-	ary->wc_discrim = ntohl(ary->wc_discrim);
-	ary->wc_nchunks = ntohl(ary->wc_nchunks);
+	nchunks = ntohl(ary->wc_nchunks);
 	if (((unsigned long)&ary->wc_array[0] +
-	     (sizeof(struct rpcrdma_write_chunk) * ary->wc_nchunks)) >
+	     (sizeof(struct rpcrdma_write_chunk) * nchunks)) >
 	    (unsigned long)vaend) {
 		dprintk("svcrdma: ary=%p, wc_nchunks=%d, vaend=%p\n",
-			ary, ary->wc_nchunks, vaend);
+			ary, nchunks, vaend);
 		return NULL;
 	}
-	for (ch_no = 0; ch_no < ary->wc_nchunks; ch_no++) {
-		u64 ch_offset;
-
-		ary->wc_array[ch_no].wc_target.rs_handle =
-			ntohl(ary->wc_array[ch_no].wc_target.rs_handle);
-		ary->wc_array[ch_no].wc_target.rs_length =
-			ntohl(ary->wc_array[ch_no].wc_target.rs_length);
-		va = (u32 *)&ary->wc_array[ch_no].wc_target.rs_offset;
-		xdr_decode_hyper(va, &ch_offset);
-		put_unaligned(ch_offset, (u64 *)va);
-	}
-
 	/*
 	 * rs_length is the 2nd 4B field in wc_target and taking its
 	 * address skips the list terminator
 	 */
-	return (u32 *)&ary->wc_array[ch_no].wc_target.rs_length;
+	return (u32 *)&ary->wc_array[nchunks].wc_target.rs_length;
 }
 
 static u32 *decode_reply_array(u32 *va, u32 *vaend)
 {
-	int ch_no;
+	int nchunks;
 	struct rpcrdma_write_array *ary =
 		(struct rpcrdma_write_array *)va;
 
@@ -164,28 +142,15 @@ static u32 *decode_reply_array(u32 *va, u32 *vaend)
 		dprintk("svcrdma: ary=%p, vaend=%p\n", ary, vaend);
 		return NULL;
 	}
-	ary->wc_discrim = ntohl(ary->wc_discrim);
-	ary->wc_nchunks = ntohl(ary->wc_nchunks);
+	nchunks = ntohl(ary->wc_nchunks);
 	if (((unsigned long)&ary->wc_array[0] +
-	     (sizeof(struct rpcrdma_write_chunk) * ary->wc_nchunks)) >
+	     (sizeof(struct rpcrdma_write_chunk) * nchunks)) >
 	    (unsigned long)vaend) {
 		dprintk("svcrdma: ary=%p, wc_nchunks=%d, vaend=%p\n",
-			ary, ary->wc_nchunks, vaend);
+			ary, nchunks, vaend);
 		return NULL;
 	}
-	for (ch_no = 0; ch_no < ary->wc_nchunks; ch_no++) {
-		u64 ch_offset;
-
-		ary->wc_array[ch_no].wc_target.rs_handle =
-			ntohl(ary->wc_array[ch_no].wc_target.rs_handle);
-		ary->wc_array[ch_no].wc_target.rs_length =
-			ntohl(ary->wc_array[ch_no].wc_target.rs_length);
-		va = (u32 *)&ary->wc_array[ch_no].wc_target.rs_offset;
-		xdr_decode_hyper(va, &ch_offset);
-		put_unaligned(ch_offset, (u64 *)va);
-	}
-
-	return (u32 *)&ary->wc_array[ch_no];
+	return (u32 *)&ary->wc_array[nchunks];
 }
 
 int svc_rdma_xdr_decode_req(struct rpcrdma_msg **rdma_req,
@@ -386,13 +351,14 @@ void svc_rdma_xdr_encode_reply_array(struct rpcrdma_write_array *ary,
 
 void svc_rdma_xdr_encode_array_chunk(struct rpcrdma_write_array *ary,
 				     int chunk_no,
-				     u32 rs_handle, u64 rs_offset,
+				     __be32 rs_handle,
+				     __be64 rs_offset,
 				     u32 write_len)
 {
 	struct rpcrdma_segment *seg = &ary->wc_array[chunk_no].wc_target;
-	seg->rs_handle = htonl(rs_handle);
+	seg->rs_handle = rs_handle;
+	seg->rs_offset = rs_offset;
 	seg->rs_length = htonl(write_len);
-	xdr_encode_hyper((u32 *) &seg->rs_offset, rs_offset);
 }
 
 void svc_rdma_xdr_encode_reply_header(struct svcxprt_rdma *xprt,

net/sunrpc/xprtrdma/svc_rdma_recvfrom.c

@@ -147,7 +147,7 @@ static int map_read_chunks(struct svcxprt_rdma *xprt,
 	page_off = 0;
 	ch = (struct rpcrdma_read_chunk *)&rmsgp->rm_body.rm_chunks[0];
 	ch_no = 0;
-	ch_bytes = ch->rc_target.rs_length;
+	ch_bytes = ntohl(ch->rc_target.rs_length);
 	head->arg.head[0] = rqstp->rq_arg.head[0];
 	head->arg.tail[0] = rqstp->rq_arg.tail[0];
 	head->arg.pages = &head->pages[head->count];
@@ -183,7 +183,7 @@ static int map_read_chunks(struct svcxprt_rdma *xprt,
 			ch_no++;
 			ch++;
 			chl_map->ch[ch_no].start = sge_no;
-			ch_bytes = ch->rc_target.rs_length;
+			ch_bytes = ntohl(ch->rc_target.rs_length);
 			/* If bytes remaining account for next chunk */
 			if (byte_count) {
 				head->arg.page_len += ch_bytes;
@@ -281,11 +281,12 @@ static int fast_reg_read_chunks(struct svcxprt_rdma *xprt,
 	offset = 0;
 	ch = (struct rpcrdma_read_chunk *)&rmsgp->rm_body.rm_chunks[0];
 	for (ch_no = 0; ch_no < ch_count; ch_no++) {
+		int len = ntohl(ch->rc_target.rs_length);
 		rpl_map->sge[ch_no].iov_base = frmr->kva + offset;
-		rpl_map->sge[ch_no].iov_len = ch->rc_target.rs_length;
+		rpl_map->sge[ch_no].iov_len = len;
 		chl_map->ch[ch_no].count = 1;
 		chl_map->ch[ch_no].start = ch_no;
-		offset += ch->rc_target.rs_length;
+		offset += len;
 		ch++;
 	}
 
@@ -316,7 +317,7 @@ static int rdma_set_ctxt_sge(struct svcxprt_rdma *xprt,
 	for (i = 0; i < count; i++) {
 		ctxt->sge[i].length = 0; /* in case map fails */
 		if (!frmr) {
-			BUG_ON(0 == virt_to_page(vec[i].iov_base));
+			BUG_ON(!virt_to_page(vec[i].iov_base));
 			off = (unsigned long)vec[i].iov_base & ~PAGE_MASK;
 			ctxt->sge[i].addr =
 				ib_dma_map_page(xprt->sc_cm_id->device,
@@ -426,6 +427,7 @@ static int rdma_read_xdr(struct svcxprt_rdma *xprt,
 
 	for (ch = (struct rpcrdma_read_chunk *)&rmsgp->rm_body.rm_chunks[0];
 	     ch->rc_discrim != 0; ch++, ch_no++) {
+		u64 rs_offset;
 next_sge:
 		ctxt = svc_rdma_get_context(xprt);
 		ctxt->direction = DMA_FROM_DEVICE;
@@ -440,10 +442,10 @@ static int rdma_read_xdr(struct svcxprt_rdma *xprt,
 		read_wr.opcode = IB_WR_RDMA_READ;
 		ctxt->wr_op = read_wr.opcode;
 		read_wr.send_flags = IB_SEND_SIGNALED;
-		read_wr.wr.rdma.rkey = ch->rc_target.rs_handle;
-		read_wr.wr.rdma.remote_addr =
-			get_unaligned(&(ch->rc_target.rs_offset)) +
-			sgl_offset;
+		read_wr.wr.rdma.rkey = ntohl(ch->rc_target.rs_handle);
+		xdr_decode_hyper((__be32 *)&ch->rc_target.rs_offset,
+				 &rs_offset);
+		read_wr.wr.rdma.remote_addr = rs_offset + sgl_offset;
 		read_wr.sg_list = ctxt->sge;
 		read_wr.num_sge =
 			rdma_read_max_sge(xprt, chl_map->ch[ch_no].count);

net/sunrpc/xprtrdma/svc_rdma_sendto.c

@@ -409,21 +409,21 @@ static int send_write_chunks(struct svcxprt_rdma *xprt,
 		u64 rs_offset;
 
 		arg_ch = &arg_ary->wc_array[chunk_no].wc_target;
-		write_len = min(xfer_len, arg_ch->rs_length);
+		write_len = min(xfer_len, ntohl(arg_ch->rs_length));
 
 		/* Prepare the response chunk given the length actually
 		 * written */
-		rs_offset = get_unaligned(&(arg_ch->rs_offset));
+		xdr_decode_hyper((__be32 *)&arg_ch->rs_offset, &rs_offset);
 		svc_rdma_xdr_encode_array_chunk(res_ary, chunk_no,
 						arg_ch->rs_handle,
-					    rs_offset,
+						arg_ch->rs_offset,
 						write_len);
 		chunk_off = 0;
 		while (write_len) {
 			int this_write;
 			this_write = min(write_len, max_write);
 			ret = send_write(xprt, rqstp,
-					 arg_ch->rs_handle,
+					 ntohl(arg_ch->rs_handle),
 					 rs_offset + chunk_off,
 					 xdr_off,
 					 this_write,
@@ -457,6 +457,7 @@ static int send_reply_chunks(struct svcxprt_rdma *xprt,
 	u32 xdr_off;
 	int chunk_no;
 	int chunk_off;
+	int nchunks;
 	struct rpcrdma_segment *ch;
 	struct rpcrdma_write_array *arg_ary;
 	struct rpcrdma_write_array *res_ary;
@@ -476,18 +477,19 @@ static int send_reply_chunks(struct svcxprt_rdma *xprt,
 		max_write = xprt->sc_max_sge * PAGE_SIZE;
 
 	/* xdr offset starts at RPC message */
+	nchunks = ntohl(arg_ary->wc_nchunks);
 	for (xdr_off = 0, chunk_no = 0;
-	     xfer_len && chunk_no < arg_ary->wc_nchunks;
+	     xfer_len && chunk_no < nchunks;
 	     chunk_no++) {
 		u64 rs_offset;
 		ch = &arg_ary->wc_array[chunk_no].wc_target;
-		write_len = min(xfer_len, ch->rs_length);
+		write_len = min(xfer_len, htonl(ch->rs_length));
 
 		/* Prepare the reply chunk given the length actually
 		 * written */
-		rs_offset = get_unaligned(&(ch->rs_offset));
+		xdr_decode_hyper((__be32 *)&ch->rs_offset, &rs_offset);
 		svc_rdma_xdr_encode_array_chunk(res_ary, chunk_no,
-					    ch->rs_handle, rs_offset,
+						ch->rs_handle, ch->rs_offset,
 						write_len);
 		chunk_off = 0;
 		while (write_len) {
@@ -495,7 +497,7 @@ static int send_reply_chunks(struct svcxprt_rdma *xprt,
 
 			this_write = min(write_len, max_write);
 			ret = send_write(xprt, rqstp,
-					 ch->rs_handle,
+					 ntohl(ch->rs_handle),
 					 rs_offset + chunk_off,
 					 xdr_off,
 					 this_write,

net/sunrpc/xprtrdma/svc_rdma_transport.c

@@ -51,6 +51,7 @@
 #include <rdma/rdma_cm.h>
 #include <linux/sunrpc/svc_rdma.h>
 #include <linux/export.h>
+#include "xprt_rdma.h"
 
 #define RPCDBG_FACILITY	RPCDBG_SVCXPRT
 
@@ -90,12 +91,6 @@ struct svc_xprt_class svc_rdma_class = {
 	.xcl_max_payload = RPCSVC_MAXPAYLOAD_TCP,
 };
 
-/* WR context cache. Created in svc_rdma.c  */
-extern struct kmem_cache *svc_rdma_ctxt_cachep;
-
-/* Workqueue created in svc_rdma.c */
-extern struct workqueue_struct *svc_rdma_wq;
-
 struct svc_rdma_op_ctxt *svc_rdma_get_context(struct svcxprt_rdma *xprt)
 {
 	struct svc_rdma_op_ctxt *ctxt;
@@ -150,9 +145,6 @@ void svc_rdma_put_context(struct svc_rdma_op_ctxt *ctxt, int free_pages)
 	atomic_dec(&xprt->sc_ctxt_used);
 }
 
-/* Temporary NFS request map cache. Created in svc_rdma.c  */
-extern struct kmem_cache *svc_rdma_map_cachep;
-
 /*
  * Temporary NFS req mappings are shared across all transport
  * instances. These are short lived and should be bounded by the number

net/sunrpc/xprtrdma/xprt_rdma.h

@@ -343,4 +343,11 @@ void rpcrdma_reply_handler(struct rpcrdma_rep *);
  */
 int rpcrdma_marshal_req(struct rpc_rqst *);
 
+/* Temporary NFS request map cache. Created in svc_rdma.c  */
+extern struct kmem_cache *svc_rdma_map_cachep;
+/* WR context cache. Created in svc_rdma.c  */
+extern struct kmem_cache *svc_rdma_ctxt_cachep;
+/* Workqueue created in svc_rdma.c */
+extern struct workqueue_struct *svc_rdma_wq;
+
 #endif				/* _LINUX_SUNRPC_XPRT_RDMA_H */

net/sunrpc/xprtsock.c

@@ -2475,6 +2475,7 @@ static struct rpc_xprt_ops xs_tcp_ops = {
 static struct rpc_xprt_ops bc_tcp_ops = {
 	.reserve_xprt		= xprt_reserve_xprt,
 	.release_xprt		= xprt_release_xprt,
+	.rpcbind		= xs_local_rpcbind,
 	.buf_alloc		= bc_malloc,
 	.buf_free		= bc_free,
 	.send_request		= bc_send_request,