Merge commit 'linus/master' into HEAD

Conflicts:
	MAINTAINERS

Signed-off-by: Vegard Nossum <vegard.nossum@gmail.com>

80 files changed, 3001 insertions, 2459 deletions

diff --git a/fs/Kconfig b/fs/Kconfig
index 9f7270f36b2a..525da2e8f73b 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -62,6 +62,16 @@ source "fs/autofs/Kconfig"
 source "fs/autofs4/Kconfig"
 source "fs/fuse/Kconfig"
 
+config CUSE
+	tristate "Character device in Userpace support"
+	depends on FUSE_FS
+	help
+	  This FUSE extension allows character devices to be
+	  implemented in userspace.
+
+	  If you want to develop or use userspace character device
+	  based on CUSE, answer Y or M.
+
 config GENERIC_ACL
 	bool
 	select FS_POSIX_ACL
diff --git a/fs/bio.c b/fs/bio.c
index 59000215e59b..5f80848c320c 100644
--- a/fs/bio.c
+++ b/fs/bio.c
@@ -358,9 +358,9 @@ static void bio_kmalloc_destructor(struct bio *bio)
  *
  *   If %__GFP_WAIT is set, then bio_alloc will always be able to allocate
  *   a bio. This is due to the mempool guarantees. To make this work, callers
- *   must never allocate more than 1 bio at the time from this pool. Callers
+ *   must never allocate more than 1 bio at a time from this pool. Callers
  *   that need to allocate more than 1 bio must always submit the previously
- *   allocate bio for IO before attempting to allocate a new one. Failure to
+ *   allocated bio for IO before attempting to allocate a new one. Failure to
  *   do so can cause livelocks under memory pressure.
  *
  **/
diff --git a/fs/compat.c b/fs/compat.c
index 6aefb776dfeb..cdd51a3a7c53 100644
--- a/fs/compat.c
+++ b/fs/compat.c
@@ -471,7 +471,7 @@ asmlinkage long compat_sys_fcntl64(unsigned int fd, unsigned int cmd,
 		ret = sys_fcntl(fd, cmd, (unsigned long)&f);
 		set_fs(old_fs);
 		if (cmd == F_GETLK && ret == 0) {
-			/* GETLK was successfule and we need to return the data...
+			/* GETLK was successful and we need to return the data...
 			 * but it needs to fit in the compat structure.
 			 * l_start shouldn't be too big, unless the original
 			 * start + end is greater than COMPAT_OFF_T_MAX, in which
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index b83f6bcfa51a..0aac371bff0b 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -1765,7 +1765,7 @@ static int do_i2c_smbus_ioctl(unsigned int fd, unsigned int cmd, unsigned long a
 
 /* Since old style bridge ioctl's endup using SIOCDEVPRIVATE
  * for some operations; this forces use of the newer bridge-utils that
- * use compatiable ioctls
+ * use compatible ioctls
  */
 static int old_bridge_ioctl(unsigned int fd, unsigned int cmd, unsigned long arg)
 {
diff --git a/fs/configfs/configfs_internal.h b/fs/configfs/configfs_internal.h
index 762d287123ca..da6061a6df40 100644
--- a/fs/configfs/configfs_internal.h
+++ b/fs/configfs/configfs_internal.h
@@ -39,6 +39,9 @@ struct configfs_dirent {
 	umode_t			s_mode;
 	struct dentry		* s_dentry;
 	struct iattr		* s_iattr;
+#ifdef CONFIG_LOCKDEP
+	int			s_depth;
+#endif
 };
 
 #define CONFIGFS_ROOT		0x0001
diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c
index 05373db21a4e..8e48b52205aa 100644
--- a/fs/configfs/dir.c
+++ b/fs/configfs/dir.c
@@ -78,11 +78,97 @@ static const struct dentry_operations configfs_dentry_ops = {
 	.d_delete	= configfs_d_delete,
 };
 
+#ifdef CONFIG_LOCKDEP
+
+/*
+ * Helpers to make lockdep happy with our recursive locking of default groups'
+ * inodes (see configfs_attach_group() and configfs_detach_group()).
+ * We put default groups i_mutexes in separate classes according to their depth
+ * from the youngest non-default group ancestor.
+ *
+ * For a non-default group A having default groups A/B, A/C, and A/C/D, default
+ * groups A/B and A/C will have their inode's mutex in class
+ * default_group_class[0], and default group A/C/D will be in
+ * default_group_class[1].
+ *
+ * The lock classes are declared and assigned in inode.c, according to the
+ * s_depth value.
+ * The s_depth value is initialized to -1, adjusted to >= 0 when attaching
+ * default groups, and reset to -1 when all default groups are attached. During
+ * attachment, if configfs_create() sees s_depth > 0, the lock class of the new
+ * inode's mutex is set to default_group_class[s_depth - 1].
+ */
+
+static void configfs_init_dirent_depth(struct configfs_dirent *sd)
+{
+	sd->s_depth = -1;
+}
+
+static void configfs_set_dir_dirent_depth(struct configfs_dirent *parent_sd,
+					  struct configfs_dirent *sd)
+{
+	int parent_depth = parent_sd->s_depth;
+
+	if (parent_depth >= 0)
+		sd->s_depth = parent_depth + 1;
+}
+
+static void
+configfs_adjust_dir_dirent_depth_before_populate(struct configfs_dirent *sd)
+{
+	/*
+	 * item's i_mutex class is already setup, so s_depth is now only
+	 * used to set new sub-directories s_depth, which is always done
+	 * with item's i_mutex locked.
+	 */
+	/*
+	 *  sd->s_depth == -1 iff we are a non default group.
+	 *  else (we are a default group) sd->s_depth > 0 (see
+	 *  create_dir()).
+	 */
+	if (sd->s_depth == -1)
+		/*
+		 * We are a non default group and we are going to create
+		 * default groups.
+		 */
+		sd->s_depth = 0;
+}
+
+static void
+configfs_adjust_dir_dirent_depth_after_populate(struct configfs_dirent *sd)
+{
+	/* We will not create default groups anymore. */
+	sd->s_depth = -1;
+}
+
+#else /* CONFIG_LOCKDEP */
+
+static void configfs_init_dirent_depth(struct configfs_dirent *sd)
+{
+}
+
+static void configfs_set_dir_dirent_depth(struct configfs_dirent *parent_sd,
+					  struct configfs_dirent *sd)
+{
+}
+
+static void
+configfs_adjust_dir_dirent_depth_before_populate(struct configfs_dirent *sd)
+{
+}
+
+static void
+configfs_adjust_dir_dirent_depth_after_populate(struct configfs_dirent *sd)
+{
+}
+
+#endif /* CONFIG_LOCKDEP */
+
 /*
  * Allocates a new configfs_dirent and links it to the parent configfs_dirent
  */
-static struct configfs_dirent *configfs_new_dirent(struct configfs_dirent * parent_sd,
-						void * element)
+static struct configfs_dirent *configfs_new_dirent(struct configfs_dirent *parent_sd,
+						   void *element, int type)
 {
 	struct configfs_dirent * sd;
 
@@ -94,6 +180,8 @@ static struct configfs_dirent *configfs_new_dirent(struct configfs_dirent * pare
 	INIT_LIST_HEAD(&sd->s_links);
 	INIT_LIST_HEAD(&sd->s_children);
 	sd->s_element = element;
+	sd->s_type = type;
+	configfs_init_dirent_depth(sd);
 	spin_lock(&configfs_dirent_lock);
 	if (parent_sd->s_type & CONFIGFS_USET_DROPPING) {
 		spin_unlock(&configfs_dirent_lock);
@@ -138,12 +226,11 @@ int configfs_make_dirent(struct configfs_dirent * parent_sd,
 {
 	struct configfs_dirent * sd;
 
-	sd = configfs_new_dirent(parent_sd, element);
+	sd = configfs_new_dirent(parent_sd, element, type);
 	if (IS_ERR(sd))
 		return PTR_ERR(sd);
 
 	sd->s_mode = mode;
-	sd->s_type = type;
 	sd->s_dentry = dentry;
 	if (dentry) {
 		dentry->d_fsdata = configfs_get(sd);
@@ -187,6 +274,7 @@ static int create_dir(struct config_item * k, struct dentry * p,
 		error = configfs_make_dirent(p->d_fsdata, d, k, mode,
 					     CONFIGFS_DIR | CONFIGFS_USET_CREATING);
 	if (!error) {
+		configfs_set_dir_dirent_depth(p->d_fsdata, d->d_fsdata);
 		error = configfs_create(d, mode, init_dir);
 		if (!error) {
 			inc_nlink(p->d_inode);
@@ -789,11 +877,13 @@ static int configfs_attach_group(struct config_item *parent_item,
 		 * error, as rmdir() would.
 		 */
 		mutex_lock_nested(&dentry->d_inode->i_mutex, I_MUTEX_CHILD);
+		configfs_adjust_dir_dirent_depth_before_populate(sd);
 		ret = populate_groups(to_config_group(item));
 		if (ret) {
 			configfs_detach_item(item);
 			dentry->d_inode->i_flags |= S_DEAD;
 		}
+		configfs_adjust_dir_dirent_depth_after_populate(sd);
 		mutex_unlock(&dentry->d_inode->i_mutex);
 		if (ret)
 			d_delete(dentry);
@@ -916,11 +1006,11 @@ static int configfs_dump(struct configfs_dirent *sd, int level)
  * Note, btw, that this can be called at *any* time, even when a configfs
  * subsystem isn't registered, or when configfs is loading or unloading.
  * Just like configfs_register_subsystem().  So we take the same
- * precautions.  We pin the filesystem.  We lock each i_mutex _in_order_
- * on our way down the tree.  If we can find the target item in the
+ * precautions.  We pin the filesystem.  We lock configfs_dirent_lock.
+ * If we can find the target item in the
  * configfs tree, it must be part of the subsystem tree as well, so we
- * do not need the subsystem semaphore.  Holding the i_mutex chain locks
- * out mkdir() and rmdir(), who might be racing us.
+ * do not need the subsystem semaphore.  Holding configfs_dirent_lock helps
+ * locking out mkdir() and rmdir(), who might be racing us.
  */
 
 /*
@@ -933,17 +1023,21 @@ static int configfs_dump(struct configfs_dirent *sd, int level)
  * do that so we can unlock it if we find nothing.
  *
  * Here we do a depth-first search of the dentry hierarchy looking for
- * our object.  We take i_mutex on each step of the way down.  IT IS
- * ESSENTIAL THAT i_mutex LOCKING IS ORDERED.  If we come back up a branch,
- * we'll drop the i_mutex.
+ * our object.
+ * We deliberately ignore items tagged as dropping since they are virtually
+ * dead, as well as items in the middle of attachment since they virtually
+ * do not exist yet. This completes the locking out of racing mkdir() and
+ * rmdir().
+ * Note: subdirectories in the middle of attachment start with s_type =
+ * CONFIGFS_DIR|CONFIGFS_USET_CREATING set by create_dir().  When
+ * CONFIGFS_USET_CREATING is set, we ignore the item.  The actual set of
+ * s_type is in configfs_new_dirent(), which has configfs_dirent_lock.
  *
- * If the target is not found, -ENOENT is bubbled up and we have released
- * all locks.  If the target was found, the locks will be cleared by
- * configfs_depend_rollback().
+ * If the target is not found, -ENOENT is bubbled up.
  *
  * This adds a requirement that all config_items be unique!
  *
- * This is recursive because the locking traversal is tricky.  There isn't
+ * This is recursive.  There isn't
  * much on the stack, though, so folks that need this function - be careful
  * about your stack!  Patches will be accepted to make it iterative.
  */
@@ -955,13 +1049,13 @@ static int configfs_depend_prep(struct dentry *origin,
 
 	BUG_ON(!origin || !sd);
 
-	/* Lock this guy on the way down */
-	mutex_lock(&sd->s_dentry->d_inode->i_mutex);
 	if (sd->s_element == target)  /* Boo-yah */
 		goto out;
 
 	list_for_each_entry(child_sd, &sd->s_children, s_sibling) {
-		if (child_sd->s_type & CONFIGFS_DIR) {
+		if ((child_sd->s_type & CONFIGFS_DIR) &&
+		    !(child_sd->s_type & CONFIGFS_USET_DROPPING) &&
+		    !(child_sd->s_type & CONFIGFS_USET_CREATING)) {
 			ret = configfs_depend_prep(child_sd->s_dentry,
 						   target);
 			if (!ret)
@@ -970,33 +1064,12 @@ static int configfs_depend_prep(struct dentry *origin,
 	}
 
 	/* We looped all our children and didn't find target */
-	mutex_unlock(&sd->s_dentry->d_inode->i_mutex);
 	ret = -ENOENT;
 
 out:
 	return ret;
 }
 
-/*
- * This is ONLY called if configfs_depend_prep() did its job.  So we can
- * trust the entire path from item back up to origin.
- *
- * We walk backwards from item, unlocking each i_mutex.  We finish by
- * unlocking origin.
- */
-static void configfs_depend_rollback(struct dentry *origin,
-				     struct config_item *item)
-{
-	struct dentry *dentry = item->ci_dentry;
-
-	while (dentry != origin) {
-		mutex_unlock(&dentry->d_inode->i_mutex);
-		dentry = dentry->d_parent;
-	}
-
-	mutex_unlock(&origin->d_inode->i_mutex);
-}
-
 int configfs_depend_item(struct configfs_subsystem *subsys,
 			 struct config_item *target)
 {
@@ -1037,17 +1110,21 @@ int configfs_depend_item(struct configfs_subsystem *subsys,
 
 	/* Ok, now we can trust subsys/s_item */
 
-	/* Scan the tree, locking i_mutex recursively, return 0 if found */
+	spin_lock(&configfs_dirent_lock);
+	/* Scan the tree, return 0 if found */
 	ret = configfs_depend_prep(subsys_sd->s_dentry, target);
 	if (ret)
-		goto out_unlock_fs;
+		goto out_unlock_dirent_lock;
 
-	/* We hold all i_mutexes from the subsystem down to the target */
+	/*
+	 * We are sure that the item is not about to be removed by rmdir(), and
+	 * not in the middle of attachment by mkdir().
+	 */
 	p = target->ci_dentry->d_fsdata;
 	p->s_dependent_count += 1;
 
-	configfs_depend_rollback(subsys_sd->s_dentry, target);
-
+out_unlock_dirent_lock:
+	spin_unlock(&configfs_dirent_lock);
 out_unlock_fs:
 	mutex_unlock(&configfs_sb->s_root->d_inode->i_mutex);
 
@@ -1072,10 +1149,10 @@ void configfs_undepend_item(struct configfs_subsystem *subsys,
 	struct configfs_dirent *sd;
 
 	/*
-	 * Since we can trust everything is pinned, we just need i_mutex
-	 * on the item.
+	 * Since we can trust everything is pinned, we just need
+	 * configfs_dirent_lock.
 	 */
-	mutex_lock(&target->ci_dentry->d_inode->i_mutex);
+	spin_lock(&configfs_dirent_lock);
 
 	sd = target->ci_dentry->d_fsdata;
 	BUG_ON(sd->s_dependent_count < 1);
@@ -1086,7 +1163,7 @@ void configfs_undepend_item(struct configfs_subsystem *subsys,
 	 * After this unlock, we cannot trust the item to stay alive!
 	 * DO NOT REFERENCE item after this unlock.
 	 */
-	mutex_unlock(&target->ci_dentry->d_inode->i_mutex);
+	spin_unlock(&configfs_dirent_lock);
 }
 EXPORT_SYMBOL(configfs_undepend_item);
 
@@ -1286,13 +1363,6 @@ static int configfs_rmdir(struct inode *dir, struct dentry *dentry)
 	if (sd->s_type & CONFIGFS_USET_DEFAULT)
 		return -EPERM;
 
-	/*
-	 * Here's where we check for dependents.  We're protected by
-	 * i_mutex.
-	 */
-	if (sd->s_dependent_count)
-		return -EBUSY;
-
 	/* Get a working ref until we have the child */
 	parent_item = configfs_get_config_item(dentry->d_parent);
 	subsys = to_config_group(parent_item)->cg_subsys;
@@ -1316,9 +1386,17 @@ static int configfs_rmdir(struct inode *dir, struct dentry *dentry)
 
 		mutex_lock(&configfs_symlink_mutex);
 		spin_lock(&configfs_dirent_lock);
-		ret = configfs_detach_prep(dentry, &wait_mutex);
-		if (ret)
-			configfs_detach_rollback(dentry);
+		/*
+		 * Here's where we check for dependents.  We're protected by
+		 * configfs_dirent_lock.
+		 * If no dependent, atomically tag the item as dropping.
+		 */
+		ret = sd->s_dependent_count ? -EBUSY : 0;
+		if (!ret) {
+			ret = configfs_detach_prep(dentry, &wait_mutex);
+			if (ret)
+				configfs_detach_rollback(dentry);
+		}
 		spin_unlock(&configfs_dirent_lock);
 		mutex_unlock(&configfs_symlink_mutex);
 
@@ -1429,7 +1507,7 @@ static int configfs_dir_open(struct inode *inode, struct file *file)
 	 */
 	err = -ENOENT;
 	if (configfs_dirent_is_ready(parent_sd)) {
-		file->private_data = configfs_new_dirent(parent_sd, NULL);
+		file->private_data = configfs_new_dirent(parent_sd, NULL, 0);
 		if (IS_ERR(file->private_data))
 			err = PTR_ERR(file->private_data);
 		else
diff --git a/fs/configfs/inode.c b/fs/configfs/inode.c
index 5d349d38e056..4921e7426d95 100644
--- a/fs/configfs/inode.c
+++ b/fs/configfs/inode.c
@@ -33,10 +33,15 @@
 #include <linux/backing-dev.h>
 #include <linux/capability.h>
 #include <linux/sched.h>
+#include <linux/lockdep.h>
 
 #include <linux/configfs.h>
 #include "configfs_internal.h"
 
+#ifdef CONFIG_LOCKDEP
+static struct lock_class_key default_group_class[MAX_LOCK_DEPTH];
+#endif
+
 extern struct super_block * configfs_sb;
 
 static const struct address_space_operations configfs_aops = {
@@ -150,6 +155,38 @@ struct inode * configfs_new_inode(mode_t mode, struct configfs_dirent * sd)
 	return inode;
 }
 
+#ifdef CONFIG_LOCKDEP
+
+static void configfs_set_inode_lock_class(struct configfs_dirent *sd,
+					  struct inode *inode)
+{
+	int depth = sd->s_depth;
+
+	if (depth > 0) {
+		if (depth <= ARRAY_SIZE(default_group_class)) {
+			lockdep_set_class(&inode->i_mutex,
+					  &default_group_class[depth - 1]);
+		} else {
+			/*
+			 * In practice the maximum level of locking depth is
+			 * already reached. Just inform about possible reasons.
+			 */
+			printk(KERN_INFO "configfs: Too many levels of inodes"
+			       " for the locking correctness validator.\n");
+			printk(KERN_INFO "Spurious warnings may appear.\n");
+		}
+	}
+}
+
+#else /* CONFIG_LOCKDEP */
+
+static void configfs_set_inode_lock_class(struct configfs_dirent *sd,
+					  struct inode *inode)
+{
+}
+
+#endif /* CONFIG_LOCKDEP */
+
 int configfs_create(struct dentry * dentry, int mode, int (*init)(struct inode *))
 {
 	int error = 0;
@@ -162,6 +199,7 @@ int configfs_create(struct dentry * dentry, int mode, int (*init)(struct inode *
 					struct inode *p_inode = dentry->d_parent->d_inode;
 					p_inode->i_mtime = p_inode->i_ctime = CURRENT_TIME;
 				}
+				configfs_set_inode_lock_class(sd, inode);
 				goto Proceed;
 			}
 			else
diff --git a/fs/dlm/dir.c b/fs/dlm/dir.c
index 858fba14aaa6..c4dfa1dcc86f 100644
--- a/fs/dlm/dir.c
+++ b/fs/dlm/dir.c
@@ -49,7 +49,8 @@ static struct dlm_direntry *get_free_de(struct dlm_ls *ls, int len)
 	spin_unlock(&ls->ls_recover_list_lock);
 
 	if (!found)
-		de = kzalloc(sizeof(struct dlm_direntry) + len, GFP_KERNEL);
+		de = kzalloc(sizeof(struct dlm_direntry) + len,
+			     ls->ls_allocation);
 	return de;
 }
 
@@ -211,7 +212,7 @@ int dlm_recover_directory(struct dlm_ls *ls)
 
 	dlm_dir_clear(ls);
 
-	last_name = kmalloc(DLM_RESNAME_MAXLEN, GFP_KERNEL);
+	last_name = kmalloc(DLM_RESNAME_MAXLEN, ls->ls_allocation);
 	if (!last_name)
 		goto out;
 
@@ -322,7 +323,7 @@ static int get_entry(struct dlm_ls *ls, int nodeid, char *name,
 	if (namelen > DLM_RESNAME_MAXLEN)
 		return -EINVAL;
 
-	de = kzalloc(sizeof(struct dlm_direntry) + namelen, GFP_KERNEL);
+	de = kzalloc(sizeof(struct dlm_direntry) + namelen, ls->ls_allocation);
 	if (!de)
 		return -ENOMEM;
 
diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c
index cd8e2df3c295..d489fcc86713 100644
--- a/fs/dlm/lockspace.c
+++ b/fs/dlm/lockspace.c
@@ -384,7 +384,7 @@ static void threads_stop(void)
 	dlm_astd_stop();
 }
 
-static int new_lockspace(char *name, int namelen, void **lockspace,
+static int new_lockspace(const char *name, int namelen, void **lockspace,
 			 uint32_t flags, int lvblen)
 {
 	struct dlm_ls *ls;
@@ -419,16 +419,14 @@ static int new_lockspace(char *name, int namelen, void **lockspace,
 			break;
 		}
 		ls->ls_create_count++;
-		module_put(THIS_MODULE);
-		error = 1; /* not an error, return 0 */
+		*lockspace = ls;
+		error = 1;
 		break;
 	}
 	spin_unlock(&lslist_lock);
 
-	if (error < 0)
-		goto out;
 	if (error)
-		goto ret_zero;
+		goto out;
 
 	error = -ENOMEM;
 
@@ -583,7 +581,6 @@ static int new_lockspace(char *name, int namelen, void **lockspace,
 	dlm_create_debug_file(ls);
 
 	log_debug(ls, "join complete");
- ret_zero:
 	*lockspace = ls;
 	return 0;
 
@@ -614,7 +611,7 @@ static int new_lockspace(char *name, int namelen, void **lockspace,
 	return error;
 }
 
-int dlm_new_lockspace(char *name, int namelen, void **lockspace,
+int dlm_new_lockspace(const char *name, int namelen, void **lockspace,
 		      uint32_t flags, int lvblen)
 {
 	int error = 0;
@@ -628,7 +625,9 @@ int dlm_new_lockspace(char *name, int namelen, void **lockspace,
 	error = new_lockspace(name, namelen, lockspace, flags, lvblen);
 	if (!error)
 		ls_count++;
-	else if (!ls_count)
+	if (error > 0)
+		error = 0;
+	if (!ls_count)
 		threads_stop();
  out:
 	mutex_unlock(&ls_lock);
diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c
index 609108a83267..cdb580a9c7a2 100644
--- a/fs/dlm/lowcomms.c
+++ b/fs/dlm/lowcomms.c
@@ -309,6 +309,20 @@ static void lowcomms_state_change(struct sock *sk)
 		lowcomms_write_space(sk);
 }
 
+int dlm_lowcomms_connect_node(int nodeid)
+{
+	struct connection *con;
+
+	if (nodeid == dlm_our_nodeid())
+		return 0;
+
+	con = nodeid2con(nodeid, GFP_NOFS);
+	if (!con)
+		return -ENOMEM;
+	lowcomms_connect_sock(con);
+	return 0;
+}
+
 /* Make a socket active */
 static int add_sock(struct socket *sock, struct connection *con)
 {
@@ -486,7 +500,7 @@ static void process_sctp_notification(struct connection *con,
 				return;
 			}
 
-			new_con = nodeid2con(nodeid, GFP_KERNEL);
+			new_con = nodeid2con(nodeid, GFP_NOFS);
 			if (!new_con)
 				return;
 
@@ -722,7 +736,7 @@ static int tcp_accept_from_sock(struct connection *con)
 	 *  the same time and the connections cross on the wire.
 	 *  In this case we store the incoming one in "othercon"
 	 */
-	newcon = nodeid2con(nodeid, GFP_KERNEL);
+	newcon = nodeid2con(nodeid, GFP_NOFS);
 	if (!newcon) {
 		result = -ENOMEM;
 		goto accept_err;
@@ -732,7 +746,7 @@ static int tcp_accept_from_sock(struct connection *con)
 		struct connection *othercon = newcon->othercon;
 
 		if (!othercon) {
-			othercon = kmem_cache_zalloc(con_cache, GFP_KERNEL);
+			othercon = kmem_cache_zalloc(con_cache, GFP_NOFS);
 			if (!othercon) {
 				log_print("failed to allocate incoming socket");
 				mutex_unlock(&newcon->sock_mutex);
@@ -1421,7 +1435,7 @@ static int work_start(void)
 static void stop_conn(struct connection *con)
 {
 	con->flags |= 0x0F;
-	if (con->sock)
+	if (con->sock && con->sock->sk)
 		con->sock->sk->sk_user_data = NULL;
 }
 
diff --git a/fs/dlm/lowcomms.h b/fs/dlm/lowcomms.h
index a9a9618c0d3f..1311e6426287 100644
--- a/fs/dlm/lowcomms.h
+++ b/fs/dlm/lowcomms.h
@@ -2,7 +2,7 @@
 *******************************************************************************
 **
 **  Copyright (C) Sistina Software, Inc.  1997-2003  All rights reserved.
-**  Copyright (C) 2004-2005 Red Hat, Inc.  All rights reserved.
+**  Copyright (C) 2004-2009 Red Hat, Inc.  All rights reserved.
 **
 **  This copyrighted material is made available to anyone wishing to use,
 **  modify, copy, or redistribute it subject to the terms and conditions
@@ -19,6 +19,7 @@ void dlm_lowcomms_stop(void);
 int dlm_lowcomms_close(int nodeid);
 void *dlm_lowcomms_get_buffer(int nodeid, int len, gfp_t allocation, char **ppc);
 void dlm_lowcomms_commit_buffer(void *mh);
+int dlm_lowcomms_connect_node(int nodeid);
 
 #endif				/* __LOWCOMMS_DOT_H__ */
 
diff --git a/fs/dlm/member.c b/fs/dlm/member.c
index 26133f05ae3a..b128775913b2 100644
--- a/fs/dlm/member.c
+++ b/fs/dlm/member.c
@@ -1,7 +1,7 @@
 /******************************************************************************
 *******************************************************************************
 **
-**  Copyright (C) 2005-2008 Red Hat, Inc.  All rights reserved.
+**  Copyright (C) 2005-2009 Red Hat, Inc.  All rights reserved.
 **
 **  This copyrighted material is made available to anyone wishing to use,
 **  modify, copy, or redistribute it subject to the terms and conditions
@@ -17,6 +17,7 @@
 #include "recover.h"
 #include "rcom.h"
 #include "config.h"
+#include "lowcomms.h"
 
 static void add_ordered_member(struct dlm_ls *ls, struct dlm_member *new)
 {
@@ -45,9 +46,9 @@ static void add_ordered_member(struct dlm_ls *ls, struct dlm_member *new)
 static int dlm_add_member(struct dlm_ls *ls, int nodeid)
 {
 	struct dlm_member *memb;
-	int w;
+	int w, error;
 
-	memb = kzalloc(sizeof(struct dlm_member), GFP_KERNEL);
+	memb = kzalloc(sizeof(struct dlm_member), ls->ls_allocation);
 	if (!memb)
 		return -ENOMEM;
 
@@ -57,6 +58,12 @@ static int dlm_add_member(struct dlm_ls *ls, int nodeid)
 		return w;
 	}
 
+	error = dlm_lowcomms_connect_node(nodeid);
+	if (error < 0) {
+		kfree(memb);
+		return error;
+	}
+
 	memb->nodeid = nodeid;
 	memb->weight = w;
 	add_ordered_member(ls, memb);
@@ -136,7 +143,7 @@ static void make_member_array(struct dlm_ls *ls)
 
 	ls->ls_total_weight = total;
 
-	array = kmalloc(sizeof(int) * total, GFP_KERNEL);
+	array = kmalloc(sizeof(int) * total, ls->ls_allocation);
 	if (!array)
 		return;
 
@@ -219,7 +226,7 @@ int dlm_recover_members(struct dlm_ls *ls, struct dlm_recover *rv, int *neg_out)
 			continue;
 		log_debug(ls, "new nodeid %d is a re-added member", rv->new[i]);
 
-		memb = kzalloc(sizeof(struct dlm_member), GFP_KERNEL);
+		memb = kzalloc(sizeof(struct dlm_member), ls->ls_allocation);
 		if (!memb)
 			return -ENOMEM;
 		memb->nodeid = rv->new[i];
@@ -334,7 +341,7 @@ int dlm_ls_start(struct dlm_ls *ls)
 	int *ids = NULL, *new = NULL;
 	int error, ids_count = 0, new_count = 0;
 
-	rv = kzalloc(sizeof(struct dlm_recover), GFP_KERNEL);
+	rv = kzalloc(sizeof(struct dlm_recover), ls->ls_allocation);
 	if (!rv)
 		return -ENOMEM;
 
diff --git a/fs/dlm/requestqueue.c b/fs/dlm/requestqueue.c
index daa4183fbb84..7a2307c08911 100644
--- a/fs/dlm/requestqueue.c
+++ b/fs/dlm/requestqueue.c
@@ -35,7 +35,7 @@ void dlm_add_requestqueue(struct dlm_ls *ls, int nodeid, struct dlm_message *ms)
 	struct rq_entry *e;
 	int length = ms->m_header.h_length - sizeof(struct dlm_message);
 
-	e = kmalloc(sizeof(struct rq_entry) + length, GFP_KERNEL);
+	e = kmalloc(sizeof(struct rq_entry) + length, ls->ls_allocation);
 	if (!e) {
 		log_print("dlm_add_requestqueue: out of memory len %d", length);
 		return;
diff --git a/fs/eventfd.c b/fs/eventfd.c
index 2a701d593d35..3f0e1974abdc 100644
--- a/fs/eventfd.c
+++ b/fs/eventfd.c
@@ -16,6 +16,7 @@
 #include <linux/anon_inodes.h>
 #include <linux/eventfd.h>
 #include <linux/syscalls.h>
+#include <linux/module.h>
 
 struct eventfd_ctx {
 	wait_queue_head_t wqh;
@@ -56,6 +57,7 @@ int eventfd_signal(struct file *file, int n)
 
 	return n;
 }
+EXPORT_SYMBOL_GPL(eventfd_signal);
 
 static int eventfd_release(struct inode *inode, struct file *file)
 {
@@ -197,6 +199,7 @@ struct file *eventfd_fget(int fd)
 
 	return file;
 }
+EXPORT_SYMBOL_GPL(eventfd_fget);
 
 SYSCALL_DEFINE2(eventfd2, unsigned int, count, int, flags)
 {
diff --git a/fs/exofs/common.h b/fs/exofs/common.h
index b1512c4bb8c7..24667eedc023 100644
--- a/fs/exofs/common.h
+++ b/fs/exofs/common.h
@@ -175,10 +175,4 @@ int exofs_async_op(struct osd_request *or,
 
 int extract_attr_from_req(struct osd_request *or, struct osd_attr *attr);
 
-int osd_req_read_kern(struct osd_request *or,
-	const struct osd_obj_id *obj, u64 offset, void *buff, u64 len);
-
-int osd_req_write_kern(struct osd_request *or,
-	const struct osd_obj_id *obj, u64 offset, void *buff, u64 len);
-
 #endif /*ifndef __EXOFS_COM_H__*/
diff --git a/fs/exofs/inode.c b/fs/exofs/inode.c
index ba8d9fab4693..77d0a295eb1c 100644
--- a/fs/exofs/inode.c
+++ b/fs/exofs/inode.c
@@ -59,10 +59,9 @@ static void _pcol_init(struct page_collect *pcol, unsigned expected_pages,
 		struct inode *inode)
 {
 	struct exofs_sb_info *sbi = inode->i_sb->s_fs_info;
-	struct request_queue *req_q = sbi->s_dev->scsi_device->request_queue;
 
 	pcol->sbi = sbi;
-	pcol->req_q = req_q;
+	pcol->req_q = osd_request_queue(sbi->s_dev);
 	pcol->inode = inode;
 	pcol->expected_pages = expected_pages;
 
@@ -266,7 +265,7 @@ static int read_exec(struct page_collect *pcol, bool is_sync)
 		goto err;
 	}
 
-	osd_req_read(or, &obj, pcol->bio, i_start);
+	osd_req_read(or, &obj, i_start, pcol->bio, pcol->length);
 
 	if (is_sync) {
 		exofs_sync_op(or, pcol->sbi->s_timeout, oi->i_cred);
@@ -522,7 +521,8 @@ static int write_exec(struct page_collect *pcol)
 
 	*pcol_copy = *pcol;
 
-	osd_req_write(or, &obj, pcol_copy->bio, i_start);
+	pcol_copy->bio->bi_rw |= (1 << BIO_RW); /* FIXME: bio_set_dir() */
+	osd_req_write(or, &obj, i_start, pcol_copy->bio, pcol_copy->length);
 	ret = exofs_async_op(or, writepages_done, pcol_copy, oi->i_cred);
 	if (unlikely(ret)) {
 		EXOFS_ERR("write_exec: exofs_async_op() Faild\n");
diff --git a/fs/exofs/osd.c b/fs/exofs/osd.c
index 06ca92672eb5..b3d2ccb87aaa 100644
--- a/fs/exofs/osd.c
+++ b/fs/exofs/osd.c
@@ -125,29 +125,3 @@ int extract_attr_from_req(struct osd_request *or, struct osd_attr *attr)
 
 	return -EIO;
 }
-
-int osd_req_read_kern(struct osd_request *or,
-	const struct osd_obj_id *obj, u64 offset, void* buff, u64 len)
-{
-	struct request_queue *req_q = or->osd_dev->scsi_device->request_queue;
-	struct bio *bio = bio_map_kern(req_q, buff, len, GFP_KERNEL);
-
-	if (!bio)
-		return -ENOMEM;
-
-	osd_req_read(or, obj, bio, offset);
-	return 0;
-}
-
-int osd_req_write_kern(struct osd_request *or,
-	const struct osd_obj_id *obj, u64 offset, void* buff, u64 len)
-{
-	struct request_queue *req_q = or->osd_dev->scsi_device->request_queue;
-	struct bio *bio = bio_map_kern(req_q, buff, len, GFP_KERNEL);
-
-	if (!bio)
-		return -ENOMEM;
-
-	osd_req_write(or, obj, bio, offset);
-	return 0;
-}
diff --git a/fs/ext2/ext2.h b/fs/ext2/ext2.h
index b2bbf45039e0..f2e5811936d0 100644
--- a/fs/ext2/ext2.h
+++ b/fs/ext2/ext2.h
@@ -27,7 +27,7 @@ struct ext2_inode_info {
 	/*
 	 * i_block_group is the number of the block group which contains
 	 * this file's inode.  Constant across the lifetime of the inode,
-	 * it is ued for making block allocation decisions - we try to
+	 * it is used for making block allocation decisions - we try to
 	 * place a file's data blocks near its inode block, and new inodes
 	 * near to their parent directory's inode.
 	 */
diff --git a/fs/fuse/Makefile b/fs/fuse/Makefile
index 72437065f6ad..e95eeb445e58 100644
--- a/fs/fuse/Makefile
+++ b/fs/fuse/Makefile
@@ -3,5 +3,6 @@
 #
 
 obj-$(CONFIG_FUSE_FS) += fuse.o
+obj-$(CONFIG_CUSE) += cuse.o
 
 fuse-objs := dev.o dir.o file.o inode.o control.o
diff --git a/fs/fuse/cuse.c b/fs/fuse/cuse.c
new file mode 100644
index 000000000000..de792dcf3274
--- /dev/null
+++ b/fs/fuse/cuse.c
@@ -0,0 +1,610 @@
+/*
+ * CUSE: Character device in Userspace
+ *
+ * Copyright (C) 2008-2009  SUSE Linux Products GmbH
+ * Copyright (C) 2008-2009  Tejun Heo <tj@kernel.org>
+ *
+ * This file is released under the GPLv2.
+ *
+ * CUSE enables character devices to be implemented from userland much
+ * like FUSE allows filesystems.  On initialization /dev/cuse is
+ * created.  By opening the file and replying to the CUSE_INIT request
+ * userland CUSE server can create a character device.  After that the
+ * operation is very similar to FUSE.
+ *
+ * A CUSE instance involves the following objects.
+ *
+ * cuse_conn	: contains fuse_conn and serves as bonding structure
+ * channel	: file handle connected to the userland CUSE server
+ * cdev		: the implemented character device
+ * dev		: generic device for cdev
+ *
+ * Note that 'channel' is what 'dev' is in FUSE.  As CUSE deals with
+ * devices, it's called 'channel' to reduce confusion.
+ *
+ * channel determines when the character device dies.  When channel is
+ * closed, everything begins to destruct.  The cuse_conn is taken off
+ * the lookup table preventing further access from cdev, cdev and
+ * generic device are removed and the base reference of cuse_conn is
+ * put.
+ *
+ * On each open, the matching cuse_conn is looked up and if found an
+ * additional reference is taken which is released when the file is
+ * closed.
+ */
+
+#include <linux/fuse.h>
+#include <linux/cdev.h>
+#include <linux/device.h>
+#include <linux/file.h>
+#include <linux/fs.h>
+#include <linux/kdev_t.h>
+#include <linux/kthread.h>
+#include <linux/list.h>
+#include <linux/magic.h>
+#include <linux/miscdevice.h>
+#include <linux/mutex.h>
+#include <linux/spinlock.h>
+#include <linux/stat.h>
+
+#include "fuse_i.h"
+
+#define CUSE_CONNTBL_LEN	64
+
+struct cuse_conn {
+	struct list_head	list;	/* linked on cuse_conntbl */
+	struct fuse_conn	fc;	/* fuse connection */
+	struct cdev		*cdev;	/* associated character device */
+	struct device		*dev;	/* device representing @cdev */
+
+	/* init parameters, set once during initialization */
+	bool			unrestricted_ioctl;
+};
+
+static DEFINE_SPINLOCK(cuse_lock);		/* protects cuse_conntbl */
+static struct list_head cuse_conntbl[CUSE_CONNTBL_LEN];
+static struct class *cuse_class;
+
+static struct cuse_conn *fc_to_cc(struct fuse_conn *fc)
+{
+	return container_of(fc, struct cuse_conn, fc);
+}
+
+static struct list_head *cuse_conntbl_head(dev_t devt)
+{
+	return &cuse_conntbl[(MAJOR(devt) + MINOR(devt)) % CUSE_CONNTBL_LEN];
+}
+
+
+/**************************************************************************
+ * CUSE frontend operations
+ *
+ * These are file operations for the character device.
+ *
+ * On open, CUSE opens a file from the FUSE mnt and stores it to
+ * private_data of the open file.  All other ops call FUSE ops on the
+ * FUSE file.
+ */
+
+static ssize_t cuse_read(struct file *file, char __user *buf, size_t count,
+			 loff_t *ppos)
+{
+	loff_t pos = 0;
+
+	return fuse_direct_io(file, buf, count, &pos, 0);
+}
+
+static ssize_t cuse_write(struct file *file, const char __user *buf,
+			  size_t count, loff_t *ppos)
+{
+	loff_t pos = 0;
+	/*
+	 * No locking or generic_write_checks(), the server is
+	 * responsible for locking and sanity checks.
+	 */
+	return fuse_direct_io(file, buf, count, &pos, 1);
+}
+
+static int cuse_open(struct inode *inode, struct file *file)
+{
+	dev_t devt = inode->i_cdev->dev;
+	struct cuse_conn *cc = NULL, *pos;
+	int rc;
+
+	/* look up and get the connection */
+	spin_lock(&cuse_lock);
+	list_for_each_entry(pos, cuse_conntbl_head(devt), list)
+		if (pos->dev->devt == devt) {
+			fuse_conn_get(&pos->fc);
+			cc = pos;
+			break;
+		}
+	spin_unlock(&cuse_lock);
+
+	/* dead? */
+	if (!cc)
+		return -ENODEV;
+
+	/*
+	 * Generic permission check is already done against the chrdev
+	 * file, proceed to open.
+	 */
+	rc = fuse_do_open(&cc->fc, 0, file, 0);
+	if (rc)
+		fuse_conn_put(&cc->fc);
+	return rc;
+}
+
+static int cuse_release(struct inode *inode, struct file *file)
+{
+	struct fuse_file *ff = file->private_data;
+	struct fuse_conn *fc = ff->fc;
+
+	fuse_sync_release(ff, file->f_flags);
+	fuse_conn_put(fc);
+
+	return 0;
+}
+
+static long cuse_file_ioctl(struct file *file, unsigned int cmd,
+			    unsigned long arg)
+{
+	struct fuse_file *ff = file->private_data;
+	struct cuse_conn *cc = fc_to_cc(ff->fc);
+	unsigned int flags = 0;
+
+	if (cc->unrestricted_ioctl)
+		flags |= FUSE_IOCTL_UNRESTRICTED;
+
+	return fuse_do_ioctl(file, cmd, arg, flags);
+}
+
+static long cuse_file_compat_ioctl(struct file *file, unsigned int cmd,
+				   unsigned long arg)
+{
+	struct fuse_file *ff = file->private_data;
+	struct cuse_conn *cc = fc_to_cc(ff->fc);
+	unsigned int flags = FUSE_IOCTL_COMPAT;
+
+	if (cc->unrestricted_ioctl)
+		flags |= FUSE_IOCTL_UNRESTRICTED;
+
+	return fuse_do_ioctl(file, cmd, arg, flags);
+}
+
+static const struct file_operations cuse_frontend_fops = {
+	.owner			= THIS_MODULE,
+	.read			= cuse_read,
+	.write			= cuse_write,
+	.open			= cuse_open,
+	.release		= cuse_release,
+	.unlocked_ioctl		= cuse_file_ioctl,
+	.compat_ioctl		= cuse_file_compat_ioctl,
+	.poll			= fuse_file_poll,
+};
+
+
+/**************************************************************************
+ * CUSE channel initialization and destruction
+ */
+
+struct cuse_devinfo {
+	const char		*name;
+};
+
+/**
+ * cuse_parse_one - parse one key=value pair
+ * @pp: i/o parameter for the current position
+ * @end: points to one past the end of the packed string
+ * @keyp: out parameter for key
+ * @valp: out parameter for value
+ *
+ * *@pp points to packed strings - "key0=val0\0key1=val1\0" which ends
+ * at @end - 1.  This function parses one pair and set *@keyp to the
+ * start of the key and *@valp to the start of the value.  Note that
+ * the original string is modified such that the key string is
+ * terminated with '\0'.  *@pp is updated to point to the next string.
+ *
+ * RETURNS:
+ * 1 on successful parse, 0 on EOF, -errno on failure.
+ */
+static int cuse_parse_one(char **pp, char *end, char **keyp, char **valp)
+{
+	char *p = *pp;
+	char *key, *val;
+
+	while (p < end && *p == '\0')
+		p++;
+	if (p == end)
+		return 0;
+
+	if (end[-1] != '\0') {
+		printk(KERN_ERR "CUSE: info not properly terminated\n");
+		return -EINVAL;
+	}
+
+	key = val = p;
+	p += strlen(p);
+
+	if (valp) {
+		strsep(&val, "=");
+		if (!val)
+			val = key + strlen(key);
+		key = strstrip(key);
+		val = strstrip(val);
+	} else
+		key = strstrip(key);
+
+	if (!strlen(key)) {
+		printk(KERN_ERR "CUSE: zero length info key specified\n");
+		return -EINVAL;
+	}
+
+	*pp = p;
+	*keyp = key;
+	if (valp)
+		*valp = val;
+
+	return 1;
+}
+
+/**
+ * cuse_parse_dev_info - parse device info
+ * @p: device info string
+ * @len: length of device info string
+ * @devinfo: out parameter for parsed device info
+ *
+ * Parse @p to extract device info and store it into @devinfo.  String
+ * pointed to by @p is modified by parsing and @devinfo points into
+ * them, so @p shouldn't be freed while @devinfo is in use.
+ *
+ * RETURNS:
+ * 0 on success, -errno on failure.
+ */
+static int cuse_parse_devinfo(char *p, size_t len, struct cuse_devinfo *devinfo)
+{
+	char *end = p + len;
+	char *key, *val;
+	int rc;
+
+	while (true) {
+		rc = cuse_parse_one(&p, end, &key, &val);
+		if (rc < 0)
+			return rc;
+		if (!rc)
+			break;
+		if (strcmp(key, "DEVNAME") == 0)
+			devinfo->name = val;
+		else
+			printk(KERN_WARNING "CUSE: unknown device info \"%s\"\n",
+			       key);
+	}
+
+	if (!devinfo->name || !strlen(devinfo->name)) {
+		printk(KERN_ERR "CUSE: DEVNAME unspecified\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static void cuse_gendev_release(struct device *dev)
+{
+	kfree(dev);
+}
+
+/**
+ * cuse_process_init_reply - finish initializing CUSE channel
+ *
+ * This function creates the character device and sets up all the
+ * required data structures for it.  Please read the comment at the
+ * top of this file for high level overview.
+ */
+static void cuse_process_init_reply(struct fuse_conn *fc, struct fuse_req *req)
+{
+	struct cuse_conn *cc = fc_to_cc(fc);
+	struct cuse_init_out *arg = &req->misc.cuse_init_out;
+	struct page *page = req->pages[0];
+	struct cuse_devinfo devinfo = { };
+	struct device *dev;
+	struct cdev *cdev;
+	dev_t devt;
+	int rc;
+
+	if (req->out.h.error ||
+	    arg->major != FUSE_KERNEL_VERSION || arg->minor < 11) {
+		goto err;
+	}
+
+	fc->minor = arg->minor;
+	fc->max_read = max_t(unsigned, arg->max_read, 4096);
+	fc->max_write = max_t(unsigned, arg->max_write, 4096);
+
+	/* parse init reply */
+	cc->unrestricted_ioctl = arg->flags & CUSE_UNRESTRICTED_IOCTL;
+
+	rc = cuse_parse_devinfo(page_address(page), req->out.args[1].size,
+				&devinfo);
+	if (rc)
+		goto err;
+
+	/* determine and reserve devt */
+	devt = MKDEV(arg->dev_major, arg->dev_minor);
+	if (!MAJOR(devt))
+		rc = alloc_chrdev_region(&devt, MINOR(devt), 1, devinfo.name);
+	else
+		rc = register_chrdev_region(devt, 1, devinfo.name);
+	if (rc) {
+		printk(KERN_ERR "CUSE: failed to register chrdev region\n");
+		goto err;
+	}
+
+	/* devt determined, create device */
+	rc = -ENOMEM;
+	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
+	if (!dev)
+		goto err_region;
+
+	device_initialize(dev);
+	dev_set_uevent_suppress(dev, 1);
+	dev->class = cuse_class;
+	dev->devt = devt;
+	dev->release = cuse_gendev_release;
+	dev_set_drvdata(dev, cc);
+	dev_set_name(dev, "%s", devinfo.name);
+
+	rc = device_add(dev);
+	if (rc)
+		goto err_device;
+
+	/* register cdev */
+	rc = -ENOMEM;
+	cdev = cdev_alloc();
+	if (!cdev)
+		goto err_device;
+
+	cdev->owner = THIS_MODULE;
+	cdev->ops = &cuse_frontend_fops;
+
+	rc = cdev_add(cdev, devt, 1);
+	if (rc)
+		goto err_cdev;
+
+	cc->dev = dev;
+	cc->cdev = cdev;
+
+	/* make the device available */
+	spin_lock(&cuse_lock);
+	list_add(&cc->list, cuse_conntbl_head(devt));
+	spin_unlock(&cuse_lock);
+
+	/* announce device availability */
+	dev_set_uevent_suppress(dev, 0);
+	kobject_uevent(&dev->kobj, KOBJ_ADD);
+out:
+	__free_page(page);
+	return;
+
+err_cdev:
+	cdev_del(cdev);
+err_device:
+	put_device(dev);
+err_region:
+	unregister_chrdev_region(devt, 1);
+err:
+	fc->conn_error = 1;
+	goto out;
+}
+
+static int cuse_send_init(struct cuse_conn *cc)
+{
+	int rc;
+	struct fuse_req *req;
+	struct page *page;
+	struct fuse_conn *fc = &cc->fc;
+	struct cuse_init_in *arg;
+
+	BUILD_BUG_ON(CUSE_INIT_INFO_MAX > PAGE_SIZE);
+
+	req = fuse_get_req(fc);
+	if (IS_ERR(req)) {
+		rc = PTR_ERR(req);
+		goto err;
+	}
+
+	rc = -ENOMEM;
+	page = alloc_page(GFP_KERNEL | __GFP_ZERO);
+	if (!page)
+		goto err_put_req;
+
+	arg = &req->misc.cuse_init_in;
+	arg->major = FUSE_KERNEL_VERSION;
+	arg->minor = FUSE_KERNEL_MINOR_VERSION;
+	arg->flags |= CUSE_UNRESTRICTED_IOCTL;
+	req->in.h.opcode = CUSE_INIT;
+	req->in.numargs = 1;
+	req->in.args[0].size = sizeof(struct cuse_init_in);
+	req->in.args[0].value = arg;
+	req->out.numargs = 2;
+	req->out.args[0].size = sizeof(struct cuse_init_out);
+	req->out.args[0].value = &req->misc.cuse_init_out;
+	req->out.args[1].size = CUSE_INIT_INFO_MAX;
+	req->out.argvar = 1;
+	req->out.argpages = 1;
+	req->pages[0] = page;
+	req->num_pages = 1;
+	req->end = cuse_process_init_reply;
+	fuse_request_send_background(fc, req);
+
+	return 0;
+
+err_put_req:
+	fuse_put_request(fc, req);
+err:
+	return rc;
+}
+
+static void cuse_fc_release(struct fuse_conn *fc)
+{
+	struct cuse_conn *cc = fc_to_cc(fc);
+	kfree(cc);
+}
+
+/**
+ * cuse_channel_open - open method for /dev/cuse
+ * @inode: inode for /dev/cuse
+ * @file: file struct being opened
+ *
+ * Userland CUSE server can create a CUSE device by opening /dev/cuse
+ * and replying to the initilaization request kernel sends.  This
+ * function is responsible for handling CUSE device initialization.
+ * Because the fd opened by this function is used during
+ * initialization, this function only creates cuse_conn and sends
+ * init.  The rest is delegated to a kthread.
+ *
+ * RETURNS:
+ * 0 on success, -errno on failure.
+ */
+static int cuse_channel_open(struct inode *inode, struct file *file)
+{
+	struct cuse_conn *cc;
+	int rc;
+
+	/* set up cuse_conn */
+	cc = kzalloc(sizeof(*cc), GFP_KERNEL);
+	if (!cc)
+		return -ENOMEM;
+
+	fuse_conn_init(&cc->fc);
+
+	INIT_LIST_HEAD(&cc->list);
+	cc->fc.release = cuse_fc_release;
+
+	cc->fc.connected = 1;
+	cc->fc.blocked = 0;
+	rc = cuse_send_init(cc);
+	if (rc) {
+		fuse_conn_put(&cc->fc);
+		return rc;
+	}
+	file->private_data = &cc->fc;	/* channel owns base reference to cc */
+
+	return 0;
+}
+
+/**
+ * cuse_channel_release - release method for /dev/cuse
+ * @inode: inode for /dev/cuse
+ * @file: file struct being closed
+ *
+ * Disconnect the channel, deregister CUSE device and initiate
+ * destruction by putting the default reference.
+ *
+ * RETURNS:
+ * 0 on success, -errno on failure.
+ */
+static int cuse_channel_release(struct inode *inode, struct file *file)
+{
+	struct cuse_conn *cc = fc_to_cc(file->private_data);
+	int rc;
+
+	/* remove from the conntbl, no more access from this point on */
+	spin_lock(&cuse_lock);
+	list_del_init(&cc->list);
+	spin_unlock(&cuse_lock);
+
+	/* remove device */
+	if (cc->dev)
+		device_unregister(cc->dev);
+	if (cc->cdev) {
+		unregister_chrdev_region(cc->cdev->dev, 1);
+		cdev_del(cc->cdev);
+	}
+
+	/* kill connection and shutdown channel */
+	fuse_conn_kill(&cc->fc);
+	rc = fuse_dev_release(inode, file);	/* puts the base reference */
+
+	return rc;
+}
+
+static struct file_operations cuse_channel_fops; /* initialized during init */
+
+
+/**************************************************************************
+ * Misc stuff and module initializatiion
+ *
+ * CUSE exports the same set of attributes to sysfs as fusectl.
+ */
+
+static ssize_t cuse_class_waiting_show(struct device *dev,
+				       struct device_attribute *attr, char *buf)
+{
+	struct cuse_conn *cc = dev_get_drvdata(dev);
+
+	return sprintf(buf, "%d\n", atomic_read(&cc->fc.num_waiting));
+}
+
+static ssize_t cuse_class_abort_store(struct device *dev,
+				      struct device_attribute *attr,
+				      const char *buf, size_t count)
+{
+	struct cuse_conn *cc = dev_get_drvdata(dev);
+
+	fuse_abort_conn(&cc->fc);
+	return count;
+}
+
+static struct device_attribute cuse_class_dev_attrs[] = {
+	__ATTR(waiting, S_IFREG | 0400, cuse_class_waiting_show, NULL),
+	__ATTR(abort, S_IFREG | 0200, NULL, cuse_class_abort_store),
+	{ }
+};
+
+static struct miscdevice cuse_miscdev = {
+	.minor		= MISC_DYNAMIC_MINOR,
+	.name		= "cuse",
+	.fops		= &cuse_channel_fops,
+};
+
+static int __init cuse_init(void)
+{
+	int i, rc;
+
+	/* init conntbl */
+	for (i = 0; i < CUSE_CONNTBL_LEN; i++)
+		INIT_LIST_HEAD(&cuse_conntbl[i]);
+
+	/* inherit and extend fuse_dev_operations */
+	cuse_channel_fops		= fuse_dev_operations;
+	cuse_channel_fops.owner		= THIS_MODULE;
+	cuse_channel_fops.open		= cuse_channel_open;
+	cuse_channel_fops.release	= cuse_channel_release;
+
+	cuse_class = class_create(THIS_MODULE, "cuse");
+	if (IS_ERR(cuse_class))
+		return PTR_ERR(cuse_class);
+
+	cuse_class->dev_attrs = cuse_class_dev_attrs;
+
+	rc = misc_register(&cuse_miscdev);
+	if (rc) {
+		class_destroy(cuse_class);
+		return rc;
+	}
+
+	return 0;
+}
+
+static void __exit cuse_exit(void)
+{
+	misc_deregister(&cuse_miscdev);
+	class_destroy(cuse_class);
+}
+
+module_init(cuse_init);
+module_exit(cuse_exit);
+
+MODULE_AUTHOR("Tejun Heo <tj@kernel.org>");
+MODULE_DESCRIPTION("Character device in Userspace");
+MODULE_LICENSE("GPL");
diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c
index ba76b68c52ff..8fed2ed12f38 100644
--- a/fs/fuse/dev.c
+++ b/fs/fuse/dev.c
@@ -46,6 +46,7 @@ struct fuse_req *fuse_request_alloc(void)
 		fuse_request_init(req);
 	return req;
 }
+EXPORT_SYMBOL_GPL(fuse_request_alloc);
 
 struct fuse_req *fuse_request_alloc_nofs(void)
 {
@@ -124,6 +125,7 @@ struct fuse_req *fuse_get_req(struct fuse_conn *fc)
 	atomic_dec(&fc->num_waiting);
 	return ERR_PTR(err);
 }
+EXPORT_SYMBOL_GPL(fuse_get_req);
 
 /*
  * Return request in fuse_file->reserved_req.  However that may
@@ -208,6 +210,7 @@ void fuse_put_request(struct fuse_conn *fc, struct fuse_req *req)
 			fuse_request_free(req);
 	}
 }
+EXPORT_SYMBOL_GPL(fuse_put_request);
 
 static unsigned len_args(unsigned numargs, struct fuse_arg *args)
 {
@@ -282,7 +285,7 @@ __releases(&fc->lock)
 			wake_up_all(&fc->blocked_waitq);
 		}
 		if (fc->num_background == FUSE_CONGESTION_THRESHOLD &&
-		    fc->connected) {
+		    fc->connected && fc->bdi_initialized) {
 			clear_bdi_congested(&fc->bdi, READ);
 			clear_bdi_congested(&fc->bdi, WRITE);
 		}
@@ -400,6 +403,7 @@ void fuse_request_send(struct fuse_conn *fc, struct fuse_req *req)
 	}
 	spin_unlock(&fc->lock);
 }
+EXPORT_SYMBOL_GPL(fuse_request_send);
 
 static void fuse_request_send_nowait_locked(struct fuse_conn *fc,
 					    struct fuse_req *req)
@@ -408,7 +412,8 @@ static void fuse_request_send_nowait_locked(struct fuse_conn *fc,
 	fc->num_background++;
 	if (fc->num_background == FUSE_MAX_BACKGROUND)
 		fc->blocked = 1;
-	if (fc->num_background == FUSE_CONGESTION_THRESHOLD) {
+	if (fc->num_background == FUSE_CONGESTION_THRESHOLD &&
+	    fc->bdi_initialized) {
 		set_bdi_congested(&fc->bdi, READ);
 		set_bdi_congested(&fc->bdi, WRITE);
 	}
@@ -439,6 +444,7 @@ void fuse_request_send_background(struct fuse_conn *fc, struct fuse_req *req)
 	req->isreply = 1;
 	fuse_request_send_nowait(fc, req);
 }
+EXPORT_SYMBOL_GPL(fuse_request_send_background);
 
 /*
  * Called under fc->lock
@@ -1105,8 +1111,9 @@ void fuse_abort_conn(struct fuse_conn *fc)
 	}
 	spin_unlock(&fc->lock);
 }
+EXPORT_SYMBOL_GPL(fuse_abort_conn);
 
-static int fuse_dev_release(struct inode *inode, struct file *file)
+int fuse_dev_release(struct inode *inode, struct file *file)
 {
 	struct fuse_conn *fc = fuse_get_conn(file);
 	if (fc) {
@@ -1120,6 +1127,7 @@ static int fuse_dev_release(struct inode *inode, struct file *file)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(fuse_dev_release);
 
 static int fuse_dev_fasync(int fd, struct file *file, int on)
 {
@@ -1142,6 +1150,7 @@ const struct file_operations fuse_dev_operations = {
 	.release	= fuse_dev_release,
 	.fasync		= fuse_dev_fasync,
 };
+EXPORT_SYMBOL_GPL(fuse_dev_operations);
 
 static struct miscdevice fuse_miscdevice = {
 	.minor = FUSE_MINOR,
diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index 8b8eebc5614b..b3089a083d30 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -362,19 +362,6 @@ static struct dentry *fuse_lookup(struct inode *dir, struct dentry *entry,
 }
 
 /*
- * Synchronous release for the case when something goes wrong in CREATE_OPEN
- */
-static void fuse_sync_release(struct fuse_conn *fc, struct fuse_file *ff,
-			      u64 nodeid, int flags)
-{
-	fuse_release_fill(ff, nodeid, flags, FUSE_RELEASE);
-	ff->reserved_req->force = 1;
-	fuse_request_send(fc, ff->reserved_req);
-	fuse_put_request(fc, ff->reserved_req);
-	kfree(ff);
-}
-
-/*
  * Atomic create+open operation
  *
  * If the filesystem doesn't support this, then fall back to separate
@@ -445,12 +432,14 @@ static int fuse_create_open(struct inode *dir, struct dentry *entry, int mode,
 		goto out_free_ff;
 
 	fuse_put_request(fc, req);
+	ff->fh = outopen.fh;
+	ff->nodeid = outentry.nodeid;
+	ff->open_flags = outopen.open_flags;
 	inode = fuse_iget(dir->i_sb, outentry.nodeid, outentry.generation,
 			  &outentry.attr, entry_attr_timeout(&outentry), 0);
 	if (!inode) {
 		flags &= ~(O_CREAT | O_EXCL | O_TRUNC);
-		ff->fh = outopen.fh;
-		fuse_sync_release(fc, ff, outentry.nodeid, flags);
+		fuse_sync_release(ff, flags);
 		fuse_send_forget(fc, forget_req, outentry.nodeid, 1);
 		return -ENOMEM;
 	}
@@ -460,11 +449,11 @@ static int fuse_create_open(struct inode *dir, struct dentry *entry, int mode,
 	fuse_invalidate_attr(dir);
 	file = lookup_instantiate_filp(nd, entry, generic_file_open);
 	if (IS_ERR(file)) {
-		ff->fh = outopen.fh;
-		fuse_sync_release(fc, ff, outentry.nodeid, flags);
+		fuse_sync_release(ff, flags);
 		return PTR_ERR(file);
 	}
-	fuse_finish_open(inode, file, ff, &outopen);
+	file->private_data = fuse_file_get(ff);
+	fuse_finish_open(inode, file);
 	return 0;
 
  out_free_ff:
@@ -1035,7 +1024,7 @@ static int fuse_readdir(struct file *file, void *dstbuf, filldir_t filldir)
 	req->out.argpages = 1;
 	req->num_pages = 1;
 	req->pages[0] = page;
-	fuse_read_fill(req, file, inode, file->f_pos, PAGE_SIZE, FUSE_READDIR);
+	fuse_read_fill(req, file, file->f_pos, PAGE_SIZE, FUSE_READDIR);
 	fuse_request_send(fc, req);
 	nbytes = req->out.args[0].size;
 	err = req->out.h.error;
@@ -1101,12 +1090,14 @@ static void fuse_put_link(struct dentry *dentry, struct nameidata *nd, void *c)
 
 static int fuse_dir_open(struct inode *inode, struct file *file)
 {
-	return fuse_open_common(inode, file, 1);
+	return fuse_open_common(inode, file, true);
 }
 
 static int fuse_dir_release(struct inode *inode, struct file *file)
 {
-	return fuse_release_common(inode, file, 1);
+	fuse_release_common(file, FUSE_RELEASEDIR);
+
+	return 0;
 }
 
 static int fuse_dir_fsync(struct file *file, struct dentry *de, int datasync)
diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index 06f30e965676..fce6ce694fde 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -12,13 +12,13 @@
 #include <linux/slab.h>
 #include <linux/kernel.h>
 #include <linux/sched.h>
+#include <linux/module.h>
 
 static const struct file_operations fuse_direct_io_file_operations;
 
-static int fuse_send_open(struct inode *inode, struct file *file, int isdir,
-			  struct fuse_open_out *outargp)
+static int fuse_send_open(struct fuse_conn *fc, u64 nodeid, struct file *file,
+			  int opcode, struct fuse_open_out *outargp)
 {
-	struct fuse_conn *fc = get_fuse_conn(inode);
 	struct fuse_open_in inarg;
 	struct fuse_req *req;
 	int err;
@@ -31,8 +31,8 @@ static int fuse_send_open(struct inode *inode, struct file *file, int isdir,
 	inarg.flags = file->f_flags & ~(O_CREAT | O_EXCL | O_NOCTTY);
 	if (!fc->atomic_o_trunc)
 		inarg.flags &= ~O_TRUNC;
-	req->in.h.opcode = isdir ? FUSE_OPENDIR : FUSE_OPEN;
-	req->in.h.nodeid = get_node_id(inode);
+	req->in.h.opcode = opcode;
+	req->in.h.nodeid = nodeid;
 	req->in.numargs = 1;
 	req->in.args[0].size = sizeof(inarg);
 	req->in.args[0].value = &inarg;
@@ -49,22 +49,27 @@ static int fuse_send_open(struct inode *inode, struct file *file, int isdir,
 struct fuse_file *fuse_file_alloc(struct fuse_conn *fc)
 {
 	struct fuse_file *ff;
+
 	ff = kmalloc(sizeof(struct fuse_file), GFP_KERNEL);
-	if (ff) {
-		ff->reserved_req = fuse_request_alloc();
-		if (!ff->reserved_req) {
-			kfree(ff);
-			return NULL;
-		} else {
-			INIT_LIST_HEAD(&ff->write_entry);
-			atomic_set(&ff->count, 0);
-			spin_lock(&fc->lock);
-			ff->kh = ++fc->khctr;
-			spin_unlock(&fc->lock);
-		}
-		RB_CLEAR_NODE(&ff->polled_node);
-		init_waitqueue_head(&ff->poll_wait);
+	if (unlikely(!ff))
+		return NULL;
+
+	ff->fc = fc;
+	ff->reserved_req = fuse_request_alloc();
+	if (unlikely(!ff->reserved_req)) {
+		kfree(ff);
+		return NULL;
 	}
+
+	INIT_LIST_HEAD(&ff->write_entry);
+	atomic_set(&ff->count, 0);
+	RB_CLEAR_NODE(&ff->polled_node);
+	init_waitqueue_head(&ff->poll_wait);
+
+	spin_lock(&fc->lock);
+	ff->kh = ++fc->khctr;
+	spin_unlock(&fc->lock);
+
 	return ff;
 }
 
@@ -74,7 +79,7 @@ void fuse_file_free(struct fuse_file *ff)
 	kfree(ff);
 }
 
-static struct fuse_file *fuse_file_get(struct fuse_file *ff)
+struct fuse_file *fuse_file_get(struct fuse_file *ff)
 {
 	atomic_inc(&ff->count);
 	return ff;
@@ -82,40 +87,65 @@ static struct fuse_file *fuse_file_get(struct fuse_file *ff)
 
 static void fuse_release_end(struct fuse_conn *fc, struct fuse_req *req)
 {
-	dput(req->misc.release.dentry);
-	mntput(req->misc.release.vfsmount);
+	path_put(&req->misc.release.path);
 }
 
 static void fuse_file_put(struct fuse_file *ff)
 {
 	if (atomic_dec_and_test(&ff->count)) {
 		struct fuse_req *req = ff->reserved_req;
-		struct inode *inode = req->misc.release.dentry->d_inode;
-		struct fuse_conn *fc = get_fuse_conn(inode);
+
 		req->end = fuse_release_end;
-		fuse_request_send_background(fc, req);
+		fuse_request_send_background(ff->fc, req);
 		kfree(ff);
 	}
 }
 
-void fuse_finish_open(struct inode *inode, struct file *file,
-		      struct fuse_file *ff, struct fuse_open_out *outarg)
+int fuse_do_open(struct fuse_conn *fc, u64 nodeid, struct file *file,
+		 bool isdir)
 {
-	if (outarg->open_flags & FOPEN_DIRECT_IO)
+	struct fuse_open_out outarg;
+	struct fuse_file *ff;
+	int err;
+	int opcode = isdir ? FUSE_OPENDIR : FUSE_OPEN;
+
+	ff = fuse_file_alloc(fc);
+	if (!ff)
+		return -ENOMEM;
+
+	err = fuse_send_open(fc, nodeid, file, opcode, &outarg);
+	if (err) {
+		fuse_file_free(ff);
+		return err;
+	}
+
+	if (isdir)
+		outarg.open_flags &= ~FOPEN_DIRECT_IO;
+
+	ff->fh = outarg.fh;
+	ff->nodeid = nodeid;
+	ff->open_flags = outarg.open_flags;
+	file->private_data = fuse_file_get(ff);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(fuse_do_open);
+
+void fuse_finish_open(struct inode *inode, struct file *file)
+{
+	struct fuse_file *ff = file->private_data;
+
+	if (ff->open_flags & FOPEN_DIRECT_IO)
 		file->f_op = &fuse_direct_io_file_operations;
-	if (!(outarg->open_flags & FOPEN_KEEP_CACHE))
+	if (!(ff->open_flags & FOPEN_KEEP_CACHE))
 		invalidate_inode_pages2(inode->i_mapping);
-	if (outarg->open_flags & FOPEN_NONSEEKABLE)
+	if (ff->open_flags & FOPEN_NONSEEKABLE)
 		nonseekable_open(inode, file);
-	ff->fh = outarg->fh;
-	file->private_data = fuse_file_get(ff);
 }
 
-int fuse_open_common(struct inode *inode, struct file *file, int isdir)
+int fuse_open_common(struct inode *inode, struct file *file, bool isdir)
 {
 	struct fuse_conn *fc = get_fuse_conn(inode);
-	struct fuse_open_out outarg;
-	struct fuse_file *ff;
 	int err;
 
 	/* VFS checks this, but only _after_ ->open() */
@@ -126,78 +156,85 @@ int fuse_open_common(struct inode *inode, struct file *file, int isdir)
 	if (err)
 		return err;
 
-	ff = fuse_file_alloc(fc);
-	if (!ff)
-		return -ENOMEM;
-
-	err = fuse_send_open(inode, file, isdir, &outarg);
+	err = fuse_do_open(fc, get_node_id(inode), file, isdir);
 	if (err)
-		fuse_file_free(ff);
-	else {
-		if (isdir)
-			outarg.open_flags &= ~FOPEN_DIRECT_IO;
-		fuse_finish_open(inode, file, ff, &outarg);
-	}
+		return err;
 
-	return err;
+	fuse_finish_open(inode, file);
+
+	return 0;
 }
 
-void fuse_release_fill(struct fuse_file *ff, u64 nodeid, int flags, int opcode)
+static void fuse_prepare_release(struct fuse_file *ff, int flags, int opcode)
 {
+	struct fuse_conn *fc = ff->fc;
 	struct fuse_req *req = ff->reserved_req;
 	struct fuse_release_in *inarg = &req->misc.release.in;
 
+	spin_lock(&fc->lock);
+	list_del(&ff->write_entry);
+	if (!RB_EMPTY_NODE(&ff->polled_node))
+		rb_erase(&ff->polled_node, &fc->polled_files);
+	spin_unlock(&fc->lock);
+
+	wake_up_interruptible_sync(&ff->poll_wait);
+
 	inarg->fh = ff->fh;
 	inarg->flags = flags;
 	req->in.h.opcode = opcode;
-	req->in.h.nodeid = nodeid;
+	req->in.h.nodeid = ff->nodeid;
 	req->in.numargs = 1;
 	req->in.args[0].size = sizeof(struct fuse_release_in);
 	req->in.args[0].value = inarg;
 }
 
-int fuse_release_common(struct inode *inode, struct file *file, int isdir)
+void fuse_release_common(struct file *file, int opcode)
 {
-	struct fuse_file *ff = file->private_data;
-	if (ff) {
-		struct fuse_conn *fc = get_fuse_conn(inode);
-		struct fuse_req *req = ff->reserved_req;
-
-		fuse_release_fill(ff, get_node_id(inode), file->f_flags,
-				  isdir ? FUSE_RELEASEDIR : FUSE_RELEASE);
+	struct fuse_file *ff;
+	struct fuse_req *req;
 
-		/* Hold vfsmount and dentry until release is finished */
-		req->misc.release.vfsmount = mntget(file->f_path.mnt);
-		req->misc.release.dentry = dget(file->f_path.dentry);
+	ff = file->private_data;
+	if (unlikely(!ff))
+		return;
 
-		spin_lock(&fc->lock);
-		list_del(&ff->write_entry);
-		if (!RB_EMPTY_NODE(&ff->polled_node))
-			rb_erase(&ff->polled_node, &fc->polled_files);
-		spin_unlock(&fc->lock);
+	req = ff->reserved_req;
+	fuse_prepare_release(ff, file->f_flags, opcode);
 
-		wake_up_interruptible_sync(&ff->poll_wait);
-		/*
-		 * Normally this will send the RELEASE request,
-		 * however if some asynchronous READ or WRITE requests
-		 * are outstanding, the sending will be delayed
-		 */
-		fuse_file_put(ff);
-	}
+	/* Hold vfsmount and dentry until release is finished */
+	path_get(&file->f_path);
+	req->misc.release.path = file->f_path;
 
-	/* Return value is ignored by VFS */
-	return 0;
+	/*
+	 * Normally this will send the RELEASE request, however if
+	 * some asynchronous READ or WRITE requests are outstanding,
+	 * the sending will be delayed.
+	 */
+	fuse_file_put(ff);
 }
 
 static int fuse_open(struct inode *inode, struct file *file)
 {
-	return fuse_open_common(inode, file, 0);
+	return fuse_open_common(inode, file, false);
 }
 
 static int fuse_release(struct inode *inode, struct file *file)
 {
-	return fuse_release_common(inode, file, 0);
+	fuse_release_common(file, FUSE_RELEASE);
+
+	/* return value is ignored by VFS */
+	return 0;
+}
+
+void fuse_sync_release(struct fuse_file *ff, int flags)
+{
+	WARN_ON(atomic_read(&ff->count) > 1);
+	fuse_prepare_release(ff, flags, FUSE_RELEASE);
+	ff->reserved_req->force = 1;
+	fuse_request_send(ff->fc, ff->reserved_req);
+	fuse_put_request(ff->fc, ff->reserved_req);
+	kfree(ff);
 }
+EXPORT_SYMBOL_GPL(fuse_sync_release);
 
 /*
  * Scramble the ID space with XTEA, so that the value of the files_struct
@@ -371,8 +408,8 @@ static int fuse_fsync(struct file *file, struct dentry *de, int datasync)
 	return fuse_fsync_common(file, de, datasync, 0);
 }
 
-void fuse_read_fill(struct fuse_req *req, struct file *file,
-		    struct inode *inode, loff_t pos, size_t count, int opcode)
+void fuse_read_fill(struct fuse_req *req, struct file *file, loff_t pos,
+		    size_t count, int opcode)
 {
 	struct fuse_read_in *inarg = &req->misc.read.in;
 	struct fuse_file *ff = file->private_data;
@@ -382,7 +419,7 @@ void fuse_read_fill(struct fuse_req *req, struct file *file,
 	inarg->size = count;
 	inarg->flags = file->f_flags;
 	req->in.h.opcode = opcode;
-	req->in.h.nodeid = get_node_id(inode);
+	req->in.h.nodeid = ff->nodeid;
 	req->in.numargs = 1;
 	req->in.args[0].size = sizeof(struct fuse_read_in);
 	req->in.args[0].value = inarg;
@@ -392,12 +429,12 @@ void fuse_read_fill(struct fuse_req *req, struct file *file,
 }
 
 static size_t fuse_send_read(struct fuse_req *req, struct file *file,
-			     struct inode *inode, loff_t pos, size_t count,
-			     fl_owner_t owner)
+			     loff_t pos, size_t count, fl_owner_t owner)
 {
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_file *ff = file->private_data;
+	struct fuse_conn *fc = ff->fc;
 
-	fuse_read_fill(req, file, inode, pos, count, FUSE_READ);
+	fuse_read_fill(req, file, pos, count, FUSE_READ);
 	if (owner != NULL) {
 		struct fuse_read_in *inarg = &req->misc.read.in;
 
@@ -455,7 +492,7 @@ static int fuse_readpage(struct file *file, struct page *page)
 	req->out.argpages = 1;
 	req->num_pages = 1;
 	req->pages[0] = page;
-	num_read = fuse_send_read(req, file, inode, pos, count, NULL);
+	num_read = fuse_send_read(req, file, pos, count, NULL);
 	err = req->out.h.error;
 	fuse_put_request(fc, req);
 
@@ -504,19 +541,18 @@ static void fuse_readpages_end(struct fuse_conn *fc, struct fuse_req *req)
 		fuse_file_put(req->ff);
 }
 
-static void fuse_send_readpages(struct fuse_req *req, struct file *file,
-				struct inode *inode)
+static void fuse_send_readpages(struct fuse_req *req, struct file *file)
 {
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_file *ff = file->private_data;
+	struct fuse_conn *fc = ff->fc;
 	loff_t pos = page_offset(req->pages[0]);
 	size_t count = req->num_pages << PAGE_CACHE_SHIFT;
 
 	req->out.argpages = 1;
 	req->out.page_zeroing = 1;
-	fuse_read_fill(req, file, inode, pos, count, FUSE_READ);
+	fuse_read_fill(req, file, pos, count, FUSE_READ);
 	req->misc.read.attr_ver = fuse_get_attr_version(fc);
 	if (fc->async_read) {
-		struct fuse_file *ff = file->private_data;
 		req->ff = fuse_file_get(ff);
 		req->end = fuse_readpages_end;
 		fuse_request_send_background(fc, req);
@@ -546,7 +582,7 @@ static int fuse_readpages_fill(void *_data, struct page *page)
 	    (req->num_pages == FUSE_MAX_PAGES_PER_REQ ||
 	     (req->num_pages + 1) * PAGE_CACHE_SIZE > fc->max_read ||
 	     req->pages[req->num_pages - 1]->index + 1 != page->index)) {
-		fuse_send_readpages(req, data->file, inode);
+		fuse_send_readpages(req, data->file);
 		data->req = req = fuse_get_req(fc);
 		if (IS_ERR(req)) {
 			unlock_page(page);
@@ -580,7 +616,7 @@ static int fuse_readpages(struct file *file, struct address_space *mapping,
 	err = read_cache_pages(mapping, pages, fuse_readpages_fill, &data);
 	if (!err) {
 		if (data.req->num_pages)
-			fuse_send_readpages(data.req, file, inode);
+			fuse_send_readpages(data.req, file);
 		else
 			fuse_put_request(fc, data.req);
 	}
@@ -607,24 +643,19 @@ static ssize_t fuse_file_aio_read(struct kiocb *iocb, const struct iovec *iov,
 	return generic_file_aio_read(iocb, iov, nr_segs, pos);
 }
 
-static void fuse_write_fill(struct fuse_req *req, struct file *file,
-			    struct fuse_file *ff, struct inode *inode,
-			    loff_t pos, size_t count, int writepage)
+static void fuse_write_fill(struct fuse_req *req, struct fuse_file *ff,
+			    loff_t pos, size_t count)
 {
-	struct fuse_conn *fc = get_fuse_conn(inode);
 	struct fuse_write_in *inarg = &req->misc.write.in;
 	struct fuse_write_out *outarg = &req->misc.write.out;
 
-	memset(inarg, 0, sizeof(struct fuse_write_in));
 	inarg->fh = ff->fh;
 	inarg->offset = pos;
 	inarg->size = count;
-	inarg->write_flags = writepage ? FUSE_WRITE_CACHE : 0;
-	inarg->flags = file ? file->f_flags : 0;
 	req->in.h.opcode = FUSE_WRITE;
-	req->in.h.nodeid = get_node_id(inode);
+	req->in.h.nodeid = ff->nodeid;
 	req->in.numargs = 2;
-	if (fc->minor < 9)
+	if (ff->fc->minor < 9)
 		req->in.args[0].size = FUSE_COMPAT_WRITE_IN_SIZE;
 	else
 		req->in.args[0].size = sizeof(struct fuse_write_in);
@@ -636,13 +667,15 @@ static void fuse_write_fill(struct fuse_req *req, struct file *file,
 }
 
 static size_t fuse_send_write(struct fuse_req *req, struct file *file,
-			      struct inode *inode, loff_t pos, size_t count,
-			      fl_owner_t owner)
+			      loff_t pos, size_t count, fl_owner_t owner)
 {
-	struct fuse_conn *fc = get_fuse_conn(inode);
-	fuse_write_fill(req, file, file->private_data, inode, pos, count, 0);
+	struct fuse_file *ff = file->private_data;
+	struct fuse_conn *fc = ff->fc;
+	struct fuse_write_in *inarg = &req->misc.write.in;
+
+	fuse_write_fill(req, ff, pos, count);
+	inarg->flags = file->f_flags;
 	if (owner != NULL) {
-		struct fuse_write_in *inarg = &req->misc.write.in;
 		inarg->write_flags |= FUSE_WRITE_LOCKOWNER;
 		inarg->lock_owner = fuse_lock_owner_id(fc, owner);
 	}
@@ -700,7 +733,7 @@ static int fuse_buffered_write(struct file *file, struct inode *inode,
 	req->num_pages = 1;
 	req->pages[0] = page;
 	req->page_offset = offset;
-	nres = fuse_send_write(req, file, inode, pos, count, NULL);
+	nres = fuse_send_write(req, file, pos, count, NULL);
 	err = req->out.h.error;
 	fuse_put_request(fc, req);
 	if (!err && !nres)
@@ -741,7 +774,7 @@ static size_t fuse_send_write_pages(struct fuse_req *req, struct file *file,
 	for (i = 0; i < req->num_pages; i++)
 		fuse_wait_on_page_writeback(inode, req->pages[i]->index);
 
-	res = fuse_send_write(req, file, inode, pos, count, NULL);
+	res = fuse_send_write(req, file, pos, count, NULL);
 
 	offset = req->page_offset;
 	count = res;
@@ -979,25 +1012,23 @@ static int fuse_get_user_pages(struct fuse_req *req, const char __user *buf,
 	return 0;
 }
 
-static ssize_t fuse_direct_io(struct file *file, const char __user *buf,
-			      size_t count, loff_t *ppos, int write)
+ssize_t fuse_direct_io(struct file *file, const char __user *buf,
+		       size_t count, loff_t *ppos, int write)
 {
-	struct inode *inode = file->f_path.dentry->d_inode;
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_file *ff = file->private_data;
+	struct fuse_conn *fc = ff->fc;
 	size_t nmax = write ? fc->max_write : fc->max_read;
 	loff_t pos = *ppos;
 	ssize_t res = 0;
 	struct fuse_req *req;
 
-	if (is_bad_inode(inode))
-		return -EIO;
-
 	req = fuse_get_req(fc);
 	if (IS_ERR(req))
 		return PTR_ERR(req);
 
 	while (count) {
 		size_t nres;
+		fl_owner_t owner = current->files;
 		size_t nbytes = min(count, nmax);
 		int err = fuse_get_user_pages(req, buf, &nbytes, write);
 		if (err) {
@@ -1006,11 +1037,10 @@ static ssize_t fuse_direct_io(struct file *file, const char __user *buf,
 		}
 
 		if (write)
-			nres = fuse_send_write(req, file, inode, pos, nbytes,
-					       current->files);
+			nres = fuse_send_write(req, file, pos, nbytes, owner);
 		else
-			nres = fuse_send_read(req, file, inode, pos, nbytes,
-					      current->files);
+			nres = fuse_send_read(req, file, pos, nbytes, owner);
+
 		fuse_release_user_pages(req, !write);
 		if (req->out.h.error) {
 			if (!res)
@@ -1034,20 +1064,27 @@ static ssize_t fuse_direct_io(struct file *file, const char __user *buf,
 		}
 	}
 	fuse_put_request(fc, req);
-	if (res > 0) {
-		if (write)
-			fuse_write_update_size(inode, pos);
+	if (res > 0)
 		*ppos = pos;
-	}
-	fuse_invalidate_attr(inode);
 
 	return res;
 }
+EXPORT_SYMBOL_GPL(fuse_direct_io);
 
 static ssize_t fuse_direct_read(struct file *file, char __user *buf,
 				     size_t count, loff_t *ppos)
 {
-	return fuse_direct_io(file, buf, count, ppos, 0);
+	ssize_t res;
+	struct inode *inode = file->f_path.dentry->d_inode;
+
+	if (is_bad_inode(inode))
+		return -EIO;
+
+	res = fuse_direct_io(file, buf, count, ppos, 0);
+
+	fuse_invalidate_attr(inode);
+
+	return res;
 }
 
 static ssize_t fuse_direct_write(struct file *file, const char __user *buf,
@@ -1055,12 +1092,22 @@ static ssize_t fuse_direct_write(struct file *file, const char __user *buf,
 {
 	struct inode *inode = file->f_path.dentry->d_inode;
 	ssize_t res;
+
+	if (is_bad_inode(inode))
+		return -EIO;
+
 	/* Don't allow parallel writes to the same file */
 	mutex_lock(&inode->i_mutex);
 	res = generic_write_checks(file, ppos, &count, 0);
-	if (!res)
+	if (!res) {
 		res = fuse_direct_io(file, buf, count, ppos, 1);
+		if (res > 0)
+			fuse_write_update_size(inode, *ppos);
+	}
 	mutex_unlock(&inode->i_mutex);
+
+	fuse_invalidate_attr(inode);
+
 	return res;
 }
 
@@ -1177,9 +1224,10 @@ static int fuse_writepage_locked(struct page *page)
 	req->ff = fuse_file_get(ff);
 	spin_unlock(&fc->lock);
 
-	fuse_write_fill(req, NULL, ff, inode, page_offset(page), 0, 1);
+	fuse_write_fill(req, ff, page_offset(page), 0);
 
 	copy_highpage(tmp_page, page);
+	req->misc.write.in.write_flags |= FUSE_WRITE_CACHE;
 	req->in.argpages = 1;
 	req->num_pages = 1;
 	req->pages[0] = tmp_page;
@@ -1603,12 +1651,11 @@ static int fuse_ioctl_copy_user(struct page **pages, struct iovec *iov,
  * limits ioctl data transfers to well-formed ioctls and is the forced
  * behavior for all FUSE servers.
  */
-static long fuse_file_do_ioctl(struct file *file, unsigned int cmd,
-			       unsigned long arg, unsigned int flags)
+long fuse_do_ioctl(struct file *file, unsigned int cmd, unsigned long arg,
+		   unsigned int flags)
 {
-	struct inode *inode = file->f_dentry->d_inode;
 	struct fuse_file *ff = file->private_data;
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_conn *fc = ff->fc;
 	struct fuse_ioctl_in inarg = {
 		.fh = ff->fh,
 		.cmd = cmd,
@@ -1627,13 +1674,6 @@ static long fuse_file_do_ioctl(struct file *file, unsigned int cmd,
 	/* assume all the iovs returned by client always fits in a page */
 	BUILD_BUG_ON(sizeof(struct iovec) * FUSE_IOCTL_MAX_IOV > PAGE_SIZE);
 
-	if (!fuse_allow_task(fc, current))
-		return -EACCES;
-
-	err = -EIO;
-	if (is_bad_inode(inode))
-		goto out;
-
 	err = -ENOMEM;
 	pages = kzalloc(sizeof(pages[0]) * FUSE_MAX_PAGES_PER_REQ, GFP_KERNEL);
 	iov_page = alloc_page(GFP_KERNEL);
@@ -1694,7 +1734,7 @@ static long fuse_file_do_ioctl(struct file *file, unsigned int cmd,
 
 	/* okay, let's send it to the client */
 	req->in.h.opcode = FUSE_IOCTL;
-	req->in.h.nodeid = get_node_id(inode);
+	req->in.h.nodeid = ff->nodeid;
 	req->in.numargs = 1;
 	req->in.args[0].size = sizeof(inarg);
 	req->in.args[0].value = &inarg;
@@ -1777,17 +1817,33 @@ static long fuse_file_do_ioctl(struct file *file, unsigned int cmd,
 
 	return err ? err : outarg.result;
 }
+EXPORT_SYMBOL_GPL(fuse_do_ioctl);
+
+static long fuse_file_ioctl_common(struct file *file, unsigned int cmd,
+				   unsigned long arg, unsigned int flags)
+{
+	struct inode *inode = file->f_dentry->d_inode;
+	struct fuse_conn *fc = get_fuse_conn(inode);
+
+	if (!fuse_allow_task(fc, current))
+		return -EACCES;
+
+	if (is_bad_inode(inode))
+		return -EIO;
+
+	return fuse_do_ioctl(file, cmd, arg, flags);
+}
 
 static long fuse_file_ioctl(struct file *file, unsigned int cmd,
 			    unsigned long arg)
 {
-	return fuse_file_do_ioctl(file, cmd, arg, 0);
+	return fuse_file_ioctl_common(file, cmd, arg, 0);
 }
 
 static long fuse_file_compat_ioctl(struct file *file, unsigned int cmd,
 				   unsigned long arg)
 {
-	return fuse_file_do_ioctl(file, cmd, arg, FUSE_IOCTL_COMPAT);
+	return fuse_file_ioctl_common(file, cmd, arg, FUSE_IOCTL_COMPAT);
 }
 
 /*
@@ -1841,11 +1897,10 @@ static void fuse_register_polled_file(struct fuse_conn *fc,
 	spin_unlock(&fc->lock);
 }
 
-static unsigned fuse_file_poll(struct file *file, poll_table *wait)
+unsigned fuse_file_poll(struct file *file, poll_table *wait)
 {
-	struct inode *inode = file->f_dentry->d_inode;
 	struct fuse_file *ff = file->private_data;
-	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_conn *fc = ff->fc;
 	struct fuse_poll_in inarg = { .fh = ff->fh, .kh = ff->kh };
 	struct fuse_poll_out outarg;
 	struct fuse_req *req;
@@ -1870,7 +1925,7 @@ static unsigned fuse_file_poll(struct file *file, poll_table *wait)
 		return PTR_ERR(req);
 
 	req->in.h.opcode = FUSE_POLL;
-	req->in.h.nodeid = get_node_id(inode);
+	req->in.h.nodeid = ff->nodeid;
 	req->in.numargs = 1;
 	req->in.args[0].size = sizeof(inarg);
 	req->in.args[0].value = &inarg;
@@ -1889,6 +1944,7 @@ static unsigned fuse_file_poll(struct file *file, poll_table *wait)
 	}
 	return POLLERR;
 }
+EXPORT_SYMBOL_GPL(fuse_file_poll);
 
 /*
  * This is called from fuse_handle_notify() on FUSE_NOTIFY_POLL and
diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h
index 6fc5aedaa0d5..aaf2f9ff970e 100644
--- a/fs/fuse/fuse_i.h
+++ b/fs/fuse/fuse_i.h
@@ -97,8 +97,13 @@ struct fuse_inode {
 	struct list_head writepages;
 };
 
+struct fuse_conn;
+
 /** FUSE specific file data */
 struct fuse_file {
+	/** Fuse connection for this file */
+	struct fuse_conn *fc;
+
 	/** Request reserved for flush and release */
 	struct fuse_req *reserved_req;
 
@@ -108,9 +113,15 @@ struct fuse_file {
 	/** File handle used by userspace */
 	u64 fh;
 
+	/** Node id of this file */
+	u64 nodeid;
+
 	/** Refcount */
 	atomic_t count;
 
+	/** FOPEN_* flags returned by open */
+	u32 open_flags;
+
 	/** Entry on inode's write_files list */
 	struct list_head write_entry;
 
@@ -185,8 +196,6 @@ enum fuse_req_state {
 	FUSE_REQ_FINISHED
 };
 
-struct fuse_conn;
-
 /**
  * A request to the client
  */
@@ -248,11 +257,12 @@ struct fuse_req {
 		struct fuse_forget_in forget_in;
 		struct {
 			struct fuse_release_in in;
-			struct vfsmount *vfsmount;
-			struct dentry *dentry;
+			struct path path;
 		} release;
 		struct fuse_init_in init_in;
 		struct fuse_init_out init_out;
+		struct cuse_init_in cuse_init_in;
+		struct cuse_init_out cuse_init_out;
 		struct {
 			struct fuse_read_in in;
 			u64 attr_ver;
@@ -386,6 +396,9 @@ struct fuse_conn {
 	/** Filesystem supports NFS exporting.  Only set in INIT */
 	unsigned export_support:1;
 
+	/** Set if bdi is valid */
+	unsigned bdi_initialized:1;
+
 	/*
 	 * The following bitfields are only for optimization purposes
 	 * and hence races in setting them will not cause malfunction
@@ -515,25 +528,24 @@ void fuse_send_forget(struct fuse_conn *fc, struct fuse_req *req,
  * Initialize READ or READDIR request
  */
 void fuse_read_fill(struct fuse_req *req, struct file *file,
-		    struct inode *inode, loff_t pos, size_t count, int opcode);
+		    loff_t pos, size_t count, int opcode);
 
 /**
  * Send OPEN or OPENDIR request
  */
-int fuse_open_common(struct inode *inode, struct file *file, int isdir);
+int fuse_open_common(struct inode *inode, struct file *file, bool isdir);
 
 struct fuse_file *fuse_file_alloc(struct fuse_conn *fc);
+struct fuse_file *fuse_file_get(struct fuse_file *ff);
 void fuse_file_free(struct fuse_file *ff);
-void fuse_finish_open(struct inode *inode, struct file *file,
-		      struct fuse_file *ff, struct fuse_open_out *outarg);
+void fuse_finish_open(struct inode *inode, struct file *file);
 
-/** Fill in ff->reserved_req with a RELEASE request */
-void fuse_release_fill(struct fuse_file *ff, u64 nodeid, int flags, int opcode);
+void fuse_sync_release(struct fuse_file *ff, int flags);
 
 /**
  * Send RELEASE or RELEASEDIR request
  */
-int fuse_release_common(struct inode *inode, struct file *file, int isdir);
+void fuse_release_common(struct file *file, int opcode);
 
 /**
  * Send FSYNC or FSYNCDIR request
@@ -652,10 +664,12 @@ void fuse_invalidate_entry_cache(struct dentry *entry);
  */
 struct fuse_conn *fuse_conn_get(struct fuse_conn *fc);
 
+void fuse_conn_kill(struct fuse_conn *fc);
+
 /**
  * Initialize fuse_conn
  */
-int fuse_conn_init(struct fuse_conn *fc, struct super_block *sb);
+void fuse_conn_init(struct fuse_conn *fc);
 
 /**
  * Release reference to fuse_conn
@@ -694,4 +708,13 @@ void fuse_release_nowrite(struct inode *inode);
 
 u64 fuse_get_attr_version(struct fuse_conn *fc);
 
+int fuse_do_open(struct fuse_conn *fc, u64 nodeid, struct file *file,
+		 bool isdir);
+ssize_t fuse_direct_io(struct file *file, const char __user *buf,
+		       size_t count, loff_t *ppos, int write);
+long fuse_do_ioctl(struct file *file, unsigned int cmd, unsigned long arg,
+		   unsigned int flags);
+unsigned fuse_file_poll(struct file *file, poll_table *wait);
+int fuse_dev_release(struct inode *inode, struct file *file);
+
 #endif /* _FS_FUSE_I_H */
diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index 91f7c85f1ffd..f0df55a52929 100644
--- a/fs/fuse/inode.c
+++ b/fs/fuse/inode.c
@@ -277,11 +277,14 @@ static void fuse_send_destroy(struct fuse_conn *fc)
 	}
 }
 
-static void fuse_put_super(struct super_block *sb)
+static void fuse_bdi_destroy(struct fuse_conn *fc)
 {
-	struct fuse_conn *fc = get_fuse_conn_super(sb);
+	if (fc->bdi_initialized)
+		bdi_destroy(&fc->bdi);
+}
 
-	fuse_send_destroy(fc);
+void fuse_conn_kill(struct fuse_conn *fc)
+{
 	spin_lock(&fc->lock);
 	fc->connected = 0;
 	fc->blocked = 0;
@@ -295,7 +298,16 @@ static void fuse_put_super(struct super_block *sb)
 	list_del(&fc->entry);
 	fuse_ctl_remove_conn(fc);
 	mutex_unlock(&fuse_mutex);
-	bdi_destroy(&fc->bdi);
+	fuse_bdi_destroy(fc);
+}
+EXPORT_SYMBOL_GPL(fuse_conn_kill);
+
+static void fuse_put_super(struct super_block *sb)
+{
+	struct fuse_conn *fc = get_fuse_conn_super(sb);
+
+	fuse_send_destroy(fc);
+	fuse_conn_kill(fc);
 	fuse_conn_put(fc);
 }
 
@@ -466,10 +478,8 @@ static int fuse_show_options(struct seq_file *m, struct vfsmount *mnt)
 	return 0;
 }
 
-int fuse_conn_init(struct fuse_conn *fc, struct super_block *sb)
+void fuse_conn_init(struct fuse_conn *fc)
 {
-	int err;
-
 	memset(fc, 0, sizeof(*fc));
 	spin_lock_init(&fc->lock);
 	mutex_init(&fc->inst_mutex);
@@ -484,49 +494,12 @@ int fuse_conn_init(struct fuse_conn *fc, struct super_block *sb)
 	INIT_LIST_HEAD(&fc->bg_queue);
 	INIT_LIST_HEAD(&fc->entry);
 	atomic_set(&fc->num_waiting, 0);
-	fc->bdi.ra_pages = (VM_MAX_READAHEAD * 1024) / PAGE_CACHE_SIZE;
-	fc->bdi.unplug_io_fn = default_unplug_io_fn;
-	/* fuse does it's own writeback accounting */
-	fc->bdi.capabilities = BDI_CAP_NO_ACCT_WB;
 	fc->khctr = 0;
 	fc->polled_files = RB_ROOT;
-	fc->dev = sb->s_dev;
-	err = bdi_init(&fc->bdi);
-	if (err)
-		goto error_mutex_destroy;
-	if (sb->s_bdev) {
-		err = bdi_register(&fc->bdi, NULL, "%u:%u-fuseblk",
-				   MAJOR(fc->dev), MINOR(fc->dev));
-	} else {
-		err = bdi_register_dev(&fc->bdi, fc->dev);
-	}
-	if (err)
-		goto error_bdi_destroy;
-	/*
-	 * For a single fuse filesystem use max 1% of dirty +
-	 * writeback threshold.
-	 *
-	 * This gives about 1M of write buffer for memory maps on a
-	 * machine with 1G and 10% dirty_ratio, which should be more
-	 * than enough.
-	 *
-	 * Privileged users can raise it by writing to
-	 *
-	 *    /sys/class/bdi/<bdi>/max_ratio
-	 */
-	bdi_set_max_ratio(&fc->bdi, 1);
 	fc->reqctr = 0;
 	fc->blocked = 1;
 	fc->attr_version = 1;
 	get_random_bytes(&fc->scramble_key, sizeof(fc->scramble_key));
-
-	return 0;
-
- error_bdi_destroy:
-	bdi_destroy(&fc->bdi);
- error_mutex_destroy:
-	mutex_destroy(&fc->inst_mutex);
-	return err;
 }
 EXPORT_SYMBOL_GPL(fuse_conn_init);
 
@@ -539,12 +512,14 @@ void fuse_conn_put(struct fuse_conn *fc)
 		fc->release(fc);
 	}
 }
+EXPORT_SYMBOL_GPL(fuse_conn_put);
 
 struct fuse_conn *fuse_conn_get(struct fuse_conn *fc)
 {
 	atomic_inc(&fc->count);
 	return fc;
 }
+EXPORT_SYMBOL_GPL(fuse_conn_get);
 
 static struct inode *fuse_get_root_inode(struct super_block *sb, unsigned mode)
 {
@@ -797,6 +772,48 @@ static void fuse_free_conn(struct fuse_conn *fc)
 	kfree(fc);
 }
 
+static int fuse_bdi_init(struct fuse_conn *fc, struct super_block *sb)
+{
+	int err;
+
+	fc->bdi.ra_pages = (VM_MAX_READAHEAD * 1024) / PAGE_CACHE_SIZE;
+	fc->bdi.unplug_io_fn = default_unplug_io_fn;
+	/* fuse does it's own writeback accounting */
+	fc->bdi.capabilities = BDI_CAP_NO_ACCT_WB;
+
+	err = bdi_init(&fc->bdi);
+	if (err)
+		return err;
+
+	fc->bdi_initialized = 1;
+
+	if (sb->s_bdev) {
+		err =  bdi_register(&fc->bdi, NULL, "%u:%u-fuseblk",
+				    MAJOR(fc->dev), MINOR(fc->dev));
+	} else {
+		err = bdi_register_dev(&fc->bdi, fc->dev);
+	}
+
+	if (err)
+		return err;
+
+	/*
+	 * For a single fuse filesystem use max 1% of dirty +
+	 * writeback threshold.
+	 *
+	 * This gives about 1M of write buffer for memory maps on a
+	 * machine with 1G and 10% dirty_ratio, which should be more
+	 * than enough.
+	 *
+	 * Privileged users can raise it by writing to
+	 *
+	 *    /sys/class/bdi/<bdi>/max_ratio
+	 */
+	bdi_set_max_ratio(&fc->bdi, 1);
+
+	return 0;
+}
+
 static int fuse_fill_super(struct super_block *sb, void *data, int silent)
 {
 	struct fuse_conn *fc;
@@ -843,11 +860,12 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent)
 	if (!fc)
 		goto err_fput;
 
-	err = fuse_conn_init(fc, sb);
-	if (err) {
-		kfree(fc);
-		goto err_fput;
-	}
+	fuse_conn_init(fc);
+
+	fc->dev = sb->s_dev;
+	err = fuse_bdi_init(fc, sb);
+	if (err)
+		goto err_put_conn;
 
 	fc->release = fuse_free_conn;
 	fc->flags = d.flags;
@@ -911,7 +929,7 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent)
  err_put_root:
 	dput(root_dentry);
  err_put_conn:
-	bdi_destroy(&fc->bdi);
+	fuse_bdi_destroy(fc);
 	fuse_conn_put(fc);
  err_fput:
 	fput(file);
diff --git a/fs/gfs2/Makefile b/fs/gfs2/Makefile
index d53a9bea1c2f..3da2f1f4f738 100644
--- a/fs/gfs2/Makefile
+++ b/fs/gfs2/Makefile
@@ -1,3 +1,4 @@
+EXTRA_CFLAGS := -I$(src)
 obj-$(CONFIG_GFS2_FS) += gfs2.o
 gfs2-y := acl.o bmap.o dir.o eaops.o eattr.o glock.o \
 	glops.o inode.o log.o lops.o main.o meta_io.o \
diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c
index 329763530dc0..6d47379e794b 100644
--- a/fs/gfs2/bmap.c
+++ b/fs/gfs2/bmap.c
@@ -25,6 +25,7 @@
 #include "trans.h"
 #include "dir.h"
 #include "util.h"
+#include "trace_gfs2.h"
 
 /* This doesn't need to be that large as max 64 bit pointers in a 4k
  * block is 512, so __u16 is fine for that. It saves stack space to
@@ -589,6 +590,7 @@ int gfs2_block_map(struct inode *inode, sector_t lblock,
 	clear_buffer_mapped(bh_map);
 	clear_buffer_new(bh_map);
 	clear_buffer_boundary(bh_map);
+	trace_gfs2_bmap(ip, bh_map, lblock, create, 1);
 	if (gfs2_is_dir(ip)) {
 		bsize = sdp->sd_jbsize;
 		arr = sdp->sd_jheightsize;
@@ -623,6 +625,7 @@ int gfs2_block_map(struct inode *inode, sector_t lblock,
 	ret = 0;
 out:
 	release_metapath(&mp);
+	trace_gfs2_bmap(ip, bh_map, lblock, create, ret);
 	bmap_unlock(ip, create);
 	return ret;
 
diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c
index 2bf62bcc5181..297421c0427a 100644
--- a/fs/gfs2/glock.c
+++ b/fs/gfs2/glock.c
@@ -39,6 +39,8 @@
 #include "super.h"
 #include "util.h"
 #include "bmap.h"
+#define CREATE_TRACE_POINTS
+#include "trace_gfs2.h"
 
 struct gfs2_gl_hash_bucket {
         struct hlist_head hb_list;
@@ -155,7 +157,7 @@ static void glock_free(struct gfs2_glock *gl)
 
 	if (aspace)
 		gfs2_aspace_put(aspace);
-
+	trace_gfs2_glock_put(gl);
 	sdp->sd_lockstruct.ls_ops->lm_put_lock(gfs2_glock_cachep, gl);
 }
 
@@ -317,14 +319,17 @@ restart:
 						return 2;
 					gh->gh_error = ret;
 					list_del_init(&gh->gh_list);
+					trace_gfs2_glock_queue(gh, 0);
 					gfs2_holder_wake(gh);
 					goto restart;
 				}
 				set_bit(HIF_HOLDER, &gh->gh_iflags);
+				trace_gfs2_promote(gh, 1);
 				gfs2_holder_wake(gh);
 				goto restart;
 			}
 			set_bit(HIF_HOLDER, &gh->gh_iflags);
+			trace_gfs2_promote(gh, 0);
 			gfs2_holder_wake(gh);
 			continue;
 		}
@@ -354,6 +359,7 @@ static inline void do_error(struct gfs2_glock *gl, const int ret)
 		else
 			continue;
 		list_del_init(&gh->gh_list);
+		trace_gfs2_glock_queue(gh, 0);
 		gfs2_holder_wake(gh);
 	}
 }
@@ -422,6 +428,7 @@ static void finish_xmote(struct gfs2_glock *gl, unsigned int ret)
 	int rv;
 
 	spin_lock(&gl->gl_spin);
+	trace_gfs2_glock_state_change(gl, state);
 	state_change(gl, state);
 	gh = find_first_waiter(gl);
 
@@ -851,6 +858,7 @@ static void handle_callback(struct gfs2_glock *gl, unsigned int state,
 			gl->gl_demote_state != state) {
 		gl->gl_demote_state = LM_ST_UNLOCKED;
 	}
+	trace_gfs2_demote_rq(gl);
 }
 
 /**
@@ -936,6 +944,7 @@ fail:
 			goto do_cancel;
 		return;
 	}
+	trace_gfs2_glock_queue(gh, 1);
 	list_add_tail(&gh->gh_list, insert_pt);
 do_cancel:
 	gh = list_entry(gl->gl_holders.next, struct gfs2_holder, gh_list);
@@ -1032,6 +1041,7 @@ void gfs2_glock_dq(struct gfs2_holder *gh)
 		    !test_bit(GLF_DEMOTE, &gl->gl_flags))
 			fast_path = 1;
 	}
+	trace_gfs2_glock_queue(gh, 0);
 	spin_unlock(&gl->gl_spin);
 	if (likely(fast_path))
 		return;
diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c
index f2e449c595b4..13c6237c5f67 100644
--- a/fs/gfs2/log.c
+++ b/fs/gfs2/log.c
@@ -28,6 +28,7 @@
 #include "meta_io.h"
 #include "util.h"
 #include "dir.h"
+#include "trace_gfs2.h"
 
 #define PULL 1
 
@@ -313,6 +314,7 @@ int gfs2_log_reserve(struct gfs2_sbd *sdp, unsigned int blks)
 		gfs2_log_lock(sdp);
 	}
 	atomic_sub(blks, &sdp->sd_log_blks_free);
+	trace_gfs2_log_blocks(sdp, -blks);
 	gfs2_log_unlock(sdp);
 	mutex_unlock(&sdp->sd_log_reserve_mutex);
 
@@ -333,6 +335,7 @@ void gfs2_log_release(struct gfs2_sbd *sdp, unsigned int blks)
 
 	gfs2_log_lock(sdp);
 	atomic_add(blks, &sdp->sd_log_blks_free);
+	trace_gfs2_log_blocks(sdp, blks);
 	gfs2_assert_withdraw(sdp,
 			     atomic_read(&sdp->sd_log_blks_free) <= sdp->sd_jdesc->jd_blocks);
 	gfs2_log_unlock(sdp);
@@ -558,6 +561,7 @@ static void log_pull_tail(struct gfs2_sbd *sdp, unsigned int new_tail)
 
 	gfs2_log_lock(sdp);
 	atomic_add(dist, &sdp->sd_log_blks_free);
+	trace_gfs2_log_blocks(sdp, dist);
 	gfs2_assert_withdraw(sdp, atomic_read(&sdp->sd_log_blks_free) <= sdp->sd_jdesc->jd_blocks);
 	gfs2_log_unlock(sdp);
 
@@ -715,6 +719,7 @@ void __gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl)
 		up_write(&sdp->sd_log_flush_lock);
 		return;
 	}
+	trace_gfs2_log_flush(sdp, 1);
 
 	ai = kzalloc(sizeof(struct gfs2_ail), GFP_NOFS | __GFP_NOFAIL);
 	INIT_LIST_HEAD(&ai->ai_ail1_list);
@@ -746,6 +751,7 @@ void __gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl)
 	else if (sdp->sd_log_tail != current_tail(sdp) && !sdp->sd_log_idle){
 		gfs2_log_lock(sdp);
 		atomic_dec(&sdp->sd_log_blks_free); /* Adjust for unreserved buffer */
+		trace_gfs2_log_blocks(sdp, -1);
 		gfs2_log_unlock(sdp);
 		log_write_header(sdp, 0, PULL);
 	}
@@ -763,7 +769,7 @@ void __gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl)
 		ai = NULL;
 	}
 	gfs2_log_unlock(sdp);
-
+	trace_gfs2_log_flush(sdp, 0);
 	up_write(&sdp->sd_log_flush_lock);
 
 	kfree(ai);
@@ -787,6 +793,7 @@ static void log_refund(struct gfs2_sbd *sdp, struct gfs2_trans *tr)
 	gfs2_assert_withdraw(sdp, sdp->sd_log_blks_reserved + tr->tr_reserved >= reserved);
 	unused = sdp->sd_log_blks_reserved - reserved + tr->tr_reserved;
 	atomic_add(unused, &sdp->sd_log_blks_free);
+	trace_gfs2_log_blocks(sdp, unused);
 	gfs2_assert_withdraw(sdp, atomic_read(&sdp->sd_log_blks_free) <=
 			     sdp->sd_jdesc->jd_blocks);
 	sdp->sd_log_blks_reserved = reserved;
diff --git a/fs/gfs2/lops.c b/fs/gfs2/lops.c
index 00315f50fa46..9969ff062c5b 100644
--- a/fs/gfs2/lops.c
+++ b/fs/gfs2/lops.c
@@ -27,6 +27,7 @@
 #include "rgrp.h"
 #include "trans.h"
 #include "util.h"
+#include "trace_gfs2.h"
 
 /**
  * gfs2_pin - Pin a buffer in memory
@@ -53,6 +54,7 @@ static void gfs2_pin(struct gfs2_sbd *sdp, struct buffer_head *bh)
 	if (bd->bd_ail)
 		list_move(&bd->bd_ail_st_list, &bd->bd_ail->ai_ail2_list);
 	get_bh(bh);
+	trace_gfs2_pin(bd, 1);
 }
 
 /**
@@ -89,6 +91,7 @@ static void gfs2_unpin(struct gfs2_sbd *sdp, struct buffer_head *bh,
 	bd->bd_ail = ai;
 	list_add(&bd->bd_ail_st_list, &ai->ai_ail1_list);
 	clear_bit(GLF_LFLUSH, &bd->bd_gl->gl_flags);
+	trace_gfs2_pin(bd, 0);
 	gfs2_log_unlock(sdp);
 	unlock_buffer(bh);
 }
diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index cc34f271b3e7..7bc3c45cd676 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -33,6 +33,7 @@
 #include "log.h"
 #include "quota.h"
 #include "dir.h"
+#include "trace_gfs2.h"
 
 #define DO 0
 #define UNDO 1
@@ -775,6 +776,7 @@ static int init_journal(struct gfs2_sbd *sdp, int undo)
 		/* Map the extents for this journal's blocks */
 		map_journal_extents(sdp);
 	}
+	trace_gfs2_log_blocks(sdp, atomic_read(&sdp->sd_log_blks_free));
 
 	if (sdp->sd_lockstruct.ls_first) {
 		unsigned int x;
diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
index de3239731db8..daa4ae341a29 100644
--- a/fs/gfs2/rgrp.c
+++ b/fs/gfs2/rgrp.c
@@ -29,6 +29,7 @@
 #include "util.h"
 #include "log.h"
 #include "inode.h"
+#include "trace_gfs2.h"
 
 #define BFITNOENT ((u32)~0)
 #define NO_BLOCK ((u64)~0)
@@ -1519,7 +1520,7 @@ int gfs2_alloc_block(struct gfs2_inode *ip, u64 *bn, unsigned int *n)
 	spin_lock(&sdp->sd_rindex_spin);
 	rgd->rd_free_clone -= *n;
 	spin_unlock(&sdp->sd_rindex_spin);
-
+	trace_gfs2_block_alloc(ip, block, *n, GFS2_BLKST_USED);
 	*bn = block;
 	return 0;
 
@@ -1571,7 +1572,7 @@ u64 gfs2_alloc_di(struct gfs2_inode *dip, u64 *generation)
 	spin_lock(&sdp->sd_rindex_spin);
 	rgd->rd_free_clone--;
 	spin_unlock(&sdp->sd_rindex_spin);
-
+	trace_gfs2_block_alloc(dip, block, 1, GFS2_BLKST_DINODE);
 	return block;
 }
 
@@ -1591,7 +1592,7 @@ void gfs2_free_data(struct gfs2_inode *ip, u64 bstart, u32 blen)
 	rgd = rgblk_free(sdp, bstart, blen, GFS2_BLKST_FREE);
 	if (!rgd)
 		return;
-
+	trace_gfs2_block_alloc(ip, bstart, blen, GFS2_BLKST_FREE);
 	rgd->rd_free += blen;
 
 	gfs2_trans_add_bh(rgd->rd_gl, rgd->rd_bits[0].bi_bh, 1);
@@ -1619,7 +1620,7 @@ void gfs2_free_meta(struct gfs2_inode *ip, u64 bstart, u32 blen)
 	rgd = rgblk_free(sdp, bstart, blen, GFS2_BLKST_FREE);
 	if (!rgd)
 		return;
-
+	trace_gfs2_block_alloc(ip, bstart, blen, GFS2_BLKST_FREE);
 	rgd->rd_free += blen;
 
 	gfs2_trans_add_bh(rgd->rd_gl, rgd->rd_bits[0].bi_bh, 1);
@@ -1642,6 +1643,7 @@ void gfs2_unlink_di(struct inode *inode)
 	rgd = rgblk_free(sdp, blkno, 1, GFS2_BLKST_UNLINKED);
 	if (!rgd)
 		return;
+	trace_gfs2_block_alloc(ip, blkno, 1, GFS2_BLKST_UNLINKED);
 	gfs2_trans_add_bh(rgd->rd_gl, rgd->rd_bits[0].bi_bh, 1);
 	gfs2_rgrp_out(rgd, rgd->rd_bits[0].bi_bh->b_data);
 	gfs2_trans_add_rg(rgd);
@@ -1673,6 +1675,7 @@ static void gfs2_free_uninit_di(struct gfs2_rgrpd *rgd, u64 blkno)
 void gfs2_free_di(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip)
 {
 	gfs2_free_uninit_di(rgd, ip->i_no_addr);
+	trace_gfs2_block_alloc(ip, ip->i_no_addr, 1, GFS2_BLKST_FREE);
 	gfs2_quota_change(ip, -1, ip->i_inode.i_uid, ip->i_inode.i_gid);
 	gfs2_meta_wipe(ip, ip->i_no_addr, 1);
 }
diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c
index c8930b31cdf0..0a6801336470 100644
--- a/fs/gfs2/super.c
+++ b/fs/gfs2/super.c
@@ -719,8 +719,6 @@ static void gfs2_put_super(struct super_block *sb)
 	int error;
 	struct gfs2_jdesc *jd;
 
-	lock_kernel();
-
 	/*  Unfreeze the filesystem, if we need to  */
 
 	mutex_lock(&sdp->sd_freeze_lock);
@@ -787,8 +785,6 @@ restart:
 
 	/*  At this point, we're through participating in the lockspace  */
 	gfs2_sys_fs_del(sdp);
-
-	unlock_kernel();
 }
 
 /**
diff --git a/fs/gfs2/trace_gfs2.h b/fs/gfs2/trace_gfs2.h
new file mode 100644
index 000000000000..98d6ef1c1dc0
--- /dev/null
+++ b/fs/gfs2/trace_gfs2.h
@@ -0,0 +1,407 @@
+#if !defined(_TRACE_GFS2_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_GFS2_H
+
+#include <linux/tracepoint.h>
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM gfs2
+#define TRACE_INCLUDE_FILE trace_gfs2
+
+#include <linux/fs.h>
+#include <linux/buffer_head.h>
+#include <linux/dlmconstants.h>
+#include <linux/gfs2_ondisk.h>
+#include "incore.h"
+#include "glock.h"
+
+#define dlm_state_name(nn) { DLM_LOCK_##nn, #nn }
+#define glock_trace_name(x) __print_symbolic(x,		\
+			    dlm_state_name(IV),		\
+			    dlm_state_name(NL),		\
+			    dlm_state_name(CR),		\
+			    dlm_state_name(CW),		\
+			    dlm_state_name(PR),		\
+			    dlm_state_name(PW),		\
+			    dlm_state_name(EX))
+
+#define block_state_name(x) __print_symbolic(x,			\
+			    { GFS2_BLKST_FREE, "free" },	\
+			    { GFS2_BLKST_USED, "used" },	\
+			    { GFS2_BLKST_DINODE, "dinode" },	\
+			    { GFS2_BLKST_UNLINKED, "unlinked" })
+
+#define show_glock_flags(flags) __print_flags(flags, "",	\
+	{(1UL << GLF_LOCK),			"l" },		\
+	{(1UL << GLF_DEMOTE),			"D" },		\
+	{(1UL << GLF_PENDING_DEMOTE),		"d" },		\
+	{(1UL << GLF_DEMOTE_IN_PROGRESS),	"p" },		\
+	{(1UL << GLF_DIRTY),			"y" },		\
+	{(1UL << GLF_LFLUSH),			"f" },		\
+	{(1UL << GLF_INVALIDATE_IN_PROGRESS),	"i" },		\
+	{(1UL << GLF_REPLY_PENDING),		"r" },		\
+	{(1UL << GLF_INITIAL),			"I" },		\
+	{(1UL << GLF_FROZEN),			"F" })
+
+#ifndef NUMPTY
+#define NUMPTY
+static inline u8 glock_trace_state(unsigned int state)
+{
+	switch(state) {
+	case LM_ST_SHARED:
+		return DLM_LOCK_PR;
+	case LM_ST_DEFERRED:
+		return DLM_LOCK_CW;
+	case LM_ST_EXCLUSIVE:
+		return DLM_LOCK_EX;
+	}
+	return DLM_LOCK_NL;
+}
+#endif
+
+/* Section 1 - Locking
+ *
+ * Objectives:
+ * Latency: Remote demote request to state change
+ * Latency: Local lock request to state change
+ * Latency: State change to lock grant
+ * Correctness: Ordering of local lock state vs. I/O requests
+ * Correctness: Responses to remote demote requests
+ */
+
+/* General glock state change (DLM lock request completes) */
+TRACE_EVENT(gfs2_glock_state_change,
+
+	TP_PROTO(const struct gfs2_glock *gl, unsigned int new_state),
+
+	TP_ARGS(gl, new_state),
+
+	TP_STRUCT__entry(
+		__field(	dev_t,	dev			)
+		__field(	u64,	glnum			)
+		__field(	u32,	gltype			)
+		__field(	u8,	cur_state		)
+		__field(	u8,	new_state		)
+		__field(	u8,	dmt_state		)
+		__field(	u8,	tgt_state		)
+		__field(	unsigned long,	flags		)
+	),
+
+	TP_fast_assign(
+		__entry->dev		= gl->gl_sbd->sd_vfs->s_dev;
+		__entry->glnum		= gl->gl_name.ln_number;
+		__entry->gltype		= gl->gl_name.ln_type;
+		__entry->cur_state	= glock_trace_state(gl->gl_state);
+		__entry->new_state	= glock_trace_state(new_state);
+		__entry->tgt_state	= glock_trace_state(gl->gl_target);
+		__entry->dmt_state	= glock_trace_state(gl->gl_demote_state);
+		__entry->flags		= gl->gl_flags;
+	),
+
+	TP_printk("%u,%u glock %d:%lld state %s to %s tgt:%s dmt:%s flags:%s",
+		  MAJOR(__entry->dev), MINOR(__entry->dev), __entry->gltype,
+		 (unsigned long long)__entry->glnum,
+		  glock_trace_name(__entry->cur_state),
+		  glock_trace_name(__entry->new_state),
+		  glock_trace_name(__entry->tgt_state),
+		  glock_trace_name(__entry->dmt_state),
+		  show_glock_flags(__entry->flags))
+);
+
+/* State change -> unlocked, glock is being deallocated */
+TRACE_EVENT(gfs2_glock_put,
+
+	TP_PROTO(const struct gfs2_glock *gl),
+
+	TP_ARGS(gl),
+
+	TP_STRUCT__entry(
+		__field(        dev_t,  dev                     )
+		__field(	u64,	glnum			)
+		__field(	u32,	gltype			)
+		__field(	u8,	cur_state		)
+		__field(	unsigned long,	flags		)
+	),
+
+	TP_fast_assign(
+		__entry->dev		= gl->gl_sbd->sd_vfs->s_dev;
+		__entry->gltype		= gl->gl_name.ln_type;
+		__entry->glnum		= gl->gl_name.ln_number;
+		__entry->cur_state	= glock_trace_state(gl->gl_state);
+		__entry->flags		= gl->gl_flags;
+	),
+
+	TP_printk("%u,%u glock %d:%lld state %s => %s flags:%s",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+                  __entry->gltype, (unsigned long long)__entry->glnum,
+                  glock_trace_name(__entry->cur_state),
+		  glock_trace_name(DLM_LOCK_IV),
+		  show_glock_flags(__entry->flags))
+
+);
+
+/* Callback (local or remote) requesting lock demotion */
+TRACE_EVENT(gfs2_demote_rq,
+
+	TP_PROTO(const struct gfs2_glock *gl),
+
+	TP_ARGS(gl),
+
+	TP_STRUCT__entry(
+		__field(        dev_t,  dev                     )
+		__field(	u64,	glnum			)
+		__field(	u32,	gltype			)
+		__field(	u8,	cur_state		)
+		__field(	u8,	dmt_state		)
+		__field(	unsigned long,	flags		)
+	),
+
+	TP_fast_assign(
+		__entry->dev		= gl->gl_sbd->sd_vfs->s_dev;
+		__entry->gltype		= gl->gl_name.ln_type;
+		__entry->glnum		= gl->gl_name.ln_number;
+		__entry->cur_state	= glock_trace_state(gl->gl_state);
+		__entry->dmt_state	= glock_trace_state(gl->gl_demote_state);
+		__entry->flags		= gl->gl_flags;
+	),
+
+	TP_printk("%u,%u glock %d:%lld demote %s to %s flags:%s",
+		  MAJOR(__entry->dev), MINOR(__entry->dev), __entry->gltype,
+		  (unsigned long long)__entry->glnum,
+                  glock_trace_name(__entry->cur_state),
+                  glock_trace_name(__entry->dmt_state),
+		  show_glock_flags(__entry->flags))
+
+);
+
+/* Promotion/grant of a glock */
+TRACE_EVENT(gfs2_promote,
+
+	TP_PROTO(const struct gfs2_holder *gh, int first),
+
+	TP_ARGS(gh, first),
+
+	TP_STRUCT__entry(
+		__field(        dev_t,  dev                     )
+		__field(	u64,	glnum			)
+		__field(	u32,	gltype			)
+		__field(	int,	first			)
+		__field(	u8,	state			)
+	),
+
+	TP_fast_assign(
+		__entry->dev	= gh->gh_gl->gl_sbd->sd_vfs->s_dev;
+		__entry->glnum	= gh->gh_gl->gl_name.ln_number;
+		__entry->gltype	= gh->gh_gl->gl_name.ln_type;
+		__entry->first	= first;
+		__entry->state	= glock_trace_state(gh->gh_state);
+	),
+
+	TP_printk("%u,%u glock %u:%llu promote %s %s",
+		  MAJOR(__entry->dev), MINOR(__entry->dev), __entry->gltype,
+		  (unsigned long long)__entry->glnum,
+		  __entry->first ? "first": "other",
+		  glock_trace_name(__entry->state))
+);
+
+/* Queue/dequeue a lock request */
+TRACE_EVENT(gfs2_glock_queue,
+
+	TP_PROTO(const struct gfs2_holder *gh, int queue),
+
+	TP_ARGS(gh, queue),
+
+	TP_STRUCT__entry(
+		__field(        dev_t,  dev                     )
+		__field(	u64,	glnum			)
+		__field(	u32,	gltype			)
+		__field(	int,	queue			)
+		__field(	u8,	state			)
+	),
+
+	TP_fast_assign(
+		__entry->dev	= gh->gh_gl->gl_sbd->sd_vfs->s_dev;
+		__entry->glnum	= gh->gh_gl->gl_name.ln_number;
+		__entry->gltype	= gh->gh_gl->gl_name.ln_type;
+		__entry->queue	= queue;
+		__entry->state	= glock_trace_state(gh->gh_state);
+	),
+
+	TP_printk("%u,%u glock %u:%llu %squeue %s",
+		  MAJOR(__entry->dev), MINOR(__entry->dev), __entry->gltype,
+		  (unsigned long long)__entry->glnum,
+		  __entry->queue ? "" : "de",
+		  glock_trace_name(__entry->state))
+);
+
+/* Section 2 - Log/journal
+ *
+ * Objectives:
+ * Latency: Log flush time
+ * Correctness: pin/unpin vs. disk I/O ordering
+ * Performance: Log usage stats
+ */
+
+/* Pin/unpin a block in the log */
+TRACE_EVENT(gfs2_pin,
+
+	TP_PROTO(const struct gfs2_bufdata *bd, int pin),
+
+	TP_ARGS(bd, pin),
+
+	TP_STRUCT__entry(
+		__field(        dev_t,  dev                     )
+		__field(	int,	pin			)
+		__field(	u32,	len			)
+		__field(	sector_t,	block		)
+		__field(	u64,	ino			)
+	),
+
+	TP_fast_assign(
+		__entry->dev		= bd->bd_gl->gl_sbd->sd_vfs->s_dev;
+		__entry->pin		= pin;
+		__entry->len		= bd->bd_bh->b_size;
+		__entry->block		= bd->bd_bh->b_blocknr;
+		__entry->ino		= bd->bd_gl->gl_name.ln_number;
+	),
+
+	TP_printk("%u,%u log %s %llu/%lu inode %llu",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->pin ? "pin" : "unpin",
+		  (unsigned long long)__entry->block,
+		  (unsigned long)__entry->len,
+		  (unsigned long long)__entry->ino)
+);
+
+/* Flushing the log */
+TRACE_EVENT(gfs2_log_flush,
+
+	TP_PROTO(const struct gfs2_sbd *sdp, int start),
+
+	TP_ARGS(sdp, start),
+
+	TP_STRUCT__entry(
+		__field(        dev_t,  dev                     )
+		__field(	int,	start			)
+		__field(	u64,	log_seq			)
+	),
+
+	TP_fast_assign(
+		__entry->dev            = sdp->sd_vfs->s_dev;
+		__entry->start		= start;
+		__entry->log_seq	= sdp->sd_log_sequence;
+	),
+
+	TP_printk("%u,%u log flush %s %llu",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->start ? "start" : "end",
+		  (unsigned long long)__entry->log_seq)
+);
+
+/* Reserving/releasing blocks in the log */
+TRACE_EVENT(gfs2_log_blocks,
+
+	TP_PROTO(const struct gfs2_sbd *sdp, int blocks),
+
+	TP_ARGS(sdp, blocks),
+
+	TP_STRUCT__entry(
+		__field(        dev_t,  dev                     )
+		__field(	int,	blocks			)
+	),
+
+	TP_fast_assign(
+		__entry->dev		= sdp->sd_vfs->s_dev;
+		__entry->blocks		= blocks;
+	),
+
+	TP_printk("%u,%u log reserve %d", MAJOR(__entry->dev),
+		  MINOR(__entry->dev), __entry->blocks)
+);
+
+/* Section 3 - bmap
+ *
+ * Objectives:
+ * Latency: Bmap request time
+ * Performance: Block allocator tracing
+ * Correctness: Test of disard generation vs. blocks allocated
+ */
+
+/* Map an extent of blocks, possibly a new allocation */
+TRACE_EVENT(gfs2_bmap,
+
+	TP_PROTO(const struct gfs2_inode *ip, const struct buffer_head *bh,
+		sector_t lblock, int create, int errno),
+
+	TP_ARGS(ip, bh, lblock, create, errno),
+
+	TP_STRUCT__entry(
+		__field(        dev_t,  dev                     )
+		__field(	sector_t, lblock		)
+		__field(	sector_t, pblock		)
+		__field(	u64,	inum			)
+		__field(	unsigned long, state		)
+		__field(	u32,	len			)
+		__field(	int,	create			)
+		__field(	int,	errno			)
+	),
+
+	TP_fast_assign(
+		__entry->dev            = ip->i_gl->gl_sbd->sd_vfs->s_dev;
+		__entry->lblock		= lblock;
+		__entry->pblock		= buffer_mapped(bh) ?  bh->b_blocknr : 0;
+		__entry->inum		= ip->i_no_addr;
+		__entry->state		= bh->b_state;
+		__entry->len		= bh->b_size;
+		__entry->create		= create;
+		__entry->errno		= errno;
+	),
+
+	TP_printk("%u,%u bmap %llu map %llu/%lu to %llu flags:%08lx %s %d",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  (unsigned long long)__entry->inum,
+		  (unsigned long long)__entry->lblock,
+		  (unsigned long)__entry->len,
+		  (unsigned long long)__entry->pblock,
+		  __entry->state, __entry->create ? "create " : "nocreate",
+		  __entry->errno)
+);
+
+/* Keep track of blocks as they are allocated/freed */
+TRACE_EVENT(gfs2_block_alloc,
+
+	TP_PROTO(const struct gfs2_inode *ip, u64 block, unsigned len,
+		u8 block_state),
+
+	TP_ARGS(ip, block, len, block_state),
+
+	TP_STRUCT__entry(
+		__field(        dev_t,  dev                     )
+		__field(	u64,	start			)
+		__field(	u64,	inum			)
+		__field(	u32,	len			)
+		__field(	u8,	block_state		)
+	),
+
+	TP_fast_assign(
+		__entry->dev		= ip->i_gl->gl_sbd->sd_vfs->s_dev;
+		__entry->start		= block;
+		__entry->inum		= ip->i_no_addr;
+		__entry->len		= len;
+		__entry->block_state	= block_state;
+	),
+
+	TP_printk("%u,%u bmap %llu alloc %llu/%lu %s",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  (unsigned long long)__entry->inum,
+		  (unsigned long long)__entry->start,
+		  (unsigned long)__entry->len,
+		  block_state_name(__entry->block_state))
+);
+
+#endif /* _TRACE_GFS2_H */
+
+/* This part must be outside protection */
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH .
+#include <trace/define_trace.h>
+
diff --git a/fs/inode.c b/fs/inode.c
index a88baebf77cf..f643be565df8 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -1408,7 +1408,7 @@ EXPORT_SYMBOL(touch_atime);
  *	for writeback.  Note that this function is meant exclusively for
  *	usage in the file write path of filesystems, and filesystems may
  *	choose to explicitly ignore update via this function with the
- *	S_NOCTIME inode flag, e.g. for network filesystem where these
+ *	S_NOCMTIME inode flag, e.g. for network filesystem where these
  *	timestamps are handled by the server.
  */
 
diff --git a/fs/partitions/check.c b/fs/partitions/check.c
index 0af36085eb28..1a9c7878f864 100644
--- a/fs/partitions/check.c
+++ b/fs/partitions/check.c
@@ -556,27 +556,49 @@ int rescan_partitions(struct gendisk *disk, struct block_device *bdev)
 
 	/* add partitions */
 	for (p = 1; p < state->limit; p++) {
-		sector_t size = state->parts[p].size;
-		sector_t from = state->parts[p].from;
+		sector_t size, from;
+try_scan:
+		size = state->parts[p].size;
 		if (!size)
 			continue;
+
+		from = state->parts[p].from;
 		if (from >= get_capacity(disk)) {
 			printk(KERN_WARNING
 			       "%s: p%d ignored, start %llu is behind the end of the disk\n",
 			       disk->disk_name, p, (unsigned long long) from);
 			continue;
 		}
+
 		if (from + size > get_capacity(disk)) {
-			/*
-			 * we can not ignore partitions of broken tables
-			 * created by for example camera firmware, but we
-			 * limit them to the end of the disk to avoid
-			 * creating invalid block devices
-			 */
+			struct block_device_operations *bdops = disk->fops;
+			unsigned long long capacity;
+
 			printk(KERN_WARNING
-			       "%s: p%d size %llu limited to end of disk\n",
+			       "%s: p%d size %llu exceeds device capacity, ",
 			       disk->disk_name, p, (unsigned long long) size);
-			size = get_capacity(disk) - from;
+
+			if (bdops->set_capacity &&
+			    (disk->flags & GENHD_FL_NATIVE_CAPACITY) == 0) {
+				printk(KERN_CONT "enabling native capacity\n");
+				capacity = bdops->set_capacity(disk, ~0ULL);
+				disk->flags |= GENHD_FL_NATIVE_CAPACITY;
+				if (capacity > get_capacity(disk)) {
+					set_capacity(disk, capacity);
+					check_disk_size_change(disk, bdev);
+					bdev->bd_invalidated = 0;
+				}
+				goto try_scan;
+			} else {
+				/*
+				 * we can not ignore partitions of broken tables
+				 * created by for example camera firmware, but
+				 * we limit them to the end of the disk to avoid
+				 * creating invalid block devices
+				 */
+				printk(KERN_CONT "limited to end of disk\n");
+				size = get_capacity(disk) - from;
+			}
 		}
 		part = add_partition(disk, p, from, size,
 				     state->parts[p].flags);
diff --git a/fs/xfs/Kconfig b/fs/xfs/Kconfig
index 29228f5899cd..480f28127f09 100644
--- a/fs/xfs/Kconfig
+++ b/fs/xfs/Kconfig
@@ -39,6 +39,7 @@ config XFS_QUOTA
 config XFS_POSIX_ACL
 	bool "XFS POSIX ACL support"
 	depends on XFS_FS
+	select FS_POSIX_ACL
 	help
 	  POSIX Access Control Lists (ACLs) support permissions for users and
 	  groups beyond the owner/group/world scheme.
diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile
index 60f107e47fe9..7a59daed1782 100644
--- a/fs/xfs/Makefile
+++ b/fs/xfs/Makefile
@@ -40,7 +40,7 @@ xfs-$(CONFIG_PROC_FS)		+= quota/xfs_qm_stats.o
 endif
 
 xfs-$(CONFIG_XFS_RT)		+= xfs_rtalloc.o
-xfs-$(CONFIG_XFS_POSIX_ACL)	+= xfs_acl.o
+xfs-$(CONFIG_XFS_POSIX_ACL)	+= $(XFS_LINUX)/xfs_acl.o
 xfs-$(CONFIG_PROC_FS)		+= $(XFS_LINUX)/xfs_stats.o
 xfs-$(CONFIG_SYSCTL)		+= $(XFS_LINUX)/xfs_sysctl.o
 xfs-$(CONFIG_COMPAT)		+= $(XFS_LINUX)/xfs_ioctl32.o
@@ -88,8 +88,7 @@ xfs-y				+= xfs_alloc.o \
 				   xfs_utils.o \
 				   xfs_vnodeops.o \
 				   xfs_rw.o \
-				   xfs_dmops.o \
-				   xfs_qmops.o
+				   xfs_dmops.o
 
 xfs-$(CONFIG_XFS_TRACE)		+= xfs_btree_trace.o \
 				   xfs_dir2_trace.o
diff --git a/fs/xfs/linux-2.6/xfs_acl.c b/fs/xfs/linux-2.6/xfs_acl.c
new file mode 100644
index 000000000000..1e9d1246eebc
--- /dev/null
+++ b/fs/xfs/linux-2.6/xfs_acl.c
@@ -0,0 +1,523 @@
+/*
+ * Copyright (c) 2008, Christoph Hellwig
+ * All Rights Reserved.
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it would 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 the Free Software Foundation,
+ * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+#include "xfs.h"
+#include "xfs_acl.h"
+#include "xfs_attr.h"
+#include "xfs_bmap_btree.h"
+#include "xfs_inode.h"
+#include "xfs_vnodeops.h"
+#include <linux/xattr.h>
+#include <linux/posix_acl_xattr.h>
+
+
+#define XFS_ACL_NOT_CACHED	((void *)-1)
+
+/*
+ * Locking scheme:
+ *  - all ACL updates are protected by inode->i_mutex, which is taken before
+ *    calling into this file.
+ *  - access and updates to the ip->i_acl and ip->i_default_acl pointers are
+ *    protected by inode->i_lock.
+ */
+
+STATIC struct posix_acl *
+xfs_acl_from_disk(struct xfs_acl *aclp)
+{
+	struct posix_acl_entry *acl_e;
+	struct posix_acl *acl;
+	struct xfs_acl_entry *ace;
+	int count, i;
+
+	count = be32_to_cpu(aclp->acl_cnt);
+
+	acl = posix_acl_alloc(count, GFP_KERNEL);
+	if (!acl)
+		return ERR_PTR(-ENOMEM);
+
+	for (i = 0; i < count; i++) {
+		acl_e = &acl->a_entries[i];
+		ace = &aclp->acl_entry[i];
+
+		/*
+		 * The tag is 32 bits on disk and 16 bits in core.
+		 *
+		 * Because every access to it goes through the core
+		 * format first this is not a problem.
+		 */
+		acl_e->e_tag = be32_to_cpu(ace->ae_tag);
+		acl_e->e_perm = be16_to_cpu(ace->ae_perm);
+
+		switch (acl_e->e_tag) {
+		case ACL_USER:
+		case ACL_GROUP:
+			acl_e->e_id = be32_to_cpu(ace->ae_id);
+			break;
+		case ACL_USER_OBJ:
+		case ACL_GROUP_OBJ:
+		case ACL_MASK:
+		case ACL_OTHER:
+			acl_e->e_id = ACL_UNDEFINED_ID;
+			break;
+		default:
+			goto fail;
+		}
+	}
+	return acl;
+
+fail:
+	posix_acl_release(acl);
+	return ERR_PTR(-EINVAL);
+}
+
+STATIC void
+xfs_acl_to_disk(struct xfs_acl *aclp, const struct posix_acl *acl)
+{
+	const struct posix_acl_entry *acl_e;
+	struct xfs_acl_entry *ace;
+	int i;
+
+	aclp->acl_cnt = cpu_to_be32(acl->a_count);
+	for (i = 0; i < acl->a_count; i++) {
+		ace = &aclp->acl_entry[i];
+		acl_e = &acl->a_entries[i];
+
+		ace->ae_tag = cpu_to_be32(acl_e->e_tag);
+		ace->ae_id = cpu_to_be32(acl_e->e_id);
+		ace->ae_perm = cpu_to_be16(acl_e->e_perm);
+	}
+}
+
+/*
+ * Update the cached ACL pointer in the inode.
+ *
+ * Because we don't hold any locks while reading/writing the attribute
+ * from/to disk another thread could have raced and updated the cached
+ * ACL value before us. In that case we release the previous cached value
+ * and update it with our new value.
+ */
+STATIC void
+xfs_update_cached_acl(struct inode *inode, struct posix_acl **p_acl,
+		struct posix_acl *acl)
+{
+	spin_lock(&inode->i_lock);
+	if (*p_acl && *p_acl != XFS_ACL_NOT_CACHED)
+		posix_acl_release(*p_acl);
+	*p_acl = posix_acl_dup(acl);
+	spin_unlock(&inode->i_lock);
+}
+
+struct posix_acl *
+xfs_get_acl(struct inode *inode, int type)
+{
+	struct xfs_inode *ip = XFS_I(inode);
+	struct posix_acl *acl = NULL, **p_acl;
+	struct xfs_acl *xfs_acl;
+	int len = sizeof(struct xfs_acl);
+	char *ea_name;
+	int error;
+
+	switch (type) {
+	case ACL_TYPE_ACCESS:
+		ea_name = SGI_ACL_FILE;
+		p_acl = &ip->i_acl;
+		break;
+	case ACL_TYPE_DEFAULT:
+		ea_name = SGI_ACL_DEFAULT;
+		p_acl = &ip->i_default_acl;
+		break;
+	default:
+		return ERR_PTR(-EINVAL);
+	}
+
+	spin_lock(&inode->i_lock);
+	if (*p_acl != XFS_ACL_NOT_CACHED)
+		acl = posix_acl_dup(*p_acl);
+	spin_unlock(&inode->i_lock);
+
+	/*
+	 * If we have a cached ACLs value just return it, not need to
+	 * go out to the disk.
+	 */
+	if (acl)
+		return acl;
+
+	xfs_acl = kzalloc(sizeof(struct xfs_acl), GFP_KERNEL);
+	if (!xfs_acl)
+		return ERR_PTR(-ENOMEM);
+
+	error = -xfs_attr_get(ip, ea_name, (char *)xfs_acl, &len, ATTR_ROOT);
+	if (error) {
+		/*
+		 * If the attribute doesn't exist make sure we have a negative
+		 * cache entry, for any other error assume it is transient and
+		 * leave the cache entry as XFS_ACL_NOT_CACHED.
+		 */
+		if (error == -ENOATTR) {
+			acl = NULL;
+			goto out_update_cache;
+		}
+		goto out;
+	}
+
+	acl = xfs_acl_from_disk(xfs_acl);
+	if (IS_ERR(acl))
+		goto out;
+
+ out_update_cache:
+	xfs_update_cached_acl(inode, p_acl, acl);
+ out:
+	kfree(xfs_acl);
+	return acl;
+}
+
+STATIC int
+xfs_set_acl(struct inode *inode, int type, struct posix_acl *acl)
+{
+	struct xfs_inode *ip = XFS_I(inode);
+	struct posix_acl **p_acl;
+	char *ea_name;
+	int error;
+
+	if (S_ISLNK(inode->i_mode))
+		return -EOPNOTSUPP;
+
+	switch (type) {
+	case ACL_TYPE_ACCESS:
+		ea_name = SGI_ACL_FILE;
+		p_acl = &ip->i_acl;
+		break;
+	case ACL_TYPE_DEFAULT:
+		if (!S_ISDIR(inode->i_mode))
+			return acl ? -EACCES : 0;
+		ea_name = SGI_ACL_DEFAULT;
+		p_acl = &ip->i_default_acl;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	if (acl) {
+		struct xfs_acl *xfs_acl;
+		int len;
+
+		xfs_acl = kzalloc(sizeof(struct xfs_acl), GFP_KERNEL);
+		if (!xfs_acl)
+			return -ENOMEM;
+
+		xfs_acl_to_disk(xfs_acl, acl);
+		len = sizeof(struct xfs_acl) -
+			(sizeof(struct xfs_acl_entry) *
+			 (XFS_ACL_MAX_ENTRIES - acl->a_count));
+
+		error = -xfs_attr_set(ip, ea_name, (char *)xfs_acl,
+				len, ATTR_ROOT);
+
+		kfree(xfs_acl);
+	} else {
+		/*
+		 * A NULL ACL argument means we want to remove the ACL.
+		 */
+		error = -xfs_attr_remove(ip, ea_name, ATTR_ROOT);
+
+		/*
+		 * If the attribute didn't exist to start with that's fine.
+		 */
+		if (error == -ENOATTR)
+			error = 0;
+	}
+
+	if (!error)
+		xfs_update_cached_acl(inode, p_acl, acl);
+	return error;
+}
+
+int
+xfs_check_acl(struct inode *inode, int mask)
+{
+	struct xfs_inode *ip = XFS_I(inode);
+	struct posix_acl *acl;
+	int error = -EAGAIN;
+
+	xfs_itrace_entry(ip);
+
+	/*
+	 * If there is no attribute fork no ACL exists on this inode and
+	 * we can skip the whole exercise.
+	 */
+	if (!XFS_IFORK_Q(ip))
+		return -EAGAIN;
+
+	acl = xfs_get_acl(inode, ACL_TYPE_ACCESS);
+	if (IS_ERR(acl))
+		return PTR_ERR(acl);
+	if (acl) {
+		error = posix_acl_permission(inode, acl, mask);
+		posix_acl_release(acl);
+	}
+
+	return error;
+}
+
+static int
+xfs_set_mode(struct inode *inode, mode_t mode)
+{
+	int error = 0;
+
+	if (mode != inode->i_mode) {
+		struct iattr iattr;
+
+		iattr.ia_valid = ATTR_MODE;
+		iattr.ia_mode = mode;
+
+		error = -xfs_setattr(XFS_I(inode), &iattr, XFS_ATTR_NOACL);
+	}
+
+	return error;
+}
+
+static int
+xfs_acl_exists(struct inode *inode, char *name)
+{
+	int len = sizeof(struct xfs_acl);
+
+	return (xfs_attr_get(XFS_I(inode), name, NULL, &len,
+			    ATTR_ROOT|ATTR_KERNOVAL) == 0);
+}
+
+int
+posix_acl_access_exists(struct inode *inode)
+{
+	return xfs_acl_exists(inode, SGI_ACL_FILE);
+}
+
+int
+posix_acl_default_exists(struct inode *inode)
+{
+	if (!S_ISDIR(inode->i_mode))
+		return 0;
+	return xfs_acl_exists(inode, SGI_ACL_DEFAULT);
+}
+
+/*
+ * No need for i_mutex because the inode is not yet exposed to the VFS.
+ */
+int
+xfs_inherit_acl(struct inode *inode, struct posix_acl *default_acl)
+{
+	struct posix_acl *clone;
+	mode_t mode;
+	int error = 0, inherit = 0;
+
+	if (S_ISDIR(inode->i_mode)) {
+		error = xfs_set_acl(inode, ACL_TYPE_DEFAULT, default_acl);
+		if (error)
+			return error;
+	}
+
+	clone = posix_acl_clone(default_acl, GFP_KERNEL);
+	if (!clone)
+		return -ENOMEM;
+
+	mode = inode->i_mode;
+	error = posix_acl_create_masq(clone, &mode);
+	if (error < 0)
+		goto out_release_clone;
+
+	/*
+	 * If posix_acl_create_masq returns a positive value we need to
+	 * inherit a permission that can't be represented using the Unix
+	 * mode bits and we actually need to set an ACL.
+	 */
+	if (error > 0)
+		inherit = 1;
+
+	error = xfs_set_mode(inode, mode);
+	if (error)
+		goto out_release_clone;
+
+	if (inherit)
+		error = xfs_set_acl(inode, ACL_TYPE_ACCESS, clone);
+
+ out_release_clone:
+	posix_acl_release(clone);
+	return error;
+}
+
+int
+xfs_acl_chmod(struct inode *inode)
+{
+	struct posix_acl *acl, *clone;
+	int error;
+
+	if (S_ISLNK(inode->i_mode))
+		return -EOPNOTSUPP;
+
+	acl = xfs_get_acl(inode, ACL_TYPE_ACCESS);
+	if (IS_ERR(acl) || !acl)
+		return PTR_ERR(acl);
+
+	clone = posix_acl_clone(acl, GFP_KERNEL);
+	posix_acl_release(acl);
+	if (!clone)
+		return -ENOMEM;
+
+	error = posix_acl_chmod_masq(clone, inode->i_mode);
+	if (!error)
+		error = xfs_set_acl(inode, ACL_TYPE_ACCESS, clone);
+
+	posix_acl_release(clone);
+	return error;
+}
+
+void
+xfs_inode_init_acls(struct xfs_inode *ip)
+{
+	/*
+	 * No need for locking, inode is not live yet.
+	 */
+	ip->i_acl = XFS_ACL_NOT_CACHED;
+	ip->i_default_acl = XFS_ACL_NOT_CACHED;
+}
+
+void
+xfs_inode_clear_acls(struct xfs_inode *ip)
+{
+	/*
+	 * No need for locking here, the inode is not live anymore
+	 * and just about to be freed.
+	 */
+	if (ip->i_acl != XFS_ACL_NOT_CACHED)
+		posix_acl_release(ip->i_acl);
+	if (ip->i_default_acl != XFS_ACL_NOT_CACHED)
+		posix_acl_release(ip->i_default_acl);
+}
+
+
+/*
+ * System xattr handlers.
+ *
+ * Currently Posix ACLs are the only system namespace extended attribute
+ * handlers supported by XFS, so we just implement the handlers here.
+ * If we ever support other system extended attributes this will need
+ * some refactoring.
+ */
+
+static int
+xfs_decode_acl(const char *name)
+{
+	if (strcmp(name, "posix_acl_access") == 0)
+		return ACL_TYPE_ACCESS;
+	else if (strcmp(name, "posix_acl_default") == 0)
+		return ACL_TYPE_DEFAULT;
+	return -EINVAL;
+}
+
+static int
+xfs_xattr_system_get(struct inode *inode, const char *name,
+		void *value, size_t size)
+{
+	struct posix_acl *acl;
+	int type, error;
+
+	type = xfs_decode_acl(name);
+	if (type < 0)
+		return type;
+
+	acl = xfs_get_acl(inode, type);
+	if (IS_ERR(acl))
+		return PTR_ERR(acl);
+	if (acl == NULL)
+		return -ENODATA;
+
+	error = posix_acl_to_xattr(acl, value, size);
+	posix_acl_release(acl);
+
+	return error;
+}
+
+static int
+xfs_xattr_system_set(struct inode *inode, const char *name,
+		const void *value, size_t size, int flags)
+{
+	struct posix_acl *acl = NULL;
+	int error = 0, type;
+
+	type = xfs_decode_acl(name);
+	if (type < 0)
+		return type;
+	if (flags & XATTR_CREATE)
+		return -EINVAL;
+	if (type == ACL_TYPE_DEFAULT && !S_ISDIR(inode->i_mode))
+		return value ? -EACCES : 0;
+	if ((current_fsuid() != inode->i_uid) && !capable(CAP_FOWNER))
+		return -EPERM;
+
+	if (!value)
+		goto set_acl;
+
+	acl = posix_acl_from_xattr(value, size);
+	if (!acl) {
+		/*
+		 * acl_set_file(3) may request that we set default ACLs with
+		 * zero length -- defend (gracefully) against that here.
+		 */
+		goto out;
+	}
+	if (IS_ERR(acl)) {
+		error = PTR_ERR(acl);
+		goto out;
+	}
+
+	error = posix_acl_valid(acl);
+	if (error)
+		goto out_release;
+
+	error = -EINVAL;
+	if (acl->a_count > XFS_ACL_MAX_ENTRIES)
+		goto out_release;
+
+	if (type == ACL_TYPE_ACCESS) {
+		mode_t mode = inode->i_mode;
+		error = posix_acl_equiv_mode(acl, &mode);
+
+		if (error <= 0) {
+			posix_acl_release(acl);
+			acl = NULL;
+
+			if (error < 0)
+				return error;
+		}
+
+		error = xfs_set_mode(inode, mode);
+		if (error)
+			goto out_release;
+	}
+
+ set_acl:
+	error = xfs_set_acl(inode, type, acl);
+ out_release:
+	posix_acl_release(acl);
+ out:
+	return error;
+}
+
+struct xattr_handler xfs_xattr_system_handler = {
+	.prefix	= XATTR_SYSTEM_PREFIX,
+	.get	= xfs_xattr_system_get,
+	.set	= xfs_xattr_system_set,
+};
diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c
index 34eaab608e6e..5bb523d7f37e 100644
--- a/fs/xfs/linux-2.6/xfs_ioctl.c
+++ b/fs/xfs/linux-2.6/xfs_ioctl.c
@@ -41,7 +41,6 @@
 #include "xfs_itable.h"
 #include "xfs_error.h"
 #include "xfs_rw.h"
-#include "xfs_acl.h"
 #include "xfs_attr.h"
 #include "xfs_bmap.h"
 #include "xfs_buf_item.h"
@@ -899,7 +898,8 @@ xfs_ioctl_setattr(
 	struct xfs_mount	*mp = ip->i_mount;
 	struct xfs_trans	*tp;
 	unsigned int		lock_flags = 0;
-	struct xfs_dquot	*udqp = NULL, *gdqp = NULL;
+	struct xfs_dquot	*udqp = NULL;
+	struct xfs_dquot	*gdqp = NULL;
 	struct xfs_dquot	*olddquot = NULL;
 	int			code;
 
@@ -919,7 +919,7 @@ xfs_ioctl_setattr(
 	 * because the i_*dquot fields will get updated anyway.
 	 */
 	if (XFS_IS_QUOTA_ON(mp) && (mask & FSX_PROJID)) {
-		code = XFS_QM_DQVOPALLOC(mp, ip, ip->i_d.di_uid,
+		code = xfs_qm_vop_dqalloc(ip, ip->i_d.di_uid,
 					 ip->i_d.di_gid, fa->fsx_projid,
 					 XFS_QMOPT_PQUOTA, &udqp, &gdqp);
 		if (code)
@@ -954,10 +954,11 @@ xfs_ioctl_setattr(
 	 * Do a quota reservation only if projid is actually going to change.
 	 */
 	if (mask & FSX_PROJID) {
-		if (XFS_IS_PQUOTA_ON(mp) &&
+		if (XFS_IS_QUOTA_RUNNING(mp) &&
+		    XFS_IS_PQUOTA_ON(mp) &&
 		    ip->i_d.di_projid != fa->fsx_projid) {
 			ASSERT(tp);
-			code = XFS_QM_DQVOPCHOWNRESV(mp, tp, ip, udqp, gdqp,
+			code = xfs_qm_vop_chown_reserve(tp, ip, udqp, gdqp,
 						capable(CAP_FOWNER) ?
 						XFS_QMOPT_FORCE_RES : 0);
 			if (code)	/* out of quota */
@@ -1059,8 +1060,8 @@ xfs_ioctl_setattr(
 		 * in the transaction.
 		 */
 		if (ip->i_d.di_projid != fa->fsx_projid) {
-			if (XFS_IS_PQUOTA_ON(mp)) {
-				olddquot = XFS_QM_DQVOPCHOWN(mp, tp, ip,
+			if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_PQUOTA_ON(mp)) {
+				olddquot = xfs_qm_vop_chown(tp, ip,
 							&ip->i_gdquot, gdqp);
 			}
 			ip->i_d.di_projid = fa->fsx_projid;
@@ -1106,9 +1107,9 @@ xfs_ioctl_setattr(
 	/*
 	 * Release any dquot(s) the inode had kept before chown.
 	 */
-	XFS_QM_DQRELE(mp, olddquot);
-	XFS_QM_DQRELE(mp, udqp);
-	XFS_QM_DQRELE(mp, gdqp);
+	xfs_qm_dqrele(olddquot);
+	xfs_qm_dqrele(udqp);
+	xfs_qm_dqrele(gdqp);
 
 	if (code)
 		return code;
@@ -1122,8 +1123,8 @@ xfs_ioctl_setattr(
 	return 0;
 
  error_return:
-	XFS_QM_DQRELE(mp, udqp);
-	XFS_QM_DQRELE(mp, gdqp);
+	xfs_qm_dqrele(udqp);
+	xfs_qm_dqrele(gdqp);
 	xfs_trans_cancel(tp, 0);
 	if (lock_flags)
 		xfs_iunlock(ip, lock_flags);
diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c
index 6075382336d7..58973bb46038 100644
--- a/fs/xfs/linux-2.6/xfs_iops.c
+++ b/fs/xfs/linux-2.6/xfs_iops.c
@@ -17,6 +17,7 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
+#include "xfs_acl.h"
 #include "xfs_bit.h"
 #include "xfs_log.h"
 #include "xfs_inum.h"
@@ -51,6 +52,7 @@
 #include <linux/capability.h>
 #include <linux/xattr.h>
 #include <linux/namei.h>
+#include <linux/posix_acl.h>
 #include <linux/security.h>
 #include <linux/falloc.h>
 #include <linux/fiemap.h>
@@ -202,9 +204,8 @@ xfs_vn_mknod(
 {
 	struct inode	*inode;
 	struct xfs_inode *ip = NULL;
-	xfs_acl_t	*default_acl = NULL;
+	struct posix_acl *default_acl = NULL;
 	struct xfs_name	name;
-	int (*test_default_acl)(struct inode *) = _ACL_DEFAULT_EXISTS;
 	int		error;
 
 	/*
@@ -219,18 +220,14 @@ xfs_vn_mknod(
 		rdev = 0;
 	}
 
-	if (test_default_acl && test_default_acl(dir)) {
-		if (!_ACL_ALLOC(default_acl)) {
-			return -ENOMEM;
-		}
-		if (!_ACL_GET_DEFAULT(dir, default_acl)) {
-			_ACL_FREE(default_acl);
-			default_acl = NULL;
-		}
-	}
+	if (IS_POSIXACL(dir)) {
+		default_acl = xfs_get_acl(dir, ACL_TYPE_DEFAULT);
+		if (IS_ERR(default_acl))
+			return -PTR_ERR(default_acl);
 
-	if (IS_POSIXACL(dir) && !default_acl)
-		mode &= ~current_umask();
+		if (!default_acl)
+			mode &= ~current_umask();
+	}
 
 	xfs_dentry_to_name(&name, dentry);
 	error = xfs_create(XFS_I(dir), &name, mode, rdev, &ip, NULL);
@@ -244,10 +241,10 @@ xfs_vn_mknod(
 		goto out_cleanup_inode;
 
 	if (default_acl) {
-		error = _ACL_INHERIT(inode, mode, default_acl);
+		error = -xfs_inherit_acl(inode, default_acl);
 		if (unlikely(error))
 			goto out_cleanup_inode;
-		_ACL_FREE(default_acl);
+		posix_acl_release(default_acl);
 	}
 
 
@@ -257,8 +254,7 @@ xfs_vn_mknod(
  out_cleanup_inode:
 	xfs_cleanup_inode(dir, inode, dentry);
  out_free_acl:
-	if (default_acl)
-		_ACL_FREE(default_acl);
+	posix_acl_release(default_acl);
 	return -error;
 }
 
@@ -488,26 +484,6 @@ xfs_vn_put_link(
 		kfree(s);
 }
 
-#ifdef CONFIG_XFS_POSIX_ACL
-STATIC int
-xfs_check_acl(
-	struct inode		*inode,
-	int			mask)
-{
-	struct xfs_inode	*ip = XFS_I(inode);
-	int			error;
-
-	xfs_itrace_entry(ip);
-
-	if (XFS_IFORK_Q(ip)) {
-		error = xfs_acl_iaccess(ip, mask, NULL);
-		if (error != -1)
-			return -error;
-	}
-
-	return -EAGAIN;
-}
-
 STATIC int
 xfs_vn_permission(
 	struct inode		*inode,
@@ -515,9 +491,6 @@ xfs_vn_permission(
 {
 	return generic_permission(inode, mask, xfs_check_acl);
 }
-#else
-#define xfs_vn_permission NULL
-#endif
 
 STATIC int
 xfs_vn_getattr(
diff --git a/fs/xfs/linux-2.6/xfs_lrw.c b/fs/xfs/linux-2.6/xfs_lrw.c
index 9142192ccbe6..7078974a6eee 100644
--- a/fs/xfs/linux-2.6/xfs_lrw.c
+++ b/fs/xfs/linux-2.6/xfs_lrw.c
@@ -42,7 +42,6 @@
 #include "xfs_error.h"
 #include "xfs_itable.h"
 #include "xfs_rw.h"
-#include "xfs_acl.h"
 #include "xfs_attr.h"
 #include "xfs_inode_item.h"
 #include "xfs_buf_item.h"
diff --git a/fs/xfs/linux-2.6/xfs_quotaops.c b/fs/xfs/linux-2.6/xfs_quotaops.c
index 94d9a633d3d9..cb6e2cca214f 100644
--- a/fs/xfs/linux-2.6/xfs_quotaops.c
+++ b/fs/xfs/linux-2.6/xfs_quotaops.c
@@ -50,9 +50,11 @@ xfs_fs_quota_sync(
 {
 	struct xfs_mount	*mp = XFS_M(sb);
 
+	if (sb->s_flags & MS_RDONLY)
+		return -EROFS;
 	if (!XFS_IS_QUOTA_RUNNING(mp))
 		return -ENOSYS;
-	return -xfs_sync_inodes(mp, SYNC_DELWRI);
+	return -xfs_sync_data(mp, 0);
 }
 
 STATIC int
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index 08d6bd9a3947..2e09efbca8db 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -43,7 +43,6 @@
 #include "xfs_itable.h"
 #include "xfs_fsops.h"
 #include "xfs_rw.h"
-#include "xfs_acl.h"
 #include "xfs_attr.h"
 #include "xfs_buf_item.h"
 #include "xfs_utils.h"
@@ -405,6 +404,14 @@ xfs_parseargs(
 		return EINVAL;
 	}
 
+#ifndef CONFIG_XFS_QUOTA
+	if (XFS_IS_QUOTA_RUNNING(mp)) {
+		cmn_err(CE_WARN,
+			"XFS: quota support not available in this kernel.");
+		return EINVAL;
+	}
+#endif
+
 	if ((mp->m_qflags & (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE)) &&
 	    (mp->m_qflags & (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE))) {
 		cmn_err(CE_WARN,
@@ -1063,7 +1070,18 @@ xfs_fs_put_super(
 	int			unmount_event_flags = 0;
 
 	xfs_syncd_stop(mp);
-	xfs_sync_inodes(mp, SYNC_ATTR|SYNC_DELWRI);
+
+	if (!(sb->s_flags & MS_RDONLY)) {
+		/*
+		 * XXX(hch): this should be SYNC_WAIT.
+		 *
+		 * Or more likely not needed at all because the VFS is already
+		 * calling ->sync_fs after shutting down all filestem
+		 * operations and just before calling ->put_super.
+		 */
+		xfs_sync_data(mp, 0);
+		xfs_sync_attr(mp, 0);
+	}
 
 #ifdef HAVE_DMAPI
 	if (mp->m_flags & XFS_MOUNT_DMAPI) {
@@ -1098,7 +1116,6 @@ xfs_fs_put_super(
 	xfs_freesb(mp);
 	xfs_icsb_destroy_counters(mp);
 	xfs_close_devices(mp);
-	xfs_qmops_put(mp);
 	xfs_dmops_put(mp);
 	xfs_free_fsname(mp);
 	kfree(mp);
@@ -1158,6 +1175,7 @@ xfs_fs_statfs(
 {
 	struct xfs_mount	*mp = XFS_M(dentry->d_sb);
 	xfs_sb_t		*sbp = &mp->m_sb;
+	struct xfs_inode	*ip = XFS_I(dentry->d_inode);
 	__uint64_t		fakeinos, id;
 	xfs_extlen_t		lsize;
 
@@ -1186,7 +1204,10 @@ xfs_fs_statfs(
 	statp->f_ffree = statp->f_files - (sbp->sb_icount - sbp->sb_ifree);
 	spin_unlock(&mp->m_sb_lock);
 
-	XFS_QM_DQSTATVFS(XFS_I(dentry->d_inode), statp);
+	if ((ip->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) ||
+	    ((mp->m_qflags & (XFS_PQUOTA_ACCT|XFS_OQUOTA_ENFD))) ==
+			      (XFS_PQUOTA_ACCT|XFS_OQUOTA_ENFD))
+		xfs_qm_statvfs(ip, statp);
 	return 0;
 }
 
@@ -1394,16 +1415,13 @@ xfs_fs_fill_super(
 	error = xfs_dmops_get(mp);
 	if (error)
 		goto out_free_fsname;
-	error = xfs_qmops_get(mp);
-	if (error)
-		goto out_put_dmops;
 
 	if (silent)
 		flags |= XFS_MFSI_QUIET;
 
 	error = xfs_open_devices(mp);
 	if (error)
-		goto out_put_qmops;
+		goto out_put_dmops;
 
 	if (xfs_icsb_init_counters(mp))
 		mp->m_flags |= XFS_MOUNT_NO_PERCPU_SB;
@@ -1471,8 +1489,6 @@ xfs_fs_fill_super(
  out_destroy_counters:
 	xfs_icsb_destroy_counters(mp);
 	xfs_close_devices(mp);
- out_put_qmops:
-	xfs_qmops_put(mp);
  out_put_dmops:
 	xfs_dmops_put(mp);
  out_free_fsname:
@@ -1706,18 +1722,8 @@ xfs_init_zones(void)
 	if (!xfs_ili_zone)
 		goto out_destroy_inode_zone;
 
-#ifdef CONFIG_XFS_POSIX_ACL
-	xfs_acl_zone = kmem_zone_init(sizeof(xfs_acl_t), "xfs_acl");
-	if (!xfs_acl_zone)
-		goto out_destroy_ili_zone;
-#endif
-
 	return 0;
 
-#ifdef CONFIG_XFS_POSIX_ACL
- out_destroy_ili_zone:
-#endif
-	kmem_zone_destroy(xfs_ili_zone);
  out_destroy_inode_zone:
 	kmem_zone_destroy(xfs_inode_zone);
  out_destroy_efi_zone:
@@ -1751,9 +1757,6 @@ xfs_init_zones(void)
 STATIC void
 xfs_destroy_zones(void)
 {
-#ifdef CONFIG_XFS_POSIX_ACL
-	kmem_zone_destroy(xfs_acl_zone);
-#endif
 	kmem_zone_destroy(xfs_ili_zone);
 	kmem_zone_destroy(xfs_inode_zone);
 	kmem_zone_destroy(xfs_efi_zone);
diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c
index f7ba76633c29..b619d6b8ca43 100644
--- a/fs/xfs/linux-2.6/xfs_sync.c
+++ b/fs/xfs/linux-2.6/xfs_sync.c
@@ -43,166 +43,267 @@
 #include "xfs_buf_item.h"
 #include "xfs_inode_item.h"
 #include "xfs_rw.h"
+#include "xfs_quota.h"
 
 #include <linux/kthread.h>
 #include <linux/freezer.h>
 
-/*
- * Sync all the inodes in the given AG according to the
- * direction given by the flags.
- */
-STATIC int
-xfs_sync_inodes_ag(
-	xfs_mount_t	*mp,
-	int		ag,
-	int		flags)
-{
-	xfs_perag_t	*pag = &mp->m_perag[ag];
-	int		nr_found;
-	uint32_t	first_index = 0;
-	int		error = 0;
-	int		last_error = 0;
 
-	do {
-		struct inode	*inode;
-		xfs_inode_t	*ip = NULL;
-		int		lock_flags = XFS_ILOCK_SHARED;
+STATIC xfs_inode_t *
+xfs_inode_ag_lookup(
+	struct xfs_mount	*mp,
+	struct xfs_perag	*pag,
+	uint32_t		*first_index,
+	int			tag)
+{
+	int			nr_found;
+	struct xfs_inode	*ip;
 
-		/*
-		 * use a gang lookup to find the next inode in the tree
-		 * as the tree is sparse and a gang lookup walks to find
-		 * the number of objects requested.
-		 */
-		read_lock(&pag->pag_ici_lock);
+	/*
+	 * use a gang lookup to find the next inode in the tree
+	 * as the tree is sparse and a gang lookup walks to find
+	 * the number of objects requested.
+	 */
+	read_lock(&pag->pag_ici_lock);
+	if (tag == XFS_ICI_NO_TAG) {
 		nr_found = radix_tree_gang_lookup(&pag->pag_ici_root,
-				(void**)&ip, first_index, 1);
+				(void **)&ip, *first_index, 1);
+	} else {
+		nr_found = radix_tree_gang_lookup_tag(&pag->pag_ici_root,
+				(void **)&ip, *first_index, 1, tag);
+	}
+	if (!nr_found)
+		goto unlock;
 
-		if (!nr_found) {
-			read_unlock(&pag->pag_ici_lock);
-			break;
-		}
+	/*
+	 * Update the index for the next lookup. Catch overflows
+	 * into the next AG range which can occur if we have inodes
+	 * in the last block of the AG and we are currently
+	 * pointing to the last inode.
+	 */
+	*first_index = XFS_INO_TO_AGINO(mp, ip->i_ino + 1);
+	if (*first_index < XFS_INO_TO_AGINO(mp, ip->i_ino))
+		goto unlock;
 
-		/*
-		 * Update the index for the next lookup. Catch overflows
-		 * into the next AG range which can occur if we have inodes
-		 * in the last block of the AG and we are currently
-		 * pointing to the last inode.
-		 */
-		first_index = XFS_INO_TO_AGINO(mp, ip->i_ino + 1);
-		if (first_index < XFS_INO_TO_AGINO(mp, ip->i_ino)) {
-			read_unlock(&pag->pag_ici_lock);
-			break;
-		}
+	return ip;
 
-		/* nothing to sync during shutdown */
-		if (XFS_FORCED_SHUTDOWN(mp)) {
-			read_unlock(&pag->pag_ici_lock);
-			return 0;
-		}
+unlock:
+	read_unlock(&pag->pag_ici_lock);
+	return NULL;
+}
 
-		/*
-		 * If we can't get a reference on the inode, it must be
-		 * in reclaim. Leave it for the reclaim code to flush.
-		 */
-		inode = VFS_I(ip);
-		if (!igrab(inode)) {
-			read_unlock(&pag->pag_ici_lock);
-			continue;
-		}
-		read_unlock(&pag->pag_ici_lock);
+STATIC int
+xfs_inode_ag_walk(
+	struct xfs_mount	*mp,
+	xfs_agnumber_t		ag,
+	int			(*execute)(struct xfs_inode *ip,
+					   struct xfs_perag *pag, int flags),
+	int			flags,
+	int			tag)
+{
+	struct xfs_perag	*pag = &mp->m_perag[ag];
+	uint32_t		first_index;
+	int			last_error = 0;
+	int			skipped;
 
-		/* avoid new or bad inodes */
-		if (is_bad_inode(inode) ||
-		    xfs_iflags_test(ip, XFS_INEW)) {
-			IRELE(ip);
-			continue;
-		}
+restart:
+	skipped = 0;
+	first_index = 0;
+	do {
+		int		error = 0;
+		xfs_inode_t	*ip;
 
-		/*
-		 * If we have to flush data or wait for I/O completion
-		 * we need to hold the iolock.
-		 */
-		if (flags & SYNC_DELWRI) {
-			if (VN_DIRTY(inode)) {
-				if (flags & SYNC_TRYLOCK) {
-					if (xfs_ilock_nowait(ip, XFS_IOLOCK_SHARED))
-						lock_flags |= XFS_IOLOCK_SHARED;
-				} else {
-					xfs_ilock(ip, XFS_IOLOCK_SHARED);
-					lock_flags |= XFS_IOLOCK_SHARED;
-				}
-				if (lock_flags & XFS_IOLOCK_SHARED) {
-					error = xfs_flush_pages(ip, 0, -1,
-							(flags & SYNC_WAIT) ? 0
-								: XFS_B_ASYNC,
-							FI_NONE);
-				}
-			}
-			if (VN_CACHED(inode) && (flags & SYNC_IOWAIT))
-				xfs_ioend_wait(ip);
-		}
-		xfs_ilock(ip, XFS_ILOCK_SHARED);
-
-		if ((flags & SYNC_ATTR) && !xfs_inode_clean(ip)) {
-			if (flags & SYNC_WAIT) {
-				xfs_iflock(ip);
-				if (!xfs_inode_clean(ip))
-					error = xfs_iflush(ip, XFS_IFLUSH_SYNC);
-				else
-					xfs_ifunlock(ip);
-			} else if (xfs_iflock_nowait(ip)) {
-				if (!xfs_inode_clean(ip))
-					error = xfs_iflush(ip, XFS_IFLUSH_DELWRI);
-				else
-					xfs_ifunlock(ip);
-			}
-		}
-		xfs_iput(ip, lock_flags);
+		ip = xfs_inode_ag_lookup(mp, pag, &first_index, tag);
+		if (!ip)
+			break;
 
+		error = execute(ip, pag, flags);
+		if (error == EAGAIN) {
+			skipped++;
+			continue;
+		}
 		if (error)
 			last_error = error;
 		/*
 		 * bail out if the filesystem is corrupted.
 		 */
 		if (error == EFSCORRUPTED)
-			return XFS_ERROR(error);
+			break;
 
-	} while (nr_found);
+	} while (1);
+
+	if (skipped) {
+		delay(1);
+		goto restart;
+	}
 
+	xfs_put_perag(mp, pag);
 	return last_error;
 }
 
 int
-xfs_sync_inodes(
-	xfs_mount_t	*mp,
-	int		flags)
+xfs_inode_ag_iterator(
+	struct xfs_mount	*mp,
+	int			(*execute)(struct xfs_inode *ip,
+					   struct xfs_perag *pag, int flags),
+	int			flags,
+	int			tag)
 {
-	int		error;
-	int		last_error;
-	int		i;
-	int		lflags = XFS_LOG_FORCE;
+	int			error = 0;
+	int			last_error = 0;
+	xfs_agnumber_t		ag;
 
-	if (mp->m_flags & XFS_MOUNT_RDONLY)
-		return 0;
-	error = 0;
-	last_error = 0;
+	for (ag = 0; ag < mp->m_sb.sb_agcount; ag++) {
+		if (!mp->m_perag[ag].pag_ici_init)
+			continue;
+		error = xfs_inode_ag_walk(mp, ag, execute, flags, tag);
+		if (error) {
+			last_error = error;
+			if (error == EFSCORRUPTED)
+				break;
+		}
+	}
+	return XFS_ERROR(last_error);
+}
+
+/* must be called with pag_ici_lock held and releases it */
+int
+xfs_sync_inode_valid(
+	struct xfs_inode	*ip,
+	struct xfs_perag	*pag)
+{
+	struct inode		*inode = VFS_I(ip);
+
+	/* nothing to sync during shutdown */
+	if (XFS_FORCED_SHUTDOWN(ip->i_mount)) {
+		read_unlock(&pag->pag_ici_lock);
+		return EFSCORRUPTED;
+	}
+
+	/*
+	 * If we can't get a reference on the inode, it must be in reclaim.
+	 * Leave it for the reclaim code to flush. Also avoid inodes that
+	 * haven't been fully initialised.
+	 */
+	if (!igrab(inode)) {
+		read_unlock(&pag->pag_ici_lock);
+		return ENOENT;
+	}
+	read_unlock(&pag->pag_ici_lock);
+
+	if (is_bad_inode(inode) || xfs_iflags_test(ip, XFS_INEW)) {
+		IRELE(ip);
+		return ENOENT;
+	}
+
+	return 0;
+}
+
+STATIC int
+xfs_sync_inode_data(
+	struct xfs_inode	*ip,
+	struct xfs_perag	*pag,
+	int			flags)
+{
+	struct inode		*inode = VFS_I(ip);
+	struct address_space *mapping = inode->i_mapping;
+	int			error = 0;
+
+	error = xfs_sync_inode_valid(ip, pag);
+	if (error)
+		return error;
+
+	if (!mapping_tagged(mapping, PAGECACHE_TAG_DIRTY))
+		goto out_wait;
+
+	if (!xfs_ilock_nowait(ip, XFS_IOLOCK_SHARED)) {
+		if (flags & SYNC_TRYLOCK)
+			goto out_wait;
+		xfs_ilock(ip, XFS_IOLOCK_SHARED);
+	}
+
+	error = xfs_flush_pages(ip, 0, -1, (flags & SYNC_WAIT) ?
+				0 : XFS_B_ASYNC, FI_NONE);
+	xfs_iunlock(ip, XFS_IOLOCK_SHARED);
 
+ out_wait:
 	if (flags & SYNC_WAIT)
-		lflags |= XFS_LOG_SYNC;
+		xfs_ioend_wait(ip);
+	IRELE(ip);
+	return error;
+}
 
-	for (i = 0; i < mp->m_sb.sb_agcount; i++) {
-		if (!mp->m_perag[i].pag_ici_init)
-			continue;
-		error = xfs_sync_inodes_ag(mp, i, flags);
-		if (error)
-			last_error = error;
-		if (error == EFSCORRUPTED)
-			break;
+STATIC int
+xfs_sync_inode_attr(
+	struct xfs_inode	*ip,
+	struct xfs_perag	*pag,
+	int			flags)
+{
+	int			error = 0;
+
+	error = xfs_sync_inode_valid(ip, pag);
+	if (error)
+		return error;
+
+	xfs_ilock(ip, XFS_ILOCK_SHARED);
+	if (xfs_inode_clean(ip))
+		goto out_unlock;
+	if (!xfs_iflock_nowait(ip)) {
+		if (!(flags & SYNC_WAIT))
+			goto out_unlock;
+		xfs_iflock(ip);
 	}
-	if (flags & SYNC_DELWRI)
-		xfs_log_force(mp, 0, lflags);
 
-	return XFS_ERROR(last_error);
+	if (xfs_inode_clean(ip)) {
+		xfs_ifunlock(ip);
+		goto out_unlock;
+	}
+
+	error = xfs_iflush(ip, (flags & SYNC_WAIT) ?
+			   XFS_IFLUSH_SYNC : XFS_IFLUSH_DELWRI);
+
+ out_unlock:
+	xfs_iunlock(ip, XFS_ILOCK_SHARED);
+	IRELE(ip);
+	return error;
+}
+
+/*
+ * Write out pagecache data for the whole filesystem.
+ */
+int
+xfs_sync_data(
+	struct xfs_mount	*mp,
+	int			flags)
+{
+	int			error;
+
+	ASSERT((flags & ~(SYNC_TRYLOCK|SYNC_WAIT)) == 0);
+
+	error = xfs_inode_ag_iterator(mp, xfs_sync_inode_data, flags,
+				      XFS_ICI_NO_TAG);
+	if (error)
+		return XFS_ERROR(error);
+
+	xfs_log_force(mp, 0,
+		      (flags & SYNC_WAIT) ?
+		       XFS_LOG_FORCE | XFS_LOG_SYNC :
+		       XFS_LOG_FORCE);
+	return 0;
+}
+
+/*
+ * Write out inode metadata (attributes) for the whole filesystem.
+ */
+int
+xfs_sync_attr(
+	struct xfs_mount	*mp,
+	int			flags)
+{
+	ASSERT((flags & ~SYNC_WAIT) == 0);
+
+	return xfs_inode_ag_iterator(mp, xfs_sync_inode_attr, flags,
+				     XFS_ICI_NO_TAG);
 }
 
 STATIC int
@@ -252,7 +353,7 @@ xfs_sync_fsdata(
 	 * If this is xfssyncd() then only sync the superblock if we can
 	 * lock it without sleeping and it is not pinned.
 	 */
-	if (flags & SYNC_BDFLUSH) {
+	if (flags & SYNC_TRYLOCK) {
 		ASSERT(!(flags & SYNC_WAIT));
 
 		bp = xfs_getsb(mp, XFS_BUF_TRYLOCK);
@@ -316,13 +417,13 @@ xfs_quiesce_data(
 	int error;
 
 	/* push non-blocking */
-	xfs_sync_inodes(mp, SYNC_DELWRI|SYNC_BDFLUSH);
-	XFS_QM_DQSYNC(mp, SYNC_BDFLUSH);
+	xfs_sync_data(mp, 0);
+	xfs_qm_sync(mp, SYNC_TRYLOCK);
 	xfs_filestream_flush(mp);
 
 	/* push and block */
-	xfs_sync_inodes(mp, SYNC_DELWRI|SYNC_WAIT|SYNC_IOWAIT);
-	XFS_QM_DQSYNC(mp, SYNC_WAIT);
+	xfs_sync_data(mp, SYNC_WAIT);
+	xfs_qm_sync(mp, SYNC_WAIT);
 
 	/* write superblock and hoover up shutdown errors */
 	error = xfs_sync_fsdata(mp, 0);
@@ -341,7 +442,7 @@ xfs_quiesce_fs(
 	int	count = 0, pincount;
 
 	xfs_flush_buftarg(mp->m_ddev_targp, 0);
-	xfs_reclaim_inodes(mp, 0, XFS_IFLUSH_DELWRI_ELSE_ASYNC);
+	xfs_reclaim_inodes(mp, XFS_IFLUSH_DELWRI_ELSE_ASYNC);
 
 	/*
 	 * This loop must run at least twice.  The first instance of the loop
@@ -350,7 +451,7 @@ xfs_quiesce_fs(
 	 * logged before we can write the unmount record.
 	 */
 	do {
-		xfs_sync_inodes(mp, SYNC_ATTR|SYNC_WAIT);
+		xfs_sync_attr(mp, SYNC_WAIT);
 		pincount = xfs_flush_buftarg(mp->m_ddev_targp, 1);
 		if (!pincount) {
 			delay(50);
@@ -433,8 +534,8 @@ xfs_flush_inodes_work(
 	void		*arg)
 {
 	struct inode	*inode = arg;
-	xfs_sync_inodes(mp, SYNC_DELWRI | SYNC_TRYLOCK);
-	xfs_sync_inodes(mp, SYNC_DELWRI | SYNC_TRYLOCK | SYNC_IOWAIT);
+	xfs_sync_data(mp, SYNC_TRYLOCK);
+	xfs_sync_data(mp, SYNC_TRYLOCK | SYNC_WAIT);
 	iput(inode);
 }
 
@@ -465,10 +566,10 @@ xfs_sync_worker(
 
 	if (!(mp->m_flags & XFS_MOUNT_RDONLY)) {
 		xfs_log_force(mp, (xfs_lsn_t)0, XFS_LOG_FORCE);
-		xfs_reclaim_inodes(mp, 0, XFS_IFLUSH_DELWRI_ELSE_ASYNC);
+		xfs_reclaim_inodes(mp, XFS_IFLUSH_DELWRI_ELSE_ASYNC);
 		/* dgc: errors ignored here */
-		error = XFS_QM_DQSYNC(mp, SYNC_BDFLUSH);
-		error = xfs_sync_fsdata(mp, SYNC_BDFLUSH);
+		error = xfs_qm_sync(mp, SYNC_TRYLOCK);
+		error = xfs_sync_fsdata(mp, SYNC_TRYLOCK);
 		if (xfs_log_need_covered(mp))
 			error = xfs_commit_dummy_trans(mp, XFS_LOG_FORCE);
 	}
@@ -569,7 +670,7 @@ xfs_reclaim_inode(
 			xfs_ifunlock(ip);
 			xfs_iunlock(ip, XFS_ILOCK_EXCL);
 		}
-		return 1;
+		return -EAGAIN;
 	}
 	__xfs_iflags_set(ip, XFS_IRECLAIM);
 	spin_unlock(&ip->i_flags_lock);
@@ -654,101 +755,27 @@ xfs_inode_clear_reclaim_tag(
 	xfs_put_perag(mp, pag);
 }
 
-
-STATIC void
-xfs_reclaim_inodes_ag(
-	xfs_mount_t	*mp,
-	int		ag,
-	int		noblock,
-	int		mode)
+STATIC int
+xfs_reclaim_inode_now(
+	struct xfs_inode	*ip,
+	struct xfs_perag	*pag,
+	int			flags)
 {
-	xfs_inode_t	*ip = NULL;
-	xfs_perag_t	*pag = &mp->m_perag[ag];
-	int		nr_found;
-	uint32_t	first_index;
-	int		skipped;
-
-restart:
-	first_index = 0;
-	skipped = 0;
-	do {
-		/*
-		 * use a gang lookup to find the next inode in the tree
-		 * as the tree is sparse and a gang lookup walks to find
-		 * the number of objects requested.
-		 */
-		read_lock(&pag->pag_ici_lock);
-		nr_found = radix_tree_gang_lookup_tag(&pag->pag_ici_root,
-					(void**)&ip, first_index, 1,
-					XFS_ICI_RECLAIM_TAG);
-
-		if (!nr_found) {
-			read_unlock(&pag->pag_ici_lock);
-			break;
-		}
-
-		/*
-		 * Update the index for the next lookup. Catch overflows
-		 * into the next AG range which can occur if we have inodes
-		 * in the last block of the AG and we are currently
-		 * pointing to the last inode.
-		 */
-		first_index = XFS_INO_TO_AGINO(mp, ip->i_ino + 1);
-		if (first_index < XFS_INO_TO_AGINO(mp, ip->i_ino)) {
-			read_unlock(&pag->pag_ici_lock);
-			break;
-		}
-
-		/* ignore if already under reclaim */
-		if (xfs_iflags_test(ip, XFS_IRECLAIM)) {
-			read_unlock(&pag->pag_ici_lock);
-			continue;
-		}
-
-		if (noblock) {
-			if (!xfs_ilock_nowait(ip, XFS_ILOCK_EXCL)) {
-				read_unlock(&pag->pag_ici_lock);
-				continue;
-			}
-			if (xfs_ipincount(ip) ||
-			    !xfs_iflock_nowait(ip)) {
-				xfs_iunlock(ip, XFS_ILOCK_EXCL);
-				read_unlock(&pag->pag_ici_lock);
-				continue;
-			}
-		}
+	/* ignore if already under reclaim */
+	if (xfs_iflags_test(ip, XFS_IRECLAIM)) {
 		read_unlock(&pag->pag_ici_lock);
-
-		/*
-		 * hmmm - this is an inode already in reclaim. Do
-		 * we even bother catching it here?
-		 */
-		if (xfs_reclaim_inode(ip, noblock, mode))
-			skipped++;
-	} while (nr_found);
-
-	if (skipped) {
-		delay(1);
-		goto restart;
+		return 0;
 	}
-	return;
+	read_unlock(&pag->pag_ici_lock);
 
+	return xfs_reclaim_inode(ip, 0, flags);
 }
 
 int
 xfs_reclaim_inodes(
 	xfs_mount_t	*mp,
-	int		 noblock,
 	int		mode)
 {
-	int		i;
-
-	for (i = 0; i < mp->m_sb.sb_agcount; i++) {
-		if (!mp->m_perag[i].pag_ici_init)
-			continue;
-		xfs_reclaim_inodes_ag(mp, i, noblock, mode);
-	}
-	return 0;
+	return xfs_inode_ag_iterator(mp, xfs_reclaim_inode_now, mode,
+					XFS_ICI_RECLAIM_TAG);
 }
-
-
diff --git a/fs/xfs/linux-2.6/xfs_sync.h b/fs/xfs/linux-2.6/xfs_sync.h
index 308d5bf6dfbd..2a10301c99c7 100644
--- a/fs/xfs/linux-2.6/xfs_sync.h
+++ b/fs/xfs/linux-2.6/xfs_sync.h
@@ -29,17 +29,14 @@ typedef struct xfs_sync_work {
 	struct completion	*w_completion;
 } xfs_sync_work_t;
 
-#define SYNC_ATTR		0x0001	/* sync attributes */
-#define SYNC_DELWRI		0x0002	/* look at delayed writes */
-#define SYNC_WAIT		0x0004	/* wait for i/o to complete */
-#define SYNC_BDFLUSH		0x0008	/* BDFLUSH is calling -- don't block */
-#define SYNC_IOWAIT		0x0010  /* wait for all I/O to complete */
-#define SYNC_TRYLOCK		0x0020  /* only try to lock inodes */
+#define SYNC_WAIT		0x0001	/* wait for i/o to complete */
+#define SYNC_TRYLOCK		0x0002  /* only try to lock inodes */
 
 int xfs_syncd_init(struct xfs_mount *mp);
 void xfs_syncd_stop(struct xfs_mount *mp);
 
-int xfs_sync_inodes(struct xfs_mount *mp, int flags);
+int xfs_sync_attr(struct xfs_mount *mp, int flags);
+int xfs_sync_data(struct xfs_mount *mp, int flags);
 int xfs_sync_fsdata(struct xfs_mount *mp, int flags);
 
 int xfs_quiesce_data(struct xfs_mount *mp);
@@ -48,10 +45,16 @@ void xfs_quiesce_attr(struct xfs_mount *mp);
 void xfs_flush_inodes(struct xfs_inode *ip);
 
 int xfs_reclaim_inode(struct xfs_inode *ip, int locked, int sync_mode);
-int xfs_reclaim_inodes(struct xfs_mount *mp, int noblock, int mode);
+int xfs_reclaim_inodes(struct xfs_mount *mp, int mode);
 
 void xfs_inode_set_reclaim_tag(struct xfs_inode *ip);
 void xfs_inode_clear_reclaim_tag(struct xfs_inode *ip);
 void __xfs_inode_clear_reclaim_tag(struct xfs_mount *mp, struct xfs_perag *pag,
 				struct xfs_inode *ip);
+
+int xfs_sync_inode_valid(struct xfs_inode *ip, struct xfs_perag *pag);
+int xfs_inode_ag_iterator(struct xfs_mount *mp,
+	int (*execute)(struct xfs_inode *ip, struct xfs_perag *pag, int flags),
+	int flags, int tag);
+
 #endif
diff --git a/fs/xfs/linux-2.6/xfs_xattr.c b/fs/xfs/linux-2.6/xfs_xattr.c
index 964621fde6ed..497c7fb75cc1 100644
--- a/fs/xfs/linux-2.6/xfs_xattr.c
+++ b/fs/xfs/linux-2.6/xfs_xattr.c
@@ -29,67 +29,6 @@
 #include <linux/xattr.h>
 
 
-/*
- * ACL handling.  Should eventually be moved into xfs_acl.c
- */
-
-static int
-xfs_decode_acl(const char *name)
-{
-	if (strcmp(name, "posix_acl_access") == 0)
-		return _ACL_TYPE_ACCESS;
-	else if (strcmp(name, "posix_acl_default") == 0)
-		return _ACL_TYPE_DEFAULT;
-	return -EINVAL;
-}
-
-/*
- * Get system extended attributes which at the moment only
- * includes Posix ACLs.
- */
-static int
-xfs_xattr_system_get(struct inode *inode, const char *name,
-		void *buffer, size_t size)
-{
-	int acl;
-
-	acl = xfs_decode_acl(name);
-	if (acl < 0)
-		return acl;
-
-	return xfs_acl_vget(inode, buffer, size, acl);
-}
-
-static int
-xfs_xattr_system_set(struct inode *inode, const char *name,
-		const void *value, size_t size, int flags)
-{
-	int acl;
-
-	acl = xfs_decode_acl(name);
-	if (acl < 0)
-		return acl;
-	if (flags & XATTR_CREATE)
-		return -EINVAL;
-
-	if (!value)
-		return xfs_acl_vremove(inode, acl);
-
-	return xfs_acl_vset(inode, (void *)value, size, acl);
-}
-
-static struct xattr_handler xfs_xattr_system_handler = {
-	.prefix	= XATTR_SYSTEM_PREFIX,
-	.get	= xfs_xattr_system_get,
-	.set	= xfs_xattr_system_set,
-};
-
-
-/*
- * Real xattr handling.  The only difference between the namespaces is
- * a flag passed to the low-level attr code.
- */
-
 static int
 __xfs_xattr_get(struct inode *inode, const char *name,
 		void *value, size_t size, int xflags)
@@ -199,7 +138,9 @@ struct xattr_handler *xfs_xattr_handlers[] = {
 	&xfs_xattr_user_handler,
 	&xfs_xattr_trusted_handler,
 	&xfs_xattr_security_handler,
+#ifdef CONFIG_XFS_POSIX_ACL
 	&xfs_xattr_system_handler,
+#endif
 	NULL
 };
 
@@ -310,7 +251,7 @@ xfs_vn_listxattr(struct dentry *dentry, char *data, size_t size)
 	/*
 	 * Then add the two synthetic ACL attributes.
 	 */
-	if (xfs_acl_vhasacl_access(inode)) {
+	if (posix_acl_access_exists(inode)) {
 		error = list_one_attr(POSIX_ACL_XATTR_ACCESS,
 				strlen(POSIX_ACL_XATTR_ACCESS) + 1,
 				data, size, &context.count);
@@ -318,7 +259,7 @@ xfs_vn_listxattr(struct dentry *dentry, char *data, size_t size)
 			return error;
 	}
 
-	if (xfs_acl_vhasacl_default(inode)) {
+	if (posix_acl_default_exists(inode)) {
 		error = list_one_attr(POSIX_ACL_XATTR_DEFAULT,
 				strlen(POSIX_ACL_XATTR_DEFAULT) + 1,
 				data, size, &context.count);
diff --git a/fs/xfs/quota/xfs_dquot.c b/fs/xfs/quota/xfs_dquot.c
index e4babcc63423..2f3f2229eaaf 100644
--- a/fs/xfs/quota/xfs_dquot.c
+++ b/fs/xfs/quota/xfs_dquot.c
@@ -42,7 +42,6 @@
 #include "xfs_error.h"
 #include "xfs_itable.h"
 #include "xfs_rw.h"
-#include "xfs_acl.h"
 #include "xfs_attr.h"
 #include "xfs_buf_item.h"
 #include "xfs_trans_space.h"
@@ -1194,7 +1193,9 @@ void
 xfs_qm_dqrele(
 	xfs_dquot_t	*dqp)
 {
-	ASSERT(dqp);
+	if (!dqp)
+		return;
+
 	xfs_dqtrace_entry(dqp, "DQRELE");
 
 	xfs_dqlock(dqp);
diff --git a/fs/xfs/quota/xfs_dquot.h b/fs/xfs/quota/xfs_dquot.h
index de0f402ddb4c..6533ead9b889 100644
--- a/fs/xfs/quota/xfs_dquot.h
+++ b/fs/xfs/quota/xfs_dquot.h
@@ -181,7 +181,6 @@ extern void		xfs_qm_adjust_dqlimits(xfs_mount_t *,
 extern int		xfs_qm_dqget(xfs_mount_t *, xfs_inode_t *,
 					xfs_dqid_t, uint, uint, xfs_dquot_t **);
 extern void		xfs_qm_dqput(xfs_dquot_t *);
-extern void		xfs_qm_dqrele(xfs_dquot_t *);
 extern void		xfs_dqlock(xfs_dquot_t *);
 extern void		xfs_dqlock2(xfs_dquot_t *, xfs_dquot_t *);
 extern void		xfs_dqunlock(xfs_dquot_t *);
diff --git a/fs/xfs/quota/xfs_dquot_item.c b/fs/xfs/quota/xfs_dquot_item.c
index 1728f6a7c4f5..d0d4a9a0bbd7 100644
--- a/fs/xfs/quota/xfs_dquot_item.c
+++ b/fs/xfs/quota/xfs_dquot_item.c
@@ -42,7 +42,6 @@
 #include "xfs_error.h"
 #include "xfs_itable.h"
 #include "xfs_rw.h"
-#include "xfs_acl.h"
 #include "xfs_attr.h"
 #include "xfs_buf_item.h"
 #include "xfs_trans_priv.h"
diff --git a/fs/xfs/quota/xfs_qm.c b/fs/xfs/quota/xfs_qm.c
index 5b6695049e00..45b1bfef7388 100644
--- a/fs/xfs/quota/xfs_qm.c
+++ b/fs/xfs/quota/xfs_qm.c
@@ -42,7 +42,6 @@
 #include "xfs_error.h"
 #include "xfs_bmap.h"
 #include "xfs_rw.h"
-#include "xfs_acl.h"
 #include "xfs_attr.h"
 #include "xfs_buf_item.h"
 #include "xfs_trans_space.h"
@@ -287,11 +286,13 @@ xfs_qm_rele_quotafs_ref(
  * Just destroy the quotainfo structure.
  */
 void
-xfs_qm_unmount_quotadestroy(
-	xfs_mount_t	*mp)
+xfs_qm_unmount(
+	struct xfs_mount	*mp)
 {
-	if (mp->m_quotainfo)
+	if (mp->m_quotainfo) {
+		xfs_qm_dqpurge_all(mp, XFS_QMOPT_QUOTALL | XFS_QMOPT_UMOUNTING);
 		xfs_qm_destroy_quotainfo(mp);
+	}
 }
 
 
@@ -385,8 +386,13 @@ xfs_qm_mount_quotas(
 	if (error) {
 		xfs_fs_cmn_err(CE_WARN, mp,
 			"Failed to initialize disk quotas.");
+		return;
 	}
-	return;
+
+#ifdef QUOTADEBUG
+	if (XFS_IS_QUOTA_ON(mp))
+		xfs_qm_internalqcheck(mp);
+#endif
 }
 
 /*
@@ -774,12 +780,11 @@ xfs_qm_dqattach_grouphint(
  * Given a locked inode, attach dquot(s) to it, taking U/G/P-QUOTAON
  * into account.
  * If XFS_QMOPT_DQALLOC, the dquot(s) will be allocated if needed.
- * If XFS_QMOPT_ILOCKED, then inode sent is already locked EXCL.
  * Inode may get unlocked and relocked in here, and the caller must deal with
  * the consequences.
  */
 int
-xfs_qm_dqattach(
+xfs_qm_dqattach_locked(
 	xfs_inode_t	*ip,
 	uint		flags)
 {
@@ -787,17 +792,14 @@ xfs_qm_dqattach(
 	uint		nquotas = 0;
 	int		error = 0;
 
-	if ((! XFS_IS_QUOTA_ON(mp)) ||
-	    (! XFS_NOT_DQATTACHED(mp, ip)) ||
-	    (ip->i_ino == mp->m_sb.sb_uquotino) ||
-	    (ip->i_ino == mp->m_sb.sb_gquotino))
+	if (!XFS_IS_QUOTA_RUNNING(mp) ||
+	    !XFS_IS_QUOTA_ON(mp) ||
+	    !XFS_NOT_DQATTACHED(mp, ip) ||
+	    ip->i_ino == mp->m_sb.sb_uquotino ||
+	    ip->i_ino == mp->m_sb.sb_gquotino)
 		return 0;
 
-	ASSERT((flags & XFS_QMOPT_ILOCKED) == 0 ||
-	       xfs_isilocked(ip, XFS_ILOCK_EXCL));
-
-	if (! (flags & XFS_QMOPT_ILOCKED))
-		xfs_ilock(ip, XFS_ILOCK_EXCL);
+	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 
 	if (XFS_IS_UQUOTA_ON(mp)) {
 		error = xfs_qm_dqattach_one(ip, ip->i_d.di_uid, XFS_DQ_USER,
@@ -849,8 +851,7 @@ xfs_qm_dqattach(
 		xfs_qm_dqattach_grouphint(ip->i_udquot, ip->i_gdquot);
 	}
 
-      done:
-
+ done:
 #ifdef QUOTADEBUG
 	if (! error) {
 		if (XFS_IS_UQUOTA_ON(mp))
@@ -858,15 +859,22 @@ xfs_qm_dqattach(
 		if (XFS_IS_OQUOTA_ON(mp))
 			ASSERT(ip->i_gdquot);
 	}
+	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 #endif
+	return error;
+}
 
-	if (! (flags & XFS_QMOPT_ILOCKED))
-		xfs_iunlock(ip, XFS_ILOCK_EXCL);
+int
+xfs_qm_dqattach(
+	struct xfs_inode	*ip,
+	uint			flags)
+{
+	int			error;
+
+	xfs_ilock(ip, XFS_ILOCK_EXCL);
+	error = xfs_qm_dqattach_locked(ip, flags);
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 
-#ifdef QUOTADEBUG
-	else
-		ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
-#endif
 	return error;
 }
 
@@ -896,11 +904,6 @@ xfs_qm_dqdetach(
 	}
 }
 
-/*
- * This is called to sync quotas. We can be told to use non-blocking
- * semantics by either the SYNC_BDFLUSH flag or the absence of the
- * SYNC_WAIT flag.
- */
 int
 xfs_qm_sync(
 	xfs_mount_t	*mp,
@@ -909,17 +912,13 @@ xfs_qm_sync(
 	int		recl, restarts;
 	xfs_dquot_t	*dqp;
 	uint		flush_flags;
-	boolean_t	nowait;
 	int		error;
 
-	if (! XFS_IS_QUOTA_ON(mp))
+	if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp))
 		return 0;
 
+	flush_flags = (flags & SYNC_WAIT) ? XFS_QMOPT_SYNC : XFS_QMOPT_DELWRI;
 	restarts = 0;
-	/*
-	 * We won't block unless we are asked to.
-	 */
-	nowait = (boolean_t)(flags & SYNC_BDFLUSH || (flags & SYNC_WAIT) == 0);
 
   again:
 	xfs_qm_mplist_lock(mp);
@@ -939,18 +938,10 @@ xfs_qm_sync(
 		 * don't 'seem' to be dirty. ie. don't acquire dqlock.
 		 * This is very similar to what xfs_sync does with inodes.
 		 */
-		if (flags & SYNC_BDFLUSH) {
-			if (! XFS_DQ_IS_DIRTY(dqp))
+		if (flags & SYNC_TRYLOCK) {
+			if (!XFS_DQ_IS_DIRTY(dqp))
 				continue;
-		}
-
-		if (nowait) {
-			/*
-			 * Try to acquire the dquot lock. We are NOT out of
-			 * lock order, but we just don't want to wait for this
-			 * lock, unless somebody wanted us to.
-			 */
-			if (! xfs_qm_dqlock_nowait(dqp))
+			if (!xfs_qm_dqlock_nowait(dqp))
 				continue;
 		} else {
 			xfs_dqlock(dqp);
@@ -967,7 +958,7 @@ xfs_qm_sync(
 		/* XXX a sentinel would be better */
 		recl = XFS_QI_MPLRECLAIMS(mp);
 		if (!xfs_dqflock_nowait(dqp)) {
-			if (nowait) {
+			if (flags & SYNC_TRYLOCK) {
 				xfs_dqunlock(dqp);
 				continue;
 			}
@@ -985,7 +976,6 @@ xfs_qm_sync(
 		 * Let go of the mplist lock. We don't want to hold it
 		 * across a disk write
 		 */
-		flush_flags = (nowait) ? XFS_QMOPT_DELWRI : XFS_QMOPT_SYNC;
 		xfs_qm_mplist_unlock(mp);
 		xfs_dqtrace_entry(dqp, "XQM_SYNC: DQFLUSH");
 		error = xfs_qm_dqflush(dqp, flush_flags);
@@ -2319,20 +2309,20 @@ xfs_qm_write_sb_changes(
  */
 int
 xfs_qm_vop_dqalloc(
-	xfs_mount_t	*mp,
-	xfs_inode_t	*ip,
-	uid_t		uid,
-	gid_t		gid,
-	prid_t		prid,
-	uint		flags,
-	xfs_dquot_t	**O_udqpp,
-	xfs_dquot_t	**O_gdqpp)
+	struct xfs_inode	*ip,
+	uid_t			uid,
+	gid_t			gid,
+	prid_t			prid,
+	uint			flags,
+	struct xfs_dquot	**O_udqpp,
+	struct xfs_dquot	**O_gdqpp)
 {
-	int		error;
-	xfs_dquot_t	*uq, *gq;
-	uint		lockflags;
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_dquot	*uq, *gq;
+	int			error;
+	uint			lockflags;
 
-	if (!XFS_IS_QUOTA_ON(mp))
+	if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp))
 		return 0;
 
 	lockflags = XFS_ILOCK_EXCL;
@@ -2346,8 +2336,8 @@ xfs_qm_vop_dqalloc(
 	 * if necessary. The dquot(s) will not be locked.
 	 */
 	if (XFS_NOT_DQATTACHED(mp, ip)) {
-		if ((error = xfs_qm_dqattach(ip, XFS_QMOPT_DQALLOC |
-					    XFS_QMOPT_ILOCKED))) {
+		error = xfs_qm_dqattach_locked(ip, XFS_QMOPT_DQALLOC);
+		if (error) {
 			xfs_iunlock(ip, lockflags);
 			return error;
 		}
@@ -2469,6 +2459,7 @@ xfs_qm_vop_chown(
 	uint		bfield = XFS_IS_REALTIME_INODE(ip) ?
 				 XFS_TRANS_DQ_RTBCOUNT : XFS_TRANS_DQ_BCOUNT;
 
+
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 	ASSERT(XFS_IS_QUOTA_RUNNING(ip->i_mount));
 
@@ -2508,13 +2499,13 @@ xfs_qm_vop_chown_reserve(
 	xfs_dquot_t	*gdqp,
 	uint		flags)
 {
-	int		error;
-	xfs_mount_t	*mp;
+	xfs_mount_t	*mp = ip->i_mount;
 	uint		delblks, blkflags, prjflags = 0;
 	xfs_dquot_t	*unresudq, *unresgdq, *delblksudq, *delblksgdq;
+	int		error;
+
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED));
-	mp = ip->i_mount;
 	ASSERT(XFS_IS_QUOTA_RUNNING(mp));
 
 	delblks = ip->i_delayed_blks;
@@ -2582,28 +2573,23 @@ xfs_qm_vop_chown_reserve(
 
 int
 xfs_qm_vop_rename_dqattach(
-	xfs_inode_t	**i_tab)
+	struct xfs_inode	**i_tab)
 {
-	xfs_inode_t	*ip;
-	int		i;
-	int		error;
+	struct xfs_mount	*mp = i_tab[0]->i_mount;
+	int			i;
 
-	ip = i_tab[0];
-
-	if (! XFS_IS_QUOTA_ON(ip->i_mount))
+	if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp))
 		return 0;
 
-	if (XFS_NOT_DQATTACHED(ip->i_mount, ip)) {
-		error = xfs_qm_dqattach(ip, 0);
-		if (error)
-			return error;
-	}
-	for (i = 1; (i < 4 && i_tab[i]); i++) {
+	for (i = 0; (i < 4 && i_tab[i]); i++) {
+		struct xfs_inode	*ip = i_tab[i];
+		int			error;
+
 		/*
 		 * Watch out for duplicate entries in the table.
 		 */
-		if ((ip = i_tab[i]) != i_tab[i-1]) {
-			if (XFS_NOT_DQATTACHED(ip->i_mount, ip)) {
+		if (i == 0 || ip != i_tab[i-1]) {
+			if (XFS_NOT_DQATTACHED(mp, ip)) {
 				error = xfs_qm_dqattach(ip, 0);
 				if (error)
 					return error;
@@ -2614,17 +2600,19 @@ xfs_qm_vop_rename_dqattach(
 }
 
 void
-xfs_qm_vop_dqattach_and_dqmod_newinode(
-	xfs_trans_t	*tp,
-	xfs_inode_t	*ip,
-	xfs_dquot_t	*udqp,
-	xfs_dquot_t	*gdqp)
+xfs_qm_vop_create_dqattach(
+	struct xfs_trans	*tp,
+	struct xfs_inode	*ip,
+	struct xfs_dquot	*udqp,
+	struct xfs_dquot	*gdqp)
 {
-	if (!XFS_IS_QUOTA_ON(tp->t_mountp))
+	struct xfs_mount	*mp = tp->t_mountp;
+
+	if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp))
 		return;
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
-	ASSERT(XFS_IS_QUOTA_RUNNING(tp->t_mountp));
+	ASSERT(XFS_IS_QUOTA_RUNNING(mp));
 
 	if (udqp) {
 		xfs_dqlock(udqp);
@@ -2632,7 +2620,7 @@ xfs_qm_vop_dqattach_and_dqmod_newinode(
 		xfs_dqunlock(udqp);
 		ASSERT(ip->i_udquot == NULL);
 		ip->i_udquot = udqp;
-		ASSERT(XFS_IS_UQUOTA_ON(tp->t_mountp));
+		ASSERT(XFS_IS_UQUOTA_ON(mp));
 		ASSERT(ip->i_d.di_uid == be32_to_cpu(udqp->q_core.d_id));
 		xfs_trans_mod_dquot(tp, udqp, XFS_TRANS_DQ_ICOUNT, 1);
 	}
@@ -2642,8 +2630,8 @@ xfs_qm_vop_dqattach_and_dqmod_newinode(
 		xfs_dqunlock(gdqp);
 		ASSERT(ip->i_gdquot == NULL);
 		ip->i_gdquot = gdqp;
-		ASSERT(XFS_IS_OQUOTA_ON(tp->t_mountp));
-		ASSERT((XFS_IS_GQUOTA_ON(tp->t_mountp) ?
+		ASSERT(XFS_IS_OQUOTA_ON(mp));
+		ASSERT((XFS_IS_GQUOTA_ON(mp) ?
 			ip->i_d.di_gid : ip->i_d.di_projid) ==
 				be32_to_cpu(gdqp->q_core.d_id));
 		xfs_trans_mod_dquot(tp, gdqp, XFS_TRANS_DQ_ICOUNT, 1);
diff --git a/fs/xfs/quota/xfs_qm.h b/fs/xfs/quota/xfs_qm.h
index a371954cae1b..495564b8af38 100644
--- a/fs/xfs/quota/xfs_qm.h
+++ b/fs/xfs/quota/xfs_qm.h
@@ -127,8 +127,6 @@ typedef struct xfs_quotainfo {
 } xfs_quotainfo_t;
 
 
-extern xfs_dqtrxops_t	xfs_trans_dquot_ops;
-
 extern void	xfs_trans_mod_dquot(xfs_trans_t *, xfs_dquot_t *, uint, long);
 extern int	xfs_trans_reserve_quota_bydquots(xfs_trans_t *, xfs_mount_t *,
 			xfs_dquot_t *, xfs_dquot_t *, long, long, uint);
@@ -159,17 +157,11 @@ typedef struct xfs_dquot_acct {
 #define XFS_QM_RTBWARNLIMIT	5
 
 extern void		xfs_qm_destroy_quotainfo(xfs_mount_t *);
-extern void		xfs_qm_mount_quotas(xfs_mount_t *);
 extern int		xfs_qm_quotacheck(xfs_mount_t *);
-extern void		xfs_qm_unmount_quotadestroy(xfs_mount_t *);
-extern void		xfs_qm_unmount_quotas(xfs_mount_t *);
 extern int		xfs_qm_write_sb_changes(xfs_mount_t *, __int64_t);
-extern int		xfs_qm_sync(xfs_mount_t *, int);
 
 /* dquot stuff */
 extern boolean_t	xfs_qm_dqalloc_incore(xfs_dquot_t **);
-extern int		xfs_qm_dqattach(xfs_inode_t *, uint);
-extern void		xfs_qm_dqdetach(xfs_inode_t *);
 extern int		xfs_qm_dqpurge_all(xfs_mount_t *, uint);
 extern void		xfs_qm_dqrele_all_inodes(xfs_mount_t *, uint);
 
@@ -183,19 +175,6 @@ extern int		xfs_qm_scall_getqstat(xfs_mount_t *, fs_quota_stat_t *);
 extern int		xfs_qm_scall_quotaon(xfs_mount_t *, uint);
 extern int		xfs_qm_scall_quotaoff(xfs_mount_t *, uint);
 
-/* vop stuff */
-extern int		xfs_qm_vop_dqalloc(xfs_mount_t *, xfs_inode_t *,
-					uid_t, gid_t, prid_t, uint,
-					xfs_dquot_t **, xfs_dquot_t **);
-extern void		xfs_qm_vop_dqattach_and_dqmod_newinode(
-					xfs_trans_t *, xfs_inode_t *,
-					xfs_dquot_t *, xfs_dquot_t *);
-extern int		xfs_qm_vop_rename_dqattach(xfs_inode_t **);
-extern xfs_dquot_t *	xfs_qm_vop_chown(xfs_trans_t *, xfs_inode_t *,
-					xfs_dquot_t **, xfs_dquot_t *);
-extern int		xfs_qm_vop_chown_reserve(xfs_trans_t *, xfs_inode_t *,
-					xfs_dquot_t *, xfs_dquot_t *, uint);
-
 /* list stuff */
 extern void		xfs_qm_freelist_append(xfs_frlist_t *, xfs_dquot_t *);
 extern void		xfs_qm_freelist_unlink(xfs_dquot_t *);
diff --git a/fs/xfs/quota/xfs_qm_bhv.c b/fs/xfs/quota/xfs_qm_bhv.c
index 63037c689a4b..a5346630dfae 100644
--- a/fs/xfs/quota/xfs_qm_bhv.c
+++ b/fs/xfs/quota/xfs_qm_bhv.c
@@ -42,7 +42,6 @@
 #include "xfs_rtalloc.h"
 #include "xfs_error.h"
 #include "xfs_rw.h"
-#include "xfs_acl.h"
 #include "xfs_attr.h"
 #include "xfs_buf_item.h"
 #include "xfs_qm.h"
@@ -84,7 +83,7 @@ xfs_fill_statvfs_from_dquot(
  * return a statvfs of the project, not the entire filesystem.
  * This makes such trees appear as if they are filesystems in themselves.
  */
-STATIC void
+void
 xfs_qm_statvfs(
 	xfs_inode_t		*ip,
 	struct kstatfs		*statp)
@@ -92,20 +91,13 @@ xfs_qm_statvfs(
 	xfs_mount_t		*mp = ip->i_mount;
 	xfs_dquot_t		*dqp;
 
-	if (!(ip->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) ||
-	    !((mp->m_qflags & (XFS_PQUOTA_ACCT|XFS_OQUOTA_ENFD))) ==
-	    		      (XFS_PQUOTA_ACCT|XFS_OQUOTA_ENFD))
-		return;
-
 	if (!xfs_qm_dqget(mp, NULL, ip->i_d.di_projid, XFS_DQ_PROJ, 0, &dqp)) {
-		xfs_disk_dquot_t	*dp = &dqp->q_core;
-
-		xfs_fill_statvfs_from_dquot(statp, dp);
+		xfs_fill_statvfs_from_dquot(statp, &dqp->q_core);
 		xfs_qm_dqput(dqp);
 	}
 }
 
-STATIC int
+int
 xfs_qm_newmount(
 	xfs_mount_t	*mp,
 	uint		*needquotamount,
@@ -114,9 +106,6 @@ xfs_qm_newmount(
 	uint		quotaondisk;
 	uint		uquotaondisk = 0, gquotaondisk = 0, pquotaondisk = 0;
 
-	*quotaflags = 0;
-	*needquotamount = B_FALSE;
-
 	quotaondisk = xfs_sb_version_hasquota(&mp->m_sb) &&
 				(mp->m_sb.sb_qflags & XFS_ALL_QUOTA_ACCT);
 
@@ -179,66 +168,6 @@ xfs_qm_newmount(
 	return 0;
 }
 
-STATIC int
-xfs_qm_endmount(
-	xfs_mount_t	*mp,
-	uint		needquotamount,
-	uint		quotaflags)
-{
-	if (needquotamount) {
-		ASSERT(mp->m_qflags == 0);
-		mp->m_qflags = quotaflags;
-		xfs_qm_mount_quotas(mp);
-	}
-
-#if defined(DEBUG) && defined(XFS_LOUD_RECOVERY)
-	if (! (XFS_IS_QUOTA_ON(mp)))
-		xfs_fs_cmn_err(CE_NOTE, mp, "Disk quotas not turned on");
-	else
-		xfs_fs_cmn_err(CE_NOTE, mp, "Disk quotas turned on");
-#endif
-
-#ifdef QUOTADEBUG
-	if (XFS_IS_QUOTA_ON(mp) && xfs_qm_internalqcheck(mp))
-		cmn_err(CE_WARN, "XFS: mount internalqcheck failed");
-#endif
-
-	return 0;
-}
-
-STATIC void
-xfs_qm_dqrele_null(
-	xfs_dquot_t	*dq)
-{
-	/*
-	 * Called from XFS, where we always check first for a NULL dquot.
-	 */
-	if (!dq)
-		return;
-	xfs_qm_dqrele(dq);
-}
-
-
-struct xfs_qmops xfs_qmcore_xfs = {
-	.xfs_qminit		= xfs_qm_newmount,
-	.xfs_qmdone		= xfs_qm_unmount_quotadestroy,
-	.xfs_qmmount		= xfs_qm_endmount,
-	.xfs_qmunmount		= xfs_qm_unmount_quotas,
-	.xfs_dqrele		= xfs_qm_dqrele_null,
-	.xfs_dqattach		= xfs_qm_dqattach,
-	.xfs_dqdetach		= xfs_qm_dqdetach,
-	.xfs_dqpurgeall		= xfs_qm_dqpurge_all,
-	.xfs_dqvopalloc		= xfs_qm_vop_dqalloc,
-	.xfs_dqvopcreate	= xfs_qm_vop_dqattach_and_dqmod_newinode,
-	.xfs_dqvoprename	= xfs_qm_vop_rename_dqattach,
-	.xfs_dqvopchown		= xfs_qm_vop_chown,
-	.xfs_dqvopchownresv	= xfs_qm_vop_chown_reserve,
-	.xfs_dqstatvfs		= xfs_qm_statvfs,
-	.xfs_dqsync		= xfs_qm_sync,
-	.xfs_dqtrxops		= &xfs_trans_dquot_ops,
-};
-EXPORT_SYMBOL(xfs_qmcore_xfs);
-
 void __init
 xfs_qm_init(void)
 {
diff --git a/fs/xfs/quota/xfs_qm_stats.c b/fs/xfs/quota/xfs_qm_stats.c
index 709f5f545cf5..21b08c0396a1 100644
--- a/fs/xfs/quota/xfs_qm_stats.c
+++ b/fs/xfs/quota/xfs_qm_stats.c
@@ -42,7 +42,6 @@
 #include "xfs_rtalloc.h"
 #include "xfs_error.h"
 #include "xfs_rw.h"
-#include "xfs_acl.h"
 #include "xfs_attr.h"
 #include "xfs_buf_item.h"
 #include "xfs_qm.h"
diff --git a/fs/xfs/quota/xfs_qm_syscalls.c b/fs/xfs/quota/xfs_qm_syscalls.c
index c7b66f6506ce..4e4276b956e8 100644
--- a/fs/xfs/quota/xfs_qm_syscalls.c
+++ b/fs/xfs/quota/xfs_qm_syscalls.c
@@ -45,7 +45,6 @@
 #include "xfs_rtalloc.h"
 #include "xfs_error.h"
 #include "xfs_rw.h"
-#include "xfs_acl.h"
 #include "xfs_attr.h"
 #include "xfs_buf_item.h"
 #include "xfs_utils.h"
@@ -847,105 +846,55 @@ xfs_qm_export_flags(
 }
 
 
-/*
- * Release all the dquots on the inodes in an AG.
- */
-STATIC void
-xfs_qm_dqrele_inodes_ag(
-	xfs_mount_t	*mp,
-	int		ag,
-	uint		flags)
+STATIC int
+xfs_dqrele_inode(
+	struct xfs_inode	*ip,
+	struct xfs_perag	*pag,
+	int			flags)
 {
-	xfs_inode_t	*ip = NULL;
-	xfs_perag_t	*pag = &mp->m_perag[ag];
-	int		first_index = 0;
-	int		nr_found;
-
-	do {
-		/*
-		 * use a gang lookup to find the next inode in the tree
-		 * as the tree is sparse and a gang lookup walks to find
-		 * the number of objects requested.
-		 */
-		read_lock(&pag->pag_ici_lock);
-		nr_found = radix_tree_gang_lookup(&pag->pag_ici_root,
-				(void**)&ip, first_index, 1);
-
-		if (!nr_found) {
-			read_unlock(&pag->pag_ici_lock);
-			break;
-		}
-
-		/*
-		 * Update the index for the next lookup. Catch overflows
-		 * into the next AG range which can occur if we have inodes
-		 * in the last block of the AG and we are currently
-		 * pointing to the last inode.
-		 */
-		first_index = XFS_INO_TO_AGINO(mp, ip->i_ino + 1);
-		if (first_index < XFS_INO_TO_AGINO(mp, ip->i_ino)) {
-			read_unlock(&pag->pag_ici_lock);
-			break;
-		}
-
-		/* skip quota inodes */
-		if (ip == XFS_QI_UQIP(mp) || ip == XFS_QI_GQIP(mp)) {
-			ASSERT(ip->i_udquot == NULL);
-			ASSERT(ip->i_gdquot == NULL);
-			read_unlock(&pag->pag_ici_lock);
-			continue;
-		}
+	int			error;
 
-		/*
-		 * If we can't get a reference on the inode, it must be
-		 * in reclaim. Leave it for the reclaim code to flush.
-		 */
-		if (!igrab(VFS_I(ip))) {
-			read_unlock(&pag->pag_ici_lock);
-			continue;
-		}
+	/* skip quota inodes */
+	if (ip == XFS_QI_UQIP(ip->i_mount) || ip == XFS_QI_GQIP(ip->i_mount)) {
+		ASSERT(ip->i_udquot == NULL);
+		ASSERT(ip->i_gdquot == NULL);
 		read_unlock(&pag->pag_ici_lock);
+		return 0;
+	}
 
-		/* avoid new inodes though we shouldn't find any here */
-		if (xfs_iflags_test(ip, XFS_INEW)) {
-			IRELE(ip);
-			continue;
-		}
+	error = xfs_sync_inode_valid(ip, pag);
+	if (error)
+		return error;
 
-		xfs_ilock(ip, XFS_ILOCK_EXCL);
-		if ((flags & XFS_UQUOTA_ACCT) && ip->i_udquot) {
-			xfs_qm_dqrele(ip->i_udquot);
-			ip->i_udquot = NULL;
-		}
-		if (flags & (XFS_PQUOTA_ACCT|XFS_GQUOTA_ACCT) &&
-		    ip->i_gdquot) {
-			xfs_qm_dqrele(ip->i_gdquot);
-			ip->i_gdquot = NULL;
-		}
-		xfs_iput(ip, XFS_ILOCK_EXCL);
+	xfs_ilock(ip, XFS_ILOCK_EXCL);
+	if ((flags & XFS_UQUOTA_ACCT) && ip->i_udquot) {
+		xfs_qm_dqrele(ip->i_udquot);
+		ip->i_udquot = NULL;
+	}
+	if (flags & (XFS_PQUOTA_ACCT|XFS_GQUOTA_ACCT) && ip->i_gdquot) {
+		xfs_qm_dqrele(ip->i_gdquot);
+		ip->i_gdquot = NULL;
+	}
+	xfs_iput(ip, XFS_ILOCK_EXCL);
+	IRELE(ip);
 
-	} while (nr_found);
+	return 0;
 }
 
+
 /*
  * Go thru all the inodes in the file system, releasing their dquots.
+ *
  * Note that the mount structure gets modified to indicate that quotas are off
- * AFTER this, in the case of quotaoff. This also gets called from
- * xfs_rootumount.
+ * AFTER this, in the case of quotaoff.
  */
 void
 xfs_qm_dqrele_all_inodes(
 	struct xfs_mount *mp,
 	uint		 flags)
 {
-	int		i;
-
 	ASSERT(mp->m_quotainfo);
-	for (i = 0; i < mp->m_sb.sb_agcount; i++) {
-		if (!mp->m_perag[i].pag_ici_init)
-			continue;
-		xfs_qm_dqrele_inodes_ag(mp, i, flags);
-	}
+	xfs_inode_ag_iterator(mp, xfs_dqrele_inode, flags, XFS_ICI_NO_TAG);
 }
 
 /*------------------------------------------------------------------------*/
diff --git a/fs/xfs/quota/xfs_trans_dquot.c b/fs/xfs/quota/xfs_trans_dquot.c
index 447173bcf96d..97ac9640be98 100644
--- a/fs/xfs/quota/xfs_trans_dquot.c
+++ b/fs/xfs/quota/xfs_trans_dquot.c
@@ -42,7 +42,6 @@
 #include "xfs_rtalloc.h"
 #include "xfs_error.h"
 #include "xfs_rw.h"
-#include "xfs_acl.h"
 #include "xfs_attr.h"
 #include "xfs_buf_item.h"
 #include "xfs_trans_priv.h"
@@ -111,7 +110,7 @@ xfs_trans_log_dquot(
  * Carry forward whatever is left of the quota blk reservation to
  * the spanky new transaction
  */
-STATIC void
+void
 xfs_trans_dup_dqinfo(
 	xfs_trans_t	*otp,
 	xfs_trans_t	*ntp)
@@ -167,19 +166,17 @@ xfs_trans_dup_dqinfo(
 /*
  * Wrap around mod_dquot to account for both user and group quotas.
  */
-STATIC void
+void
 xfs_trans_mod_dquot_byino(
 	xfs_trans_t	*tp,
 	xfs_inode_t	*ip,
 	uint		field,
 	long		delta)
 {
-	xfs_mount_t	*mp;
-
-	ASSERT(tp);
-	mp = tp->t_mountp;
+	xfs_mount_t	*mp = tp->t_mountp;
 
-	if (!XFS_IS_QUOTA_ON(mp) ||
+	if (!XFS_IS_QUOTA_RUNNING(mp) ||
+	    !XFS_IS_QUOTA_ON(mp) ||
 	    ip->i_ino == mp->m_sb.sb_uquotino ||
 	    ip->i_ino == mp->m_sb.sb_gquotino)
 		return;
@@ -229,6 +226,7 @@ xfs_trans_mod_dquot(
 	xfs_dqtrx_t	*qtrx;
 
 	ASSERT(tp);
+	ASSERT(XFS_IS_QUOTA_RUNNING(tp->t_mountp));
 	qtrx = NULL;
 
 	if (tp->t_dqinfo == NULL)
@@ -346,7 +344,7 @@ xfs_trans_dqlockedjoin(
  * Unreserve just the reservations done by this transaction.
  * dquot is still left locked at exit.
  */
-STATIC void
+void
 xfs_trans_apply_dquot_deltas(
 	xfs_trans_t		*tp)
 {
@@ -357,7 +355,7 @@ xfs_trans_apply_dquot_deltas(
 	long			totalbdelta;
 	long			totalrtbdelta;
 
-	if (! (tp->t_flags & XFS_TRANS_DQ_DIRTY))
+	if (!(tp->t_flags & XFS_TRANS_DQ_DIRTY))
 		return;
 
 	ASSERT(tp->t_dqinfo);
@@ -531,7 +529,7 @@ xfs_trans_apply_dquot_deltas(
  * we simply throw those away, since that's the expected behavior
  * when a transaction is curtailed without a commit.
  */
-STATIC void
+void
 xfs_trans_unreserve_and_mod_dquots(
 	xfs_trans_t		*tp)
 {
@@ -768,7 +766,7 @@ xfs_trans_reserve_quota_bydquots(
 {
 	int		resvd = 0, error;
 
-	if (!XFS_IS_QUOTA_ON(mp))
+	if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp))
 		return 0;
 
 	if (tp && tp->t_dqinfo == NULL)
@@ -811,18 +809,17 @@ xfs_trans_reserve_quota_bydquots(
  * This doesn't change the actual usage, just the reservation.
  * The inode sent in is locked.
  */
-STATIC int
+int
 xfs_trans_reserve_quota_nblks(
-	xfs_trans_t	*tp,
-	xfs_mount_t	*mp,
-	xfs_inode_t	*ip,
-	long		nblks,
-	long		ninos,
-	uint		flags)
+	struct xfs_trans	*tp,
+	struct xfs_inode	*ip,
+	long			nblks,
+	long			ninos,
+	uint			flags)
 {
-	int		error;
+	struct xfs_mount	*mp = ip->i_mount;
 
-	if (!XFS_IS_QUOTA_ON(mp))
+	if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp))
 		return 0;
 	if (XFS_IS_PQUOTA_ON(mp))
 		flags |= XFS_QMOPT_ENOSPC;
@@ -831,7 +828,6 @@ xfs_trans_reserve_quota_nblks(
 	ASSERT(ip->i_ino != mp->m_sb.sb_gquotino);
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
-	ASSERT(XFS_IS_QUOTA_RUNNING(ip->i_mount));
 	ASSERT((flags & ~(XFS_QMOPT_FORCE_RES | XFS_QMOPT_ENOSPC)) ==
 				XFS_TRANS_DQ_RES_RTBLKS ||
 	       (flags & ~(XFS_QMOPT_FORCE_RES | XFS_QMOPT_ENOSPC)) ==
@@ -840,11 +836,9 @@ xfs_trans_reserve_quota_nblks(
 	/*
 	 * Reserve nblks against these dquots, with trans as the mediator.
 	 */
-	error = xfs_trans_reserve_quota_bydquots(tp, mp,
-						 ip->i_udquot, ip->i_gdquot,
-						 nblks, ninos,
-						 flags);
-	return error;
+	return xfs_trans_reserve_quota_bydquots(tp, mp,
+						ip->i_udquot, ip->i_gdquot,
+						nblks, ninos, flags);
 }
 
 /*
@@ -895,25 +889,15 @@ STATIC void
 xfs_trans_alloc_dqinfo(
 	xfs_trans_t	*tp)
 {
-	(tp)->t_dqinfo = kmem_zone_zalloc(xfs_Gqm->qm_dqtrxzone, KM_SLEEP);
+	tp->t_dqinfo = kmem_zone_zalloc(xfs_Gqm->qm_dqtrxzone, KM_SLEEP);
 }
 
-STATIC void
+void
 xfs_trans_free_dqinfo(
 	xfs_trans_t	*tp)
 {
 	if (!tp->t_dqinfo)
 		return;
-	kmem_zone_free(xfs_Gqm->qm_dqtrxzone, (tp)->t_dqinfo);
-	(tp)->t_dqinfo = NULL;
+	kmem_zone_free(xfs_Gqm->qm_dqtrxzone, tp->t_dqinfo);
+	tp->t_dqinfo = NULL;
 }
-
-xfs_dqtrxops_t	xfs_trans_dquot_ops = {
-	.qo_dup_dqinfo			= xfs_trans_dup_dqinfo,
-	.qo_free_dqinfo			= xfs_trans_free_dqinfo,
-	.qo_mod_dquot_byino		= xfs_trans_mod_dquot_byino,
-	.qo_apply_dquot_deltas		= xfs_trans_apply_dquot_deltas,
-	.qo_reserve_quota_nblks		= xfs_trans_reserve_quota_nblks,
-	.qo_reserve_quota_bydquots	= xfs_trans_reserve_quota_bydquots,
-	.qo_unreserve_and_mod_dquots	= xfs_trans_unreserve_and_mod_dquots,
-};
diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c
deleted file mode 100644
index a8cdd73999a4..000000000000
--- a/fs/xfs/xfs_acl.c
+++ /dev/null
@@ -1,874 +0,0 @@
-/*
- * Copyright (c) 2001-2002,2005 Silicon Graphics, Inc.
- * All Rights Reserved.
- *
- * 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.
- *
- * This program is distributed in the hope that it would 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 the Free Software Foundation,
- * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
- */
-#include "xfs.h"
-#include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_bit.h"
-#include "xfs_inum.h"
-#include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
-#include "xfs_dinode.h"
-#include "xfs_inode.h"
-#include "xfs_btree.h"
-#include "xfs_acl.h"
-#include "xfs_attr.h"
-#include "xfs_vnodeops.h"
-
-#include <linux/capability.h>
-#include <linux/posix_acl_xattr.h>
-
-STATIC int	xfs_acl_setmode(struct inode *, xfs_acl_t *, int *);
-STATIC void     xfs_acl_filter_mode(mode_t, xfs_acl_t *);
-STATIC void	xfs_acl_get_endian(xfs_acl_t *);
-STATIC int	xfs_acl_access(uid_t, gid_t, xfs_acl_t *, mode_t, cred_t *);
-STATIC int	xfs_acl_invalid(xfs_acl_t *);
-STATIC void	xfs_acl_sync_mode(mode_t, xfs_acl_t *);
-STATIC void	xfs_acl_get_attr(struct inode *, xfs_acl_t *, int, int, int *);
-STATIC void	xfs_acl_set_attr(struct inode *, xfs_acl_t *, int, int *);
-STATIC int	xfs_acl_allow_set(struct inode *, int);
-
-kmem_zone_t *xfs_acl_zone;
-
-
-/*
- * Test for existence of access ACL attribute as efficiently as possible.
- */
-int
-xfs_acl_vhasacl_access(
-	struct inode	*vp)
-{
-	int		error;
-
-	xfs_acl_get_attr(vp, NULL, _ACL_TYPE_ACCESS, ATTR_KERNOVAL, &error);
-	return (error == 0);
-}
-
-/*
- * Test for existence of default ACL attribute as efficiently as possible.
- */
-int
-xfs_acl_vhasacl_default(
-	struct inode	*vp)
-{
-	int		error;
-
-	if (!S_ISDIR(vp->i_mode))
-		return 0;
-	xfs_acl_get_attr(vp, NULL, _ACL_TYPE_DEFAULT, ATTR_KERNOVAL, &error);
-	return (error == 0);
-}
-
-/*
- * Convert from extended attribute representation to in-memory for XFS.
- */
-STATIC int
-posix_acl_xattr_to_xfs(
-	posix_acl_xattr_header	*src,
-	size_t			size,
-	xfs_acl_t		*dest)
-{
-	posix_acl_xattr_entry	*src_entry;
-	xfs_acl_entry_t		*dest_entry;
-	int			n;
-
-	if (!src || !dest)
-		return EINVAL;
-
-	if (size < sizeof(posix_acl_xattr_header))
-		return EINVAL;
-
-	if (src->a_version != cpu_to_le32(POSIX_ACL_XATTR_VERSION))
-		return EOPNOTSUPP;
-
-	memset(dest, 0, sizeof(xfs_acl_t));
-	dest->acl_cnt = posix_acl_xattr_count(size);
-	if (dest->acl_cnt < 0 || dest->acl_cnt > XFS_ACL_MAX_ENTRIES)
-		return EINVAL;
-
-	/*
-	 * acl_set_file(3) may request that we set default ACLs with
-	 * zero length -- defend (gracefully) against that here.
-	 */
-	if (!dest->acl_cnt)
-		return 0;
-
-	src_entry = (posix_acl_xattr_entry *)((char *)src + sizeof(*src));
-	dest_entry = &dest->acl_entry[0];
-
-	for (n = 0; n < dest->acl_cnt; n++, src_entry++, dest_entry++) {
-		dest_entry->ae_perm = le16_to_cpu(src_entry->e_perm);
-		if (_ACL_PERM_INVALID(dest_entry->ae_perm))
-			return EINVAL;
-		dest_entry->ae_tag  = le16_to_cpu(src_entry->e_tag);
-		switch(dest_entry->ae_tag) {
-		case ACL_USER:
-		case ACL_GROUP:
-			dest_entry->ae_id = le32_to_cpu(src_entry->e_id);
-			break;
-		case ACL_USER_OBJ:
-		case ACL_GROUP_OBJ:
-		case ACL_MASK:
-		case ACL_OTHER:
-			dest_entry->ae_id = ACL_UNDEFINED_ID;
-			break;
-		default:
-			return EINVAL;
-		}
-	}
-	if (xfs_acl_invalid(dest))
-		return EINVAL;
-
-	return 0;
-}
-
-/*
- * Comparison function called from xfs_sort().
- * Primary key is ae_tag, secondary key is ae_id.
- */
-STATIC int
-xfs_acl_entry_compare(
-	const void	*va,
-	const void	*vb)
-{
-	xfs_acl_entry_t	*a = (xfs_acl_entry_t *)va,
-			*b = (xfs_acl_entry_t *)vb;
-
-	if (a->ae_tag == b->ae_tag)
-		return (a->ae_id - b->ae_id);
-	return (a->ae_tag - b->ae_tag);
-}
-
-/*
- * Convert from in-memory XFS to extended attribute representation.
- */
-STATIC int
-posix_acl_xfs_to_xattr(
-	xfs_acl_t		*src,
-	posix_acl_xattr_header	*dest,
-	size_t			size)
-{
-	int			n;
-	size_t			new_size = posix_acl_xattr_size(src->acl_cnt);
-	posix_acl_xattr_entry	*dest_entry;
-	xfs_acl_entry_t		*src_entry;
-
-	if (size < new_size)
-		return -ERANGE;
-
-	/* Need to sort src XFS ACL by <ae_tag,ae_id> */
-	xfs_sort(src->acl_entry, src->acl_cnt, sizeof(src->acl_entry[0]),
-		 xfs_acl_entry_compare);
-
-	dest->a_version = cpu_to_le32(POSIX_ACL_XATTR_VERSION);
-	dest_entry = &dest->a_entries[0];
-	src_entry = &src->acl_entry[0];
-	for (n = 0; n < src->acl_cnt; n++, dest_entry++, src_entry++) {
-		dest_entry->e_perm = cpu_to_le16(src_entry->ae_perm);
-		if (_ACL_PERM_INVALID(src_entry->ae_perm))
-			return -EINVAL;
-		dest_entry->e_tag  = cpu_to_le16(src_entry->ae_tag);
-		switch (src_entry->ae_tag) {
-		case ACL_USER:
-		case ACL_GROUP:
-			dest_entry->e_id = cpu_to_le32(src_entry->ae_id);
-				break;
-		case ACL_USER_OBJ:
-		case ACL_GROUP_OBJ:
-		case ACL_MASK:
-		case ACL_OTHER:
-			dest_entry->e_id = cpu_to_le32(ACL_UNDEFINED_ID);
-			break;
-		default:
-			return -EINVAL;
-		}
-	}
-	return new_size;
-}
-
-int
-xfs_acl_vget(
-	struct inode	*vp,
-	void		*acl,
-	size_t		size,
-	int		kind)
-{
-	int			error;
-	xfs_acl_t		*xfs_acl = NULL;
-	posix_acl_xattr_header	*ext_acl = acl;
-	int			flags = 0;
-
-	if(size) {
-		if (!(_ACL_ALLOC(xfs_acl))) {
-			error = ENOMEM;
-			goto out;
-		}
-		memset(xfs_acl, 0, sizeof(xfs_acl_t));
-	} else
-		flags = ATTR_KERNOVAL;
-
-	xfs_acl_get_attr(vp, xfs_acl, kind, flags, &error);
-	if (error)
-		goto out;
-
-	if (!size) {
-		error = -posix_acl_xattr_size(XFS_ACL_MAX_ENTRIES);
-	} else {
-		if (xfs_acl_invalid(xfs_acl)) {
-			error = EINVAL;
-			goto out;
-		}
-		if (kind == _ACL_TYPE_ACCESS)
-			xfs_acl_sync_mode(XFS_I(vp)->i_d.di_mode, xfs_acl);
-		error = -posix_acl_xfs_to_xattr(xfs_acl, ext_acl, size);
-	}
-out:
-	if(xfs_acl)
-		_ACL_FREE(xfs_acl);
-	return -error;
-}
-
-int
-xfs_acl_vremove(
-	struct inode	*vp,
-	int		kind)
-{
-	int		error;
-
-	error = xfs_acl_allow_set(vp, kind);
-	if (!error) {
-		error = xfs_attr_remove(XFS_I(vp),
-						kind == _ACL_TYPE_DEFAULT?
-						SGI_ACL_DEFAULT: SGI_ACL_FILE,
-						ATTR_ROOT);
-		if (error == ENOATTR)
-			error = 0;	/* 'scool */
-	}
-	return -error;
-}
-
-int
-xfs_acl_vset(
-	struct inode		*vp,
-	void			*acl,
-	size_t			size,
-	int			kind)
-{
-	posix_acl_xattr_header	*ext_acl = acl;
-	xfs_acl_t		*xfs_acl;
-	int			error;
-	int			basicperms = 0; /* more than std unix perms? */
-
-	if (!acl)
-		return -EINVAL;
-
-	if (!(_ACL_ALLOC(xfs_acl)))
-		return -ENOMEM;
-
-	error = posix_acl_xattr_to_xfs(ext_acl, size, xfs_acl);
-	if (error) {
-		_ACL_FREE(xfs_acl);
-		return -error;
-	}
-	if (!xfs_acl->acl_cnt) {
-		_ACL_FREE(xfs_acl);
-		return 0;
-	}
-
-	error = xfs_acl_allow_set(vp, kind);
-
-	/* Incoming ACL exists, set file mode based on its value */
-	if (!error && kind == _ACL_TYPE_ACCESS)
-		error = xfs_acl_setmode(vp, xfs_acl, &basicperms);
-
-	if (error)
-		goto out;
-
-	/*
-	 * If we have more than std unix permissions, set up the actual attr.
-	 * Otherwise, delete any existing attr.  This prevents us from
-	 * having actual attrs for permissions that can be stored in the
-	 * standard permission bits.
-	 */
-	if (!basicperms) {
-		xfs_acl_set_attr(vp, xfs_acl, kind, &error);
-	} else {
-		error = -xfs_acl_vremove(vp, _ACL_TYPE_ACCESS);
-	}
-
-out:
-	_ACL_FREE(xfs_acl);
-	return -error;
-}
-
-int
-xfs_acl_iaccess(
-	xfs_inode_t	*ip,
-	mode_t		mode,
-	cred_t		*cr)
-{
-	xfs_acl_t	*acl;
-	int		rval;
-	struct xfs_name	acl_name = {SGI_ACL_FILE, SGI_ACL_FILE_SIZE};
-
-	if (!(_ACL_ALLOC(acl)))
-		return -1;
-
-	/* If the file has no ACL return -1. */
-	rval = sizeof(xfs_acl_t);
-	if (xfs_attr_fetch(ip, &acl_name, (char *)acl, &rval, ATTR_ROOT)) {
-		_ACL_FREE(acl);
-		return -1;
-	}
-	xfs_acl_get_endian(acl);
-
-	/* If the file has an empty ACL return -1. */
-	if (acl->acl_cnt == XFS_ACL_NOT_PRESENT) {
-		_ACL_FREE(acl);
-		return -1;
-	}
-
-	/* Synchronize ACL with mode bits */
-	xfs_acl_sync_mode(ip->i_d.di_mode, acl);
-
-	rval = xfs_acl_access(ip->i_d.di_uid, ip->i_d.di_gid, acl, mode, cr);
-	_ACL_FREE(acl);
-	return rval;
-}
-
-STATIC int
-xfs_acl_allow_set(
-	struct inode	*vp,
-	int		kind)
-{
-	if (vp->i_flags & (S_IMMUTABLE|S_APPEND))
-		return EPERM;
-	if (kind == _ACL_TYPE_DEFAULT && !S_ISDIR(vp->i_mode))
-		return ENOTDIR;
-	if (vp->i_sb->s_flags & MS_RDONLY)
-		return EROFS;
-	if (XFS_I(vp)->i_d.di_uid != current_fsuid() && !capable(CAP_FOWNER))
-		return EPERM;
-	return 0;
-}
-
-/*
- * Note: cr is only used here for the capability check if the ACL test fails.
- *       It is not used to find out the credentials uid or groups etc, as was
- *       done in IRIX. It is assumed that the uid and groups for the current
- *       thread are taken from "current" instead of the cr parameter.
- */
-STATIC int
-xfs_acl_access(
-	uid_t		fuid,
-	gid_t		fgid,
-	xfs_acl_t	*fap,
-	mode_t		md,
-	cred_t		*cr)
-{
-	xfs_acl_entry_t	matched;
-	int		i, allows;
-	int		maskallows = -1;	/* true, but not 1, either */
-	int		seen_userobj = 0;
-
-	matched.ae_tag = 0;	/* Invalid type */
-	matched.ae_perm = 0;
-
-	for (i = 0; i < fap->acl_cnt; i++) {
-		/*
-		 * Break out if we've got a user_obj entry or
-		 * a user entry and the mask (and have processed USER_OBJ)
-		 */
-		if (matched.ae_tag == ACL_USER_OBJ)
-			break;
-		if (matched.ae_tag == ACL_USER) {
-			if (maskallows != -1 && seen_userobj)
-				break;
-			if (fap->acl_entry[i].ae_tag != ACL_MASK &&
-			    fap->acl_entry[i].ae_tag != ACL_USER_OBJ)
-				continue;
-		}
-		/* True if this entry allows the requested access */
-		allows = ((fap->acl_entry[i].ae_perm & md) == md);
-
-		switch (fap->acl_entry[i].ae_tag) {
-		case ACL_USER_OBJ:
-			seen_userobj = 1;
-			if (fuid != current_fsuid())
-				continue;
-			matched.ae_tag = ACL_USER_OBJ;
-			matched.ae_perm = allows;
-			break;
-		case ACL_USER:
-			if (fap->acl_entry[i].ae_id != current_fsuid())
-				continue;
-			matched.ae_tag = ACL_USER;
-			matched.ae_perm = allows;
-			break;
-		case ACL_GROUP_OBJ:
-			if ((matched.ae_tag == ACL_GROUP_OBJ ||
-			    matched.ae_tag == ACL_GROUP) && !allows)
-				continue;
-			if (!in_group_p(fgid))
-				continue;
-			matched.ae_tag = ACL_GROUP_OBJ;
-			matched.ae_perm = allows;
-			break;
-		case ACL_GROUP:
-			if ((matched.ae_tag == ACL_GROUP_OBJ ||
-			    matched.ae_tag == ACL_GROUP) && !allows)
-				continue;
-			if (!in_group_p(fap->acl_entry[i].ae_id))
-				continue;
-			matched.ae_tag = ACL_GROUP;
-			matched.ae_perm = allows;
-			break;
-		case ACL_MASK:
-			maskallows = allows;
-			break;
-		case ACL_OTHER:
-			if (matched.ae_tag != 0)
-				continue;
-			matched.ae_tag = ACL_OTHER;
-			matched.ae_perm = allows;
-			break;
-		}
-	}
-	/*
-	 * First possibility is that no matched entry allows access.
-	 * The capability to override DAC may exist, so check for it.
-	 */
-	switch (matched.ae_tag) {
-	case ACL_OTHER:
-	case ACL_USER_OBJ:
-		if (matched.ae_perm)
-			return 0;
-		break;
-	case ACL_USER:
-	case ACL_GROUP_OBJ:
-	case ACL_GROUP:
-		if (maskallows && matched.ae_perm)
-			return 0;
-		break;
-	case 0:
-		break;
-	}
-
-	/* EACCES tells generic_permission to check for capability overrides */
-	return EACCES;
-}
-
-/*
- * ACL validity checker.
- *   This acl validation routine checks each ACL entry read in makes sense.
- */
-STATIC int
-xfs_acl_invalid(
-	xfs_acl_t	*aclp)
-{
-	xfs_acl_entry_t	*entry, *e;
-	int		user = 0, group = 0, other = 0, mask = 0;
-	int		mask_required = 0;
-	int		i, j;
-
-	if (!aclp)
-		goto acl_invalid;
-
-	if (aclp->acl_cnt > XFS_ACL_MAX_ENTRIES)
-		goto acl_invalid;
-
-	for (i = 0; i < aclp->acl_cnt; i++) {
-		entry = &aclp->acl_entry[i];
-		switch (entry->ae_tag) {
-		case ACL_USER_OBJ:
-			if (user++)
-				goto acl_invalid;
-			break;
-		case ACL_GROUP_OBJ:
-			if (group++)
-				goto acl_invalid;
-			break;
-		case ACL_OTHER:
-			if (other++)
-				goto acl_invalid;
-			break;
-		case ACL_USER:
-		case ACL_GROUP:
-			for (j = i + 1; j < aclp->acl_cnt; j++) {
-				e = &aclp->acl_entry[j];
-				if (e->ae_id == entry->ae_id &&
-				    e->ae_tag == entry->ae_tag)
-					goto acl_invalid;
-			}
-			mask_required++;
-			break;
-		case ACL_MASK:
-			if (mask++)
-				goto acl_invalid;
-			break;
-		default:
-			goto acl_invalid;
-		}
-	}
-	if (!user || !group || !other || (mask_required && !mask))
-		goto acl_invalid;
-	else
-		return 0;
-acl_invalid:
-	return EINVAL;
-}
-
-/*
- * Do ACL endian conversion.
- */
-STATIC void
-xfs_acl_get_endian(
-	xfs_acl_t	*aclp)
-{
-	xfs_acl_entry_t	*ace, *end;
-
-	INT_SET(aclp->acl_cnt, ARCH_CONVERT, aclp->acl_cnt);
-	end = &aclp->acl_entry[0]+aclp->acl_cnt;
-	for (ace = &aclp->acl_entry[0]; ace < end; ace++) {
-		INT_SET(ace->ae_tag, ARCH_CONVERT, ace->ae_tag);
-		INT_SET(ace->ae_id, ARCH_CONVERT, ace->ae_id);
-		INT_SET(ace->ae_perm, ARCH_CONVERT, ace->ae_perm);
-	}
-}
-
-/*
- * Get the ACL from the EA and do endian conversion.
- */
-STATIC void
-xfs_acl_get_attr(
-	struct inode	*vp,
-	xfs_acl_t	*aclp,
-	int		kind,
-	int		flags,
-	int		*error)
-{
-	int		len = sizeof(xfs_acl_t);
-
-	ASSERT((flags & ATTR_KERNOVAL) ? (aclp == NULL) : 1);
-	flags |= ATTR_ROOT;
-	*error = xfs_attr_get(XFS_I(vp),
-					kind == _ACL_TYPE_ACCESS ?
-					SGI_ACL_FILE : SGI_ACL_DEFAULT,
-					(char *)aclp, &len, flags);
-	if (*error || (flags & ATTR_KERNOVAL))
-		return;
-	xfs_acl_get_endian(aclp);
-}
-
-/*
- * Set the EA with the ACL and do endian conversion.
- */
-STATIC void
-xfs_acl_set_attr(
-	struct inode	*vp,
-	xfs_acl_t	*aclp,
-	int		kind,
-	int		*error)
-{
-	xfs_acl_entry_t	*ace, *newace, *end;
-	xfs_acl_t	*newacl;
-	int		len;
-
-	if (!(_ACL_ALLOC(newacl))) {
-		*error = ENOMEM;
-		return;
-	}
-
-	len = sizeof(xfs_acl_t) -
-	      (sizeof(xfs_acl_entry_t) * (XFS_ACL_MAX_ENTRIES - aclp->acl_cnt));
-	end = &aclp->acl_entry[0]+aclp->acl_cnt;
-	for (ace = &aclp->acl_entry[0], newace = &newacl->acl_entry[0];
-	     ace < end;
-	     ace++, newace++) {
-		INT_SET(newace->ae_tag, ARCH_CONVERT, ace->ae_tag);
-		INT_SET(newace->ae_id, ARCH_CONVERT, ace->ae_id);
-		INT_SET(newace->ae_perm, ARCH_CONVERT, ace->ae_perm);
-	}
-	INT_SET(newacl->acl_cnt, ARCH_CONVERT, aclp->acl_cnt);
-	*error = xfs_attr_set(XFS_I(vp),
-				kind == _ACL_TYPE_ACCESS ?
-				SGI_ACL_FILE: SGI_ACL_DEFAULT,
-				(char *)newacl, len, ATTR_ROOT);
-	_ACL_FREE(newacl);
-}
-
-int
-xfs_acl_vtoacl(
-	struct inode	*vp,
-	xfs_acl_t	*access_acl,
-	xfs_acl_t	*default_acl)
-{
-	int		error = 0;
-
-	if (access_acl) {
-		/*
-		 * Get the Access ACL and the mode.  If either cannot
-		 * be obtained for some reason, invalidate the access ACL.
-		 */
-		xfs_acl_get_attr(vp, access_acl, _ACL_TYPE_ACCESS, 0, &error);
-		if (error)
-			access_acl->acl_cnt = XFS_ACL_NOT_PRESENT;
-		else /* We have a good ACL and the file mode, synchronize. */
-			xfs_acl_sync_mode(XFS_I(vp)->i_d.di_mode, access_acl);
-	}
-
-	if (default_acl) {
-		xfs_acl_get_attr(vp, default_acl, _ACL_TYPE_DEFAULT, 0, &error);
-		if (error)
-			default_acl->acl_cnt = XFS_ACL_NOT_PRESENT;
-	}
-	return error;
-}
-
-/*
- * This function retrieves the parent directory's acl, processes it
- * and lets the child inherit the acl(s) that it should.
- */
-int
-xfs_acl_inherit(
-	struct inode	*vp,
-	mode_t		mode,
-	xfs_acl_t	*pdaclp)
-{
-	xfs_acl_t	*cacl;
-	int		error = 0;
-	int		basicperms = 0;
-
-	/*
-	 * If the parent does not have a default ACL, or it's an
-	 * invalid ACL, we're done.
-	 */
-	if (!vp)
-		return 0;
-	if (!pdaclp || xfs_acl_invalid(pdaclp))
-		return 0;
-
-	/*
-	 * Copy the default ACL of the containing directory to
-	 * the access ACL of the new file and use the mode that
-	 * was passed in to set up the correct initial values for
-	 * the u::,g::[m::], and o:: entries.  This is what makes
-	 * umask() "work" with ACL's.
-	 */
-
-	if (!(_ACL_ALLOC(cacl)))
-		return ENOMEM;
-
-	memcpy(cacl, pdaclp, sizeof(xfs_acl_t));
-	xfs_acl_filter_mode(mode, cacl);
-	error = xfs_acl_setmode(vp, cacl, &basicperms);
-	if (error)
-		goto out_error;
-
-	/*
-	 * Set the Default and Access ACL on the file.  The mode is already
-	 * set on the file, so we don't need to worry about that.
-	 *
-	 * If the new file is a directory, its default ACL is a copy of
-	 * the containing directory's default ACL.
-	 */
-	if (S_ISDIR(vp->i_mode))
-		xfs_acl_set_attr(vp, pdaclp, _ACL_TYPE_DEFAULT, &error);
-	if (!error && !basicperms)
-		xfs_acl_set_attr(vp, cacl, _ACL_TYPE_ACCESS, &error);
-out_error:
-	_ACL_FREE(cacl);
-	return error;
-}
-
-/*
- * Set up the correct mode on the file based on the supplied ACL.  This
- * makes sure that the mode on the file reflects the state of the
- * u::,g::[m::], and o:: entries in the ACL.  Since the mode is where
- * the ACL is going to get the permissions for these entries, we must
- * synchronize the mode whenever we set the ACL on a file.
- */
-STATIC int
-xfs_acl_setmode(
-	struct inode	*vp,
-	xfs_acl_t	*acl,
-	int		*basicperms)
-{
-	struct iattr	iattr;
-	xfs_acl_entry_t	*ap;
-	xfs_acl_entry_t	*gap = NULL;
-	int		i, nomask = 1;
-
-	*basicperms = 1;
-
-	if (acl->acl_cnt == XFS_ACL_NOT_PRESENT)
-		return 0;
-
-	/*
-	 * Copy the u::, g::, o::, and m:: bits from the ACL into the
-	 * mode.  The m:: bits take precedence over the g:: bits.
-	 */
-	iattr.ia_valid = ATTR_MODE;
-	iattr.ia_mode = XFS_I(vp)->i_d.di_mode;
-	iattr.ia_mode &= ~(S_IRWXU|S_IRWXG|S_IRWXO);
-	ap = acl->acl_entry;
-	for (i = 0; i < acl->acl_cnt; ++i) {
-		switch (ap->ae_tag) {
-		case ACL_USER_OBJ:
-			iattr.ia_mode |= ap->ae_perm << 6;
-			break;
-		case ACL_GROUP_OBJ:
-			gap = ap;
-			break;
-		case ACL_MASK:	/* more than just standard modes */
-			nomask = 0;
-			iattr.ia_mode |= ap->ae_perm << 3;
-			*basicperms = 0;
-			break;
-		case ACL_OTHER:
-			iattr.ia_mode |= ap->ae_perm;
-			break;
-		default:	/* more than just standard modes */
-			*basicperms = 0;
-			break;
-		}
-		ap++;
-	}
-
-	/* Set the group bits from ACL_GROUP_OBJ if there's no ACL_MASK */
-	if (gap && nomask)
-		iattr.ia_mode |= gap->ae_perm << 3;
-
-	return xfs_setattr(XFS_I(vp), &iattr, 0);
-}
-
-/*
- * The permissions for the special ACL entries (u::, g::[m::], o::) are
- * actually stored in the file mode (if there is both a group and a mask,
- * the group is stored in the ACL entry and the mask is stored on the file).
- * This allows the mode to remain automatically in sync with the ACL without
- * the need for a call-back to the ACL system at every point where the mode
- * could change.  This function takes the permissions from the specified mode
- * and places it in the supplied ACL.
- *
- * This implementation draws its validity from the fact that, when the ACL
- * was assigned, the mode was copied from the ACL.
- * If the mode did not change, therefore, the mode remains exactly what was
- * taken from the special ACL entries at assignment.
- * If a subsequent chmod() was done, the POSIX spec says that the change in
- * mode must cause an update to the ACL seen at user level and used for
- * access checks.  Before and after a mode change, therefore, the file mode
- * most accurately reflects what the special ACL entries should permit/deny.
- *
- * CAVEAT: If someone sets the SGI_ACL_FILE attribute directly,
- *         the existing mode bits will override whatever is in the
- *         ACL. Similarly, if there is a pre-existing ACL that was
- *         never in sync with its mode (owing to a bug in 6.5 and
- *         before), it will now magically (or mystically) be
- *         synchronized.  This could cause slight astonishment, but
- *         it is better than inconsistent permissions.
- *
- * The supplied ACL is a template that may contain any combination
- * of special entries.  These are treated as place holders when we fill
- * out the ACL.  This routine does not add or remove special entries, it
- * simply unites each special entry with its associated set of permissions.
- */
-STATIC void
-xfs_acl_sync_mode(
-	mode_t		mode,
-	xfs_acl_t	*acl)
-{
-	int		i, nomask = 1;
-	xfs_acl_entry_t	*ap;
-	xfs_acl_entry_t	*gap = NULL;
-
-	/*
-	 * Set ACL entries. POSIX1003.1eD16 requires that the MASK
-	 * be set instead of the GROUP entry, if there is a MASK.
-	 */
-	for (ap = acl->acl_entry, i = 0; i < acl->acl_cnt; ap++, i++) {
-		switch (ap->ae_tag) {
-		case ACL_USER_OBJ:
-			ap->ae_perm = (mode >> 6) & 0x7;
-			break;
-		case ACL_GROUP_OBJ:
-			gap = ap;
-			break;
-		case ACL_MASK:
-			nomask = 0;
-			ap->ae_perm = (mode >> 3) & 0x7;
-			break;
-		case ACL_OTHER:
-			ap->ae_perm = mode & 0x7;
-			break;
-		default:
-			break;
-		}
-	}
-	/* Set the ACL_GROUP_OBJ if there's no ACL_MASK */
-	if (gap && nomask)
-		gap->ae_perm = (mode >> 3) & 0x7;
-}
-
-/*
- * When inheriting an Access ACL from a directory Default ACL,
- * the ACL bits are set to the intersection of the ACL default
- * permission bits and the file permission bits in mode. If there
- * are no permission bits on the file then we must not give them
- * the ACL. This is what what makes umask() work with ACLs.
- */
-STATIC void
-xfs_acl_filter_mode(
-	mode_t		mode,
-	xfs_acl_t	*acl)
-{
-	int		i, nomask = 1;
-	xfs_acl_entry_t	*ap;
-	xfs_acl_entry_t	*gap = NULL;
-
-	/*
-	 * Set ACL entries. POSIX1003.1eD16 requires that the MASK
-	 * be merged with GROUP entry, if there is a MASK.
-	 */
-	for (ap = acl->acl_entry, i = 0; i < acl->acl_cnt; ap++, i++) {
-		switch (ap->ae_tag) {
-		case ACL_USER_OBJ:
-			ap->ae_perm &= (mode >> 6) & 0x7;
-			break;
-		case ACL_GROUP_OBJ:
-			gap = ap;
-			break;
-		case ACL_MASK:
-			nomask = 0;
-			ap->ae_perm &= (mode >> 3) & 0x7;
-			break;
-		case ACL_OTHER:
-			ap->ae_perm &= mode & 0x7;
-			break;
-		default:
-			break;
-		}
-	}
-	/* Set the ACL_GROUP_OBJ if there's no ACL_MASK */
-	if (gap && nomask)
-		gap->ae_perm &= (mode >> 3) & 0x7;
-}
diff --git a/fs/xfs/xfs_acl.h b/fs/xfs/xfs_acl.h
index 642f1db4def4..63dc1f2efad5 100644
--- a/fs/xfs/xfs_acl.h
+++ b/fs/xfs/xfs_acl.h
@@ -18,81 +18,48 @@
 #ifndef __XFS_ACL_H__
 #define __XFS_ACL_H__
 
-/*
- * Access Control Lists
- */
-typedef __uint16_t	xfs_acl_perm_t;
-typedef __int32_t	xfs_acl_tag_t;
-typedef __int32_t	xfs_acl_id_t;
+struct inode;
+struct posix_acl;
+struct xfs_inode;
 
 #define XFS_ACL_MAX_ENTRIES 25
 #define XFS_ACL_NOT_PRESENT (-1)
 
-typedef struct xfs_acl_entry {
-	xfs_acl_tag_t	ae_tag;
-	xfs_acl_id_t	ae_id;
-	xfs_acl_perm_t	ae_perm;
-} xfs_acl_entry_t;
-
-typedef struct xfs_acl {
-	__int32_t	acl_cnt;
-	xfs_acl_entry_t	acl_entry[XFS_ACL_MAX_ENTRIES];
-} xfs_acl_t;
+/* On-disk XFS access control list structure */
+struct xfs_acl {
+	__be32		acl_cnt;
+	struct xfs_acl_entry {
+		__be32	ae_tag;
+		__be32	ae_id;
+		__be16	ae_perm;
+	} acl_entry[XFS_ACL_MAX_ENTRIES];
+};
 
 /* On-disk XFS extended attribute names */
-#define SGI_ACL_FILE	"SGI_ACL_FILE"
-#define SGI_ACL_DEFAULT	"SGI_ACL_DEFAULT"
+#define SGI_ACL_FILE		"SGI_ACL_FILE"
+#define SGI_ACL_DEFAULT		"SGI_ACL_DEFAULT"
 #define SGI_ACL_FILE_SIZE	(sizeof(SGI_ACL_FILE)-1)
 #define SGI_ACL_DEFAULT_SIZE	(sizeof(SGI_ACL_DEFAULT)-1)
 
-#define _ACL_TYPE_ACCESS	1
-#define _ACL_TYPE_DEFAULT	2
-
 #ifdef CONFIG_XFS_POSIX_ACL
+extern int xfs_check_acl(struct inode *inode, int mask);
+extern struct posix_acl *xfs_get_acl(struct inode *inode, int type);
+extern int xfs_inherit_acl(struct inode *inode, struct posix_acl *default_acl);
+extern int xfs_acl_chmod(struct inode *inode);
+extern void xfs_inode_init_acls(struct xfs_inode *ip);
+extern void xfs_inode_clear_acls(struct xfs_inode *ip);
+extern int posix_acl_access_exists(struct inode *inode);
+extern int posix_acl_default_exists(struct inode *inode);
 
-struct vattr;
-struct xfs_inode;
-
-extern struct kmem_zone *xfs_acl_zone;
-#define xfs_acl_zone_init(zone, name)	\
-		(zone) = kmem_zone_init(sizeof(xfs_acl_t), (name))
-#define xfs_acl_zone_destroy(zone)	kmem_zone_destroy(zone)
-
-extern int xfs_acl_inherit(struct inode *, mode_t mode, xfs_acl_t *);
-extern int xfs_acl_iaccess(struct xfs_inode *, mode_t, cred_t *);
-extern int xfs_acl_vtoacl(struct inode *, xfs_acl_t *, xfs_acl_t *);
-extern int xfs_acl_vhasacl_access(struct inode *);
-extern int xfs_acl_vhasacl_default(struct inode *);
-extern int xfs_acl_vset(struct inode *, void *, size_t, int);
-extern int xfs_acl_vget(struct inode *, void *, size_t, int);
-extern int xfs_acl_vremove(struct inode *, int);
-
-#define _ACL_PERM_INVALID(perm)	((perm) & ~(ACL_READ|ACL_WRITE|ACL_EXECUTE))
-
-#define _ACL_INHERIT(c,m,d)	(xfs_acl_inherit(c,m,d))
-#define _ACL_GET_ACCESS(pv,pa)	(xfs_acl_vtoacl(pv,pa,NULL) == 0)
-#define _ACL_GET_DEFAULT(pv,pd)	(xfs_acl_vtoacl(pv,NULL,pd) == 0)
-#define _ACL_ACCESS_EXISTS	xfs_acl_vhasacl_access
-#define _ACL_DEFAULT_EXISTS	xfs_acl_vhasacl_default
-
-#define _ACL_ALLOC(a)		((a) = kmem_zone_alloc(xfs_acl_zone, KM_SLEEP))
-#define _ACL_FREE(a)		((a)? kmem_zone_free(xfs_acl_zone, (a)):(void)0)
-
+extern struct xattr_handler xfs_xattr_system_handler;
 #else
-#define xfs_acl_zone_init(zone,name)
-#define xfs_acl_zone_destroy(zone)
-#define xfs_acl_vset(v,p,sz,t)	(-EOPNOTSUPP)
-#define xfs_acl_vget(v,p,sz,t)	(-EOPNOTSUPP)
-#define xfs_acl_vremove(v,t)	(-EOPNOTSUPP)
-#define xfs_acl_vhasacl_access(v)	(0)
-#define xfs_acl_vhasacl_default(v)	(0)
-#define _ACL_ALLOC(a)		(1)	/* successfully allocate nothing */
-#define _ACL_FREE(a)		((void)0)
-#define _ACL_INHERIT(c,m,d)	(0)
-#define _ACL_GET_ACCESS(pv,pa)	(0)
-#define _ACL_GET_DEFAULT(pv,pd)	(0)
-#define _ACL_ACCESS_EXISTS	(NULL)
-#define _ACL_DEFAULT_EXISTS	(NULL)
-#endif
-
+# define xfs_check_acl					NULL
+# define xfs_get_acl(inode, type)			NULL
+# define xfs_inherit_acl(inode, default_acl)		0
+# define xfs_acl_chmod(inode)				0
+# define xfs_inode_init_acls(ip)
+# define xfs_inode_clear_acls(ip)
+# define posix_acl_access_exists(inode)			0
+# define posix_acl_default_exists(inode)		0
+#endif /* CONFIG_XFS_POSIX_ACL */
 #endif	/* __XFS_ACL_H__ */
diff --git a/fs/xfs/xfs_ag.h b/fs/xfs/xfs_ag.h
index c8641f713caa..f24b50b68d03 100644
--- a/fs/xfs/xfs_ag.h
+++ b/fs/xfs/xfs_ag.h
@@ -212,6 +212,8 @@ typedef struct xfs_perag
 /*
  * tags for inode radix tree
  */
+#define XFS_ICI_NO_TAG		(-1)	/* special flag for an untagged lookup
+					   in xfs_inode_ag_iterator */
 #define XFS_ICI_RECLAIM_TAG	0	/* inode is to be reclaimed */
 
 #define	XFS_AG_MAXLEVELS(mp)		((mp)->m_ag_maxlevels)
diff --git a/fs/xfs/xfs_arch.h b/fs/xfs/xfs_arch.h
index 53d5e70d1360..0902249354a0 100644
--- a/fs/xfs/xfs_arch.h
+++ b/fs/xfs/xfs_arch.h
@@ -73,28 +73,6 @@ static inline void be64_add_cpu(__be64 *a, __s64 b)
 
 #endif	/* __KERNEL__ */
 
-/* do we need conversion? */
-#define ARCH_NOCONVERT 1
-#ifdef XFS_NATIVE_HOST
-# define ARCH_CONVERT	ARCH_NOCONVERT
-#else
-# define ARCH_CONVERT	0
-#endif
-
-/* generic swapping macros */
-
-#ifndef HAVE_SWABMACROS
-#define INT_SWAP16(type,var) ((typeof(type))(__swab16((__u16)(var))))
-#define INT_SWAP32(type,var) ((typeof(type))(__swab32((__u32)(var))))
-#define INT_SWAP64(type,var) ((typeof(type))(__swab64((__u64)(var))))
-#endif
-
-#define INT_SWAP(type, var) \
-    ((sizeof(type) == 8) ? INT_SWAP64(type,var) : \
-    ((sizeof(type) == 4) ? INT_SWAP32(type,var) : \
-    ((sizeof(type) == 2) ? INT_SWAP16(type,var) : \
-    (var))))
-
 /*
  * get and set integers from potentially unaligned locations
  */
@@ -107,16 +85,6 @@ static inline void be64_add_cpu(__be64 *a, __s64 b)
 	((__u8*)(pointer))[1] = (((value)     ) & 0xff); \
     }
 
-/* does not return a value */
-#define INT_SET(reference,arch,valueref) \
-    (__builtin_constant_p(valueref) ? \
-	(void)( (reference) = ( ((arch) != ARCH_NOCONVERT) ? (INT_SWAP((reference),(valueref))) : (valueref)) ) : \
-	(void)( \
-	    ((reference) = (valueref)), \
-	    ( ((arch) != ARCH_NOCONVERT) ? (reference) = INT_SWAP((reference),(reference)) : 0 ) \
-	) \
-    )
-
 /*
  * In directories inode numbers are stored as unaligned arrays of unsigned
  * 8bit integers on disk.
diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/xfs_attr.c
index 5fde1654b430..db15feb906ff 100644
--- a/fs/xfs/xfs_attr.c
+++ b/fs/xfs/xfs_attr.c
@@ -45,7 +45,6 @@
 #include "xfs_error.h"
 #include "xfs_quota.h"
 #include "xfs_trans_space.h"
-#include "xfs_acl.h"
 #include "xfs_rw.h"
 #include "xfs_vnodeops.h"
 
@@ -249,8 +248,9 @@ xfs_attr_set_int(xfs_inode_t *dp, struct xfs_name *name,
 	/*
 	 * Attach the dquots to the inode.
 	 */
-	if ((error = XFS_QM_DQATTACH(mp, dp, 0)))
-		return (error);
+	error = xfs_qm_dqattach(dp, 0);
+	if (error)
+		return error;
 
 	/*
 	 * If the inode doesn't have an attribute fork, add one.
@@ -311,7 +311,7 @@ xfs_attr_set_int(xfs_inode_t *dp, struct xfs_name *name,
 	}
 	xfs_ilock(dp, XFS_ILOCK_EXCL);
 
-	error = XFS_TRANS_RESERVE_QUOTA_NBLKS(mp, args.trans, dp, args.total, 0,
+	error = xfs_trans_reserve_quota_nblks(args.trans, dp, args.total, 0,
 				rsvd ? XFS_QMOPT_RES_REGBLKS | XFS_QMOPT_FORCE_RES :
 				       XFS_QMOPT_RES_REGBLKS);
 	if (error) {
@@ -501,8 +501,9 @@ xfs_attr_remove_int(xfs_inode_t *dp, struct xfs_name *name, int flags)
 	/*
 	 * Attach the dquots to the inode.
 	 */
-	if ((error = XFS_QM_DQATTACH(mp, dp, 0)))
-		return (error);
+	error = xfs_qm_dqattach(dp, 0);
+	if (error)
+		return error;
 
 	/*
 	 * Start our first transaction of the day.
diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c
index ca7c6005a487..7928b9983c1d 100644
--- a/fs/xfs/xfs_bmap.c
+++ b/fs/xfs/xfs_bmap.c
@@ -2691,7 +2691,7 @@ xfs_bmap_rtalloc(
 		 * Adjust the disk quota also. This was reserved
 		 * earlier.
 		 */
-		XFS_TRANS_MOD_DQUOT_BYINO(mp, ap->tp, ap->ip,
+		xfs_trans_mod_dquot_byino(ap->tp, ap->ip,
 			ap->wasdel ? XFS_TRANS_DQ_DELRTBCOUNT :
 					XFS_TRANS_DQ_RTBCOUNT, (long) ralen);
 	} else {
@@ -2995,7 +2995,7 @@ xfs_bmap_btalloc(
 		 * Adjust the disk quota also. This was reserved
 		 * earlier.
 		 */
-		XFS_TRANS_MOD_DQUOT_BYINO(mp, ap->tp, ap->ip,
+		xfs_trans_mod_dquot_byino(ap->tp, ap->ip,
 			ap->wasdel ? XFS_TRANS_DQ_DELBCOUNT :
 					XFS_TRANS_DQ_BCOUNT,
 			(long) args.len);
@@ -3066,7 +3066,7 @@ xfs_bmap_btree_to_extents(
 		return error;
 	xfs_bmap_add_free(cbno, 1, cur->bc_private.b.flist, mp);
 	ip->i_d.di_nblocks--;
-	XFS_TRANS_MOD_DQUOT_BYINO(mp, tp, ip, XFS_TRANS_DQ_BCOUNT, -1L);
+	xfs_trans_mod_dquot_byino(tp, ip, XFS_TRANS_DQ_BCOUNT, -1L);
 	xfs_trans_binval(tp, cbp);
 	if (cur->bc_bufs[0] == cbp)
 		cur->bc_bufs[0] = NULL;
@@ -3386,7 +3386,7 @@ xfs_bmap_del_extent(
 	 * Adjust quota data.
 	 */
 	if (qfield)
-		XFS_TRANS_MOD_DQUOT_BYINO(mp, tp, ip, qfield, (long)-nblks);
+		xfs_trans_mod_dquot_byino(tp, ip, qfield, (long)-nblks);
 
 	/*
 	 * Account for change in delayed indirect blocks.
@@ -3523,7 +3523,7 @@ xfs_bmap_extents_to_btree(
 	*firstblock = cur->bc_private.b.firstblock = args.fsbno;
 	cur->bc_private.b.allocated++;
 	ip->i_d.di_nblocks++;
-	XFS_TRANS_MOD_DQUOT_BYINO(mp, tp, ip, XFS_TRANS_DQ_BCOUNT, 1L);
+	xfs_trans_mod_dquot_byino(tp, ip, XFS_TRANS_DQ_BCOUNT, 1L);
 	abp = xfs_btree_get_bufl(mp, tp, args.fsbno, 0);
 	/*
 	 * Fill in the child block.
@@ -3690,7 +3690,7 @@ xfs_bmap_local_to_extents(
 		XFS_BMAP_TRACE_POST_UPDATE("new", ip, 0, whichfork);
 		XFS_IFORK_NEXT_SET(ip, whichfork, 1);
 		ip->i_d.di_nblocks = 1;
-		XFS_TRANS_MOD_DQUOT_BYINO(args.mp, tp, ip,
+		xfs_trans_mod_dquot_byino(tp, ip,
 			XFS_TRANS_DQ_BCOUNT, 1L);
 		flags |= xfs_ilog_fext(whichfork);
 	} else {
@@ -4048,7 +4048,7 @@ xfs_bmap_add_attrfork(
 			XFS_TRANS_PERM_LOG_RES, XFS_ADDAFORK_LOG_COUNT)))
 		goto error0;
 	xfs_ilock(ip, XFS_ILOCK_EXCL);
-	error = XFS_TRANS_RESERVE_QUOTA_NBLKS(mp, tp, ip, blks, 0, rsvd ?
+	error = xfs_trans_reserve_quota_nblks(tp, ip, blks, 0, rsvd ?
 			XFS_QMOPT_RES_REGBLKS | XFS_QMOPT_FORCE_RES :
 			XFS_QMOPT_RES_REGBLKS);
 	if (error) {
@@ -4983,10 +4983,11 @@ xfs_bmapi(
 				 * adjusted later.  We return if we haven't
 				 * allocated blocks already inside this loop.
 				 */
-				if ((error = XFS_TRANS_RESERVE_QUOTA_NBLKS(
-						mp, NULL, ip, (long)alen, 0,
+				error = xfs_trans_reserve_quota_nblks(
+						NULL, ip, (long)alen, 0,
 						rt ? XFS_QMOPT_RES_RTBLKS :
-						     XFS_QMOPT_RES_REGBLKS))) {
+						     XFS_QMOPT_RES_REGBLKS);
+				if (error) {
 					if (n == 0) {
 						*nmap = 0;
 						ASSERT(cur == NULL);
@@ -5035,8 +5036,8 @@ xfs_bmapi(
 					if (XFS_IS_QUOTA_ON(mp))
 						/* unreserve the blocks now */
 						(void)
-						XFS_TRANS_UNRESERVE_QUOTA_NBLKS(
-							mp, NULL, ip,
+						xfs_trans_unreserve_quota_nblks(
+							NULL, ip,
 							(long)alen, 0, rt ?
 							XFS_QMOPT_RES_RTBLKS :
 							XFS_QMOPT_RES_REGBLKS);
@@ -5691,14 +5692,14 @@ xfs_bunmapi(
 				do_div(rtexts, mp->m_sb.sb_rextsize);
 				xfs_mod_incore_sb(mp, XFS_SBS_FREXTENTS,
 						(int64_t)rtexts, rsvd);
-				(void)XFS_TRANS_RESERVE_QUOTA_NBLKS(mp,
-					NULL, ip, -((long)del.br_blockcount), 0,
+				(void)xfs_trans_reserve_quota_nblks(NULL,
+					ip, -((long)del.br_blockcount), 0,
 					XFS_QMOPT_RES_RTBLKS);
 			} else {
 				xfs_mod_incore_sb(mp, XFS_SBS_FDBLOCKS,
 						(int64_t)del.br_blockcount, rsvd);
-				(void)XFS_TRANS_RESERVE_QUOTA_NBLKS(mp,
-					NULL, ip, -((long)del.br_blockcount), 0,
+				(void)xfs_trans_reserve_quota_nblks(NULL,
+					ip, -((long)del.br_blockcount), 0,
 					XFS_QMOPT_RES_REGBLKS);
 			}
 			ip->i_delayed_blks -= del.br_blockcount;
@@ -6085,6 +6086,7 @@ xfs_getbmap(
 			break;
 	}
 
+	kmem_free(out);
 	return error;
 }
 
diff --git a/fs/xfs/xfs_bmap_btree.c b/fs/xfs/xfs_bmap_btree.c
index 0760d352586f..5c1ade06578e 100644
--- a/fs/xfs/xfs_bmap_btree.c
+++ b/fs/xfs/xfs_bmap_btree.c
@@ -590,7 +590,7 @@ xfs_bmbt_alloc_block(
 	cur->bc_private.b.allocated++;
 	cur->bc_private.b.ip->i_d.di_nblocks++;
 	xfs_trans_log_inode(args.tp, cur->bc_private.b.ip, XFS_ILOG_CORE);
-	XFS_TRANS_MOD_DQUOT_BYINO(args.mp, args.tp, cur->bc_private.b.ip,
+	xfs_trans_mod_dquot_byino(args.tp, cur->bc_private.b.ip,
 			XFS_TRANS_DQ_BCOUNT, 1L);
 
 	new->l = cpu_to_be64(args.fsbno);
@@ -618,7 +618,7 @@ xfs_bmbt_free_block(
 	ip->i_d.di_nblocks--;
 
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
-	XFS_TRANS_MOD_DQUOT_BYINO(mp, tp, ip, XFS_TRANS_DQ_BCOUNT, -1L);
+	xfs_trans_mod_dquot_byino(tp, ip, XFS_TRANS_DQ_BCOUNT, -1L);
 	xfs_trans_binval(tp, bp);
 	return 0;
 }
diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c
index 6c87c8f304ef..edf8bdf4141f 100644
--- a/fs/xfs/xfs_filestream.c
+++ b/fs/xfs/xfs_filestream.c
@@ -542,10 +542,8 @@ xfs_filestream_associate(
 	 * waiting for the lock because someone else is waiting on the lock we
 	 * hold and we cannot drop that as we are in a transaction here.
 	 *
-	 * Lucky for us, this inversion is rarely a problem because it's a
-	 * directory inode that we are trying to lock here and that means the
-	 * only place that matters is xfs_sync_inodes() and SYNC_DELWRI is
-	 * used. i.e. freeze, remount-ro, quotasync or unmount.
+	 * Lucky for us, this inversion is not a problem because it's a
+	 * directory inode that we are trying to lock here.
 	 *
 	 * So, if we can't get the iolock without sleeping then just give up
 	 */
diff --git a/fs/xfs/xfs_fs.h b/fs/xfs/xfs_fs.h
index f7c06fac8229..c4ea51b55dce 100644
--- a/fs/xfs/xfs_fs.h
+++ b/fs/xfs/xfs_fs.h
@@ -239,10 +239,13 @@ typedef struct xfs_fsop_resblks {
  * Minimum and maximum sizes need for growth checks
  */
 #define XFS_MIN_AG_BLOCKS	64
-#define XFS_MIN_LOG_BLOCKS	512
-#define XFS_MAX_LOG_BLOCKS	(64 * 1024)
-#define XFS_MIN_LOG_BYTES	(256 * 1024)
-#define XFS_MAX_LOG_BYTES	(128 * 1024 * 1024)
+#define XFS_MIN_LOG_BLOCKS	512ULL
+#define XFS_MAX_LOG_BLOCKS	(1024 * 1024ULL)
+#define XFS_MIN_LOG_BYTES	(10 * 1024 * 1024ULL)
+
+/* keep the maximum size under 2^31 by a small amount */
+#define XFS_MAX_LOG_BYTES \
+	((2 * 1024 * 1024 * 1024ULL) - XFS_MIN_LOG_BYTES)
 
 /*
  * Structures for XFS_IOC_FSGROWFSDATA, XFS_IOC_FSGROWFSLOG & XFS_IOC_FSGROWFSRT
diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c
index 89b81eedce6a..76c540f719e4 100644
--- a/fs/xfs/xfs_iget.c
+++ b/fs/xfs/xfs_iget.c
@@ -18,6 +18,7 @@
 #include "xfs.h"
 #include "xfs_fs.h"
 #include "xfs_types.h"
+#include "xfs_acl.h"
 #include "xfs_bit.h"
 #include "xfs_log.h"
 #include "xfs_inum.h"
@@ -82,6 +83,7 @@ xfs_inode_alloc(
 	memset(&ip->i_d, 0, sizeof(xfs_icdinode_t));
 	ip->i_size = 0;
 	ip->i_new_size = 0;
+	xfs_inode_init_acls(ip);
 
 	/*
 	 * Initialize inode's trace buffers.
@@ -500,10 +502,7 @@ xfs_ireclaim(
 	 * ilock one but will still hold the iolock.
 	 */
 	xfs_ilock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
-	/*
-	 * Release dquots (and their references) if any.
-	 */
-	XFS_QM_DQDETACH(ip->i_mount, ip);
+	xfs_qm_dqdetach(ip);
 	xfs_iunlock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
 
 	switch (ip->i_d.di_mode & S_IFMT) {
@@ -561,6 +560,7 @@ xfs_ireclaim(
 	ASSERT(atomic_read(&ip->i_pincount) == 0);
 	ASSERT(!spin_is_locked(&ip->i_flags_lock));
 	ASSERT(completion_done(&ip->i_flush));
+	xfs_inode_clear_acls(ip);
 	kmem_zone_free(xfs_inode_zone, ip);
 }
 
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index 123b20c8cbf2..1f22d65fed0a 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -49,7 +49,6 @@
 #include "xfs_utils.h"
 #include "xfs_dir2_trace.h"
 #include "xfs_quota.h"
-#include "xfs_acl.h"
 #include "xfs_filestream.h"
 #include "xfs_vnodeops.h"
 
diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
index f879c1bc4b96..77016702938b 100644
--- a/fs/xfs/xfs_inode.h
+++ b/fs/xfs/xfs_inode.h
@@ -18,6 +18,7 @@
 #ifndef	__XFS_INODE_H__
 #define	__XFS_INODE_H__
 
+struct posix_acl;
 struct xfs_dinode;
 struct xfs_inode;
 
@@ -272,6 +273,11 @@ typedef struct xfs_inode {
 	/* VFS inode */
 	struct inode		i_vnode;	/* embedded VFS inode */
 
+#ifdef CONFIG_XFS_POSIX_ACL
+	struct posix_acl	*i_acl;
+	struct posix_acl	*i_default_acl;
+#endif
+
 	/* Trace buffers per inode. */
 #ifdef XFS_INODE_TRACE
 	struct ktrace		*i_trace;	/* general inode trace */
diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
index 5aaa2d7ec155..67ae5555a30a 100644
--- a/fs/xfs/xfs_iomap.c
+++ b/fs/xfs/xfs_iomap.c
@@ -42,7 +42,6 @@
 #include "xfs_error.h"
 #include "xfs_itable.h"
 #include "xfs_rw.h"
-#include "xfs_acl.h"
 #include "xfs_attr.h"
 #include "xfs_buf_item.h"
 #include "xfs_trans_space.h"
@@ -385,7 +384,7 @@ xfs_iomap_write_direct(
 	 * Make sure that the dquots are there. This doesn't hold
 	 * the ilock across a disk read.
 	 */
-	error = XFS_QM_DQATTACH(ip->i_mount, ip, XFS_QMOPT_ILOCKED);
+	error = xfs_qm_dqattach_locked(ip, 0);
 	if (error)
 		return XFS_ERROR(error);
 
@@ -444,8 +443,7 @@ xfs_iomap_write_direct(
 	if (error)
 		goto error_out;
 
-	error = XFS_TRANS_RESERVE_QUOTA_NBLKS(mp, tp, ip,
-					      qblocks, 0, quota_flag);
+	error = xfs_trans_reserve_quota_nblks(tp, ip, qblocks, 0, quota_flag);
 	if (error)
 		goto error1;
 
@@ -495,7 +493,7 @@ xfs_iomap_write_direct(
 
 error0:	/* Cancel bmap, unlock inode, unreserve quota blocks, cancel trans */
 	xfs_bmap_cancel(&free_list);
-	XFS_TRANS_UNRESERVE_QUOTA_NBLKS(mp, tp, ip, qblocks, 0, quota_flag);
+	xfs_trans_unreserve_quota_nblks(tp, ip, qblocks, 0, quota_flag);
 
 error1:	/* Just cancel transaction */
 	xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT);
@@ -582,7 +580,7 @@ xfs_iomap_write_delay(
 	 * Make sure that the dquots are there. This doesn't hold
 	 * the ilock across a disk read.
 	 */
-	error = XFS_QM_DQATTACH(mp, ip, XFS_QMOPT_ILOCKED);
+	error = xfs_qm_dqattach_locked(ip, 0);
 	if (error)
 		return XFS_ERROR(error);
 
@@ -684,7 +682,8 @@ xfs_iomap_write_allocate(
 	/*
 	 * Make sure that the dquots are there.
 	 */
-	if ((error = XFS_QM_DQATTACH(mp, ip, 0)))
+	error = xfs_qm_dqattach(ip, 0);
+	if (error)
 		return XFS_ERROR(error);
 
 	offset_fsb = XFS_B_TO_FSBT(mp, offset);
diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index 7ba450116d4f..47da2fb45377 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -1975,16 +1975,30 @@ xlog_recover_do_reg_buffer(
 		error = 0;
 		if (buf_f->blf_flags &
 		   (XFS_BLI_UDQUOT_BUF|XFS_BLI_PDQUOT_BUF|XFS_BLI_GDQUOT_BUF)) {
+			if (item->ri_buf[i].i_addr == NULL) {
+				cmn_err(CE_ALERT,
+					"XFS: NULL dquot in %s.", __func__);
+				goto next;
+			}
+			if (item->ri_buf[i].i_len < sizeof(xfs_dqblk_t)) {
+				cmn_err(CE_ALERT,
+					"XFS: dquot too small (%d) in %s.",
+					item->ri_buf[i].i_len, __func__);
+				goto next;
+			}
 			error = xfs_qm_dqcheck((xfs_disk_dquot_t *)
 					       item->ri_buf[i].i_addr,
 					       -1, 0, XFS_QMOPT_DOWARN,
 					       "dquot_buf_recover");
+			if (error)
+				goto next;
 		}
-		if (!error)
-			memcpy(xfs_buf_offset(bp,
-				(uint)bit << XFS_BLI_SHIFT),	/* dest */
-				item->ri_buf[i].i_addr,		/* source */
-				nbits<<XFS_BLI_SHIFT);		/* length */
+
+		memcpy(xfs_buf_offset(bp,
+			(uint)bit << XFS_BLI_SHIFT),	/* dest */
+			item->ri_buf[i].i_addr,		/* source */
+			nbits<<XFS_BLI_SHIFT);		/* length */
+ next:
 		i++;
 		bit += nbits;
 	}
@@ -2615,7 +2629,19 @@ xlog_recover_do_dquot_trans(
 		return (0);
 
 	recddq = (xfs_disk_dquot_t *)item->ri_buf[1].i_addr;
-	ASSERT(recddq);
+
+	if (item->ri_buf[1].i_addr == NULL) {
+		cmn_err(CE_ALERT,
+			"XFS: NULL dquot in %s.", __func__);
+		return XFS_ERROR(EIO);
+	}
+	if (item->ri_buf[1].i_len < sizeof(xfs_dqblk_t)) {
+		cmn_err(CE_ALERT,
+			"XFS: dquot too small (%d) in %s.",
+			item->ri_buf[1].i_len, __func__);
+		return XFS_ERROR(EIO);
+	}
+
 	/*
 	 * This type of quotas was turned off, so ignore this record.
 	 */
diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c
index 65a99725d0cc..5c6f092659c1 100644
--- a/fs/xfs/xfs_mount.c
+++ b/fs/xfs/xfs_mount.c
@@ -960,6 +960,53 @@ xfs_check_sizes(xfs_mount_t *mp)
 }
 
 /*
+ * Clear the quotaflags in memory and in the superblock.
+ */
+int
+xfs_mount_reset_sbqflags(
+	struct xfs_mount	*mp)
+{
+	int			error;
+	struct xfs_trans	*tp;
+
+	mp->m_qflags = 0;
+
+	/*
+	 * It is OK to look at sb_qflags here in mount path,
+	 * without m_sb_lock.
+	 */
+	if (mp->m_sb.sb_qflags == 0)
+		return 0;
+	spin_lock(&mp->m_sb_lock);
+	mp->m_sb.sb_qflags = 0;
+	spin_unlock(&mp->m_sb_lock);
+
+	/*
+	 * If the fs is readonly, let the incore superblock run
+	 * with quotas off but don't flush the update out to disk
+	 */
+	if (mp->m_flags & XFS_MOUNT_RDONLY)
+		return 0;
+
+#ifdef QUOTADEBUG
+	xfs_fs_cmn_err(CE_NOTE, mp, "Writing superblock quota changes");
+#endif
+
+	tp = xfs_trans_alloc(mp, XFS_TRANS_QM_SBCHANGE);
+	error = xfs_trans_reserve(tp, 0, mp->m_sb.sb_sectsize + 128, 0, 0,
+				      XFS_DEFAULT_LOG_COUNT);
+	if (error) {
+		xfs_trans_cancel(tp, 0);
+		xfs_fs_cmn_err(CE_ALERT, mp,
+			"xfs_mount_reset_sbqflags: Superblock update failed!");
+		return error;
+	}
+
+	xfs_mod_sb(tp, XFS_SB_QFLAGS);
+	return xfs_trans_commit(tp, 0);
+}
+
+/*
  * This function does the following on an initial mount of a file system:
  *	- reads the superblock from disk and init the mount struct
  *	- if we're a 32-bit kernel, do a size check on the superblock
@@ -976,7 +1023,8 @@ xfs_mountfs(
 	xfs_sb_t	*sbp = &(mp->m_sb);
 	xfs_inode_t	*rip;
 	__uint64_t	resblks;
-	uint		quotamount, quotaflags;
+	uint		quotamount = 0;
+	uint		quotaflags = 0;
 	int		error = 0;
 
 	xfs_mount_common(mp, sbp);
@@ -1210,9 +1258,28 @@ xfs_mountfs(
 	/*
 	 * Initialise the XFS quota management subsystem for this mount
 	 */
-	error = XFS_QM_INIT(mp, &quotamount, &quotaflags);
-	if (error)
-		goto out_rtunmount;
+	if (XFS_IS_QUOTA_RUNNING(mp)) {
+		error = xfs_qm_newmount(mp, &quotamount, &quotaflags);
+		if (error)
+			goto out_rtunmount;
+	} else {
+		ASSERT(!XFS_IS_QUOTA_ON(mp));
+
+		/*
+		 * If a file system had quotas running earlier, but decided to
+		 * mount without -o uquota/pquota/gquota options, revoke the
+		 * quotachecked license.
+		 */
+		if (mp->m_sb.sb_qflags & XFS_ALL_QUOTA_ACCT) {
+			cmn_err(CE_NOTE,
+				"XFS: resetting qflags for filesystem %s",
+				mp->m_fsname);
+
+			error = xfs_mount_reset_sbqflags(mp);
+			if (error)
+				return error;
+		}
+	}
 
 	/*
 	 * Finish recovering the file system.  This part needed to be
@@ -1228,9 +1295,19 @@ xfs_mountfs(
 	/*
 	 * Complete the quota initialisation, post-log-replay component.
 	 */
-	error = XFS_QM_MOUNT(mp, quotamount, quotaflags);
-	if (error)
-		goto out_rtunmount;
+	if (quotamount) {
+		ASSERT(mp->m_qflags == 0);
+		mp->m_qflags = quotaflags;
+
+		xfs_qm_mount_quotas(mp);
+	}
+
+#if defined(DEBUG) && defined(XFS_LOUD_RECOVERY)
+	if (XFS_IS_QUOTA_ON(mp))
+		xfs_fs_cmn_err(CE_NOTE, mp, "Disk quotas turned on");
+	else
+		xfs_fs_cmn_err(CE_NOTE, mp, "Disk quotas not turned on");
+#endif
 
 	/*
 	 * Now we are mounted, reserve a small amount of unused space for
@@ -1279,12 +1356,7 @@ xfs_unmountfs(
 	__uint64_t		resblks;
 	int			error;
 
-	/*
-	 * Release dquot that rootinode, rbmino and rsumino might be holding,
-	 * and release the quota inodes.
-	 */
-	XFS_QM_UNMOUNT(mp);
-
+	xfs_qm_unmount_quotas(mp);
 	xfs_rtunmount_inodes(mp);
 	IRELE(mp->m_rootip);
 
@@ -1299,12 +1371,9 @@ xfs_unmountfs(
 	 * need to force the log first.
 	 */
 	xfs_log_force(mp, (xfs_lsn_t)0, XFS_LOG_FORCE | XFS_LOG_SYNC);
-	xfs_reclaim_inodes(mp, 0, XFS_IFLUSH_ASYNC);
-
-	XFS_QM_DQPURGEALL(mp, XFS_QMOPT_QUOTALL | XFS_QMOPT_UMOUNTING);
+	xfs_reclaim_inodes(mp, XFS_IFLUSH_ASYNC);
 
-	if (mp->m_quotainfo)
-		XFS_QM_DONE(mp);
+	xfs_qm_unmount(mp);
 
 	/*
 	 * Flush out the log synchronously so that we know for sure
diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h
index d6a64392f983..a5122382afde 100644
--- a/fs/xfs/xfs_mount.h
+++ b/fs/xfs/xfs_mount.h
@@ -64,6 +64,8 @@ struct xfs_swapext;
 struct xfs_mru_cache;
 struct xfs_nameops;
 struct xfs_ail;
+struct xfs_quotainfo;
+
 
 /*
  * Prototypes and functions for the Data Migration subsystem.
@@ -107,86 +109,6 @@ typedef struct xfs_dmops {
 	(*(mp)->m_dm_ops->xfs_send_unmount)(mp,ip,right,mode,rval,fl)
 
 
-/*
- * Prototypes and functions for the Quota Management subsystem.
- */
-
-struct xfs_dquot;
-struct xfs_dqtrxops;
-struct xfs_quotainfo;
-
-typedef int	(*xfs_qminit_t)(struct xfs_mount *, uint *, uint *);
-typedef int	(*xfs_qmmount_t)(struct xfs_mount *, uint, uint);
-typedef void	(*xfs_qmunmount_t)(struct xfs_mount *);
-typedef void	(*xfs_qmdone_t)(struct xfs_mount *);
-typedef void	(*xfs_dqrele_t)(struct xfs_dquot *);
-typedef int	(*xfs_dqattach_t)(struct xfs_inode *, uint);
-typedef void	(*xfs_dqdetach_t)(struct xfs_inode *);
-typedef int	(*xfs_dqpurgeall_t)(struct xfs_mount *, uint);
-typedef int	(*xfs_dqvopalloc_t)(struct xfs_mount *,
-			struct xfs_inode *, uid_t, gid_t, prid_t, uint,
-			struct xfs_dquot **, struct xfs_dquot **);
-typedef void	(*xfs_dqvopcreate_t)(struct xfs_trans *, struct xfs_inode *,
-			struct xfs_dquot *, struct xfs_dquot *);
-typedef int	(*xfs_dqvoprename_t)(struct xfs_inode **);
-typedef struct xfs_dquot * (*xfs_dqvopchown_t)(
-			struct xfs_trans *, struct xfs_inode *,
-			struct xfs_dquot **, struct xfs_dquot *);
-typedef int	(*xfs_dqvopchownresv_t)(struct xfs_trans *, struct xfs_inode *,
-			struct xfs_dquot *, struct xfs_dquot *, uint);
-typedef void	(*xfs_dqstatvfs_t)(struct xfs_inode *, struct kstatfs *);
-typedef int	(*xfs_dqsync_t)(struct xfs_mount *, int flags);
-
-typedef struct xfs_qmops {
-	xfs_qminit_t		xfs_qminit;
-	xfs_qmdone_t		xfs_qmdone;
-	xfs_qmmount_t		xfs_qmmount;
-	xfs_qmunmount_t		xfs_qmunmount;
-	xfs_dqrele_t		xfs_dqrele;
-	xfs_dqattach_t		xfs_dqattach;
-	xfs_dqdetach_t		xfs_dqdetach;
-	xfs_dqpurgeall_t	xfs_dqpurgeall;
-	xfs_dqvopalloc_t	xfs_dqvopalloc;
-	xfs_dqvopcreate_t	xfs_dqvopcreate;
-	xfs_dqvoprename_t	xfs_dqvoprename;
-	xfs_dqvopchown_t	xfs_dqvopchown;
-	xfs_dqvopchownresv_t	xfs_dqvopchownresv;
-	xfs_dqstatvfs_t		xfs_dqstatvfs;
-	xfs_dqsync_t		xfs_dqsync;
-	struct xfs_dqtrxops	*xfs_dqtrxops;
-} xfs_qmops_t;
-
-#define XFS_QM_INIT(mp, mnt, fl) \
-	(*(mp)->m_qm_ops->xfs_qminit)(mp, mnt, fl)
-#define XFS_QM_MOUNT(mp, mnt, fl) \
-	(*(mp)->m_qm_ops->xfs_qmmount)(mp, mnt, fl)
-#define XFS_QM_UNMOUNT(mp) \
-	(*(mp)->m_qm_ops->xfs_qmunmount)(mp)
-#define XFS_QM_DONE(mp) \
-	(*(mp)->m_qm_ops->xfs_qmdone)(mp)
-#define XFS_QM_DQRELE(mp, dq) \
-	(*(mp)->m_qm_ops->xfs_dqrele)(dq)
-#define XFS_QM_DQATTACH(mp, ip, fl) \
-	(*(mp)->m_qm_ops->xfs_dqattach)(ip, fl)
-#define XFS_QM_DQDETACH(mp, ip) \
-	(*(mp)->m_qm_ops->xfs_dqdetach)(ip)
-#define XFS_QM_DQPURGEALL(mp, fl) \
-	(*(mp)->m_qm_ops->xfs_dqpurgeall)(mp, fl)
-#define XFS_QM_DQVOPALLOC(mp, ip, uid, gid, prid, fl, dq1, dq2) \
-	(*(mp)->m_qm_ops->xfs_dqvopalloc)(mp, ip, uid, gid, prid, fl, dq1, dq2)
-#define XFS_QM_DQVOPCREATE(mp, tp, ip, dq1, dq2) \
-	(*(mp)->m_qm_ops->xfs_dqvopcreate)(tp, ip, dq1, dq2)
-#define XFS_QM_DQVOPRENAME(mp, ip) \
-	(*(mp)->m_qm_ops->xfs_dqvoprename)(ip)
-#define XFS_QM_DQVOPCHOWN(mp, tp, ip, dqp, dq) \
-	(*(mp)->m_qm_ops->xfs_dqvopchown)(tp, ip, dqp, dq)
-#define XFS_QM_DQVOPCHOWNRESV(mp, tp, ip, dq1, dq2, fl) \
-	(*(mp)->m_qm_ops->xfs_dqvopchownresv)(tp, ip, dq1, dq2, fl)
-#define XFS_QM_DQSTATVFS(ip, statp) \
-	(*(ip)->i_mount->m_qm_ops->xfs_dqstatvfs)(ip, statp)
-#define XFS_QM_DQSYNC(mp, flags) \
-	(*(mp)->m_qm_ops->xfs_dqsync)(mp, flags)
-
 #ifdef HAVE_PERCPU_SB
 
 /*
@@ -510,8 +432,6 @@ extern int	xfs_sb_validate_fsb_count(struct xfs_sb *, __uint64_t);
 
 extern int	xfs_dmops_get(struct xfs_mount *);
 extern void	xfs_dmops_put(struct xfs_mount *);
-extern int	xfs_qmops_get(struct xfs_mount *);
-extern void	xfs_qmops_put(struct xfs_mount *);
 
 extern struct xfs_dmops xfs_dmcore_xfs;
 
diff --git a/fs/xfs/xfs_qmops.c b/fs/xfs/xfs_qmops.c
deleted file mode 100644
index e101790ea8e7..000000000000
--- a/fs/xfs/xfs_qmops.c
+++ /dev/null
@@ -1,152 +0,0 @@
-/*
- * Copyright (c) 2000-2005 Silicon Graphics, Inc.
- * All Rights Reserved.
- *
- * 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.
- *
- * This program is distributed in the hope that it would 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 the Free Software Foundation,
- * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
- */
-#include "xfs.h"
-#include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
-#include "xfs_inum.h"
-#include "xfs_trans.h"
-#include "xfs_sb.h"
-#include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_dmapi.h"
-#include "xfs_mount.h"
-#include "xfs_quota.h"
-#include "xfs_error.h"
-
-
-STATIC struct xfs_dquot *
-xfs_dqvopchown_default(
-	struct xfs_trans	*tp,
-	struct xfs_inode	*ip,
-	struct xfs_dquot	**dqp,
-	struct xfs_dquot	*dq)
-{
-	return NULL;
-}
-
-/*
- * Clear the quotaflags in memory and in the superblock.
- */
-int
-xfs_mount_reset_sbqflags(xfs_mount_t *mp)
-{
-	int			error;
-	xfs_trans_t		*tp;
-
-	mp->m_qflags = 0;
-	/*
-	 * It is OK to look at sb_qflags here in mount path,
-	 * without m_sb_lock.
-	 */
-	if (mp->m_sb.sb_qflags == 0)
-		return 0;
-	spin_lock(&mp->m_sb_lock);
-	mp->m_sb.sb_qflags = 0;
-	spin_unlock(&mp->m_sb_lock);
-
-	/*
-	 * if the fs is readonly, let the incore superblock run
-	 * with quotas off but don't flush the update out to disk
-	 */
-	if (mp->m_flags & XFS_MOUNT_RDONLY)
-		return 0;
-#ifdef QUOTADEBUG
-	xfs_fs_cmn_err(CE_NOTE, mp, "Writing superblock quota changes");
-#endif
-	tp = xfs_trans_alloc(mp, XFS_TRANS_QM_SBCHANGE);
-	if ((error = xfs_trans_reserve(tp, 0, mp->m_sb.sb_sectsize + 128, 0, 0,
-				      XFS_DEFAULT_LOG_COUNT))) {
-		xfs_trans_cancel(tp, 0);
-		xfs_fs_cmn_err(CE_ALERT, mp,
-			"xfs_mount_reset_sbqflags: Superblock update failed!");
-		return error;
-	}
-	xfs_mod_sb(tp, XFS_SB_QFLAGS);
-	error = xfs_trans_commit(tp, 0);
-	return error;
-}
-
-STATIC int
-xfs_noquota_init(
-	xfs_mount_t	*mp,
-	uint		*needquotamount,
-	uint		*quotaflags)
-{
-	int		error = 0;
-
-	*quotaflags = 0;
-	*needquotamount = B_FALSE;
-
-	ASSERT(!XFS_IS_QUOTA_ON(mp));
-
-	/*
-	 * If a file system had quotas running earlier, but decided to
-	 * mount without -o uquota/pquota/gquota options, revoke the
-	 * quotachecked license.
-	 */
-	if (mp->m_sb.sb_qflags & XFS_ALL_QUOTA_ACCT) {
-		cmn_err(CE_NOTE,
-                        "XFS resetting qflags for filesystem %s",
-                        mp->m_fsname);
-
-		error = xfs_mount_reset_sbqflags(mp);
-	}
-	return error;
-}
-
-static struct xfs_qmops xfs_qmcore_stub = {
-	.xfs_qminit		= (xfs_qminit_t) xfs_noquota_init,
-	.xfs_qmdone		= (xfs_qmdone_t) fs_noerr,
-	.xfs_qmmount		= (xfs_qmmount_t) fs_noerr,
-	.xfs_qmunmount		= (xfs_qmunmount_t) fs_noerr,
-	.xfs_dqrele		= (xfs_dqrele_t) fs_noerr,
-	.xfs_dqattach		= (xfs_dqattach_t) fs_noerr,
-	.xfs_dqdetach		= (xfs_dqdetach_t) fs_noerr,
-	.xfs_dqpurgeall		= (xfs_dqpurgeall_t) fs_noerr,
-	.xfs_dqvopalloc		= (xfs_dqvopalloc_t) fs_noerr,
-	.xfs_dqvopcreate	= (xfs_dqvopcreate_t) fs_noerr,
-	.xfs_dqvoprename	= (xfs_dqvoprename_t) fs_noerr,
-	.xfs_dqvopchown		= xfs_dqvopchown_default,
-	.xfs_dqvopchownresv	= (xfs_dqvopchownresv_t) fs_noerr,
-	.xfs_dqstatvfs		= (xfs_dqstatvfs_t) fs_noval,
-	.xfs_dqsync		= (xfs_dqsync_t) fs_noerr,
-};
-
-int
-xfs_qmops_get(struct xfs_mount *mp)
-{
-	if (XFS_IS_QUOTA_RUNNING(mp)) {
-#ifdef CONFIG_XFS_QUOTA
-		mp->m_qm_ops = &xfs_qmcore_xfs;
-#else
-		cmn_err(CE_WARN,
-			"XFS: qouta support not available in this kernel.");
-		return EINVAL;
-#endif
-	} else {
-		mp->m_qm_ops = &xfs_qmcore_stub;
-	}
-
-	return 0;
-}
-
-void
-xfs_qmops_put(struct xfs_mount *mp)
-{
-}
diff --git a/fs/xfs/xfs_quota.h b/fs/xfs/xfs_quota.h
index f5d1202dde25..3ec91ac74c2a 100644
--- a/fs/xfs/xfs_quota.h
+++ b/fs/xfs/xfs_quota.h
@@ -197,7 +197,6 @@ typedef struct xfs_qoff_logformat {
 #define XFS_QMOPT_UMOUNTING	0x0000100 /* filesys is being unmounted */
 #define XFS_QMOPT_DOLOG		0x0000200 /* log buf changes (in quotacheck) */
 #define XFS_QMOPT_DOWARN        0x0000400 /* increase warning cnt if needed */
-#define XFS_QMOPT_ILOCKED	0x0000800 /* inode is already locked (excl) */
 #define XFS_QMOPT_DQREPAIR	0x0001000 /* repair dquot if damaged */
 #define XFS_QMOPT_GQUOTA	0x0002000 /* group dquot requested */
 #define XFS_QMOPT_ENOSPC	0x0004000 /* enospc instead of edquot (prj) */
@@ -302,69 +301,79 @@ typedef struct xfs_dqtrx {
 	long		qt_delrtb_delta;  /* delayed RT blk count changes */
 } xfs_dqtrx_t;
 
-/*
- * Dquot transaction functions, used if quota is enabled.
- */
-typedef void	(*qo_dup_dqinfo_t)(struct xfs_trans *, struct xfs_trans *);
-typedef void	(*qo_mod_dquot_byino_t)(struct xfs_trans *,
-				struct xfs_inode *, uint, long);
-typedef void	(*qo_free_dqinfo_t)(struct xfs_trans *);
-typedef void	(*qo_apply_dquot_deltas_t)(struct xfs_trans *);
-typedef void	(*qo_unreserve_and_mod_dquots_t)(struct xfs_trans *);
-typedef int	(*qo_reserve_quota_nblks_t)(
-				struct xfs_trans *, struct xfs_mount *,
-				struct xfs_inode *, long, long, uint);
-typedef int	(*qo_reserve_quota_bydquots_t)(
-				struct xfs_trans *, struct xfs_mount *,
-				struct xfs_dquot *, struct xfs_dquot *,
-				long, long, uint);
-typedef struct xfs_dqtrxops {
-	qo_dup_dqinfo_t			qo_dup_dqinfo;
-	qo_free_dqinfo_t		qo_free_dqinfo;
-	qo_mod_dquot_byino_t		qo_mod_dquot_byino;
-	qo_apply_dquot_deltas_t		qo_apply_dquot_deltas;
-	qo_reserve_quota_nblks_t	qo_reserve_quota_nblks;
-	qo_reserve_quota_bydquots_t	qo_reserve_quota_bydquots;
-	qo_unreserve_and_mod_dquots_t	qo_unreserve_and_mod_dquots;
-} xfs_dqtrxops_t;
-
-#define XFS_DQTRXOP(mp, tp, op, args...) \
-		((mp)->m_qm_ops->xfs_dqtrxops ? \
-		((mp)->m_qm_ops->xfs_dqtrxops->op)(tp, ## args) : 0)
-
-#define XFS_DQTRXOP_VOID(mp, tp, op, args...) \
-		((mp)->m_qm_ops->xfs_dqtrxops ? \
-		((mp)->m_qm_ops->xfs_dqtrxops->op)(tp, ## args) : (void)0)
-
-#define XFS_TRANS_DUP_DQINFO(mp, otp, ntp) \
-	XFS_DQTRXOP_VOID(mp, otp, qo_dup_dqinfo, ntp)
-#define XFS_TRANS_FREE_DQINFO(mp, tp) \
-	XFS_DQTRXOP_VOID(mp, tp, qo_free_dqinfo)
-#define XFS_TRANS_MOD_DQUOT_BYINO(mp, tp, ip, field, delta) \
-	XFS_DQTRXOP_VOID(mp, tp, qo_mod_dquot_byino, ip, field, delta)
-#define XFS_TRANS_APPLY_DQUOT_DELTAS(mp, tp) \
-	XFS_DQTRXOP_VOID(mp, tp, qo_apply_dquot_deltas)
-#define XFS_TRANS_RESERVE_QUOTA_NBLKS(mp, tp, ip, nblks, ninos, fl) \
-	XFS_DQTRXOP(mp, tp, qo_reserve_quota_nblks, mp, ip, nblks, ninos, fl)
-#define XFS_TRANS_RESERVE_QUOTA_BYDQUOTS(mp, tp, ud, gd, nb, ni, fl) \
-	XFS_DQTRXOP(mp, tp, qo_reserve_quota_bydquots, mp, ud, gd, nb, ni, fl)
-#define XFS_TRANS_UNRESERVE_AND_MOD_DQUOTS(mp, tp) \
-	XFS_DQTRXOP_VOID(mp, tp, qo_unreserve_and_mod_dquots)
-
-#define XFS_TRANS_UNRESERVE_QUOTA_NBLKS(mp, tp, ip, nblks, ninos, flags) \
-	XFS_TRANS_RESERVE_QUOTA_NBLKS(mp, tp, ip, -(nblks), -(ninos), flags)
-#define XFS_TRANS_RESERVE_QUOTA(mp, tp, ud, gd, nb, ni, f) \
-	XFS_TRANS_RESERVE_QUOTA_BYDQUOTS(mp, tp, ud, gd, nb, ni, \
-				f | XFS_QMOPT_RES_REGBLKS)
-#define XFS_TRANS_UNRESERVE_QUOTA(mp, tp, ud, gd, nb, ni, f) \
-	XFS_TRANS_RESERVE_QUOTA_BYDQUOTS(mp, tp, ud, gd, -(nb), -(ni), \
+#ifdef CONFIG_XFS_QUOTA
+extern void xfs_trans_dup_dqinfo(struct xfs_trans *, struct xfs_trans *);
+extern void xfs_trans_free_dqinfo(struct xfs_trans *);
+extern void xfs_trans_mod_dquot_byino(struct xfs_trans *, struct xfs_inode *,
+		uint, long);
+extern void xfs_trans_apply_dquot_deltas(struct xfs_trans *);
+extern void xfs_trans_unreserve_and_mod_dquots(struct xfs_trans *);
+extern int xfs_trans_reserve_quota_nblks(struct xfs_trans *,
+		struct xfs_inode *, long, long, uint);
+extern int xfs_trans_reserve_quota_bydquots(struct xfs_trans *,
+		struct xfs_mount *, struct xfs_dquot *,
+		struct xfs_dquot *, long, long, uint);
+
+extern int xfs_qm_vop_dqalloc(struct xfs_inode *, uid_t, gid_t, prid_t, uint,
+		struct xfs_dquot **, struct xfs_dquot **);
+extern void xfs_qm_vop_create_dqattach(struct xfs_trans *, struct xfs_inode *,
+		struct xfs_dquot *, struct xfs_dquot *);
+extern int xfs_qm_vop_rename_dqattach(struct xfs_inode **);
+extern struct xfs_dquot *xfs_qm_vop_chown(struct xfs_trans *,
+		struct xfs_inode *, struct xfs_dquot **, struct xfs_dquot *);
+extern int xfs_qm_vop_chown_reserve(struct xfs_trans *, struct xfs_inode *,
+		struct xfs_dquot *, struct xfs_dquot *, uint);
+extern int xfs_qm_dqattach(struct xfs_inode *, uint);
+extern int xfs_qm_dqattach_locked(struct xfs_inode *, uint);
+extern void xfs_qm_dqdetach(struct xfs_inode *);
+extern void xfs_qm_dqrele(struct xfs_dquot *);
+extern void xfs_qm_statvfs(struct xfs_inode *, struct kstatfs *);
+extern int xfs_qm_sync(struct xfs_mount *, int);
+extern int xfs_qm_newmount(struct xfs_mount *, uint *, uint *);
+extern void xfs_qm_mount_quotas(struct xfs_mount *);
+extern void xfs_qm_unmount(struct xfs_mount *);
+extern void xfs_qm_unmount_quotas(struct xfs_mount *);
+
+#else
+static inline int
+xfs_qm_vop_dqalloc(struct xfs_inode *ip, uid_t uid, gid_t gid, prid_t prid,
+		uint flags, struct xfs_dquot **udqp, struct xfs_dquot **gdqp)
+{
+	*udqp = NULL;
+	*gdqp = NULL;
+	return 0;
+}
+#define xfs_trans_dup_dqinfo(tp, tp2)
+#define xfs_trans_free_dqinfo(tp)
+#define xfs_trans_mod_dquot_byino(tp, ip, fields, delta)
+#define xfs_trans_apply_dquot_deltas(tp)
+#define xfs_trans_unreserve_and_mod_dquots(tp)
+#define xfs_trans_reserve_quota_nblks(tp, ip, nblks, ninos, flags)	(0)
+#define xfs_trans_reserve_quota_bydquots(tp, mp, u, g, nb, ni, fl)	(0)
+#define xfs_qm_vop_create_dqattach(tp, ip, u, g)
+#define xfs_qm_vop_rename_dqattach(it)					(0)
+#define xfs_qm_vop_chown(tp, ip, old, new)				(NULL)
+#define xfs_qm_vop_chown_reserve(tp, ip, u, g, fl)			(0)
+#define xfs_qm_dqattach(ip, fl)						(0)
+#define xfs_qm_dqattach_locked(ip, fl)					(0)
+#define xfs_qm_dqdetach(ip)
+#define xfs_qm_dqrele(d)
+#define xfs_qm_statvfs(ip, s)
+#define xfs_qm_sync(mp, fl)						(0)
+#define xfs_qm_newmount(mp, a, b)					(0)
+#define xfs_qm_mount_quotas(mp)
+#define xfs_qm_unmount(mp)
+#define xfs_qm_unmount_quotas(mp)					(0)
+#endif /* CONFIG_XFS_QUOTA */
+
+#define xfs_trans_unreserve_quota_nblks(tp, ip, nblks, ninos, flags) \
+	xfs_trans_reserve_quota_nblks(tp, ip, -(nblks), -(ninos), flags)
+#define xfs_trans_reserve_quota(tp, mp, ud, gd, nb, ni, f) \
+	xfs_trans_reserve_quota_bydquots(tp, mp, ud, gd, nb, ni, \
 				f | XFS_QMOPT_RES_REGBLKS)
 
 extern int xfs_qm_dqcheck(xfs_disk_dquot_t *, xfs_dqid_t, uint, uint, char *);
 extern int xfs_mount_reset_sbqflags(struct xfs_mount *);
 
-extern struct xfs_qmops xfs_qmcore_xfs;
-
 #endif	/* __KERNEL__ */
-
 #endif	/* __XFS_QUOTA_H__ */
diff --git a/fs/xfs/xfs_rename.c b/fs/xfs/xfs_rename.c
index 58f85e9cd11d..b81deea0ce19 100644
--- a/fs/xfs/xfs_rename.c
+++ b/fs/xfs/xfs_rename.c
@@ -166,7 +166,8 @@ xfs_rename(
 	/*
 	 * Attach the dquots to the inodes
 	 */
-	if ((error = XFS_QM_DQVOPRENAME(mp, inodes))) {
+	error = xfs_qm_vop_rename_dqattach(inodes);
+	if (error) {
 		xfs_trans_cancel(tp, cancel_flags);
 		goto std_return;
 	}
diff --git a/fs/xfs/xfs_rw.c b/fs/xfs/xfs_rw.c
index 36f3a21c54d2..fea68615ed23 100644
--- a/fs/xfs/xfs_rw.c
+++ b/fs/xfs/xfs_rw.c
@@ -41,7 +41,6 @@
 #include "xfs_ialloc.h"
 #include "xfs_attr.h"
 #include "xfs_bmap.h"
-#include "xfs_acl.h"
 #include "xfs_error.h"
 #include "xfs_buf_item.h"
 #include "xfs_rw.h"
diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index bcc39d358ad3..66b849358e62 100644
--- a/fs/xfs/xfs_trans.c
+++ b/fs/xfs/xfs_trans.c
@@ -297,7 +297,7 @@ xfs_trans_dup(
 	tp->t_rtx_res = tp->t_rtx_res_used;
 	ntp->t_pflags = tp->t_pflags;
 
-	XFS_TRANS_DUP_DQINFO(tp->t_mountp, tp, ntp);
+	xfs_trans_dup_dqinfo(tp, ntp);
 
 	atomic_inc(&tp->t_mountp->m_active_trans);
 	return ntp;
@@ -829,7 +829,7 @@ shut_us_down:
 		 * means is that we have some (non-persistent) quota
 		 * reservations that need to be unreserved.
 		 */
-		XFS_TRANS_UNRESERVE_AND_MOD_DQUOTS(mp, tp);
+		xfs_trans_unreserve_and_mod_dquots(tp);
 		if (tp->t_ticket) {
 			commit_lsn = xfs_log_done(mp, tp->t_ticket,
 							NULL, log_flags);
@@ -848,10 +848,9 @@ shut_us_down:
 	/*
 	 * If we need to update the superblock, then do it now.
 	 */
-	if (tp->t_flags & XFS_TRANS_SB_DIRTY) {
+	if (tp->t_flags & XFS_TRANS_SB_DIRTY)
 		xfs_trans_apply_sb_deltas(tp);
-	}
-	XFS_TRANS_APPLY_DQUOT_DELTAS(mp, tp);
+	xfs_trans_apply_dquot_deltas(tp);
 
 	/*
 	 * Ask each log item how many log_vector entries it will
@@ -1056,7 +1055,7 @@ xfs_trans_uncommit(
 	}
 
 	xfs_trans_unreserve_and_mod_sb(tp);
-	XFS_TRANS_UNRESERVE_AND_MOD_DQUOTS(tp->t_mountp, tp);
+	xfs_trans_unreserve_and_mod_dquots(tp);
 
 	xfs_trans_free_items(tp, flags);
 	xfs_trans_free_busy(tp);
@@ -1181,7 +1180,7 @@ xfs_trans_cancel(
 	}
 #endif
 	xfs_trans_unreserve_and_mod_sb(tp);
-	XFS_TRANS_UNRESERVE_AND_MOD_DQUOTS(mp, tp);
+	xfs_trans_unreserve_and_mod_dquots(tp);
 
 	if (tp->t_ticket) {
 		if (flags & XFS_TRANS_RELEASE_LOG_RES) {
@@ -1211,7 +1210,7 @@ xfs_trans_free(
 	xfs_trans_t	*tp)
 {
 	atomic_dec(&tp->t_mountp->m_active_trans);
-	XFS_TRANS_FREE_DQINFO(tp->t_mountp, tp);
+	xfs_trans_free_dqinfo(tp);
 	kmem_zone_free(xfs_trans_zone, tp);
 }
 
diff --git a/fs/xfs/xfs_utils.c b/fs/xfs/xfs_utils.c
index 79b9e5ea5359..4d88616bde91 100644
--- a/fs/xfs/xfs_utils.c
+++ b/fs/xfs/xfs_utils.c
@@ -166,7 +166,7 @@ xfs_dir_ialloc(
 			xfs_buf_relse(ialloc_context);
 			if (dqinfo) {
 				tp->t_dqinfo = dqinfo;
-				XFS_TRANS_FREE_DQINFO(tp->t_mountp, tp);
+				xfs_trans_free_dqinfo(tp);
 			}
 			*tpp = ntp;
 			*ipp = NULL;
diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c
index 19cf90a9c762..c4eca5ed5dab 100644
--- a/fs/xfs/xfs_vnodeops.c
+++ b/fs/xfs/xfs_vnodeops.c
@@ -42,6 +42,7 @@
 #include "xfs_ialloc.h"
 #include "xfs_alloc.h"
 #include "xfs_bmap.h"
+#include "xfs_acl.h"
 #include "xfs_attr.h"
 #include "xfs_rw.h"
 #include "xfs_error.h"
@@ -118,7 +119,7 @@ xfs_setattr(
 		 */
 		ASSERT(udqp == NULL);
 		ASSERT(gdqp == NULL);
-		code = XFS_QM_DQVOPALLOC(mp, ip, uid, gid, ip->i_d.di_projid,
+		code = xfs_qm_vop_dqalloc(ip, uid, gid, ip->i_d.di_projid,
 					 qflags, &udqp, &gdqp);
 		if (code)
 			return code;
@@ -180,10 +181,11 @@ xfs_setattr(
 		 * Do a quota reservation only if uid/gid is actually
 		 * going to change.
 		 */
-		if ((XFS_IS_UQUOTA_ON(mp) && iuid != uid) ||
-		    (XFS_IS_GQUOTA_ON(mp) && igid != gid)) {
+		if (XFS_IS_QUOTA_RUNNING(mp) &&
+		    ((XFS_IS_UQUOTA_ON(mp) && iuid != uid) ||
+		     (XFS_IS_GQUOTA_ON(mp) && igid != gid))) {
 			ASSERT(tp);
-			code = XFS_QM_DQVOPCHOWNRESV(mp, tp, ip, udqp, gdqp,
+			code = xfs_qm_vop_chown_reserve(tp, ip, udqp, gdqp,
 						capable(CAP_FOWNER) ?
 						XFS_QMOPT_FORCE_RES : 0);
 			if (code)	/* out of quota */
@@ -217,7 +219,7 @@ xfs_setattr(
 		/*
 		 * Make sure that the dquots are attached to the inode.
 		 */
-		code = XFS_QM_DQATTACH(mp, ip, XFS_QMOPT_ILOCKED);
+		code = xfs_qm_dqattach_locked(ip, 0);
 		if (code)
 			goto error_return;
 
@@ -351,21 +353,21 @@ xfs_setattr(
 		 * in the transaction.
 		 */
 		if (iuid != uid) {
-			if (XFS_IS_UQUOTA_ON(mp)) {
+			if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_UQUOTA_ON(mp)) {
 				ASSERT(mask & ATTR_UID);
 				ASSERT(udqp);
-				olddquot1 = XFS_QM_DQVOPCHOWN(mp, tp, ip,
+				olddquot1 = xfs_qm_vop_chown(tp, ip,
 							&ip->i_udquot, udqp);
 			}
 			ip->i_d.di_uid = uid;
 			inode->i_uid = uid;
 		}
 		if (igid != gid) {
-			if (XFS_IS_GQUOTA_ON(mp)) {
+			if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_GQUOTA_ON(mp)) {
 				ASSERT(!XFS_IS_PQUOTA_ON(mp));
 				ASSERT(mask & ATTR_GID);
 				ASSERT(gdqp);
-				olddquot2 = XFS_QM_DQVOPCHOWN(mp, tp, ip,
+				olddquot2 = xfs_qm_vop_chown(tp, ip,
 							&ip->i_gdquot, gdqp);
 			}
 			ip->i_d.di_gid = gid;
@@ -461,13 +463,25 @@ xfs_setattr(
 	/*
 	 * Release any dquot(s) the inode had kept before chown.
 	 */
-	XFS_QM_DQRELE(mp, olddquot1);
-	XFS_QM_DQRELE(mp, olddquot2);
-	XFS_QM_DQRELE(mp, udqp);
-	XFS_QM_DQRELE(mp, gdqp);
+	xfs_qm_dqrele(olddquot1);
+	xfs_qm_dqrele(olddquot2);
+	xfs_qm_dqrele(udqp);
+	xfs_qm_dqrele(gdqp);
 
-	if (code) {
+	if (code)
 		return code;
+
+	/*
+	 * XXX(hch): Updating the ACL entries is not atomic vs the i_mode
+	 * 	     update.  We could avoid this with linked transactions
+	 * 	     and passing down the transaction pointer all the way
+	 *	     to attr_set.  No previous user of the generic
+	 * 	     Posix ACL code seems to care about this issue either.
+	 */
+	if ((mask & ATTR_MODE) && !(flags & XFS_ATTR_NOACL)) {
+		code = -xfs_acl_chmod(inode);
+		if (code)
+			return XFS_ERROR(code);
 	}
 
 	if (DM_EVENT_ENABLED(ip, DM_EVENT_ATTRIBUTE) &&
@@ -482,8 +496,8 @@ xfs_setattr(
 	commit_flags |= XFS_TRANS_ABORT;
 	/* FALLTHROUGH */
  error_return:
-	XFS_QM_DQRELE(mp, udqp);
-	XFS_QM_DQRELE(mp, gdqp);
+	xfs_qm_dqrele(udqp);
+	xfs_qm_dqrele(gdqp);
 	if (tp) {
 		xfs_trans_cancel(tp, commit_flags);
 	}
@@ -739,7 +753,8 @@ xfs_free_eofblocks(
 		/*
 		 * Attach the dquots to the inode up front.
 		 */
-		if ((error = XFS_QM_DQATTACH(mp, ip, 0)))
+		error = xfs_qm_dqattach(ip, 0);
+		if (error)
 			return error;
 
 		/*
@@ -1181,7 +1196,8 @@ xfs_inactive(
 
 	ASSERT(ip->i_d.di_nlink == 0);
 
-	if ((error = XFS_QM_DQATTACH(mp, ip, 0)))
+	error = xfs_qm_dqattach(ip, 0);
+	if (error)
 		return VN_INACTIVE_CACHE;
 
 	tp = xfs_trans_alloc(mp, XFS_TRANS_INACTIVE);
@@ -1307,7 +1323,7 @@ xfs_inactive(
 		/*
 		 * Credit the quota account(s). The inode is gone.
 		 */
-		XFS_TRANS_MOD_DQUOT_BYINO(mp, tp, ip, XFS_TRANS_DQ_ICOUNT, -1);
+		xfs_trans_mod_dquot_byino(tp, ip, XFS_TRANS_DQ_ICOUNT, -1);
 
 		/*
 		 * Just ignore errors at this point.  There is nothing we can
@@ -1323,11 +1339,11 @@ xfs_inactive(
 			xfs_fs_cmn_err(CE_NOTE, mp, "xfs_inactive: "
 				"xfs_trans_commit() returned error %d", error);
 	}
+
 	/*
 	 * Release the dquots held by inode, if any.
 	 */
-	XFS_QM_DQDETACH(mp, ip);
-
+	xfs_qm_dqdetach(ip);
 	xfs_iunlock(ip, XFS_IOLOCK_EXCL | XFS_ILOCK_EXCL);
 
  out:
@@ -1427,8 +1443,7 @@ xfs_create(
 	/*
 	 * Make sure that we have allocated dquot(s) on disk.
 	 */
-	error = XFS_QM_DQVOPALLOC(mp, dp,
-			current_fsuid(), current_fsgid(), prid,
+	error = xfs_qm_vop_dqalloc(dp, current_fsuid(), current_fsgid(), prid,
 			XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, &udqp, &gdqp);
 	if (error)
 		goto std_return;
@@ -1489,7 +1504,7 @@ xfs_create(
 	/*
 	 * Reserve disk quota and the inode.
 	 */
-	error = XFS_TRANS_RESERVE_QUOTA(mp, tp, udqp, gdqp, resblks, 1, 0);
+	error = xfs_trans_reserve_quota(tp, mp, udqp, gdqp, resblks, 1, 0);
 	if (error)
 		goto out_trans_cancel;
 
@@ -1561,7 +1576,7 @@ xfs_create(
 	 * These ids of the inode couldn't have changed since the new
 	 * inode has been locked ever since it was created.
 	 */
-	XFS_QM_DQVOPCREATE(mp, tp, ip, udqp, gdqp);
+	xfs_qm_vop_create_dqattach(tp, ip, udqp, gdqp);
 
 	/*
 	 * xfs_trans_commit normally decrements the vnode ref count
@@ -1580,8 +1595,8 @@ xfs_create(
 		goto out_dqrele;
 	}
 
-	XFS_QM_DQRELE(mp, udqp);
-	XFS_QM_DQRELE(mp, gdqp);
+	xfs_qm_dqrele(udqp);
+	xfs_qm_dqrele(gdqp);
 
 	*ipp = ip;
 
@@ -1602,8 +1617,8 @@ xfs_create(
  out_trans_cancel:
 	xfs_trans_cancel(tp, cancel_flags);
  out_dqrele:
-	XFS_QM_DQRELE(mp, udqp);
-	XFS_QM_DQRELE(mp, gdqp);
+	xfs_qm_dqrele(udqp);
+	xfs_qm_dqrele(gdqp);
 
 	if (unlock_dp_on_error)
 		xfs_iunlock(dp, XFS_ILOCK_EXCL);
@@ -1837,11 +1852,11 @@ xfs_remove(
 			return error;
 	}
 
-	error = XFS_QM_DQATTACH(mp, dp, 0);
+	error = xfs_qm_dqattach(dp, 0);
 	if (error)
 		goto std_return;
 
-	error = XFS_QM_DQATTACH(mp, ip, 0);
+	error = xfs_qm_dqattach(ip, 0);
 	if (error)
 		goto std_return;
 
@@ -2028,11 +2043,11 @@ xfs_link(
 
 	/* Return through std_return after this point. */
 
-	error = XFS_QM_DQATTACH(mp, sip, 0);
+	error = xfs_qm_dqattach(sip, 0);
 	if (error)
 		goto std_return;
 
-	error = XFS_QM_DQATTACH(mp, tdp, 0);
+	error = xfs_qm_dqattach(tdp, 0);
 	if (error)
 		goto std_return;
 
@@ -2205,8 +2220,7 @@ xfs_symlink(
 	/*
 	 * Make sure that we have allocated dquot(s) on disk.
 	 */
-	error = XFS_QM_DQVOPALLOC(mp, dp,
-			current_fsuid(), current_fsgid(), prid,
+	error = xfs_qm_vop_dqalloc(dp, current_fsuid(), current_fsgid(), prid,
 			XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, &udqp, &gdqp);
 	if (error)
 		goto std_return;
@@ -2248,7 +2262,7 @@ xfs_symlink(
 	/*
 	 * Reserve disk quota : blocks and inode.
 	 */
-	error = XFS_TRANS_RESERVE_QUOTA(mp, tp, udqp, gdqp, resblks, 1, 0);
+	error = xfs_trans_reserve_quota(tp, mp, udqp, gdqp, resblks, 1, 0);
 	if (error)
 		goto error_return;
 
@@ -2288,7 +2302,7 @@ xfs_symlink(
 	/*
 	 * Also attach the dquot(s) to it, if applicable.
 	 */
-	XFS_QM_DQVOPCREATE(mp, tp, ip, udqp, gdqp);
+	xfs_qm_vop_create_dqattach(tp, ip, udqp, gdqp);
 
 	if (resblks)
 		resblks -= XFS_IALLOC_SPACE_RES(mp);
@@ -2376,8 +2390,8 @@ xfs_symlink(
 		goto error2;
 	}
 	error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
-	XFS_QM_DQRELE(mp, udqp);
-	XFS_QM_DQRELE(mp, gdqp);
+	xfs_qm_dqrele(udqp);
+	xfs_qm_dqrele(gdqp);
 
 	/* Fall through to std_return with error = 0 or errno from
 	 * xfs_trans_commit	*/
@@ -2401,8 +2415,8 @@ std_return:
 	cancel_flags |= XFS_TRANS_ABORT;
  error_return:
 	xfs_trans_cancel(tp, cancel_flags);
-	XFS_QM_DQRELE(mp, udqp);
-	XFS_QM_DQRELE(mp, gdqp);
+	xfs_qm_dqrele(udqp);
+	xfs_qm_dqrele(gdqp);
 
 	if (unlock_dp_on_error)
 		xfs_iunlock(dp, XFS_ILOCK_EXCL);
@@ -2541,7 +2555,8 @@ xfs_alloc_file_space(
 	if (XFS_FORCED_SHUTDOWN(mp))
 		return XFS_ERROR(EIO);
 
-	if ((error = XFS_QM_DQATTACH(mp, ip, 0)))
+	error = xfs_qm_dqattach(ip, 0);
+	if (error)
 		return error;
 
 	if (len <= 0)
@@ -2628,8 +2643,8 @@ retry:
 			break;
 		}
 		xfs_ilock(ip, XFS_ILOCK_EXCL);
-		error = XFS_TRANS_RESERVE_QUOTA_NBLKS(mp, tp, ip,
-						      qblocks, 0, quota_flag);
+		error = xfs_trans_reserve_quota_nblks(tp, ip, qblocks,
+						      0, quota_flag);
 		if (error)
 			goto error1;
 
@@ -2688,7 +2703,7 @@ dmapi_enospc_check:
 
 error0:	/* Cancel bmap, unlock inode, unreserve quota blocks, cancel trans */
 	xfs_bmap_cancel(&free_list);
-	XFS_TRANS_UNRESERVE_QUOTA_NBLKS(mp, tp, ip, qblocks, 0, quota_flag);
+	xfs_trans_unreserve_quota_nblks(tp, ip, qblocks, 0, quota_flag);
 
 error1:	/* Just cancel transaction */
 	xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT);
@@ -2827,7 +2842,8 @@ xfs_free_file_space(
 
 	xfs_itrace_entry(ip);
 
-	if ((error = XFS_QM_DQATTACH(mp, ip, 0)))
+	error = xfs_qm_dqattach(ip, 0);
+	if (error)
 		return error;
 
 	error = 0;
@@ -2953,9 +2969,9 @@ xfs_free_file_space(
 			break;
 		}
 		xfs_ilock(ip, XFS_ILOCK_EXCL);
-		error = XFS_TRANS_RESERVE_QUOTA(mp, tp,
-				ip->i_udquot, ip->i_gdquot, resblks, 0,
-				XFS_QMOPT_RES_REGBLKS);
+		error = xfs_trans_reserve_quota(tp, mp,
+				ip->i_udquot, ip->i_gdquot,
+				resblks, 0, XFS_QMOPT_RES_REGBLKS);
 		if (error)
 			goto error1;
 
diff --git a/fs/xfs/xfs_vnodeops.h b/fs/xfs/xfs_vnodeops.h
index 04373c6c61ff..a9e102de71a1 100644
--- a/fs/xfs/xfs_vnodeops.h
+++ b/fs/xfs/xfs_vnodeops.h
@@ -18,6 +18,7 @@ int xfs_setattr(struct xfs_inode *ip, struct iattr *vap, int flags);
 #define	XFS_ATTR_DMI		0x01	/* invocation from a DMI function */
 #define	XFS_ATTR_NONBLOCK	0x02	/* return EAGAIN if operation would block */
 #define XFS_ATTR_NOLOCK		0x04	/* Don't grab any conflicting locks */
+#define XFS_ATTR_NOACL		0x08	/* Don't call xfs_acl_chmod */
 
 int xfs_readlink(struct xfs_inode *ip, char *link);
 int xfs_fsync(struct xfs_inode *ip);