24 files changed, 4103 insertions, 1563 deletions

diff --git a/fs/notify/Kconfig b/fs/notify/Kconfig
index 50914d7303c..22c629eedd8 100644
--- a/fs/notify/Kconfig
+++ b/fs/notify/Kconfig
@@ -1,2 +1,6 @@
+config FSNOTIFY
+	def_bool n
+
 source "fs/notify/dnotify/Kconfig"
 source "fs/notify/inotify/Kconfig"
+source "fs/notify/fanotify/Kconfig"
diff --git a/fs/notify/Makefile b/fs/notify/Makefile
index 5a95b6010ce..96d3420d024 100644
--- a/fs/notify/Makefile
+++ b/fs/notify/Makefile
@@ -1,2 +1,6 @@
+obj-$(CONFIG_FSNOTIFY)		+= fsnotify.o notification.o group.o inode_mark.o \
+				   mark.o vfsmount_mark.o fdinfo.o
+
 obj-y			+= dnotify/
 obj-y			+= inotify/
+obj-y			+= fanotify/
diff --git a/fs/notify/dnotify/Kconfig b/fs/notify/dnotify/Kconfig
index 26adf5dfa64..f9c1ca139d8 100644
--- a/fs/notify/dnotify/Kconfig
+++ b/fs/notify/dnotify/Kconfig
@@ -1,5 +1,6 @@
 config DNOTIFY
 	bool "Dnotify support"
+	select FSNOTIFY
 	default y
 	help
 	  Dnotify is a directory-based per-fd file change notification system
diff --git a/fs/notify/dnotify/dnotify.c b/fs/notify/dnotify/dnotify.c
index b0aa2cde80b..abc8cbcfe90 100644
--- a/fs/notify/dnotify/dnotify.c
+++ b/fs/notify/dnotify/dnotify.c
@@ -3,6 +3,9 @@
  *
  * Copyright (C) 2000,2001,2002 Stephen Rothwell
  *
+ * Copyright (C) 2009 Eric Paris <Red Hat Inc>
+ * dnotify was largly rewritten to use the new fsnotify infrastructure
+ *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of the GNU General Public License as published by the
  * Free Software Foundation; either version 2, or (at your option) any
@@ -21,170 +24,370 @@
 #include <linux/spinlock.h>
 #include <linux/slab.h>
 #include <linux/fdtable.h>
+#include <linux/fsnotify_backend.h>
 
 int dir_notify_enable __read_mostly = 1;
 
-static struct kmem_cache *dn_cache __read_mostly;
+static struct kmem_cache *dnotify_struct_cache __read_mostly;
+static struct kmem_cache *dnotify_mark_cache __read_mostly;
+static struct fsnotify_group *dnotify_group __read_mostly;
+
+/*
+ * dnotify will attach one of these to each inode (i_fsnotify_marks) which
+ * is being watched by dnotify.  If multiple userspace applications are watching
+ * the same directory with dnotify their information is chained in dn
+ */
+struct dnotify_mark {
+	struct fsnotify_mark fsn_mark;
+	struct dnotify_struct *dn;
+};
 
-static void redo_inode_mask(struct inode *inode)
+/*
+ * When a process starts or stops watching an inode the set of events which
+ * dnotify cares about for that inode may change.  This function runs the
+ * list of everything receiving dnotify events about this directory and calculates
+ * the set of all those events.  After it updates what dnotify is interested in
+ * it calls the fsnotify function so it can update the set of all events relevant
+ * to this inode.
+ */
+static void dnotify_recalc_inode_mask(struct fsnotify_mark *fsn_mark)
 {
-	unsigned long new_mask;
+	__u32 new_mask, old_mask;
 	struct dnotify_struct *dn;
+	struct dnotify_mark *dn_mark  = container_of(fsn_mark,
+						     struct dnotify_mark,
+						     fsn_mark);
 
+	assert_spin_locked(&fsn_mark->lock);
+
+	old_mask = fsn_mark->mask;
 	new_mask = 0;
-	for (dn = inode->i_dnotify; dn != NULL; dn = dn->dn_next)
-		new_mask |= dn->dn_mask & ~DN_MULTISHOT;
-	inode->i_dnotify_mask = new_mask;
+	for (dn = dn_mark->dn; dn != NULL; dn = dn->dn_next)
+		new_mask |= (dn->dn_mask & ~FS_DN_MULTISHOT);
+	fsnotify_set_mark_mask_locked(fsn_mark, new_mask);
+
+	if (old_mask == new_mask)
+		return;
+
+	if (fsn_mark->i.inode)
+		fsnotify_recalc_inode_mask(fsn_mark->i.inode);
+}
+
+/*
+ * Mains fsnotify call where events are delivered to dnotify.
+ * Find the dnotify mark on the relevant inode, run the list of dnotify structs
+ * on that mark and determine which of them has expressed interest in receiving
+ * events of this type.  When found send the correct process and signal and
+ * destroy the dnotify struct if it was not registered to receive multiple
+ * events.
+ */
+static int dnotify_handle_event(struct fsnotify_group *group,
+				struct inode *inode,
+				struct fsnotify_mark *inode_mark,
+				struct fsnotify_mark *vfsmount_mark,
+				u32 mask, void *data, int data_type,
+				const unsigned char *file_name, u32 cookie)
+{
+	struct dnotify_mark *dn_mark;
+	struct dnotify_struct *dn;
+	struct dnotify_struct **prev;
+	struct fown_struct *fown;
+	__u32 test_mask = mask & ~FS_EVENT_ON_CHILD;
+
+	/* not a dir, dnotify doesn't care */
+	if (!S_ISDIR(inode->i_mode))
+		return 0;
+
+	BUG_ON(vfsmount_mark);
+
+	dn_mark = container_of(inode_mark, struct dnotify_mark, fsn_mark);
+
+	spin_lock(&inode_mark->lock);
+	prev = &dn_mark->dn;
+	while ((dn = *prev) != NULL) {
+		if ((dn->dn_mask & test_mask) == 0) {
+			prev = &dn->dn_next;
+			continue;
+		}
+		fown = &dn->dn_filp->f_owner;
+		send_sigio(fown, dn->dn_fd, POLL_MSG);
+		if (dn->dn_mask & FS_DN_MULTISHOT)
+			prev = &dn->dn_next;
+		else {
+			*prev = dn->dn_next;
+			kmem_cache_free(dnotify_struct_cache, dn);
+			dnotify_recalc_inode_mask(inode_mark);
+		}
+	}
+
+	spin_unlock(&inode_mark->lock);
+
+	return 0;
 }
 
+static void dnotify_free_mark(struct fsnotify_mark *fsn_mark)
+{
+	struct dnotify_mark *dn_mark = container_of(fsn_mark,
+						    struct dnotify_mark,
+						    fsn_mark);
+
+	BUG_ON(dn_mark->dn);
+
+	kmem_cache_free(dnotify_mark_cache, dn_mark);
+}
+
+static struct fsnotify_ops dnotify_fsnotify_ops = {
+	.handle_event = dnotify_handle_event,
+};
+
+/*
+ * Called every time a file is closed.  Looks first for a dnotify mark on the
+ * inode.  If one is found run all of the ->dn structures attached to that
+ * mark for one relevant to this process closing the file and remove that
+ * dnotify_struct.  If that was the last dnotify_struct also remove the
+ * fsnotify_mark.
+ */
 void dnotify_flush(struct file *filp, fl_owner_t id)
 {
+	struct fsnotify_mark *fsn_mark;
+	struct dnotify_mark *dn_mark;
 	struct dnotify_struct *dn;
 	struct dnotify_struct **prev;
 	struct inode *inode;
 
-	inode = filp->f_path.dentry->d_inode;
+	inode = file_inode(filp);
 	if (!S_ISDIR(inode->i_mode))
 		return;
-	spin_lock(&inode->i_lock);
-	prev = &inode->i_dnotify;
+
+	fsn_mark = fsnotify_find_inode_mark(dnotify_group, inode);
+	if (!fsn_mark)
+		return;
+	dn_mark = container_of(fsn_mark, struct dnotify_mark, fsn_mark);
+
+	mutex_lock(&dnotify_group->mark_mutex);
+
+	spin_lock(&fsn_mark->lock);
+	prev = &dn_mark->dn;
 	while ((dn = *prev) != NULL) {
 		if ((dn->dn_owner == id) && (dn->dn_filp == filp)) {
 			*prev = dn->dn_next;
-			redo_inode_mask(inode);
-			kmem_cache_free(dn_cache, dn);
+			kmem_cache_free(dnotify_struct_cache, dn);
+			dnotify_recalc_inode_mask(fsn_mark);
 			break;
 		}
 		prev = &dn->dn_next;
 	}
-	spin_unlock(&inode->i_lock);
+
+	spin_unlock(&fsn_mark->lock);
+
+	/* nothing else could have found us thanks to the dnotify_groups
+	   mark_mutex */
+	if (dn_mark->dn == NULL)
+		fsnotify_destroy_mark_locked(fsn_mark, dnotify_group);
+
+	mutex_unlock(&dnotify_group->mark_mutex);
+
+	fsnotify_put_mark(fsn_mark);
 }
 
-int fcntl_dirnotify(int fd, struct file *filp, unsigned long arg)
+/* this conversion is done only at watch creation */
+static __u32 convert_arg(unsigned long arg)
+{
+	__u32 new_mask = FS_EVENT_ON_CHILD;
+
+	if (arg & DN_MULTISHOT)
+		new_mask |= FS_DN_MULTISHOT;
+	if (arg & DN_DELETE)
+		new_mask |= (FS_DELETE | FS_MOVED_FROM);
+	if (arg & DN_MODIFY)
+		new_mask |= FS_MODIFY;
+	if (arg & DN_ACCESS)
+		new_mask |= FS_ACCESS;
+	if (arg & DN_ATTRIB)
+		new_mask |= FS_ATTRIB;
+	if (arg & DN_RENAME)
+		new_mask |= FS_DN_RENAME;
+	if (arg & DN_CREATE)
+		new_mask |= (FS_CREATE | FS_MOVED_TO);
+
+	return new_mask;
+}
+
+/*
+ * If multiple processes watch the same inode with dnotify there is only one
+ * dnotify mark in inode->i_fsnotify_marks but we chain a dnotify_struct
+ * onto that mark.  This function either attaches the new dnotify_struct onto
+ * that list, or it |= the mask onto an existing dnofiy_struct.
+ */
+static int attach_dn(struct dnotify_struct *dn, struct dnotify_mark *dn_mark,
+		     fl_owner_t id, int fd, struct file *filp, __u32 mask)
 {
-	struct dnotify_struct *dn;
 	struct dnotify_struct *odn;
-	struct dnotify_struct **prev;
-	struct inode *inode;
-	fl_owner_t id = current->files;
-	struct file *f;
-	int error = 0;
 
-	if ((arg & ~DN_MULTISHOT) == 0) {
-		dnotify_flush(filp, id);
-		return 0;
-	}
-	if (!dir_notify_enable)
-		return -EINVAL;
-	inode = filp->f_path.dentry->d_inode;
-	if (!S_ISDIR(inode->i_mode))
-		return -ENOTDIR;
-	dn = kmem_cache_alloc(dn_cache, GFP_KERNEL);
-	if (dn == NULL)
-		return -ENOMEM;
-	spin_lock(&inode->i_lock);
-	prev = &inode->i_dnotify;
-	while ((odn = *prev) != NULL) {
+	odn = dn_mark->dn;
+	while (odn != NULL) {
+		/* adding more events to existing dnofiy_struct? */
 		if ((odn->dn_owner == id) && (odn->dn_filp == filp)) {
 			odn->dn_fd = fd;
-			odn->dn_mask |= arg;
-			inode->i_dnotify_mask |= arg & ~DN_MULTISHOT;
-			goto out_free;
+			odn->dn_mask |= mask;
+			return -EEXIST;
 		}
-		prev = &odn->dn_next;
+		odn = odn->dn_next;
 	}
 
-	rcu_read_lock();
-	f = fcheck(fd);
-	rcu_read_unlock();
-	/* we'd lost the race with close(), sod off silently */
-	/* note that inode->i_lock prevents reordering problems
-	 * between accesses to descriptor table and ->i_dnotify */
-	if (f != filp)
-		goto out_free;
-
-	error = __f_setown(filp, task_pid(current), PIDTYPE_PID, 0);
-	if (error)
-		goto out_free;
-
-	dn->dn_mask = arg;
+	dn->dn_mask = mask;
 	dn->dn_fd = fd;
 	dn->dn_filp = filp;
 	dn->dn_owner = id;
-	inode->i_dnotify_mask |= arg & ~DN_MULTISHOT;
-	dn->dn_next = inode->i_dnotify;
-	inode->i_dnotify = dn;
-	spin_unlock(&inode->i_lock);
-	return 0;
+	dn->dn_next = dn_mark->dn;
+	dn_mark->dn = dn;
 
-out_free:
-	spin_unlock(&inode->i_lock);
-	kmem_cache_free(dn_cache, dn);
-	return error;
+	return 0;
 }
 
-void __inode_dir_notify(struct inode *inode, unsigned long event)
+/*
+ * When a process calls fcntl to attach a dnotify watch to a directory it ends
+ * up here.  Allocate both a mark for fsnotify to add and a dnotify_struct to be
+ * attached to the fsnotify_mark.
+ */
+int fcntl_dirnotify(int fd, struct file *filp, unsigned long arg)
 {
-	struct dnotify_struct *	dn;
-	struct dnotify_struct **prev;
-	struct fown_struct *	fown;
-	int			changed = 0;
+	struct dnotify_mark *new_dn_mark, *dn_mark;
+	struct fsnotify_mark *new_fsn_mark, *fsn_mark;
+	struct dnotify_struct *dn;
+	struct inode *inode;
+	fl_owner_t id = current->files;
+	struct file *f;
+	int destroy = 0, error = 0;
+	__u32 mask;
 
-	spin_lock(&inode->i_lock);
-	prev = &inode->i_dnotify;
-	while ((dn = *prev) != NULL) {
-		if ((dn->dn_mask & event) == 0) {
-			prev = &dn->dn_next;
-			continue;
-		}
-		fown = &dn->dn_filp->f_owner;
-		send_sigio(fown, dn->dn_fd, POLL_MSG);
-		if (dn->dn_mask & DN_MULTISHOT)
-			prev = &dn->dn_next;
-		else {
-			*prev = dn->dn_next;
-			changed = 1;
-			kmem_cache_free(dn_cache, dn);
-		}
+	/* we use these to tell if we need to kfree */
+	new_fsn_mark = NULL;
+	dn = NULL;
+
+	if (!dir_notify_enable) {
+		error = -EINVAL;
+		goto out_err;
 	}
-	if (changed)
-		redo_inode_mask(inode);
-	spin_unlock(&inode->i_lock);
-}
 
-EXPORT_SYMBOL(__inode_dir_notify);
+	/* a 0 mask means we are explicitly removing the watch */
+	if ((arg & ~DN_MULTISHOT) == 0) {
+		dnotify_flush(filp, id);
+		error = 0;
+		goto out_err;
+	}
 
-/*
- * This is hopelessly wrong, but unfixable without API changes.  At
- * least it doesn't oops the kernel...
- *
- * To safely access ->d_parent we need to keep d_move away from it.  Use the
- * dentry's d_lock for this.
- */
-void dnotify_parent(struct dentry *dentry, unsigned long event)
-{
-	struct dentry *parent;
+	/* dnotify only works on directories */
+	inode = file_inode(filp);
+	if (!S_ISDIR(inode->i_mode)) {
+		error = -ENOTDIR;
+		goto out_err;
+	}
 
-	if (!dir_notify_enable)
-		return;
+	/* expect most fcntl to add new rather than augment old */
+	dn = kmem_cache_alloc(dnotify_struct_cache, GFP_KERNEL);
+	if (!dn) {
+		error = -ENOMEM;
+		goto out_err;
+	}
+
+	/* new fsnotify mark, we expect most fcntl calls to add a new mark */
+	new_dn_mark = kmem_cache_alloc(dnotify_mark_cache, GFP_KERNEL);
+	if (!new_dn_mark) {
+		error = -ENOMEM;
+		goto out_err;
+	}
+
+	/* convert the userspace DN_* "arg" to the internal FS_* defines in fsnotify */
+	mask = convert_arg(arg);
+
+	/* set up the new_fsn_mark and new_dn_mark */
+	new_fsn_mark = &new_dn_mark->fsn_mark;
+	fsnotify_init_mark(new_fsn_mark, dnotify_free_mark);
+	new_fsn_mark->mask = mask;
+	new_dn_mark->dn = NULL;
 
-	spin_lock(&dentry->d_lock);
-	parent = dentry->d_parent;
-	if (parent->d_inode->i_dnotify_mask & event) {
-		dget(parent);
-		spin_unlock(&dentry->d_lock);
-		__inode_dir_notify(parent->d_inode, event);
-		dput(parent);
+	/* this is needed to prevent the fcntl/close race described below */
+	mutex_lock(&dnotify_group->mark_mutex);
+
+	/* add the new_fsn_mark or find an old one. */
+	fsn_mark = fsnotify_find_inode_mark(dnotify_group, inode);
+	if (fsn_mark) {
+		dn_mark = container_of(fsn_mark, struct dnotify_mark, fsn_mark);
+		spin_lock(&fsn_mark->lock);
 	} else {
-		spin_unlock(&dentry->d_lock);
+		fsnotify_add_mark_locked(new_fsn_mark, dnotify_group, inode,
+					 NULL, 0);
+		spin_lock(&new_fsn_mark->lock);
+		fsn_mark = new_fsn_mark;
+		dn_mark = new_dn_mark;
+		/* we used new_fsn_mark, so don't free it */
+		new_fsn_mark = NULL;
+	}
+
+	rcu_read_lock();
+	f = fcheck(fd);
+	rcu_read_unlock();
+
+	/* if (f != filp) means that we lost a race and another task/thread
+	 * actually closed the fd we are still playing with before we grabbed
+	 * the dnotify_groups mark_mutex and fsn_mark->lock.  Since closing the
+	 * fd is the only time we clean up the marks we need to get our mark
+	 * off the list. */
+	if (f != filp) {
+		/* if we added ourselves, shoot ourselves, it's possible that
+		 * the flush actually did shoot this fsn_mark.  That's fine too
+		 * since multiple calls to destroy_mark is perfectly safe, if
+		 * we found a dn_mark already attached to the inode, just sod
+		 * off silently as the flush at close time dealt with it.
+		 */
+		if (dn_mark == new_dn_mark)
+			destroy = 1;
+		goto out;
+	}
+
+	error = __f_setown(filp, task_pid(current), PIDTYPE_PID, 0);
+	if (error) {
+		/* if we added, we must shoot */
+		if (dn_mark == new_dn_mark)
+			destroy = 1;
+		goto out;
 	}
+
+	error = attach_dn(dn, dn_mark, id, fd, filp, mask);
+	/* !error means that we attached the dn to the dn_mark, so don't free it */
+	if (!error)
+		dn = NULL;
+	/* -EEXIST means that we didn't add this new dn and used an old one.
+	 * that isn't an error (and the unused dn should be freed) */
+	else if (error == -EEXIST)
+		error = 0;
+
+	dnotify_recalc_inode_mask(fsn_mark);
+out:
+	spin_unlock(&fsn_mark->lock);
+
+	if (destroy)
+		fsnotify_destroy_mark_locked(fsn_mark, dnotify_group);
+
+	mutex_unlock(&dnotify_group->mark_mutex);
+	fsnotify_put_mark(fsn_mark);
+out_err:
+	if (new_fsn_mark)
+		fsnotify_put_mark(new_fsn_mark);
+	if (dn)
+		kmem_cache_free(dnotify_struct_cache, dn);
+	return error;
 }
-EXPORT_SYMBOL_GPL(dnotify_parent);
 
 static int __init dnotify_init(void)
 {
-	dn_cache = kmem_cache_create("dnotify_cache",
-		sizeof(struct dnotify_struct), 0, SLAB_PANIC, NULL);
+	dnotify_struct_cache = KMEM_CACHE(dnotify_struct, SLAB_PANIC);
+	dnotify_mark_cache = KMEM_CACHE(dnotify_mark, SLAB_PANIC);
+
+	dnotify_group = fsnotify_alloc_group(&dnotify_fsnotify_ops);
+	if (IS_ERR(dnotify_group))
+		panic("unable to allocate fsnotify group for dnotify\n");
 	return 0;
 }
 
diff --git a/fs/notify/fanotify/Kconfig b/fs/notify/fanotify/Kconfig
new file mode 100644
index 00000000000..e5f911bd80d
--- /dev/null
+++ b/fs/notify/fanotify/Kconfig
@@ -0,0 +1,26 @@
+config FANOTIFY
+	bool "Filesystem wide access notification"
+	select FSNOTIFY
+	select ANON_INODES
+	default n
+	---help---
+	   Say Y here to enable fanotify support.  fanotify is a file access
+	   notification system which differs from inotify in that it sends
+	   an open file descriptor to the userspace listener along with
+	   the event.
+
+	   If unsure, say Y.
+
+config FANOTIFY_ACCESS_PERMISSIONS
+	bool "fanotify permissions checking"
+	depends on FANOTIFY
+	depends on SECURITY
+	default n
+	---help---
+	   Say Y here is you want fanotify listeners to be able to make permissions
+	   decisions concerning filesystem events.  This is used by some fanotify
+	   listeners which need to scan files before allowing the system access to
+	   use those files.  This is used by some anti-malware vendors and by some
+	   hierarchical storage managent systems.
+
+	   If unsure, say N.
diff --git a/fs/notify/fanotify/Makefile b/fs/notify/fanotify/Makefile
new file mode 100644
index 00000000000..0999213e7e6
--- /dev/null
+++ b/fs/notify/fanotify/Makefile
@@ -0,0 +1 @@
+obj-$(CONFIG_FANOTIFY)		+= fanotify.o fanotify_user.o
diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c
new file mode 100644
index 00000000000..ee9cb3795c2
--- /dev/null
+++ b/fs/notify/fanotify/fanotify.c
@@ -0,0 +1,262 @@
+#include <linux/fanotify.h>
+#include <linux/fdtable.h>
+#include <linux/fsnotify_backend.h>
+#include <linux/init.h>
+#include <linux/jiffies.h>
+#include <linux/kernel.h> /* UINT_MAX */
+#include <linux/mount.h>
+#include <linux/sched.h>
+#include <linux/types.h>
+#include <linux/wait.h>
+
+#include "fanotify.h"
+
+static bool should_merge(struct fsnotify_event *old_fsn,
+			 struct fsnotify_event *new_fsn)
+{
+	struct fanotify_event_info *old, *new;
+
+	pr_debug("%s: old=%p new=%p\n", __func__, old_fsn, new_fsn);
+	old = FANOTIFY_E(old_fsn);
+	new = FANOTIFY_E(new_fsn);
+
+	if (old_fsn->inode == new_fsn->inode && old->tgid == new->tgid &&
+	    old->path.mnt == new->path.mnt &&
+	    old->path.dentry == new->path.dentry)
+		return true;
+	return false;
+}
+
+/* and the list better be locked by something too! */
+static int fanotify_merge(struct list_head *list, struct fsnotify_event *event)
+{
+	struct fsnotify_event *test_event;
+	bool do_merge = false;
+
+	pr_debug("%s: list=%p event=%p\n", __func__, list, event);
+
+#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
+	/*
+	 * Don't merge a permission event with any other event so that we know
+	 * the event structure we have created in fanotify_handle_event() is the
+	 * one we should check for permission response.
+	 */
+	if (event->mask & FAN_ALL_PERM_EVENTS)
+		return 0;
+#endif
+
+	list_for_each_entry_reverse(test_event, list, list) {
+		if (should_merge(test_event, event)) {
+			do_merge = true;
+			break;
+		}
+	}
+
+	if (!do_merge)
+		return 0;
+
+	test_event->mask |= event->mask;
+	return 1;
+}
+
+#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
+static int fanotify_get_response(struct fsnotify_group *group,
+				 struct fanotify_perm_event_info *event)
+{
+	int ret;
+
+	pr_debug("%s: group=%p event=%p\n", __func__, group, event);
+
+	wait_event(group->fanotify_data.access_waitq, event->response ||
+				atomic_read(&group->fanotify_data.bypass_perm));
+
+	if (!event->response) /* bypass_perm set */
+		return 0;
+
+	/* userspace responded, convert to something usable */
+	switch (event->response) {
+	case FAN_ALLOW:
+		ret = 0;
+		break;
+	case FAN_DENY:
+	default:
+		ret = -EPERM;
+	}
+	event->response = 0;
+
+	pr_debug("%s: group=%p event=%p about to return ret=%d\n", __func__,
+		 group, event, ret);
+	
+	return ret;
+}
+#endif
+
+static bool fanotify_should_send_event(struct fsnotify_mark *inode_mark,
+				       struct fsnotify_mark *vfsmnt_mark,
+				       u32 event_mask,
+				       void *data, int data_type)
+{
+	__u32 marks_mask, marks_ignored_mask;
+	struct path *path = data;
+
+	pr_debug("%s: inode_mark=%p vfsmnt_mark=%p mask=%x data=%p"
+		 " data_type=%d\n", __func__, inode_mark, vfsmnt_mark,
+		 event_mask, data, data_type);
+
+	/* if we don't have enough info to send an event to userspace say no */
+	if (data_type != FSNOTIFY_EVENT_PATH)
+		return false;
+
+	/* sorry, fanotify only gives a damn about files and dirs */
+	if (!S_ISREG(path->dentry->d_inode->i_mode) &&
+	    !S_ISDIR(path->dentry->d_inode->i_mode))
+		return false;
+
+	if (inode_mark && vfsmnt_mark) {
+		marks_mask = (vfsmnt_mark->mask | inode_mark->mask);
+		marks_ignored_mask = (vfsmnt_mark->ignored_mask | inode_mark->ignored_mask);
+	} else if (inode_mark) {
+		/*
+		 * if the event is for a child and this inode doesn't care about
+		 * events on the child, don't send it!
+		 */
+		if ((event_mask & FS_EVENT_ON_CHILD) &&
+		    !(inode_mark->mask & FS_EVENT_ON_CHILD))
+			return false;
+		marks_mask = inode_mark->mask;
+		marks_ignored_mask = inode_mark->ignored_mask;
+	} else if (vfsmnt_mark) {
+		marks_mask = vfsmnt_mark->mask;
+		marks_ignored_mask = vfsmnt_mark->ignored_mask;
+	} else {
+		BUG();
+	}
+
+	if (S_ISDIR(path->dentry->d_inode->i_mode) &&
+	    (marks_ignored_mask & FS_ISDIR))
+		return false;
+
+	if (event_mask & marks_mask & ~marks_ignored_mask)
+		return true;
+
+	return false;
+}
+
+struct fanotify_event_info *fanotify_alloc_event(struct inode *inode, u32 mask,
+						 struct path *path)
+{
+	struct fanotify_event_info *event;
+
+#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
+	if (mask & FAN_ALL_PERM_EVENTS) {
+		struct fanotify_perm_event_info *pevent;
+
+		pevent = kmem_cache_alloc(fanotify_perm_event_cachep,
+					  GFP_KERNEL);
+		if (!pevent)
+			return NULL;
+		event = &pevent->fae;
+		pevent->response = 0;
+		goto init;
+	}
+#endif
+	event = kmem_cache_alloc(fanotify_event_cachep, GFP_KERNEL);
+	if (!event)
+		return NULL;
+init: __maybe_unused
+	fsnotify_init_event(&event->fse, inode, mask);
+	event->tgid = get_pid(task_tgid(current));
+	if (path) {
+		event->path = *path;
+		path_get(&event->path);
+	} else {
+		event->path.mnt = NULL;
+		event->path.dentry = NULL;
+	}
+	return event;
+}
+
+static int fanotify_handle_event(struct fsnotify_group *group,
+				 struct inode *inode,
+				 struct fsnotify_mark *inode_mark,
+				 struct fsnotify_mark *fanotify_mark,
+				 u32 mask, void *data, int data_type,
+				 const unsigned char *file_name, u32 cookie)
+{
+	int ret = 0;
+	struct fanotify_event_info *event;
+	struct fsnotify_event *fsn_event;
+
+	BUILD_BUG_ON(FAN_ACCESS != FS_ACCESS);
+	BUILD_BUG_ON(FAN_MODIFY != FS_MODIFY);
+	BUILD_BUG_ON(FAN_CLOSE_NOWRITE != FS_CLOSE_NOWRITE);
+	BUILD_BUG_ON(FAN_CLOSE_WRITE != FS_CLOSE_WRITE);
+	BUILD_BUG_ON(FAN_OPEN != FS_OPEN);
+	BUILD_BUG_ON(FAN_EVENT_ON_CHILD != FS_EVENT_ON_CHILD);
+	BUILD_BUG_ON(FAN_Q_OVERFLOW != FS_Q_OVERFLOW);
+	BUILD_BUG_ON(FAN_OPEN_PERM != FS_OPEN_PERM);
+	BUILD_BUG_ON(FAN_ACCESS_PERM != FS_ACCESS_PERM);
+	BUILD_BUG_ON(FAN_ONDIR != FS_ISDIR);
+
+	if (!fanotify_should_send_event(inode_mark, fanotify_mark, mask, data,
+					data_type))
+		return 0;
+
+	pr_debug("%s: group=%p inode=%p mask=%x\n", __func__, group, inode,
+		 mask);
+
+	event = fanotify_alloc_event(inode, mask, data);
+	if (unlikely(!event))
+		return -ENOMEM;
+
+	fsn_event = &event->fse;
+	ret = fsnotify_add_notify_event(group, fsn_event, fanotify_merge);
+	if (ret) {
+		/* Permission events shouldn't be merged */
+		BUG_ON(ret == 1 && mask & FAN_ALL_PERM_EVENTS);
+		/* Our event wasn't used in the end. Free it. */
+		fsnotify_destroy_event(group, fsn_event);
+
+		return 0;
+	}
+
+#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
+	if (mask & FAN_ALL_PERM_EVENTS) {
+		ret = fanotify_get_response(group, FANOTIFY_PE(fsn_event));
+		fsnotify_destroy_event(group, fsn_event);
+	}
+#endif
+	return ret;
+}
+
+static void fanotify_free_group_priv(struct fsnotify_group *group)
+{
+	struct user_struct *user;
+
+	user = group->fanotify_data.user;
+	atomic_dec(&user->fanotify_listeners);
+	free_uid(user);
+}
+
+static void fanotify_free_event(struct fsnotify_event *fsn_event)
+{
+	struct fanotify_event_info *event;
+
+	event = FANOTIFY_E(fsn_event);
+	path_put(&event->path);
+	put_pid(event->tgid);
+#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
+	if (fsn_event->mask & FAN_ALL_PERM_EVENTS) {
+		kmem_cache_free(fanotify_perm_event_cachep,
+				FANOTIFY_PE(fsn_event));
+		return;
+	}
+#endif
+	kmem_cache_free(fanotify_event_cachep, event);
+}
+
+const struct fsnotify_ops fanotify_fsnotify_ops = {
+	.handle_event = fanotify_handle_event,
+	.free_group_priv = fanotify_free_group_priv,
+	.free_event = fanotify_free_event,
+};
diff --git a/fs/notify/fanotify/fanotify.h b/fs/notify/fanotify/fanotify.h
new file mode 100644
index 00000000000..2a5fb14115d
--- /dev/null
+++ b/fs/notify/fanotify/fanotify.h
@@ -0,0 +1,50 @@
+#include <linux/fsnotify_backend.h>
+#include <linux/path.h>
+#include <linux/slab.h>
+
+extern struct kmem_cache *fanotify_event_cachep;
+extern struct kmem_cache *fanotify_perm_event_cachep;
+
+/*
+ * Structure for normal fanotify events. It gets allocated in
+ * fanotify_handle_event() and freed when the information is retrieved by
+ * userspace
+ */
+struct fanotify_event_info {
+	struct fsnotify_event fse;
+	/*
+	 * We hold ref to this path so it may be dereferenced at any point
+	 * during this object's lifetime
+	 */
+	struct path path;
+	struct pid *tgid;
+};
+
+#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
+/*
+ * Structure for permission fanotify events. It gets allocated and freed in
+ * fanotify_handle_event() since we wait there for user response. When the
+ * information is retrieved by userspace the structure is moved from
+ * group->notification_list to group->fanotify_data.access_list to wait for
+ * user response.
+ */
+struct fanotify_perm_event_info {
+	struct fanotify_event_info fae;
+	int response;	/* userspace answer to question */
+	int fd;		/* fd we passed to userspace for this event */
+};
+
+static inline struct fanotify_perm_event_info *
+FANOTIFY_PE(struct fsnotify_event *fse)
+{
+	return container_of(fse, struct fanotify_perm_event_info, fae.fse);
+}
+#endif
+
+static inline struct fanotify_event_info *FANOTIFY_E(struct fsnotify_event *fse)
+{
+	return container_of(fse, struct fanotify_event_info, fse);
+}
+
+struct fanotify_event_info *fanotify_alloc_event(struct inode *inode, u32 mask,
+						 struct path *path);
diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c
new file mode 100644
index 00000000000..3fdc8a3e113
--- /dev/null
+++ b/fs/notify/fanotify/fanotify_user.c
@@ -0,0 +1,919 @@
+#include <linux/fanotify.h>
+#include <linux/fcntl.h>
+#include <linux/file.h>
+#include <linux/fs.h>
+#include <linux/anon_inodes.h>
+#include <linux/fsnotify_backend.h>
+#include <linux/init.h>
+#include <linux/mount.h>
+#include <linux/namei.h>
+#include <linux/poll.h>
+#include <linux/security.h>
+#include <linux/syscalls.h>
+#include <linux/slab.h>
+#include <linux/types.h>
+#include <linux/uaccess.h>
+#include <linux/compat.h>
+
+#include <asm/ioctls.h>
+
+#include "../../mount.h"
+#include "../fdinfo.h"
+#include "fanotify.h"
+
+#define FANOTIFY_DEFAULT_MAX_EVENTS	16384
+#define FANOTIFY_DEFAULT_MAX_MARKS	8192
+#define FANOTIFY_DEFAULT_MAX_LISTENERS	128
+
+/*
+ * All flags that may be specified in parameter event_f_flags of fanotify_init.
+ *
+ * Internal and external open flags are stored together in field f_flags of
+ * struct file. Only external open flags shall be allowed in event_f_flags.
+ * Internal flags like FMODE_NONOTIFY, FMODE_EXEC, FMODE_NOCMTIME shall be
+ * excluded.
+ */
+#define	FANOTIFY_INIT_ALL_EVENT_F_BITS				( \
+		O_ACCMODE	| O_APPEND	| O_NONBLOCK	| \
+		__O_SYNC	| O_DSYNC	| O_CLOEXEC     | \
+		O_LARGEFILE	| O_NOATIME	)
+
+extern const struct fsnotify_ops fanotify_fsnotify_ops;
+
+static struct kmem_cache *fanotify_mark_cache __read_mostly;
+struct kmem_cache *fanotify_event_cachep __read_mostly;
+struct kmem_cache *fanotify_perm_event_cachep __read_mostly;
+
+/*
+ * Get an fsnotify notification event if one exists and is small
+ * enough to fit in "count". Return an error pointer if the count
+ * is not large enough.
+ *
+ * Called with the group->notification_mutex held.
+ */
+static struct fsnotify_event *get_one_event(struct fsnotify_group *group,
+					    size_t count)
+{
+	BUG_ON(!mutex_is_locked(&group->notification_mutex));
+
+	pr_debug("%s: group=%p count=%zd\n", __func__, group, count);
+
+	if (fsnotify_notify_queue_is_empty(group))
+		return NULL;
+
+	if (FAN_EVENT_METADATA_LEN > count)
+		return ERR_PTR(-EINVAL);
+
+	/* held the notification_mutex the whole time, so this is the
+	 * same event we peeked above */
+	return fsnotify_remove_notify_event(group);
+}
+
+static int create_fd(struct fsnotify_group *group,
+		     struct fanotify_event_info *event,
+		     struct file **file)
+{
+	int client_fd;
+	struct file *new_file;
+
+	pr_debug("%s: group=%p event=%p\n", __func__, group, event);
+
+	client_fd = get_unused_fd();
+	if (client_fd < 0)
+		return client_fd;
+
+	/*
+	 * we need a new file handle for the userspace program so it can read even if it was
+	 * originally opened O_WRONLY.
+	 */
+	/* it's possible this event was an overflow event.  in that case dentry and mnt
+	 * are NULL;  That's fine, just don't call dentry open */
+	if (event->path.dentry && event->path.mnt)
+		new_file = dentry_open(&event->path,
+				       group->fanotify_data.f_flags | FMODE_NONOTIFY,
+				       current_cred());
+	else
+		new_file = ERR_PTR(-EOVERFLOW);
+	if (IS_ERR(new_file)) {
+		/*
+		 * we still send an event even if we can't open the file.  this
+		 * can happen when say tasks are gone and we try to open their
+		 * /proc files or we try to open a WRONLY file like in sysfs
+		 * we just send the errno to userspace since there isn't much
+		 * else we can do.
+		 */
+		put_unused_fd(client_fd);
+		client_fd = PTR_ERR(new_file);
+	} else {
+		*file = new_file;
+	}
+
+	return client_fd;
+}
+
+static int fill_event_metadata(struct fsnotify_group *group,
+			       struct fanotify_event_metadata *metadata,
+			       struct fsnotify_event *fsn_event,
+			       struct file **file)
+{
+	int ret = 0;
+	struct fanotify_event_info *event;
+
+	pr_debug("%s: group=%p metadata=%p event=%p\n", __func__,
+		 group, metadata, fsn_event);
+
+	*file = NULL;
+	event = container_of(fsn_event, struct fanotify_event_info, fse);
+	metadata->event_len = FAN_EVENT_METADATA_LEN;
+	metadata->metadata_len = FAN_EVENT_METADATA_LEN;
+	metadata->vers = FANOTIFY_METADATA_VERSION;
+	metadata->reserved = 0;
+	metadata->mask = fsn_event->mask & FAN_ALL_OUTGOING_EVENTS;
+	metadata->pid = pid_vnr(event->tgid);
+	if (unlikely(fsn_event->mask & FAN_Q_OVERFLOW))
+		metadata->fd = FAN_NOFD;
+	else {
+		metadata->fd = create_fd(group, event, file);
+		if (metadata->fd < 0)
+			ret = metadata->fd;
+	}
+
+	return ret;
+}
+
+#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
+static struct fanotify_perm_event_info *dequeue_event(
+				struct fsnotify_group *group, int fd)
+{
+	struct fanotify_perm_event_info *event, *return_e = NULL;
+
+	spin_lock(&group->fanotify_data.access_lock);
+	list_for_each_entry(event, &group->fanotify_data.access_list,
+			    fae.fse.list) {
+		if (event->fd != fd)
+			continue;
+
+		list_del_init(&event->fae.fse.list);
+		return_e = event;
+		break;
+	}
+	spin_unlock(&group->fanotify_data.access_lock);
+
+	pr_debug("%s: found return_re=%p\n", __func__, return_e);
+
+	return return_e;
+}
+
+static int process_access_response(struct fsnotify_group *group,
+				   struct fanotify_response *response_struct)
+{
+	struct fanotify_perm_event_info *event;
+	int fd = response_struct->fd;
+	int response = response_struct->response;
+
+	pr_debug("%s: group=%p fd=%d response=%d\n", __func__, group,
+		 fd, response);
+	/*
+	 * make sure the response is valid, if invalid we do nothing and either
+	 * userspace can send a valid response or we will clean it up after the
+	 * timeout
+	 */
+	switch (response) {
+	case FAN_ALLOW:
+	case FAN_DENY:
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	if (fd < 0)
+		return -EINVAL;
+
+	event = dequeue_event(group, fd);
+	if (!event)
+		return -ENOENT;
+
+	event->response = response;
+	wake_up(&group->fanotify_data.access_waitq);
+
+	return 0;
+}
+#endif
+
+static ssize_t copy_event_to_user(struct fsnotify_group *group,
+				  struct fsnotify_event *event,
+				  char __user *buf)
+{
+	struct fanotify_event_metadata fanotify_event_metadata;
+	struct file *f;
+	int fd, ret;
+
+	pr_debug("%s: group=%p event=%p\n", __func__, group, event);
+
+	ret = fill_event_metadata(group, &fanotify_event_metadata, event, &f);
+	if (ret < 0)
+		return ret;
+
+	fd = fanotify_event_metadata.fd;
+	ret = -EFAULT;
+	if (copy_to_user(buf, &fanotify_event_metadata,
+			 fanotify_event_metadata.event_len))
+		goto out_close_fd;
+
+#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
+	if (event->mask & FAN_ALL_PERM_EVENTS)
+		FANOTIFY_PE(event)->fd = fd;
+#endif
+
+	if (fd != FAN_NOFD)
+		fd_install(fd, f);
+	return fanotify_event_metadata.event_len;
+
+out_close_fd:
+	if (fd != FAN_NOFD) {
+		put_unused_fd(fd);
+		fput(f);
+	}
+	return ret;
+}
+
+/* intofiy userspace file descriptor functions */
+static unsigned int fanotify_poll(struct file *file, poll_table *wait)
+{
+	struct fsnotify_group *group = file->private_data;
+	int ret = 0;
+
+	poll_wait(file, &group->notification_waitq, wait);
+	mutex_lock(&group->notification_mutex);
+	if (!fsnotify_notify_queue_is_empty(group))
+		ret = POLLIN | POLLRDNORM;
+	mutex_unlock(&group->notification_mutex);
+
+	return ret;
+}
+
+static ssize_t fanotify_read(struct file *file, char __user *buf,
+			     size_t count, loff_t *pos)
+{
+	struct fsnotify_group *group;
+	struct fsnotify_event *kevent;
+	char __user *start;
+	int ret;
+	DEFINE_WAIT(wait);
+
+	start = buf;
+	group = file->private_data;
+
+	pr_debug("%s: group=%p\n", __func__, group);
+
+	while (1) {
+		prepare_to_wait(&group->notification_waitq, &wait, TASK_INTERRUPTIBLE);
+
+		mutex_lock(&group->notification_mutex);
+		kevent = get_one_event(group, count);
+		mutex_unlock(&group->notification_mutex);
+
+		if (IS_ERR(kevent)) {
+			ret = PTR_ERR(kevent);
+			break;
+		}
+
+		if (!kevent) {
+			ret = -EAGAIN;
+			if (file->f_flags & O_NONBLOCK)
+				break;
+
+			ret = -ERESTARTSYS;
+			if (signal_pending(current))
+				break;
+
+			if (start != buf)
+				break;
+			schedule();
+			continue;
+		}
+
+		ret = copy_event_to_user(group, kevent, buf);
+		/*
+		 * Permission events get queued to wait for response.  Other
+		 * events can be destroyed now.
+		 */
+		if (!(kevent->mask & FAN_ALL_PERM_EVENTS)) {
+			fsnotify_destroy_event(group, kevent);
+			if (ret < 0)
+				break;
+		} else {
+#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
+			if (ret < 0) {
+				FANOTIFY_PE(kevent)->response = FAN_DENY;
+				wake_up(&group->fanotify_data.access_waitq);
+				break;
+			}
+			spin_lock(&group->fanotify_data.access_lock);
+			list_add_tail(&kevent->list,
+				      &group->fanotify_data.access_list);
+			spin_unlock(&group->fanotify_data.access_lock);
+#endif
+		}
+		buf += ret;
+		count -= ret;
+	}
+
+	finish_wait(&group->notification_waitq, &wait);
+	if (start != buf && ret != -EFAULT)
+		ret = buf - start;
+	return ret;
+}
+
+static ssize_t fanotify_write(struct file *file, const char __user *buf, size_t count, loff_t *pos)
+{
+#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
+	struct fanotify_response response = { .fd = -1, .response = -1 };
+	struct fsnotify_group *group;
+	int ret;
+
+	group = file->private_data;
+
+	if (count > sizeof(response))
+		count = sizeof(response);
+
+	pr_debug("%s: group=%p count=%zu\n", __func__, group, count);
+
+	if (copy_from_user(&response, buf, count))
+		return -EFAULT;
+
+	ret = process_access_response(group, &response);
+	if (ret < 0)
+		count = ret;
+
+	return count;
+#else
+	return -EINVAL;
+#endif
+}
+
+static int fanotify_release(struct inode *ignored, struct file *file)
+{
+	struct fsnotify_group *group = file->private_data;
+
+#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
+	struct fanotify_perm_event_info *event, *next;
+
+	spin_lock(&group->fanotify_data.access_lock);
+
+	atomic_inc(&group->fanotify_data.bypass_perm);
+
+	list_for_each_entry_safe(event, next, &group->fanotify_data.access_list,
+				 fae.fse.list) {
+		pr_debug("%s: found group=%p event=%p\n", __func__, group,
+			 event);
+
+		list_del_init(&event->fae.fse.list);
+		event->response = FAN_ALLOW;
+	}
+	spin_unlock(&group->fanotify_data.access_lock);
+
+	wake_up(&group->fanotify_data.access_waitq);
+#endif
+
+	/* matches the fanotify_init->fsnotify_alloc_group */
+	fsnotify_destroy_group(group);
+
+	return 0;
+}
+
+static long fanotify_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+	struct fsnotify_group *group;
+	struct fsnotify_event *fsn_event;
+	void __user *p;
+	int ret = -ENOTTY;
+	size_t send_len = 0;
+
+	group = file->private_data;
+
+	p = (void __user *) arg;
+
+	switch (cmd) {
+	case FIONREAD:
+		mutex_lock(&group->notification_mutex);
+		list_for_each_entry(fsn_event, &group->notification_list, list)
+			send_len += FAN_EVENT_METADATA_LEN;
+		mutex_unlock(&group->notification_mutex);
+		ret = put_user(send_len, (int __user *) p);
+		break;
+	}
+
+	return ret;
+}
+
+static const struct file_operations fanotify_fops = {
+	.show_fdinfo	= fanotify_show_fdinfo,
+	.poll		= fanotify_poll,
+	.read		= fanotify_read,
+	.write		= fanotify_write,
+	.fasync		= NULL,
+	.release	= fanotify_release,
+	.unlocked_ioctl	= fanotify_ioctl,
+	.compat_ioctl	= fanotify_ioctl,
+	.llseek		= noop_llseek,
+};
+
+static void fanotify_free_mark(struct fsnotify_mark *fsn_mark)
+{
+	kmem_cache_free(fanotify_mark_cache, fsn_mark);
+}
+
+static int fanotify_find_path(int dfd, const char __user *filename,
+			      struct path *path, unsigned int flags)
+{
+	int ret;
+
+	pr_debug("%s: dfd=%d filename=%p flags=%x\n", __func__,
+		 dfd, filename, flags);
+
+	if (filename == NULL) {
+		struct fd f = fdget(dfd);
+
+		ret = -EBADF;
+		if (!f.file)
+			goto out;
+
+		ret = -ENOTDIR;
+		if ((flags & FAN_MARK_ONLYDIR) &&
+		    !(S_ISDIR(file_inode(f.file)->i_mode))) {
+			fdput(f);
+			goto out;
+		}
+
+		*path = f.file->f_path;
+		path_get(path);
+		fdput(f);
+	} else {
+		unsigned int lookup_flags = 0;
+
+		if (!(flags & FAN_MARK_DONT_FOLLOW))
+			lookup_flags |= LOOKUP_FOLLOW;
+		if (flags & FAN_MARK_ONLYDIR)
+			lookup_flags |= LOOKUP_DIRECTORY;
+
+		ret = user_path_at(dfd, filename, lookup_flags, path);
+		if (ret)
+			goto out;
+	}
+
+	/* you can only watch an inode if you have read permissions on it */
+	ret = inode_permission(path->dentry->d_inode, MAY_READ);
+	if (ret)
+		path_put(path);
+out:
+	return ret;
+}
+
+static __u32 fanotify_mark_remove_from_mask(struct fsnotify_mark *fsn_mark,
+					    __u32 mask,
+					    unsigned int flags,
+					    int *destroy)
+{
+	__u32 oldmask;
+
+	spin_lock(&fsn_mark->lock);
+	if (!(flags & FAN_MARK_IGNORED_MASK)) {
+		oldmask = fsn_mark->mask;
+		fsnotify_set_mark_mask_locked(fsn_mark, (oldmask & ~mask));
+	} else {
+		oldmask = fsn_mark->ignored_mask;
+		fsnotify_set_mark_ignored_mask_locked(fsn_mark, (oldmask & ~mask));
+	}
+	spin_unlock(&fsn_mark->lock);
+
+	*destroy = !(oldmask & ~mask);
+
+	return mask & oldmask;
+}
+
+static int fanotify_remove_vfsmount_mark(struct fsnotify_group *group,
+					 struct vfsmount *mnt, __u32 mask,
+					 unsigned int flags)
+{
+	struct fsnotify_mark *fsn_mark = NULL;
+	__u32 removed;
+	int destroy_mark;
+
+	mutex_lock(&group->mark_mutex);
+	fsn_mark = fsnotify_find_vfsmount_mark(group, mnt);
+	if (!fsn_mark) {
+		mutex_unlock(&group->mark_mutex);
+		return -ENOENT;
+	}
+
+	removed = fanotify_mark_remove_from_mask(fsn_mark, mask, flags,
+						 &destroy_mark);
+	if (destroy_mark)
+		fsnotify_destroy_mark_locked(fsn_mark, group);
+	mutex_unlock(&group->mark_mutex);
+
+	fsnotify_put_mark(fsn_mark);
+	if (removed & real_mount(mnt)->mnt_fsnotify_mask)
+		fsnotify_recalc_vfsmount_mask(mnt);
+
+	return 0;
+}
+
+static int fanotify_remove_inode_mark(struct fsnotify_group *group,
+				      struct inode *inode, __u32 mask,
+				      unsigned int flags)
+{
+	struct fsnotify_mark *fsn_mark = NULL;
+	__u32 removed;
+	int destroy_mark;
+
+	mutex_lock(&group->mark_mutex);
+	fsn_mark = fsnotify_find_inode_mark(group, inode);
+	if (!fsn_mark) {
+		mutex_unlock(&group->mark_mutex);
+		return -ENOENT;
+	}
+
+	removed = fanotify_mark_remove_from_mask(fsn_mark, mask, flags,
+						 &destroy_mark);
+	if (destroy_mark)
+		fsnotify_destroy_mark_locked(fsn_mark, group);
+	mutex_unlock(&group->mark_mutex);
+
+	/* matches the fsnotify_find_inode_mark() */
+	fsnotify_put_mark(fsn_mark);
+	if (removed & inode->i_fsnotify_mask)
+		fsnotify_recalc_inode_mask(inode);
+
+	return 0;
+}
+
+static __u32 fanotify_mark_add_to_mask(struct fsnotify_mark *fsn_mark,
+				       __u32 mask,
+				       unsigned int flags)
+{
+	__u32 oldmask = -1;
+
+	spin_lock(&fsn_mark->lock);
+	if (!(flags & FAN_MARK_IGNORED_MASK)) {
+		oldmask = fsn_mark->mask;
+		fsnotify_set_mark_mask_locked(fsn_mark, (oldmask | mask));
+	} else {
+		__u32 tmask = fsn_mark->ignored_mask | mask;
+		fsnotify_set_mark_ignored_mask_locked(fsn_mark, tmask);
+		if (flags & FAN_MARK_IGNORED_SURV_MODIFY)
+			fsn_mark->flags |= FSNOTIFY_MARK_FLAG_IGNORED_SURV_MODIFY;
+	}
+
+	if (!(flags & FAN_MARK_ONDIR)) {
+		__u32 tmask = fsn_mark->ignored_mask | FAN_ONDIR;
+		fsnotify_set_mark_ignored_mask_locked(fsn_mark, tmask);
+	}
+
+	spin_unlock(&fsn_mark->lock);
+
+	return mask & ~oldmask;
+}
+
+static struct fsnotify_mark *fanotify_add_new_mark(struct fsnotify_group *group,
+						   struct inode *inode,
+						   struct vfsmount *mnt)
+{
+	struct fsnotify_mark *mark;
+	int ret;
+
+	if (atomic_read(&group->num_marks) > group->fanotify_data.max_marks)
+		return ERR_PTR(-ENOSPC);
+
+	mark = kmem_cache_alloc(fanotify_mark_cache, GFP_KERNEL);
+	if (!mark)
+		return ERR_PTR(-ENOMEM);
+
+	fsnotify_init_mark(mark, fanotify_free_mark);
+	ret = fsnotify_add_mark_locked(mark, group, inode, mnt, 0);
+	if (ret) {
+		fsnotify_put_mark(mark);
+		return ERR_PTR(ret);
+	}
+
+	return mark;
+}
+
+
+static int fanotify_add_vfsmount_mark(struct fsnotify_group *group,
+				      struct vfsmount *mnt, __u32 mask,
+				      unsigned int flags)
+{
+	struct fsnotify_mark *fsn_mark;
+	__u32 added;
+
+	mutex_lock(&group->mark_mutex);
+	fsn_mark = fsnotify_find_vfsmount_mark(group, mnt);
+	if (!fsn_mark) {
+		fsn_mark = fanotify_add_new_mark(group, NULL, mnt);
+		if (IS_ERR(fsn_mark)) {
+			mutex_unlock(&group->mark_mutex);
+			return PTR_ERR(fsn_mark);
+		}
+	}
+	added = fanotify_mark_add_to_mask(fsn_mark, mask, flags);
+	mutex_unlock(&group->mark_mutex);
+
+	if (added & ~real_mount(mnt)->mnt_fsnotify_mask)
+		fsnotify_recalc_vfsmount_mask(mnt);
+
+	fsnotify_put_mark(fsn_mark);
+	return 0;
+}
+
+static int fanotify_add_inode_mark(struct fsnotify_group *group,
+				   struct inode *inode, __u32 mask,
+				   unsigned int flags)
+{
+	struct fsnotify_mark *fsn_mark;
+	__u32 added;
+
+	pr_debug("%s: group=%p inode=%p\n", __func__, group, inode);
+
+	/*
+	 * If some other task has this inode open for write we should not add
+	 * an ignored mark, unless that ignored mark is supposed to survive
+	 * modification changes anyway.
+	 */
+	if ((flags & FAN_MARK_IGNORED_MASK) &&
+	    !(flags & FAN_MARK_IGNORED_SURV_MODIFY) &&
+	    (atomic_read(&inode->i_writecount) > 0))
+		return 0;
+
+	mutex_lock(&group->mark_mutex);
+	fsn_mark = fsnotify_find_inode_mark(group, inode);
+	if (!fsn_mark) {
+		fsn_mark = fanotify_add_new_mark(group, inode, NULL);
+		if (IS_ERR(fsn_mark)) {
+			mutex_unlock(&group->mark_mutex);
+			return PTR_ERR(fsn_mark);
+		}
+	}
+	added = fanotify_mark_add_to_mask(fsn_mark, mask, flags);
+	mutex_unlock(&group->mark_mutex);
+
+	if (added & ~inode->i_fsnotify_mask)
+		fsnotify_recalc_inode_mask(inode);
+
+	fsnotify_put_mark(fsn_mark);
+	return 0;
+}
+
+/* fanotify syscalls */
+SYSCALL_DEFINE2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags)
+{
+	struct fsnotify_group *group;
+	int f_flags, fd;
+	struct user_struct *user;
+	struct fanotify_event_info *oevent;
+
+	pr_debug("%s: flags=%d event_f_flags=%d\n",
+		__func__, flags, event_f_flags);
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	if (flags & ~FAN_ALL_INIT_FLAGS)
+		return -EINVAL;
+
+	if (event_f_flags & ~FANOTIFY_INIT_ALL_EVENT_F_BITS)
+		return -EINVAL;
+
+	switch (event_f_flags & O_ACCMODE) {
+	case O_RDONLY:
+	case O_RDWR:
+	case O_WRONLY:
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	user = get_current_user();
+	if (atomic_read(&user->fanotify_listeners) > FANOTIFY_DEFAULT_MAX_LISTENERS) {
+		free_uid(user);
+		return -EMFILE;
+	}
+
+	f_flags = O_RDWR | FMODE_NONOTIFY;
+	if (flags & FAN_CLOEXEC)
+		f_flags |= O_CLOEXEC;
+	if (flags & FAN_NONBLOCK)
+		f_flags |= O_NONBLOCK;
+
+	/* fsnotify_alloc_group takes a ref.  Dropped in fanotify_release */
+	group = fsnotify_alloc_group(&fanotify_fsnotify_ops);
+	if (IS_ERR(group)) {
+		free_uid(user);
+		return PTR_ERR(group);
+	}
+
+	group->fanotify_data.user = user;
+	atomic_inc(&user->fanotify_listeners);
+
+	oevent = fanotify_alloc_event(NULL, FS_Q_OVERFLOW, NULL);
+	if (unlikely(!oevent)) {
+		fd = -ENOMEM;
+		goto out_destroy_group;
+	}
+	group->overflow_event = &oevent->fse;
+
+	if (force_o_largefile())
+		event_f_flags |= O_LARGEFILE;
+	group->fanotify_data.f_flags = event_f_flags;
+#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
+	spin_lock_init(&group->fanotify_data.access_lock);
+	init_waitqueue_head(&group->fanotify_data.access_waitq);
+	INIT_LIST_HEAD(&group->fanotify_data.access_list);
+	atomic_set(&group->fanotify_data.bypass_perm, 0);
+#endif
+	switch (flags & FAN_ALL_CLASS_BITS) {
+	case FAN_CLASS_NOTIF:
+		group->priority = FS_PRIO_0;
+		break;
+	case FAN_CLASS_CONTENT:
+		group->priority = FS_PRIO_1;
+		break;
+	case FAN_CLASS_PRE_CONTENT:
+		group->priority = FS_PRIO_2;
+		break;
+	default:
+		fd = -EINVAL;
+		goto out_destroy_group;
+	}
+
+	if (flags & FAN_UNLIMITED_QUEUE) {
+		fd = -EPERM;
+		if (!capable(CAP_SYS_ADMIN))
+			goto out_destroy_group;
+		group->max_events = UINT_MAX;
+	} else {
+		group->max_events = FANOTIFY_DEFAULT_MAX_EVENTS;
+	}
+
+	if (flags & FAN_UNLIMITED_MARKS) {
+		fd = -EPERM;
+		if (!capable(CAP_SYS_ADMIN))
+			goto out_destroy_group;
+		group->fanotify_data.max_marks = UINT_MAX;
+	} else {
+		group->fanotify_data.max_marks = FANOTIFY_DEFAULT_MAX_MARKS;
+	}
+
+	fd = anon_inode_getfd("[fanotify]", &fanotify_fops, group, f_flags);
+	if (fd < 0)
+		goto out_destroy_group;
+
+	return fd;
+
+out_destroy_group:
+	fsnotify_destroy_group(group);
+	return fd;
+}
+
+SYSCALL_DEFINE5(fanotify_mark, int, fanotify_fd, unsigned int, flags,
+			      __u64, mask, int, dfd,
+			      const char  __user *, pathname)
+{
+	struct inode *inode = NULL;
+	struct vfsmount *mnt = NULL;
+	struct fsnotify_group *group;
+	struct fd f;
+	struct path path;
+	int ret;
+
+	pr_debug("%s: fanotify_fd=%d flags=%x dfd=%d pathname=%p mask=%llx\n",
+		 __func__, fanotify_fd, flags, dfd, pathname, mask);
+
+	/* we only use the lower 32 bits as of right now. */
+	if (mask & ((__u64)0xffffffff << 32))
+		return -EINVAL;
+
+	if (flags & ~FAN_ALL_MARK_FLAGS)
+		return -EINVAL;
+	switch (flags & (FAN_MARK_ADD | FAN_MARK_REMOVE | FAN_MARK_FLUSH)) {
+	case FAN_MARK_ADD:		/* fallthrough */
+	case FAN_MARK_REMOVE:
+		if (!mask)
+			return -EINVAL;
+		break;
+	case FAN_MARK_FLUSH:
+		if (flags & ~(FAN_MARK_MOUNT | FAN_MARK_FLUSH))
+			return -EINVAL;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	if (mask & FAN_ONDIR) {
+		flags |= FAN_MARK_ONDIR;
+		mask &= ~FAN_ONDIR;
+	}
+
+#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
+	if (mask & ~(FAN_ALL_EVENTS | FAN_ALL_PERM_EVENTS | FAN_EVENT_ON_CHILD))
+#else
+	if (mask & ~(FAN_ALL_EVENTS | FAN_EVENT_ON_CHILD))
+#endif
+		return -EINVAL;
+
+	f = fdget(fanotify_fd);
+	if (unlikely(!f.file))
+		return -EBADF;
+
+	/* verify that this is indeed an fanotify instance */
+	ret = -EINVAL;
+	if (unlikely(f.file->f_op != &fanotify_fops))
+		goto fput_and_out;
+	group = f.file->private_data;
+
+	/*
+	 * group->priority == FS_PRIO_0 == FAN_CLASS_NOTIF.  These are not
+	 * allowed to set permissions events.
+	 */
+	ret = -EINVAL;
+	if (mask & FAN_ALL_PERM_EVENTS &&
+	    group->priority == FS_PRIO_0)
+		goto fput_and_out;
+
+	if (flags & FAN_MARK_FLUSH) {
+		ret = 0;
+		if (flags & FAN_MARK_MOUNT)
+			fsnotify_clear_vfsmount_marks_by_group(group);
+		else
+			fsnotify_clear_inode_marks_by_group(group);
+		goto fput_and_out;
+	}
+
+	ret = fanotify_find_path(dfd, pathname, &path, flags);
+	if (ret)
+		goto fput_and_out;
+
+	/* inode held in place by reference to path; group by fget on fd */
+	if (!(flags & FAN_MARK_MOUNT))
+		inode = path.dentry->d_inode;
+	else
+		mnt = path.mnt;
+
+	/* create/update an inode mark */
+	switch (flags & (FAN_MARK_ADD | FAN_MARK_REMOVE)) {
+	case FAN_MARK_ADD:
+		if (flags & FAN_MARK_MOUNT)
+			ret = fanotify_add_vfsmount_mark(group, mnt, mask, flags);
+		else
+			ret = fanotify_add_inode_mark(group, inode, mask, flags);
+		break;
+	case FAN_MARK_REMOVE:
+		if (flags & FAN_MARK_MOUNT)
+			ret = fanotify_remove_vfsmount_mark(group, mnt, mask, flags);
+		else
+			ret = fanotify_remove_inode_mark(group, inode, mask, flags);
+		break;
+	default:
+		ret = -EINVAL;
+	}
+
+	path_put(&path);
+fput_and_out:
+	fdput(f);
+	return ret;
+}
+
+#ifdef CONFIG_COMPAT
+COMPAT_SYSCALL_DEFINE6(fanotify_mark,
+				int, fanotify_fd, unsigned int, flags,
+				__u32, mask0, __u32, mask1, int, dfd,
+				const char  __user *, pathname)
+{
+	return sys_fanotify_mark(fanotify_fd, flags,
+#ifdef __BIG_ENDIAN
+				((__u64)mask0 << 32) | mask1,
+#else
+				((__u64)mask1 << 32) | mask0,
+#endif
+				 dfd, pathname);
+}
+#endif
+
+/*
+ * fanotify_user_setup - Our initialization function.  Note that we cannot return
+ * error because we have compiled-in VFS hooks.  So an (unlikely) failure here
+ * must result in panic().
+ */
+static int __init fanotify_user_setup(void)
+{
+	fanotify_mark_cache = KMEM_CACHE(fsnotify_mark, SLAB_PANIC);
+	fanotify_event_cachep = KMEM_CACHE(fanotify_event_info, SLAB_PANIC);
+#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
+	fanotify_perm_event_cachep = KMEM_CACHE(fanotify_perm_event_info,
+						SLAB_PANIC);
+#endif
+
+	return 0;
+}
+device_initcall(fanotify_user_setup);
diff --git a/fs/notify/fdinfo.c b/fs/notify/fdinfo.c
new file mode 100644
index 00000000000..238a5930cb3
--- /dev/null
+++ b/fs/notify/fdinfo.c
@@ -0,0 +1,179 @@
+#include <linux/file.h>
+#include <linux/fs.h>
+#include <linux/fsnotify_backend.h>
+#include <linux/idr.h>
+#include <linux/init.h>
+#include <linux/inotify.h>
+#include <linux/fanotify.h>
+#include <linux/kernel.h>
+#include <linux/namei.h>
+#include <linux/sched.h>
+#include <linux/types.h>
+#include <linux/seq_file.h>
+#include <linux/proc_fs.h>
+#include <linux/exportfs.h>
+
+#include "inotify/inotify.h"
+#include "../fs/mount.h"
+
+#if defined(CONFIG_PROC_FS)
+
+#if defined(CONFIG_INOTIFY_USER) || defined(CONFIG_FANOTIFY)
+
+static int show_fdinfo(struct seq_file *m, struct file *f,
+		       int (*show)(struct seq_file *m, struct fsnotify_mark *mark))
+{
+	struct fsnotify_group *group = f->private_data;
+	struct fsnotify_mark *mark;
+	int ret = 0;
+
+	mutex_lock(&group->mark_mutex);
+	list_for_each_entry(mark, &group->marks_list, g_list) {
+		ret = show(m, mark);
+		if (ret)
+			break;
+	}
+	mutex_unlock(&group->mark_mutex);
+	return ret;
+}
+
+#if defined(CONFIG_EXPORTFS)
+static int show_mark_fhandle(struct seq_file *m, struct inode *inode)
+{
+	struct {
+		struct file_handle handle;
+		u8 pad[64];
+	} f;
+	int size, ret, i;
+
+	f.handle.handle_bytes = sizeof(f.pad);
+	size = f.handle.handle_bytes >> 2;
+
+	ret = exportfs_encode_inode_fh(inode, (struct fid *)f.handle.f_handle, &size, 0);
+	if ((ret == 255) || (ret == -ENOSPC)) {
+		WARN_ONCE(1, "Can't encode file handler for inotify: %d\n", ret);
+		return 0;
+	}
+
+	f.handle.handle_type = ret;
+	f.handle.handle_bytes = size * sizeof(u32);
+
+	ret = seq_printf(m, "fhandle-bytes:%x fhandle-type:%x f_handle:",
+			 f.handle.handle_bytes, f.handle.handle_type);
+
+	for (i = 0; i < f.handle.handle_bytes; i++)
+		ret |= seq_printf(m, "%02x", (int)f.handle.f_handle[i]);
+
+	return ret;
+}
+#else
+static int show_mark_fhandle(struct seq_file *m, struct inode *inode)
+{
+	return 0;
+}
+#endif
+
+#ifdef CONFIG_INOTIFY_USER
+
+static int inotify_fdinfo(struct seq_file *m, struct fsnotify_mark *mark)
+{
+	struct inotify_inode_mark *inode_mark;
+	struct inode *inode;
+	int ret = 0;
+
+	if (!(mark->flags & (FSNOTIFY_MARK_FLAG_ALIVE | FSNOTIFY_MARK_FLAG_INODE)))
+		return 0;
+
+	inode_mark = container_of(mark, struct inotify_inode_mark, fsn_mark);
+	inode = igrab(mark->i.inode);
+	if (inode) {
+		ret = seq_printf(m, "inotify wd:%x ino:%lx sdev:%x "
+				 "mask:%x ignored_mask:%x ",
+				 inode_mark->wd, inode->i_ino,
+				 inode->i_sb->s_dev,
+				 mark->mask, mark->ignored_mask);
+		ret |= show_mark_fhandle(m, inode);
+		ret |= seq_putc(m, '\n');
+		iput(inode);
+	}
+
+	return ret;
+}
+
+int inotify_show_fdinfo(struct seq_file *m, struct file *f)
+{
+	return show_fdinfo(m, f, inotify_fdinfo);
+}
+
+#endif /* CONFIG_INOTIFY_USER */
+
+#ifdef CONFIG_FANOTIFY
+
+static int fanotify_fdinfo(struct seq_file *m, struct fsnotify_mark *mark)
+{
+	unsigned int mflags = 0;
+	struct inode *inode;
+	int ret = 0;
+
+	if (!(mark->flags & FSNOTIFY_MARK_FLAG_ALIVE))
+		return 0;
+
+	if (mark->flags & FSNOTIFY_MARK_FLAG_IGNORED_SURV_MODIFY)
+		mflags |= FAN_MARK_IGNORED_SURV_MODIFY;
+
+	if (mark->flags & FSNOTIFY_MARK_FLAG_INODE) {
+		inode = igrab(mark->i.inode);
+		if (!inode)
+			goto out;
+		ret = seq_printf(m, "fanotify ino:%lx sdev:%x "
+				 "mflags:%x mask:%x ignored_mask:%x ",
+				 inode->i_ino, inode->i_sb->s_dev,
+				 mflags, mark->mask, mark->ignored_mask);
+		ret |= show_mark_fhandle(m, inode);
+		ret |= seq_putc(m, '\n');
+		iput(inode);
+	} else if (mark->flags & FSNOTIFY_MARK_FLAG_VFSMOUNT) {
+		struct mount *mnt = real_mount(mark->m.mnt);
+
+		ret = seq_printf(m, "fanotify mnt_id:%x mflags:%x mask:%x "
+				 "ignored_mask:%x\n", mnt->mnt_id, mflags,
+				 mark->mask, mark->ignored_mask);
+	}
+out:
+	return ret;
+}
+
+int fanotify_show_fdinfo(struct seq_file *m, struct file *f)
+{
+	struct fsnotify_group *group = f->private_data;
+	unsigned int flags = 0;
+
+	switch (group->priority) {
+	case FS_PRIO_0:
+		flags |= FAN_CLASS_NOTIF;
+		break;
+	case FS_PRIO_1:
+		flags |= FAN_CLASS_CONTENT;
+		break;
+	case FS_PRIO_2:
+		flags |= FAN_CLASS_PRE_CONTENT;
+		break;
+	}
+
+	if (group->max_events == UINT_MAX)
+		flags |= FAN_UNLIMITED_QUEUE;
+
+	if (group->fanotify_data.max_marks == UINT_MAX)
+		flags |= FAN_UNLIMITED_MARKS;
+
+	seq_printf(m, "fanotify flags:%x event-flags:%x\n",
+		   flags, group->fanotify_data.f_flags);
+
+	return show_fdinfo(m, f, fanotify_fdinfo);
+}
+
+#endif /* CONFIG_FANOTIFY */
+
+#endif /* CONFIG_INOTIFY_USER || CONFIG_FANOTIFY */
+
+#endif /* CONFIG_PROC_FS */
diff --git a/fs/notify/fdinfo.h b/fs/notify/fdinfo.h
new file mode 100644
index 00000000000..556afda990e
--- /dev/null
+++ b/fs/notify/fdinfo.h
@@ -0,0 +1,27 @@
+#ifndef __FSNOTIFY_FDINFO_H__
+#define __FSNOTIFY_FDINFO_H__
+
+#include <linux/errno.h>
+#include <linux/proc_fs.h>
+
+struct seq_file;
+struct file;
+
+#ifdef CONFIG_PROC_FS
+
+#ifdef CONFIG_INOTIFY_USER
+extern int inotify_show_fdinfo(struct seq_file *m, struct file *f);
+#endif
+
+#ifdef CONFIG_FANOTIFY
+extern int fanotify_show_fdinfo(struct seq_file *m, struct file *f);
+#endif
+
+#else /* CONFIG_PROC_FS */
+
+#define inotify_show_fdinfo	NULL
+#define fanotify_show_fdinfo	NULL
+
+#endif /* CONFIG_PROC_FS */
+
+#endif /* __FSNOTIFY_FDINFO_H__ */
diff --git a/fs/notify/fsnotify.c b/fs/notify/fsnotify.c
new file mode 100644
index 00000000000..9d3e9c50066
--- /dev/null
+++ b/fs/notify/fsnotify.c
@@ -0,0 +1,293 @@
+/*
+ *  Copyright (C) 2008 Red Hat, Inc., Eric Paris <eparis@redhat.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2, or (at your option)
+ *  any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; see the file COPYING.  If not, write to
+ *  the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <linux/dcache.h>
+#include <linux/fs.h>
+#include <linux/gfp.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/mount.h>
+#include <linux/srcu.h>
+
+#include <linux/fsnotify_backend.h>
+#include "fsnotify.h"
+#include "../mount.h"
+
+/*
+ * Clear all of the marks on an inode when it is being evicted from core
+ */
+void __fsnotify_inode_delete(struct inode *inode)
+{
+	fsnotify_clear_marks_by_inode(inode);
+}
+EXPORT_SYMBOL_GPL(__fsnotify_inode_delete);
+
+void __fsnotify_vfsmount_delete(struct vfsmount *mnt)
+{
+	fsnotify_clear_marks_by_mount(mnt);
+}
+
+/*
+ * Given an inode, first check if we care what happens to our children.  Inotify
+ * and dnotify both tell their parents about events.  If we care about any event
+ * on a child we run all of our children and set a dentry flag saying that the
+ * parent cares.  Thus when an event happens on a child it can quickly tell if
+ * if there is a need to find a parent and send the event to the parent.
+ */
+void __fsnotify_update_child_dentry_flags(struct inode *inode)
+{
+	struct dentry *alias;
+	int watched;
+
+	if (!S_ISDIR(inode->i_mode))
+		return;
+
+	/* determine if the children should tell inode about their events */
+	watched = fsnotify_inode_watches_children(inode);
+
+	spin_lock(&inode->i_lock);
+	/* run all of the dentries associated with this inode.  Since this is a
+	 * directory, there damn well better only be one item on this list */
+	hlist_for_each_entry(alias, &inode->i_dentry, d_alias) {
+		struct dentry *child;
+
+		/* run all of the children of the original inode and fix their
+		 * d_flags to indicate parental interest (their parent is the
+		 * original inode) */
+		spin_lock(&alias->d_lock);
+		list_for_each_entry(child, &alias->d_subdirs, d_u.d_child) {
+			if (!child->d_inode)
+				continue;
+
+			spin_lock_nested(&child->d_lock, DENTRY_D_LOCK_NESTED);
+			if (watched)
+				child->d_flags |= DCACHE_FSNOTIFY_PARENT_WATCHED;
+			else
+				child->d_flags &= ~DCACHE_FSNOTIFY_PARENT_WATCHED;
+			spin_unlock(&child->d_lock);
+		}
+		spin_unlock(&alias->d_lock);
+	}
+	spin_unlock(&inode->i_lock);
+}
+
+/* Notify this dentry's parent about a child's events. */
+int __fsnotify_parent(struct path *path, struct dentry *dentry, __u32 mask)
+{
+	struct dentry *parent;
+	struct inode *p_inode;
+	int ret = 0;
+
+	if (!dentry)
+		dentry = path->dentry;
+
+	if (!(dentry->d_flags & DCACHE_FSNOTIFY_PARENT_WATCHED))
+		return 0;
+
+	parent = dget_parent(dentry);
+	p_inode = parent->d_inode;
+
+	if (unlikely(!fsnotify_inode_watches_children(p_inode)))
+		__fsnotify_update_child_dentry_flags(p_inode);
+	else if (p_inode->i_fsnotify_mask & mask) {
+		/* we are notifying a parent so come up with the new mask which
+		 * specifies these are events which came from a child. */
+		mask |= FS_EVENT_ON_CHILD;
+
+		if (path)
+			ret = fsnotify(p_inode, mask, path, FSNOTIFY_EVENT_PATH,
+				       dentry->d_name.name, 0);
+		else
+			ret = fsnotify(p_inode, mask, dentry->d_inode, FSNOTIFY_EVENT_INODE,
+				       dentry->d_name.name, 0);
+	}
+
+	dput(parent);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(__fsnotify_parent);
+
+static int send_to_group(struct inode *to_tell,
+			 struct fsnotify_mark *inode_mark,
+			 struct fsnotify_mark *vfsmount_mark,
+			 __u32 mask, void *data,
+			 int data_is, u32 cookie,
+			 const unsigned char *file_name)
+{
+	struct fsnotify_group *group = NULL;
+	__u32 inode_test_mask = 0;
+	__u32 vfsmount_test_mask = 0;
+
+	if (unlikely(!inode_mark && !vfsmount_mark)) {
+		BUG();
+		return 0;
+	}
+
+	/* clear ignored on inode modification */
+	if (mask & FS_MODIFY) {
+		if (inode_mark &&
+		    !(inode_mark->flags & FSNOTIFY_MARK_FLAG_IGNORED_SURV_MODIFY))
+			inode_mark->ignored_mask = 0;
+		if (vfsmount_mark &&
+		    !(vfsmount_mark->flags & FSNOTIFY_MARK_FLAG_IGNORED_SURV_MODIFY))
+			vfsmount_mark->ignored_mask = 0;
+	}
+
+	/* does the inode mark tell us to do something? */
+	if (inode_mark) {
+		group = inode_mark->group;
+		inode_test_mask = (mask & ~FS_EVENT_ON_CHILD);
+		inode_test_mask &= inode_mark->mask;
+		inode_test_mask &= ~inode_mark->ignored_mask;
+	}
+
+	/* does the vfsmount_mark tell us to do something? */
+	if (vfsmount_mark) {
+		vfsmount_test_mask = (mask & ~FS_EVENT_ON_CHILD);
+		group = vfsmount_mark->group;
+		vfsmount_test_mask &= vfsmount_mark->mask;
+		vfsmount_test_mask &= ~vfsmount_mark->ignored_mask;
+		if (inode_mark)
+			vfsmount_test_mask &= ~inode_mark->ignored_mask;
+	}
+
+	pr_debug("%s: group=%p to_tell=%p mask=%x inode_mark=%p"
+		 " inode_test_mask=%x vfsmount_mark=%p vfsmount_test_mask=%x"
+		 " data=%p data_is=%d cookie=%d\n",
+		 __func__, group, to_tell, mask, inode_mark,
+		 inode_test_mask, vfsmount_mark, vfsmount_test_mask, data,
+		 data_is, cookie);
+
+	if (!inode_test_mask && !vfsmount_test_mask)
+		return 0;
+
+	return group->ops->handle_event(group, to_tell, inode_mark,
+					vfsmount_mark, mask, data, data_is,
+					file_name, cookie);
+}
+
+/*
+ * This is the main call to fsnotify.  The VFS calls into hook specific functions
+ * in linux/fsnotify.h.  Those functions then in turn call here.  Here will call
+ * out to all of the registered fsnotify_group.  Those groups can then use the
+ * notification event in whatever means they feel necessary.
+ */
+int fsnotify(struct inode *to_tell, __u32 mask, void *data, int data_is,
+	     const unsigned char *file_name, u32 cookie)
+{
+	struct hlist_node *inode_node = NULL, *vfsmount_node = NULL;
+	struct fsnotify_mark *inode_mark = NULL, *vfsmount_mark = NULL;
+	struct fsnotify_group *inode_group, *vfsmount_group;
+	struct mount *mnt;
+	int idx, ret = 0;
+	/* global tests shouldn't care about events on child only the specific event */
+	__u32 test_mask = (mask & ~FS_EVENT_ON_CHILD);
+
+	if (data_is == FSNOTIFY_EVENT_PATH)
+		mnt = real_mount(((struct path *)data)->mnt);
+	else
+		mnt = NULL;
+
+	/*
+	 * if this is a modify event we may need to clear the ignored masks
+	 * otherwise return if neither the inode nor the vfsmount care about
+	 * this type of event.
+	 */
+	if (!(mask & FS_MODIFY) &&
+	    !(test_mask & to_tell->i_fsnotify_mask) &&
+	    !(mnt && test_mask & mnt->mnt_fsnotify_mask))
+		return 0;
+
+	idx = srcu_read_lock(&fsnotify_mark_srcu);
+
+	if ((mask & FS_MODIFY) ||
+	    (test_mask & to_tell->i_fsnotify_mask))
+		inode_node = srcu_dereference(to_tell->i_fsnotify_marks.first,
+					      &fsnotify_mark_srcu);
+
+	if (mnt && ((mask & FS_MODIFY) ||
+		    (test_mask & mnt->mnt_fsnotify_mask))) {
+		vfsmount_node = srcu_dereference(mnt->mnt_fsnotify_marks.first,
+						 &fsnotify_mark_srcu);
+		inode_node = srcu_dereference(to_tell->i_fsnotify_marks.first,
+					      &fsnotify_mark_srcu);
+	}
+
+	while (inode_node || vfsmount_node) {
+		inode_group = vfsmount_group = NULL;
+
+		if (inode_node) {
+			inode_mark = hlist_entry(srcu_dereference(inode_node, &fsnotify_mark_srcu),
+						 struct fsnotify_mark, i.i_list);
+			inode_group = inode_mark->group;
+		}
+
+		if (vfsmount_node) {
+			vfsmount_mark = hlist_entry(srcu_dereference(vfsmount_node, &fsnotify_mark_srcu),
+							struct fsnotify_mark, m.m_list);
+			vfsmount_group = vfsmount_mark->group;
+		}
+
+		if (inode_group > vfsmount_group) {
+			/* handle inode */
+			ret = send_to_group(to_tell, inode_mark, NULL, mask,
+					    data, data_is, cookie, file_name);
+			/* we didn't use the vfsmount_mark */
+			vfsmount_group = NULL;
+		} else if (vfsmount_group > inode_group) {
+			ret = send_to_group(to_tell, NULL, vfsmount_mark, mask,
+					    data, data_is, cookie, file_name);
+			inode_group = NULL;
+		} else {
+			ret = send_to_group(to_tell, inode_mark, vfsmount_mark,
+					    mask, data, data_is, cookie,
+					    file_name);
+		}
+
+		if (ret && (mask & ALL_FSNOTIFY_PERM_EVENTS))
+			goto out;
+
+		if (inode_group)
+			inode_node = srcu_dereference(inode_node->next,
+						      &fsnotify_mark_srcu);
+		if (vfsmount_group)
+			vfsmount_node = srcu_dereference(vfsmount_node->next,
+							 &fsnotify_mark_srcu);
+	}
+	ret = 0;
+out:
+	srcu_read_unlock(&fsnotify_mark_srcu, idx);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(fsnotify);
+
+static __init int fsnotify_init(void)
+{
+	int ret;
+
+	BUG_ON(hweight32(ALL_FSNOTIFY_EVENTS) != 23);
+
+	ret = init_srcu_struct(&fsnotify_mark_srcu);
+	if (ret)
+		panic("initializing fsnotify_mark_srcu");
+
+	return 0;
+}
+core_initcall(fsnotify_init);
diff --git a/fs/notify/fsnotify.h b/fs/notify/fsnotify.h
new file mode 100644
index 00000000000..85e7d2b431d
--- /dev/null
+++ b/fs/notify/fsnotify.h
@@ -0,0 +1,47 @@
+#ifndef __FS_NOTIFY_FSNOTIFY_H_
+#define __FS_NOTIFY_FSNOTIFY_H_
+
+#include <linux/list.h>
+#include <linux/fsnotify.h>
+#include <linux/srcu.h>
+#include <linux/types.h>
+
+/* destroy all events sitting in this groups notification queue */
+extern void fsnotify_flush_notify(struct fsnotify_group *group);
+
+/* protects reads of inode and vfsmount marks list */
+extern struct srcu_struct fsnotify_mark_srcu;
+
+extern void fsnotify_set_inode_mark_mask_locked(struct fsnotify_mark *fsn_mark,
+						__u32 mask);
+/* add a mark to an inode */
+extern int fsnotify_add_inode_mark(struct fsnotify_mark *mark,
+				   struct fsnotify_group *group, struct inode *inode,
+				   int allow_dups);
+/* add a mark to a vfsmount */
+extern int fsnotify_add_vfsmount_mark(struct fsnotify_mark *mark,
+				      struct fsnotify_group *group, struct vfsmount *mnt,
+				      int allow_dups);
+
+/* final kfree of a group */
+extern void fsnotify_final_destroy_group(struct fsnotify_group *group);
+
+/* vfsmount specific destruction of a mark */
+extern void fsnotify_destroy_vfsmount_mark(struct fsnotify_mark *mark);
+/* inode specific destruction of a mark */
+extern void fsnotify_destroy_inode_mark(struct fsnotify_mark *mark);
+/* run the list of all marks associated with inode and flag them to be freed */
+extern void fsnotify_clear_marks_by_inode(struct inode *inode);
+/* run the list of all marks associated with vfsmount and flag them to be freed */
+extern void fsnotify_clear_marks_by_mount(struct vfsmount *mnt);
+/*
+ * update the dentry->d_flags of all of inode's children to indicate if inode cares
+ * about events that happen to its children.
+ */
+extern void __fsnotify_update_child_dentry_flags(struct inode *inode);
+
+/* allocate and destroy and event holder to attach events to notification/access queues */
+extern struct fsnotify_event_holder *fsnotify_alloc_event_holder(void);
+extern void fsnotify_destroy_event_holder(struct fsnotify_event_holder *holder);
+
+#endif	/* __FS_NOTIFY_FSNOTIFY_H_ */
diff --git a/fs/notify/group.c b/fs/notify/group.c
new file mode 100644
index 00000000000..ad199598045
--- /dev/null
+++ b/fs/notify/group.c
@@ -0,0 +1,118 @@
+/*
+ *  Copyright (C) 2008 Red Hat, Inc., Eric Paris <eparis@redhat.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2, or (at your option)
+ *  any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; see the file COPYING.  If not, write to
+ *  the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <linux/list.h>
+#include <linux/mutex.h>
+#include <linux/slab.h>
+#include <linux/srcu.h>
+#include <linux/rculist.h>
+#include <linux/wait.h>
+
+#include <linux/fsnotify_backend.h>
+#include "fsnotify.h"
+
+#include <linux/atomic.h>
+
+/*
+ * Final freeing of a group
+ */
+void fsnotify_final_destroy_group(struct fsnotify_group *group)
+{
+	if (group->ops->free_group_priv)
+		group->ops->free_group_priv(group);
+
+	kfree(group);
+}
+
+/*
+ * Trying to get rid of a group. Remove all marks, flush all events and release
+ * the group reference.
+ * Note that another thread calling fsnotify_clear_marks_by_group() may still
+ * hold a ref to the group.
+ */
+void fsnotify_destroy_group(struct fsnotify_group *group)
+{
+	/* clear all inode marks for this group */
+	fsnotify_clear_marks_by_group(group);
+
+	synchronize_srcu(&fsnotify_mark_srcu);
+
+	/* clear the notification queue of all events */
+	fsnotify_flush_notify(group);
+
+	/*
+	 * Destroy overflow event (we cannot use fsnotify_destroy_event() as
+	 * that deliberately ignores overflow events.
+	 */
+	if (group->overflow_event)
+		group->ops->free_event(group->overflow_event);
+
+	fsnotify_put_group(group);
+}
+
+/*
+ * Get reference to a group.
+ */
+void fsnotify_get_group(struct fsnotify_group *group)
+{
+	atomic_inc(&group->refcnt);
+}
+
+/*
+ * Drop a reference to a group.  Free it if it's through.
+ */
+void fsnotify_put_group(struct fsnotify_group *group)
+{
+	if (atomic_dec_and_test(&group->refcnt))
+		fsnotify_final_destroy_group(group);
+}
+
+/*
+ * Create a new fsnotify_group and hold a reference for the group returned.
+ */
+struct fsnotify_group *fsnotify_alloc_group(const struct fsnotify_ops *ops)
+{
+	struct fsnotify_group *group;
+
+	group = kzalloc(sizeof(struct fsnotify_group), GFP_KERNEL);
+	if (!group)
+		return ERR_PTR(-ENOMEM);
+
+	/* set to 0 when there a no external references to this group */
+	atomic_set(&group->refcnt, 1);
+	atomic_set(&group->num_marks, 0);
+
+	mutex_init(&group->notification_mutex);
+	INIT_LIST_HEAD(&group->notification_list);
+	init_waitqueue_head(&group->notification_waitq);
+	group->max_events = UINT_MAX;
+
+	mutex_init(&group->mark_mutex);
+	INIT_LIST_HEAD(&group->marks_list);
+
+	group->ops = ops;
+
+	return group;
+}
+
+int fsnotify_fasync(int fd, struct file *file, int on)
+{
+	struct fsnotify_group *group = file->private_data;
+
+	return fasync_helper(fd, file, on, &group->fsn_fa) >= 0 ? 0 : -EIO;
+}
diff --git a/fs/notify/inode_mark.c b/fs/notify/inode_mark.c
new file mode 100644
index 00000000000..74825be65b7
--- /dev/null
+++ b/fs/notify/inode_mark.c
@@ -0,0 +1,321 @@
+/*
+ *  Copyright (C) 2008 Red Hat, Inc., Eric Paris <eparis@redhat.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2, or (at your option)
+ *  any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; see the file COPYING.  If not, write to
+ *  the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <linux/fs.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <linux/spinlock.h>
+
+#include <linux/atomic.h>
+
+#include <linux/fsnotify_backend.h>
+#include "fsnotify.h"
+
+#include "../internal.h"
+
+/*
+ * Recalculate the mask of events relevant to a given inode locked.
+ */
+static void fsnotify_recalc_inode_mask_locked(struct inode *inode)
+{
+	struct fsnotify_mark *mark;
+	__u32 new_mask = 0;
+
+	assert_spin_locked(&inode->i_lock);
+
+	hlist_for_each_entry(mark, &inode->i_fsnotify_marks, i.i_list)
+		new_mask |= mark->mask;
+	inode->i_fsnotify_mask = new_mask;
+}
+
+/*
+ * Recalculate the inode->i_fsnotify_mask, or the mask of all FS_* event types
+ * any notifier is interested in hearing for this inode.
+ */
+void fsnotify_recalc_inode_mask(struct inode *inode)
+{
+	spin_lock(&inode->i_lock);
+	fsnotify_recalc_inode_mask_locked(inode);
+	spin_unlock(&inode->i_lock);
+
+	__fsnotify_update_child_dentry_flags(inode);
+}
+
+void fsnotify_destroy_inode_mark(struct fsnotify_mark *mark)
+{
+	struct inode *inode = mark->i.inode;
+
+	BUG_ON(!mutex_is_locked(&mark->group->mark_mutex));
+	assert_spin_locked(&mark->lock);
+
+	spin_lock(&inode->i_lock);
+
+	hlist_del_init_rcu(&mark->i.i_list);
+	mark->i.inode = NULL;
+
+	/*
+	 * this mark is now off the inode->i_fsnotify_marks list and we
+	 * hold the inode->i_lock, so this is the perfect time to update the
+	 * inode->i_fsnotify_mask
+	 */
+	fsnotify_recalc_inode_mask_locked(inode);
+
+	spin_unlock(&inode->i_lock);
+}
+
+/*
+ * Given an inode, destroy all of the marks associated with that inode.
+ */
+void fsnotify_clear_marks_by_inode(struct inode *inode)
+{
+	struct fsnotify_mark *mark, *lmark;
+	struct hlist_node *n;
+	LIST_HEAD(free_list);
+
+	spin_lock(&inode->i_lock);
+	hlist_for_each_entry_safe(mark, n, &inode->i_fsnotify_marks, i.i_list) {
+		list_add(&mark->i.free_i_list, &free_list);
+		hlist_del_init_rcu(&mark->i.i_list);
+		fsnotify_get_mark(mark);
+	}
+	spin_unlock(&inode->i_lock);
+
+	list_for_each_entry_safe(mark, lmark, &free_list, i.free_i_list) {
+		struct fsnotify_group *group;
+
+		spin_lock(&mark->lock);
+		fsnotify_get_group(mark->group);
+		group = mark->group;
+		spin_unlock(&mark->lock);
+
+		fsnotify_destroy_mark(mark, group);
+		fsnotify_put_mark(mark);
+		fsnotify_put_group(group);
+	}
+}
+
+/*
+ * Given a group clear all of the inode marks associated with that group.
+ */
+void fsnotify_clear_inode_marks_by_group(struct fsnotify_group *group)
+{
+	fsnotify_clear_marks_by_group_flags(group, FSNOTIFY_MARK_FLAG_INODE);
+}
+
+/*
+ * given a group and inode, find the mark associated with that combination.
+ * if found take a reference to that mark and return it, else return NULL
+ */
+static struct fsnotify_mark *fsnotify_find_inode_mark_locked(
+		struct fsnotify_group *group,
+		struct inode *inode)
+{
+	struct fsnotify_mark *mark;
+
+	assert_spin_locked(&inode->i_lock);
+
+	hlist_for_each_entry(mark, &inode->i_fsnotify_marks, i.i_list) {
+		if (mark->group == group) {
+			fsnotify_get_mark(mark);
+			return mark;
+		}
+	}
+	return NULL;
+}
+
+/*
+ * given a group and inode, find the mark associated with that combination.
+ * if found take a reference to that mark and return it, else return NULL
+ */
+struct fsnotify_mark *fsnotify_find_inode_mark(struct fsnotify_group *group,
+					       struct inode *inode)
+{
+	struct fsnotify_mark *mark;
+
+	spin_lock(&inode->i_lock);
+	mark = fsnotify_find_inode_mark_locked(group, inode);
+	spin_unlock(&inode->i_lock);
+
+	return mark;
+}
+
+/*
+ * If we are setting a mark mask on an inode mark we should pin the inode
+ * in memory.
+ */
+void fsnotify_set_inode_mark_mask_locked(struct fsnotify_mark *mark,
+					 __u32 mask)
+{
+	struct inode *inode;
+
+	assert_spin_locked(&mark->lock);
+
+	if (mask &&
+	    mark->i.inode &&
+	    !(mark->flags & FSNOTIFY_MARK_FLAG_OBJECT_PINNED)) {
+		mark->flags |= FSNOTIFY_MARK_FLAG_OBJECT_PINNED;
+		inode = igrab(mark->i.inode);
+		/*
+		 * we shouldn't be able to get here if the inode wasn't
+		 * already safely held in memory.  But bug in case it
+		 * ever is wrong.
+		 */
+		BUG_ON(!inode);
+	}
+}
+
+/*
+ * Attach an initialized mark to a given inode.
+ * These marks may be used for the fsnotify backend to determine which
+ * event types should be delivered to which group and for which inodes.  These
+ * marks are ordered according to priority, highest number first, and then by
+ * the group's location in memory.
+ */
+int fsnotify_add_inode_mark(struct fsnotify_mark *mark,
+			    struct fsnotify_group *group, struct inode *inode,
+			    int allow_dups)
+{
+	struct fsnotify_mark *lmark, *last = NULL;
+	int ret = 0;
+
+	mark->flags |= FSNOTIFY_MARK_FLAG_INODE;
+
+	BUG_ON(!mutex_is_locked(&group->mark_mutex));
+	assert_spin_locked(&mark->lock);
+
+	spin_lock(&inode->i_lock);
+
+	mark->i.inode = inode;
+
+	/* is mark the first mark? */
+	if (hlist_empty(&inode->i_fsnotify_marks)) {
+		hlist_add_head_rcu(&mark->i.i_list, &inode->i_fsnotify_marks);
+		goto out;
+	}
+
+	/* should mark be in the middle of the current list? */
+	hlist_for_each_entry(lmark, &inode->i_fsnotify_marks, i.i_list) {
+		last = lmark;
+
+		if ((lmark->group == group) && !allow_dups) {
+			ret = -EEXIST;
+			goto out;
+		}
+
+		if (mark->group->priority < lmark->group->priority)
+			continue;
+
+		if ((mark->group->priority == lmark->group->priority) &&
+		    (mark->group < lmark->group))
+			continue;
+
+		hlist_add_before_rcu(&mark->i.i_list, &lmark->i.i_list);
+		goto out;
+	}
+
+	BUG_ON(last == NULL);
+	/* mark should be the last entry.  last is the current last entry */
+	hlist_add_after_rcu(&last->i.i_list, &mark->i.i_list);
+out:
+	fsnotify_recalc_inode_mask_locked(inode);
+	spin_unlock(&inode->i_lock);
+
+	return ret;
+}
+
+/**
+ * fsnotify_unmount_inodes - an sb is unmounting.  handle any watched inodes.
+ * @list: list of inodes being unmounted (sb->s_inodes)
+ *
+ * Called during unmount with no locks held, so needs to be safe against
+ * concurrent modifiers. We temporarily drop inode_sb_list_lock and CAN block.
+ */
+void fsnotify_unmount_inodes(struct list_head *list)
+{
+	struct inode *inode, *next_i, *need_iput = NULL;
+
+	spin_lock(&inode_sb_list_lock);
+	list_for_each_entry_safe(inode, next_i, list, i_sb_list) {
+		struct inode *need_iput_tmp;
+
+		/*
+		 * We cannot __iget() an inode in state I_FREEING,
+		 * I_WILL_FREE, or I_NEW which is fine because by that point
+		 * the inode cannot have any associated watches.
+		 */
+		spin_lock(&inode->i_lock);
+		if (inode->i_state & (I_FREEING|I_WILL_FREE|I_NEW)) {
+			spin_unlock(&inode->i_lock);
+			continue;
+		}
+
+		/*
+		 * If i_count is zero, the inode cannot have any watches and
+		 * doing an __iget/iput with MS_ACTIVE clear would actually
+		 * evict all inodes with zero i_count from icache which is
+		 * unnecessarily violent and may in fact be illegal to do.
+		 */
+		if (!atomic_read(&inode->i_count)) {
+			spin_unlock(&inode->i_lock);
+			continue;
+		}
+
+		need_iput_tmp = need_iput;
+		need_iput = NULL;
+
+		/* In case fsnotify_inode_delete() drops a reference. */
+		if (inode != need_iput_tmp)
+			__iget(inode);
+		else
+			need_iput_tmp = NULL;
+		spin_unlock(&inode->i_lock);
+
+		/* In case the dropping of a reference would nuke next_i. */
+		if ((&next_i->i_sb_list != list) &&
+		    atomic_read(&next_i->i_count)) {
+			spin_lock(&next_i->i_lock);
+			if (!(next_i->i_state & (I_FREEING | I_WILL_FREE))) {
+				__iget(next_i);
+				need_iput = next_i;
+			}
+			spin_unlock(&next_i->i_lock);
+		}
+
+		/*
+		 * We can safely drop inode_sb_list_lock here because we hold
+		 * references on both inode and next_i.  Also no new inodes
+		 * will be added since the umount has begun.
+		 */
+		spin_unlock(&inode_sb_list_lock);
+
+		if (need_iput_tmp)
+			iput(need_iput_tmp);
+
+		/* for each watch, send FS_UNMOUNT and then remove it */
+		fsnotify(inode, FS_UNMOUNT, inode, FSNOTIFY_EVENT_INODE, NULL, 0);
+
+		fsnotify_inode_delete(inode);
+
+		iput(inode);
+
+		spin_lock(&inode_sb_list_lock);
+	}
+	spin_unlock(&inode_sb_list_lock);
+}
diff --git a/fs/notify/inotify/Kconfig b/fs/notify/inotify/Kconfig
index 44679284102..b981fc0c837 100644
--- a/fs/notify/inotify/Kconfig
+++ b/fs/notify/inotify/Kconfig
@@ -1,26 +1,16 @@
-config INOTIFY
-	bool "Inotify file change notification support"
-	default y
-	---help---
-	  Say Y here to enable inotify support.  Inotify is a file change
-	  notification system and a replacement for dnotify.  Inotify fixes
-	  numerous shortcomings in dnotify and introduces several new features
-	  including multiple file events, one-shot support, and unmount
-	  notification.
-
-	  For more information, see <file:Documentation/filesystems/inotify.txt>
-
-	  If unsure, say Y.
-
 config INOTIFY_USER
 	bool "Inotify support for userspace"
-	depends on INOTIFY
+	select ANON_INODES
+	select FSNOTIFY
 	default y
 	---help---
 	  Say Y here to enable inotify support for userspace, including the
 	  associated system calls.  Inotify allows monitoring of both files and
 	  directories via a single open fd.  Events are read from the file
 	  descriptor, which is also select()- and poll()-able.
+	  Inotify fixes numerous shortcomings in dnotify and introduces several
+	  new features including multiple file events, one-shot support, and
+	  unmount notification.
 
 	  For more information, see <file:Documentation/filesystems/inotify.txt>
 
diff --git a/fs/notify/inotify/Makefile b/fs/notify/inotify/Makefile
index e290f3bb9d8..a380dabe09d 100644
--- a/fs/notify/inotify/Makefile
+++ b/fs/notify/inotify/Makefile
@@ -1,2 +1 @@
-obj-$(CONFIG_INOTIFY)		+= inotify.o
-obj-$(CONFIG_INOTIFY_USER)	+= inotify_user.o
+obj-$(CONFIG_INOTIFY_USER)	+= inotify_fsnotify.o inotify_user.o
diff --git a/fs/notify/inotify/inotify.c b/fs/notify/inotify/inotify.c
deleted file mode 100644
index 331f2e88e28..00000000000
--- a/fs/notify/inotify/inotify.c
+++ /dev/null
@@ -1,913 +0,0 @@
-/*
- * fs/inotify.c - inode-based file event notifications
- *
- * Authors:
- *	John McCutchan	<ttb@tentacle.dhs.org>
- *	Robert Love	<rml@novell.com>
- *
- * Kernel API added by: Amy Griffis <amy.griffis@hp.com>
- *
- * Copyright (C) 2005 John McCutchan
- * Copyright 2006 Hewlett-Packard Development Company, L.P.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the
- * Free Software Foundation; either version 2, or (at your option) any
- * later version.
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
- */
-
-#include <linux/module.h>
-#include <linux/kernel.h>
-#include <linux/spinlock.h>
-#include <linux/idr.h>
-#include <linux/slab.h>
-#include <linux/fs.h>
-#include <linux/sched.h>
-#include <linux/init.h>
-#include <linux/list.h>
-#include <linux/writeback.h>
-#include <linux/inotify.h>
-
-static atomic_t inotify_cookie;
-
-/*
- * Lock ordering:
- *
- * dentry->d_lock (used to keep d_move() away from dentry->d_parent)
- * iprune_mutex (synchronize shrink_icache_memory())
- * 	inode_lock (protects the super_block->s_inodes list)
- * 	inode->inotify_mutex (protects inode->inotify_watches and watches->i_list)
- * 		inotify_handle->mutex (protects inotify_handle and watches->h_list)
- *
- * The inode->inotify_mutex and inotify_handle->mutex and held during execution
- * of a caller's event handler.  Thus, the caller must not hold any locks
- * taken in their event handler while calling any of the published inotify
- * interfaces.
- */
-
-/*
- * Lifetimes of the three main data structures--inotify_handle, inode, and
- * inotify_watch--are managed by reference count.
- *
- * inotify_handle: Lifetime is from inotify_init() to inotify_destroy().
- * Additional references can bump the count via get_inotify_handle() and drop
- * the count via put_inotify_handle().
- *
- * inotify_watch: for inotify's purposes, lifetime is from inotify_add_watch()
- * to remove_watch_no_event().  Additional references can bump the count via
- * get_inotify_watch() and drop the count via put_inotify_watch().  The caller
- * is reponsible for the final put after receiving IN_IGNORED, or when using
- * IN_ONESHOT after receiving the first event.  Inotify does the final put if
- * inotify_destroy() is called.
- *
- * inode: Pinned so long as the inode is associated with a watch, from
- * inotify_add_watch() to the final put_inotify_watch().
- */
-
-/*
- * struct inotify_handle - represents an inotify instance
- *
- * This structure is protected by the mutex 'mutex'.
- */
-struct inotify_handle {
-	struct idr		idr;		/* idr mapping wd -> watch */
-	struct mutex		mutex;		/* protects this bad boy */
-	struct list_head	watches;	/* list of watches */
-	atomic_t		count;		/* reference count */
-	u32			last_wd;	/* the last wd allocated */
-	const struct inotify_operations *in_ops; /* inotify caller operations */
-};
-
-static inline void get_inotify_handle(struct inotify_handle *ih)
-{
-	atomic_inc(&ih->count);
-}
-
-static inline void put_inotify_handle(struct inotify_handle *ih)
-{
-	if (atomic_dec_and_test(&ih->count)) {
-		idr_destroy(&ih->idr);
-		kfree(ih);
-	}
-}
-
-/**
- * get_inotify_watch - grab a reference to an inotify_watch
- * @watch: watch to grab
- */
-void get_inotify_watch(struct inotify_watch *watch)
-{
-	atomic_inc(&watch->count);
-}
-EXPORT_SYMBOL_GPL(get_inotify_watch);
-
-int pin_inotify_watch(struct inotify_watch *watch)
-{
-	struct super_block *sb = watch->inode->i_sb;
-	spin_lock(&sb_lock);
-	if (sb->s_count >= S_BIAS) {
-		atomic_inc(&sb->s_active);
-		spin_unlock(&sb_lock);
-		atomic_inc(&watch->count);
-		return 1;
-	}
-	spin_unlock(&sb_lock);
-	return 0;
-}
-
-/**
- * put_inotify_watch - decrements the ref count on a given watch.  cleans up
- * watch references if the count reaches zero.  inotify_watch is freed by
- * inotify callers via the destroy_watch() op.
- * @watch: watch to release
- */
-void put_inotify_watch(struct inotify_watch *watch)
-{
-	if (atomic_dec_and_test(&watch->count)) {
-		struct inotify_handle *ih = watch->ih;
-
-		iput(watch->inode);
-		ih->in_ops->destroy_watch(watch);
-		put_inotify_handle(ih);
-	}
-}
-EXPORT_SYMBOL_GPL(put_inotify_watch);
-
-void unpin_inotify_watch(struct inotify_watch *watch)
-{
-	struct super_block *sb = watch->inode->i_sb;
-	put_inotify_watch(watch);
-	deactivate_super(sb);
-}
-
-/*
- * inotify_handle_get_wd - returns the next WD for use by the given handle
- *
- * Callers must hold ih->mutex.  This function can sleep.
- */
-static int inotify_handle_get_wd(struct inotify_handle *ih,
-				 struct inotify_watch *watch)
-{
-	int ret;
-
-	do {
-		if (unlikely(!idr_pre_get(&ih->idr, GFP_NOFS)))
-			return -ENOSPC;
-		ret = idr_get_new_above(&ih->idr, watch, ih->last_wd+1, &watch->wd);
-	} while (ret == -EAGAIN);
-
-	if (likely(!ret))
-		ih->last_wd = watch->wd;
-
-	return ret;
-}
-
-/*
- * inotify_inode_watched - returns nonzero if there are watches on this inode
- * and zero otherwise.  We call this lockless, we do not care if we race.
- */
-static inline int inotify_inode_watched(struct inode *inode)
-{
-	return !list_empty(&inode->inotify_watches);
-}
-
-/*
- * Get child dentry flag into synch with parent inode.
- * Flag should always be clear for negative dentrys.
- */
-static void set_dentry_child_flags(struct inode *inode, int watched)
-{
-	struct dentry *alias;
-
-	spin_lock(&dcache_lock);
-	list_for_each_entry(alias, &inode->i_dentry, d_alias) {
-		struct dentry *child;
-
-		list_for_each_entry(child, &alias->d_subdirs, d_u.d_child) {
-			if (!child->d_inode)
-				continue;
-
-			spin_lock(&child->d_lock);
-			if (watched)
-				child->d_flags |= DCACHE_INOTIFY_PARENT_WATCHED;
-			else
-				child->d_flags &=~DCACHE_INOTIFY_PARENT_WATCHED;
-			spin_unlock(&child->d_lock);
-		}
-	}
-	spin_unlock(&dcache_lock);
-}
-
-/*
- * inotify_find_handle - find the watch associated with the given inode and
- * handle
- *
- * Callers must hold inode->inotify_mutex.
- */
-static struct inotify_watch *inode_find_handle(struct inode *inode,
-					       struct inotify_handle *ih)
-{
-	struct inotify_watch *watch;
-
-	list_for_each_entry(watch, &inode->inotify_watches, i_list) {
-		if (watch->ih == ih)
-			return watch;
-	}
-
-	return NULL;
-}
-
-/*
- * remove_watch_no_event - remove watch without the IN_IGNORED event.
- *
- * Callers must hold both inode->inotify_mutex and ih->mutex.
- */
-static void remove_watch_no_event(struct inotify_watch *watch,
-				  struct inotify_handle *ih)
-{
-	list_del(&watch->i_list);
-	list_del(&watch->h_list);
-
-	if (!inotify_inode_watched(watch->inode))
-		set_dentry_child_flags(watch->inode, 0);
-
-	idr_remove(&ih->idr, watch->wd);
-}
-
-/**
- * inotify_remove_watch_locked - Remove a watch from both the handle and the
- * inode.  Sends the IN_IGNORED event signifying that the inode is no longer
- * watched.  May be invoked from a caller's event handler.
- * @ih: inotify handle associated with watch
- * @watch: watch to remove
- *
- * Callers must hold both inode->inotify_mutex and ih->mutex.
- */
-void inotify_remove_watch_locked(struct inotify_handle *ih,
-				 struct inotify_watch *watch)
-{
-	remove_watch_no_event(watch, ih);
-	ih->in_ops->handle_event(watch, watch->wd, IN_IGNORED, 0, NULL, NULL);
-}
-EXPORT_SYMBOL_GPL(inotify_remove_watch_locked);
-
-/* Kernel API for producing events */
-
-/*
- * inotify_d_instantiate - instantiate dcache entry for inode
- */
-void inotify_d_instantiate(struct dentry *entry, struct inode *inode)
-{
-	struct dentry *parent;
-
-	if (!inode)
-		return;
-
-	spin_lock(&entry->d_lock);
-	parent = entry->d_parent;
-	if (parent->d_inode && inotify_inode_watched(parent->d_inode))
-		entry->d_flags |= DCACHE_INOTIFY_PARENT_WATCHED;
-	spin_unlock(&entry->d_lock);
-}
-
-/*
- * inotify_d_move - dcache entry has been moved
- */
-void inotify_d_move(struct dentry *entry)
-{
-	struct dentry *parent;
-
-	parent = entry->d_parent;
-	if (inotify_inode_watched(parent->d_inode))
-		entry->d_flags |= DCACHE_INOTIFY_PARENT_WATCHED;
-	else
-		entry->d_flags &= ~DCACHE_INOTIFY_PARENT_WATCHED;
-}
-
-/**
- * inotify_inode_queue_event - queue an event to all watches on this inode
- * @inode: inode event is originating from
- * @mask: event mask describing this event
- * @cookie: cookie for synchronization, or zero
- * @name: filename, if any
- * @n_inode: inode associated with name
- */
-void inotify_inode_queue_event(struct inode *inode, u32 mask, u32 cookie,
-			       const char *name, struct inode *n_inode)
-{
-	struct inotify_watch *watch, *next;
-
-	if (!inotify_inode_watched(inode))
-		return;
-
-	mutex_lock(&inode->inotify_mutex);
-	list_for_each_entry_safe(watch, next, &inode->inotify_watches, i_list) {
-		u32 watch_mask = watch->mask;
-		if (watch_mask & mask) {
-			struct inotify_handle *ih= watch->ih;
-			mutex_lock(&ih->mutex);
-			if (watch_mask & IN_ONESHOT)
-				remove_watch_no_event(watch, ih);
-			ih->in_ops->handle_event(watch, watch->wd, mask, cookie,
-						 name, n_inode);
-			mutex_unlock(&ih->mutex);
-		}
-	}
-	mutex_unlock(&inode->inotify_mutex);
-}
-EXPORT_SYMBOL_GPL(inotify_inode_queue_event);
-
-/**
- * inotify_dentry_parent_queue_event - queue an event to a dentry's parent
- * @dentry: the dentry in question, we queue against this dentry's parent
- * @mask: event mask describing this event
- * @cookie: cookie for synchronization, or zero
- * @name: filename, if any
- */
-void inotify_dentry_parent_queue_event(struct dentry *dentry, u32 mask,
-				       u32 cookie, const char *name)
-{
-	struct dentry *parent;
-	struct inode *inode;
-
-	if (!(dentry->d_flags & DCACHE_INOTIFY_PARENT_WATCHED))
-		return;
-
-	spin_lock(&dentry->d_lock);
-	parent = dentry->d_parent;
-	inode = parent->d_inode;
-
-	if (inotify_inode_watched(inode)) {
-		dget(parent);
-		spin_unlock(&dentry->d_lock);
-		inotify_inode_queue_event(inode, mask, cookie, name,
-					  dentry->d_inode);
-		dput(parent);
-	} else
-		spin_unlock(&dentry->d_lock);
-}
-EXPORT_SYMBOL_GPL(inotify_dentry_parent_queue_event);
-
-/**
- * inotify_get_cookie - return a unique cookie for use in synchronizing events.
- */
-u32 inotify_get_cookie(void)
-{
-	return atomic_inc_return(&inotify_cookie);
-}
-EXPORT_SYMBOL_GPL(inotify_get_cookie);
-
-/**
- * inotify_unmount_inodes - an sb is unmounting.  handle any watched inodes.
- * @list: list of inodes being unmounted (sb->s_inodes)
- *
- * Called with inode_lock held, protecting the unmounting super block's list
- * of inodes, and with iprune_mutex held, keeping shrink_icache_memory() at bay.
- * We temporarily drop inode_lock, however, and CAN block.
- */
-void inotify_unmount_inodes(struct list_head *list)
-{
-	struct inode *inode, *next_i, *need_iput = NULL;
-
-	list_for_each_entry_safe(inode, next_i, list, i_sb_list) {
-		struct inotify_watch *watch, *next_w;
-		struct inode *need_iput_tmp;
-		struct list_head *watches;
-
-		/*
-		 * If i_count is zero, the inode cannot have any watches and
-		 * doing an __iget/iput with MS_ACTIVE clear would actually
-		 * evict all inodes with zero i_count from icache which is
-		 * unnecessarily violent and may in fact be illegal to do.
-		 */
-		if (!atomic_read(&inode->i_count))
-			continue;
-
-		/*
-		 * We cannot __iget() an inode in state I_CLEAR, I_FREEING, or
-		 * I_WILL_FREE which is fine because by that point the inode
-		 * cannot have any associated watches.
-		 */
-		if (inode->i_state & (I_CLEAR | I_FREEING | I_WILL_FREE))
-			continue;
-
-		need_iput_tmp = need_iput;
-		need_iput = NULL;
-		/* In case inotify_remove_watch_locked() drops a reference. */
-		if (inode != need_iput_tmp)
-			__iget(inode);
-		else
-			need_iput_tmp = NULL;
-		/* In case the dropping of a reference would nuke next_i. */
-		if ((&next_i->i_sb_list != list) &&
-				atomic_read(&next_i->i_count) &&
-				!(next_i->i_state & (I_CLEAR | I_FREEING |
-					I_WILL_FREE))) {
-			__iget(next_i);
-			need_iput = next_i;
-		}
-
-		/*
-		 * We can safely drop inode_lock here because we hold
-		 * references on both inode and next_i.  Also no new inodes
-		 * will be added since the umount has begun.  Finally,
-		 * iprune_mutex keeps shrink_icache_memory() away.
-		 */
-		spin_unlock(&inode_lock);
-
-		if (need_iput_tmp)
-			iput(need_iput_tmp);
-
-		/* for each watch, send IN_UNMOUNT and then remove it */
-		mutex_lock(&inode->inotify_mutex);
-		watches = &inode->inotify_watches;
-		list_for_each_entry_safe(watch, next_w, watches, i_list) {
-			struct inotify_handle *ih= watch->ih;
-			get_inotify_watch(watch);
-			mutex_lock(&ih->mutex);
-			ih->in_ops->handle_event(watch, watch->wd, IN_UNMOUNT, 0,
-						 NULL, NULL);
-			inotify_remove_watch_locked(ih, watch);
-			mutex_unlock(&ih->mutex);
-			put_inotify_watch(watch);
-		}
-		mutex_unlock(&inode->inotify_mutex);
-		iput(inode);		
-
-		spin_lock(&inode_lock);
-	}
-}
-EXPORT_SYMBOL_GPL(inotify_unmount_inodes);
-
-/**
- * inotify_inode_is_dead - an inode has been deleted, cleanup any watches
- * @inode: inode that is about to be removed
- */
-void inotify_inode_is_dead(struct inode *inode)
-{
-	struct inotify_watch *watch, *next;
-
-	mutex_lock(&inode->inotify_mutex);
-	list_for_each_entry_safe(watch, next, &inode->inotify_watches, i_list) {
-		struct inotify_handle *ih = watch->ih;
-		mutex_lock(&ih->mutex);
-		inotify_remove_watch_locked(ih, watch);
-		mutex_unlock(&ih->mutex);
-	}
-	mutex_unlock(&inode->inotify_mutex);
-}
-EXPORT_SYMBOL_GPL(inotify_inode_is_dead);
-
-/* Kernel Consumer API */
-
-/**
- * inotify_init - allocate and initialize an inotify instance
- * @ops: caller's inotify operations
- */
-struct inotify_handle *inotify_init(const struct inotify_operations *ops)
-{
-	struct inotify_handle *ih;
-
-	ih = kmalloc(sizeof(struct inotify_handle), GFP_KERNEL);
-	if (unlikely(!ih))
-		return ERR_PTR(-ENOMEM);
-
-	idr_init(&ih->idr);
-	INIT_LIST_HEAD(&ih->watches);
-	mutex_init(&ih->mutex);
-	ih->last_wd = 0;
-	ih->in_ops = ops;
-	atomic_set(&ih->count, 0);
-	get_inotify_handle(ih);
-
-	return ih;
-}
-EXPORT_SYMBOL_GPL(inotify_init);
-
-/**
- * inotify_init_watch - initialize an inotify watch
- * @watch: watch to initialize
- */
-void inotify_init_watch(struct inotify_watch *watch)
-{
-	INIT_LIST_HEAD(&watch->h_list);
-	INIT_LIST_HEAD(&watch->i_list);
-	atomic_set(&watch->count, 0);
-	get_inotify_watch(watch); /* initial get */
-}
-EXPORT_SYMBOL_GPL(inotify_init_watch);
-
-/*
- * Watch removals suck violently.  To kick the watch out we need (in this
- * order) inode->inotify_mutex and ih->mutex.  That's fine if we have
- * a hold on inode; however, for all other cases we need to make damn sure
- * we don't race with umount.  We can *NOT* just grab a reference to a
- * watch - inotify_unmount_inodes() will happily sail past it and we'll end
- * with reference to inode potentially outliving its superblock.  Ideally
- * we just want to grab an active reference to superblock if we can; that
- * will make sure we won't go into inotify_umount_inodes() until we are
- * done.  Cleanup is just deactivate_super().  However, that leaves a messy
- * case - what if we *are* racing with umount() and active references to
- * superblock can't be acquired anymore?  We can bump ->s_count, grab
- * ->s_umount, which will almost certainly wait until the superblock is shut
- * down and the watch in question is pining for fjords.  That's fine, but
- * there is a problem - we might have hit the window between ->s_active
- * getting to 0 / ->s_count - below S_BIAS (i.e. the moment when superblock
- * is past the point of no return and is heading for shutdown) and the
- * moment when deactivate_super() acquires ->s_umount.  We could just do
- * drop_super() yield() and retry, but that's rather antisocial and this
- * stuff is luser-triggerable.  OTOH, having grabbed ->s_umount and having
- * found that we'd got there first (i.e. that ->s_root is non-NULL) we know
- * that we won't race with inotify_umount_inodes().  So we could grab a
- * reference to watch and do the rest as above, just with drop_super() instead
- * of deactivate_super(), right?  Wrong.  We had to drop ih->mutex before we
- * could grab ->s_umount.  So the watch could've been gone already.
- *
- * That still can be dealt with - we need to save watch->wd, do idr_find()
- * and compare its result with our pointer.  If they match, we either have
- * the damn thing still alive or we'd lost not one but two races at once,
- * the watch had been killed and a new one got created with the same ->wd
- * at the same address.  That couldn't have happened in inotify_destroy(),
- * but inotify_rm_wd() could run into that.  Still, "new one got created"
- * is not a problem - we have every right to kill it or leave it alone,
- * whatever's more convenient.
- *
- * So we can use idr_find(...) == watch && watch->inode->i_sb == sb as
- * "grab it and kill it" check.  If it's been our original watch, we are
- * fine, if it's a newcomer - nevermind, just pretend that we'd won the
- * race and kill the fscker anyway; we are safe since we know that its
- * superblock won't be going away.
- *
- * And yes, this is far beyond mere "not very pretty"; so's the entire
- * concept of inotify to start with.
- */
-
-/**
- * pin_to_kill - pin the watch down for removal
- * @ih: inotify handle
- * @watch: watch to kill
- *
- * Called with ih->mutex held, drops it.  Possible return values:
- * 0 - nothing to do, it has died
- * 1 - remove it, drop the reference and deactivate_super()
- * 2 - remove it, drop the reference and drop_super(); we tried hard to avoid
- * that variant, since it involved a lot of PITA, but that's the best that
- * could've been done.
- */
-static int pin_to_kill(struct inotify_handle *ih, struct inotify_watch *watch)
-{
-	struct super_block *sb = watch->inode->i_sb;
-	s32 wd = watch->wd;
-
-	spin_lock(&sb_lock);
-	if (sb->s_count >= S_BIAS) {
-		atomic_inc(&sb->s_active);
-		spin_unlock(&sb_lock);
-		get_inotify_watch(watch);
-		mutex_unlock(&ih->mutex);
-		return 1;	/* the best outcome */
-	}
-	sb->s_count++;
-	spin_unlock(&sb_lock);
-	mutex_unlock(&ih->mutex); /* can't grab ->s_umount under it */
-	down_read(&sb->s_umount);
-	if (likely(!sb->s_root)) {
-		/* fs is already shut down; the watch is dead */
-		drop_super(sb);
-		return 0;
-	}
-	/* raced with the final deactivate_super() */
-	mutex_lock(&ih->mutex);
-	if (idr_find(&ih->idr, wd) != watch || watch->inode->i_sb != sb) {
-		/* the watch is dead */
-		mutex_unlock(&ih->mutex);
-		drop_super(sb);
-		return 0;
-	}
-	/* still alive or freed and reused with the same sb and wd; kill */
-	get_inotify_watch(watch);
-	mutex_unlock(&ih->mutex);
-	return 2;
-}
-
-static void unpin_and_kill(struct inotify_watch *watch, int how)
-{
-	struct super_block *sb = watch->inode->i_sb;
-	put_inotify_watch(watch);
-	switch (how) {
-	case 1:
-		deactivate_super(sb);
-		break;
-	case 2:
-		drop_super(sb);
-	}
-}
-
-/**
- * inotify_destroy - clean up and destroy an inotify instance
- * @ih: inotify handle
- */
-void inotify_destroy(struct inotify_handle *ih)
-{
-	/*
-	 * Destroy all of the watches for this handle. Unfortunately, not very
-	 * pretty.  We cannot do a simple iteration over the list, because we
-	 * do not know the inode until we iterate to the watch.  But we need to
-	 * hold inode->inotify_mutex before ih->mutex.  The following works.
-	 *
-	 * AV: it had to become even uglier to start working ;-/
-	 */
-	while (1) {
-		struct inotify_watch *watch;
-		struct list_head *watches;
-		struct super_block *sb;
-		struct inode *inode;
-		int how;
-
-		mutex_lock(&ih->mutex);
-		watches = &ih->watches;
-		if (list_empty(watches)) {
-			mutex_unlock(&ih->mutex);
-			break;
-		}
-		watch = list_first_entry(watches, struct inotify_watch, h_list);
-		sb = watch->inode->i_sb;
-		how = pin_to_kill(ih, watch);
-		if (!how)
-			continue;
-
-		inode = watch->inode;
-		mutex_lock(&inode->inotify_mutex);
-		mutex_lock(&ih->mutex);
-
-		/* make sure we didn't race with another list removal */
-		if (likely(idr_find(&ih->idr, watch->wd))) {
-			remove_watch_no_event(watch, ih);
-			put_inotify_watch(watch);
-		}
-
-		mutex_unlock(&ih->mutex);
-		mutex_unlock(&inode->inotify_mutex);
-		unpin_and_kill(watch, how);
-	}
-
-	/* free this handle: the put matching the get in inotify_init() */
-	put_inotify_handle(ih);
-}
-EXPORT_SYMBOL_GPL(inotify_destroy);
-
-/**
- * inotify_find_watch - find an existing watch for an (ih,inode) pair
- * @ih: inotify handle
- * @inode: inode to watch
- * @watchp: pointer to existing inotify_watch
- *
- * Caller must pin given inode (via nameidata).
- */
-s32 inotify_find_watch(struct inotify_handle *ih, struct inode *inode,
-		       struct inotify_watch **watchp)
-{
-	struct inotify_watch *old;
-	int ret = -ENOENT;
-
-	mutex_lock(&inode->inotify_mutex);
-	mutex_lock(&ih->mutex);
-
-	old = inode_find_handle(inode, ih);
-	if (unlikely(old)) {
-		get_inotify_watch(old); /* caller must put watch */
-		*watchp = old;
-		ret = old->wd;
-	}
-
-	mutex_unlock(&ih->mutex);
-	mutex_unlock(&inode->inotify_mutex);
-
-	return ret;
-}
-EXPORT_SYMBOL_GPL(inotify_find_watch);
-
-/**
- * inotify_find_update_watch - find and update the mask of an existing watch
- * @ih: inotify handle
- * @inode: inode's watch to update
- * @mask: mask of events to watch
- *
- * Caller must pin given inode (via nameidata).
- */
-s32 inotify_find_update_watch(struct inotify_handle *ih, struct inode *inode,
-			      u32 mask)
-{
-	struct inotify_watch *old;
-	int mask_add = 0;
-	int ret;
-
-	if (mask & IN_MASK_ADD)
-		mask_add = 1;
-
-	/* don't allow invalid bits: we don't want flags set */
-	mask &= IN_ALL_EVENTS | IN_ONESHOT;
-	if (unlikely(!mask))
-		return -EINVAL;
-
-	mutex_lock(&inode->inotify_mutex);
-	mutex_lock(&ih->mutex);
-
-	/*
-	 * Handle the case of re-adding a watch on an (inode,ih) pair that we
-	 * are already watching.  We just update the mask and return its wd.
-	 */
-	old = inode_find_handle(inode, ih);
-	if (unlikely(!old)) {
-		ret = -ENOENT;
-		goto out;
-	}
-
-	if (mask_add)
-		old->mask |= mask;
-	else
-		old->mask = mask;
-	ret = old->wd;
-out:
-	mutex_unlock(&ih->mutex);
-	mutex_unlock(&inode->inotify_mutex);
-	return ret;
-}
-EXPORT_SYMBOL_GPL(inotify_find_update_watch);
-
-/**
- * inotify_add_watch - add a watch to an inotify instance
- * @ih: inotify handle
- * @watch: caller allocated watch structure
- * @inode: inode to watch
- * @mask: mask of events to watch
- *
- * Caller must pin given inode (via nameidata).
- * Caller must ensure it only calls inotify_add_watch() once per watch.
- * Calls inotify_handle_get_wd() so may sleep.
- */
-s32 inotify_add_watch(struct inotify_handle *ih, struct inotify_watch *watch,
-		      struct inode *inode, u32 mask)
-{
-	int ret = 0;
-	int newly_watched;
-
-	/* don't allow invalid bits: we don't want flags set */
-	mask &= IN_ALL_EVENTS | IN_ONESHOT;
-	if (unlikely(!mask))
-		return -EINVAL;
-	watch->mask = mask;
-
-	mutex_lock(&inode->inotify_mutex);
-	mutex_lock(&ih->mutex);
-
-	/* Initialize a new watch */
-	ret = inotify_handle_get_wd(ih, watch);
-	if (unlikely(ret))
-		goto out;
-	ret = watch->wd;
-
-	/* save a reference to handle and bump the count to make it official */
-	get_inotify_handle(ih);
-	watch->ih = ih;
-
-	/*
-	 * Save a reference to the inode and bump the ref count to make it
-	 * official.  We hold a reference to nameidata, which makes this safe.
-	 */
-	watch->inode = igrab(inode);
-
-	/* Add the watch to the handle's and the inode's list */
-	newly_watched = !inotify_inode_watched(inode);
-	list_add(&watch->h_list, &ih->watches);
-	list_add(&watch->i_list, &inode->inotify_watches);
-	/*
-	 * Set child flags _after_ adding the watch, so there is no race
-	 * windows where newly instantiated children could miss their parent's
-	 * watched flag.
-	 */
-	if (newly_watched)
-		set_dentry_child_flags(inode, 1);
-
-out:
-	mutex_unlock(&ih->mutex);
-	mutex_unlock(&inode->inotify_mutex);
-	return ret;
-}
-EXPORT_SYMBOL_GPL(inotify_add_watch);
-
-/**
- * inotify_clone_watch - put the watch next to existing one
- * @old: already installed watch
- * @new: new watch
- *
- * Caller must hold the inotify_mutex of inode we are dealing with;
- * it is expected to remove the old watch before unlocking the inode.
- */
-s32 inotify_clone_watch(struct inotify_watch *old, struct inotify_watch *new)
-{
-	struct inotify_handle *ih = old->ih;
-	int ret = 0;
-
-	new->mask = old->mask;
-	new->ih = ih;
-
-	mutex_lock(&ih->mutex);
-
-	/* Initialize a new watch */
-	ret = inotify_handle_get_wd(ih, new);
-	if (unlikely(ret))
-		goto out;
-	ret = new->wd;
-
-	get_inotify_handle(ih);
-
-	new->inode = igrab(old->inode);
-
-	list_add(&new->h_list, &ih->watches);
-	list_add(&new->i_list, &old->inode->inotify_watches);
-out:
-	mutex_unlock(&ih->mutex);
-	return ret;
-}
-
-void inotify_evict_watch(struct inotify_watch *watch)
-{
-	get_inotify_watch(watch);
-	mutex_lock(&watch->ih->mutex);
-	inotify_remove_watch_locked(watch->ih, watch);
-	mutex_unlock(&watch->ih->mutex);
-}
-
-/**
- * inotify_rm_wd - remove a watch from an inotify instance
- * @ih: inotify handle
- * @wd: watch descriptor to remove
- *
- * Can sleep.
- */
-int inotify_rm_wd(struct inotify_handle *ih, u32 wd)
-{
-	struct inotify_watch *watch;
-	struct super_block *sb;
-	struct inode *inode;
-	int how;
-
-	mutex_lock(&ih->mutex);
-	watch = idr_find(&ih->idr, wd);
-	if (unlikely(!watch)) {
-		mutex_unlock(&ih->mutex);
-		return -EINVAL;
-	}
-	sb = watch->inode->i_sb;
-	how = pin_to_kill(ih, watch);
-	if (!how)
-		return 0;
-
-	inode = watch->inode;
-
-	mutex_lock(&inode->inotify_mutex);
-	mutex_lock(&ih->mutex);
-
-	/* make sure that we did not race */
-	if (likely(idr_find(&ih->idr, wd) == watch))
-		inotify_remove_watch_locked(ih, watch);
-
-	mutex_unlock(&ih->mutex);
-	mutex_unlock(&inode->inotify_mutex);
-	unpin_and_kill(watch, how);
-
-	return 0;
-}
-EXPORT_SYMBOL_GPL(inotify_rm_wd);
-
-/**
- * inotify_rm_watch - remove a watch from an inotify instance
- * @ih: inotify handle
- * @watch: watch to remove
- *
- * Can sleep.
- */
-int inotify_rm_watch(struct inotify_handle *ih,
-		     struct inotify_watch *watch)
-{
-	return inotify_rm_wd(ih, watch->wd);
-}
-EXPORT_SYMBOL_GPL(inotify_rm_watch);
-
-/*
- * inotify_setup - core initialization function
- */
-static int __init inotify_setup(void)
-{
-	atomic_set(&inotify_cookie, 0);
-
-	return 0;
-}
-
-module_init(inotify_setup);
diff --git a/fs/notify/inotify/inotify.h b/fs/notify/inotify/inotify.h
new file mode 100644
index 00000000000..ed855ef6f07
--- /dev/null
+++ b/fs/notify/inotify/inotify.h
@@ -0,0 +1,32 @@
+#include <linux/fsnotify_backend.h>
+#include <linux/inotify.h>
+#include <linux/slab.h> /* struct kmem_cache */
+
+struct inotify_event_info {
+	struct fsnotify_event fse;
+	int wd;
+	u32 sync_cookie;
+	int name_len;
+	char name[];
+};
+
+struct inotify_inode_mark {
+	struct fsnotify_mark fsn_mark;
+	int wd;
+};
+
+static inline struct inotify_event_info *INOTIFY_E(struct fsnotify_event *fse)
+{
+	return container_of(fse, struct inotify_event_info, fse);
+}
+
+extern void inotify_ignored_and_remove_idr(struct fsnotify_mark *fsn_mark,
+					   struct fsnotify_group *group);
+extern int inotify_handle_event(struct fsnotify_group *group,
+				struct inode *inode,
+				struct fsnotify_mark *inode_mark,
+				struct fsnotify_mark *vfsmount_mark,
+				u32 mask, void *data, int data_type,
+				const unsigned char *file_name, u32 cookie);
+
+extern const struct fsnotify_ops inotify_fsnotify_ops;
diff --git a/fs/notify/inotify/inotify_fsnotify.c b/fs/notify/inotify/inotify_fsnotify.c
new file mode 100644
index 00000000000..43ab1e1a07a
--- /dev/null
+++ b/fs/notify/inotify/inotify_fsnotify.c
@@ -0,0 +1,182 @@
+/*
+ * fs/inotify_user.c - inotify support for userspace
+ *
+ * Authors:
+ *	John McCutchan	<ttb@tentacle.dhs.org>
+ *	Robert Love	<rml@novell.com>
+ *
+ * Copyright (C) 2005 John McCutchan
+ * Copyright 2006 Hewlett-Packard Development Company, L.P.
+ *
+ * Copyright (C) 2009 Eric Paris <Red Hat Inc>
+ * inotify was largely rewriten to make use of the fsnotify infrastructure
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2, or (at your option) any
+ * later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ */
+
+#include <linux/dcache.h> /* d_unlinked */
+#include <linux/fs.h> /* struct inode */
+#include <linux/fsnotify_backend.h>
+#include <linux/inotify.h>
+#include <linux/path.h> /* struct path */
+#include <linux/slab.h> /* kmem_* */
+#include <linux/types.h>
+#include <linux/sched.h>
+
+#include "inotify.h"
+
+/*
+ * Check if 2 events contain the same information.
+ */
+static bool event_compare(struct fsnotify_event *old_fsn,
+			  struct fsnotify_event *new_fsn)
+{
+	struct inotify_event_info *old, *new;
+
+	if (old_fsn->mask & FS_IN_IGNORED)
+		return false;
+	old = INOTIFY_E(old_fsn);
+	new = INOTIFY_E(new_fsn);
+	if ((old_fsn->mask == new_fsn->mask) &&
+	    (old_fsn->inode == new_fsn->inode) &&
+	    (old->name_len == new->name_len) &&
+	    (!old->name_len || !strcmp(old->name, new->name)))
+		return true;
+	return false;
+}
+
+static int inotify_merge(struct list_head *list,
+			  struct fsnotify_event *event)
+{
+	struct fsnotify_event *last_event;
+
+	last_event = list_entry(list->prev, struct fsnotify_event, list);
+	return event_compare(last_event, event);
+}
+
+int inotify_handle_event(struct fsnotify_group *group,
+			 struct inode *inode,
+			 struct fsnotify_mark *inode_mark,
+			 struct fsnotify_mark *vfsmount_mark,
+			 u32 mask, void *data, int data_type,
+			 const unsigned char *file_name, u32 cookie)
+{
+	struct inotify_inode_mark *i_mark;
+	struct inotify_event_info *event;
+	struct fsnotify_event *fsn_event;
+	int ret;
+	int len = 0;
+	int alloc_len = sizeof(struct inotify_event_info);
+
+	BUG_ON(vfsmount_mark);
+
+	if ((inode_mark->mask & FS_EXCL_UNLINK) &&
+	    (data_type == FSNOTIFY_EVENT_PATH)) {
+		struct path *path = data;
+
+		if (d_unlinked(path->dentry))
+			return 0;
+	}
+	if (file_name) {
+		len = strlen(file_name);
+		alloc_len += len + 1;
+	}
+
+	pr_debug("%s: group=%p inode=%p mask=%x\n", __func__, group, inode,
+		 mask);
+
+	i_mark = container_of(inode_mark, struct inotify_inode_mark,
+			      fsn_mark);
+
+	event = kmalloc(alloc_len, GFP_KERNEL);
+	if (unlikely(!event))
+		return -ENOMEM;
+
+	fsn_event = &event->fse;
+	fsnotify_init_event(fsn_event, inode, mask);
+	event->wd = i_mark->wd;
+	event->sync_cookie = cookie;
+	event->name_len = len;
+	if (len)
+		strcpy(event->name, file_name);
+
+	ret = fsnotify_add_notify_event(group, fsn_event, inotify_merge);
+	if (ret) {
+		/* Our event wasn't used in the end. Free it. */
+		fsnotify_destroy_event(group, fsn_event);
+	}
+
+	if (inode_mark->mask & IN_ONESHOT)
+		fsnotify_destroy_mark(inode_mark, group);
+
+	return 0;
+}
+
+static void inotify_freeing_mark(struct fsnotify_mark *fsn_mark, struct fsnotify_group *group)
+{
+	inotify_ignored_and_remove_idr(fsn_mark, group);
+}
+
+/*
+ * This is NEVER supposed to be called.  Inotify marks should either have been
+ * removed from the idr when the watch was removed or in the
+ * fsnotify_destroy_mark_by_group() call when the inotify instance was being
+ * torn down.  This is only called if the idr is about to be freed but there
+ * are still marks in it.
+ */
+static int idr_callback(int id, void *p, void *data)
+{
+	struct fsnotify_mark *fsn_mark;
+	struct inotify_inode_mark *i_mark;
+	static bool warned = false;
+
+	if (warned)
+		return 0;
+
+	warned = true;
+	fsn_mark = p;
+	i_mark = container_of(fsn_mark, struct inotify_inode_mark, fsn_mark);
+
+	WARN(1, "inotify closing but id=%d for fsn_mark=%p in group=%p still in "
+		"idr.  Probably leaking memory\n", id, p, data);
+
+	/*
+	 * I'm taking the liberty of assuming that the mark in question is a
+	 * valid address and I'm dereferencing it.  This might help to figure
+	 * out why we got here and the panic is no worse than the original
+	 * BUG() that was here.
+	 */
+	if (fsn_mark)
+		printk(KERN_WARNING "fsn_mark->group=%p inode=%p wd=%d\n",
+			fsn_mark->group, fsn_mark->i.inode, i_mark->wd);
+	return 0;
+}
+
+static void inotify_free_group_priv(struct fsnotify_group *group)
+{
+	/* ideally the idr is empty and we won't hit the BUG in the callback */
+	idr_for_each(&group->inotify_data.idr, idr_callback, group);
+	idr_destroy(&group->inotify_data.idr);
+	atomic_dec(&group->inotify_data.user->inotify_devs);
+	free_uid(group->inotify_data.user);
+}
+
+static void inotify_free_event(struct fsnotify_event *fsn_event)
+{
+	kfree(INOTIFY_E(fsn_event));
+}
+
+const struct fsnotify_ops inotify_fsnotify_ops = {
+	.handle_event = inotify_handle_event,
+	.free_group_priv = inotify_free_group_priv,
+	.free_event = inotify_free_event,
+	.freeing_mark = inotify_freeing_mark,
+};
diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c
index bed766e435b..cc423a30a0c 100644
--- a/fs/notify/inotify/inotify_user.c
+++ b/fs/notify/inotify/inotify_user.c
@@ -8,6 +8,9 @@
  * Copyright (C) 2005 John McCutchan
  * Copyright 2006 Hewlett-Packard Development Company, L.P.
  *
+ * Copyright (C) 2009 Eric Paris <Red Hat Inc>
+ * inotify was largely rewriten to make use of the fsnotify infrastructure
+ *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of the GNU General Public License as published by the
  * Free Software Foundation; either version 2, or (at your option) any
@@ -19,94 +22,34 @@
  * General Public License for more details.
  */
 
-#include <linux/kernel.h>
-#include <linux/sched.h>
-#include <linux/slab.h>
-#include <linux/fs.h>
 #include <linux/file.h>
-#include <linux/mount.h>
-#include <linux/namei.h>
-#include <linux/poll.h>
-#include <linux/init.h>
-#include <linux/list.h>
+#include <linux/fs.h> /* struct inode */
+#include <linux/fsnotify_backend.h>
+#include <linux/idr.h>
+#include <linux/init.h> /* module_init */
 #include <linux/inotify.h>
+#include <linux/kernel.h> /* roundup() */
+#include <linux/namei.h> /* LOOKUP_FOLLOW */
+#include <linux/sched.h> /* struct user */
+#include <linux/slab.h> /* struct kmem_cache */
 #include <linux/syscalls.h>
-#include <linux/magic.h>
-
-#include <asm/ioctls.h>
+#include <linux/types.h>
+#include <linux/anon_inodes.h>
+#include <linux/uaccess.h>
+#include <linux/poll.h>
+#include <linux/wait.h>
 
-static struct kmem_cache *watch_cachep __read_mostly;
-static struct kmem_cache *event_cachep __read_mostly;
+#include "inotify.h"
+#include "../fdinfo.h"
 
-static struct vfsmount *inotify_mnt __read_mostly;
+#include <asm/ioctls.h>
 
 /* these are configurable via /proc/sys/fs/inotify/ */
 static int inotify_max_user_instances __read_mostly;
-static int inotify_max_user_watches __read_mostly;
 static int inotify_max_queued_events __read_mostly;
+static int inotify_max_user_watches __read_mostly;
 
-/*
- * Lock ordering:
- *
- * inotify_dev->up_mutex (ensures we don't re-add the same watch)
- * 	inode->inotify_mutex (protects inode's watch list)
- * 		inotify_handle->mutex (protects inotify_handle's watch list)
- * 			inotify_dev->ev_mutex (protects device's event queue)
- */
-
-/*
- * Lifetimes of the main data structures:
- *
- * inotify_device: Lifetime is managed by reference count, from
- * sys_inotify_init() until release.  Additional references can bump the count
- * via get_inotify_dev() and drop the count via put_inotify_dev().
- *
- * inotify_user_watch: Lifetime is from create_watch() to the receipt of an
- * IN_IGNORED event from inotify, or when using IN_ONESHOT, to receipt of the
- * first event, or to inotify_destroy().
- */
-
-/*
- * struct inotify_device - represents an inotify instance
- *
- * This structure is protected by the mutex 'mutex'.
- */
-struct inotify_device {
-	wait_queue_head_t 	wq;		/* wait queue for i/o */
-	struct mutex		ev_mutex;	/* protects event queue */
-	struct mutex		up_mutex;	/* synchronizes watch updates */
-	struct list_head 	events;		/* list of queued events */
-	struct user_struct	*user;		/* user who opened this dev */
-	struct inotify_handle	*ih;		/* inotify handle */
-	struct fasync_struct    *fa;            /* async notification */
-	atomic_t		count;		/* reference count */
-	unsigned int		queue_size;	/* size of the queue (bytes) */
-	unsigned int		event_count;	/* number of pending events */
-	unsigned int		max_events;	/* maximum number of events */
-};
-
-/*
- * struct inotify_kernel_event - An inotify event, originating from a watch and
- * queued for user-space.  A list of these is attached to each instance of the
- * device.  In read(), this list is walked and all events that can fit in the
- * buffer are returned.
- *
- * Protected by dev->ev_mutex of the device in which we are queued.
- */
-struct inotify_kernel_event {
-	struct inotify_event	event;	/* the user-space event */
-	struct list_head        list;	/* entry in inotify_device's list */
-	char			*name;	/* filename, if any */
-};
-
-/*
- * struct inotify_user_watch - our version of an inotify_watch, we add
- * a reference to the associated inotify_device.
- */
-struct inotify_user_watch {
-	struct inotify_device	*dev;	/* associated device */
-	struct inotify_watch	wdata;	/* inotify watch data */
-};
+static struct kmem_cache *inotify_inode_mark_cachep __read_mostly;
 
 #ifdef CONFIG_SYSCTL
 
@@ -114,257 +57,280 @@ struct inotify_user_watch {
 
 static int zero;
 
-ctl_table inotify_table[] = {
+struct ctl_table inotify_table[] = {
 	{
-		.ctl_name	= INOTIFY_MAX_USER_INSTANCES,
 		.procname	= "max_user_instances",
 		.data		= &inotify_max_user_instances,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
-		.proc_handler	= &proc_dointvec_minmax,
-		.strategy	= &sysctl_intvec,
+		.proc_handler	= proc_dointvec_minmax,
 		.extra1		= &zero,
 	},
 	{
-		.ctl_name	= INOTIFY_MAX_USER_WATCHES,
 		.procname	= "max_user_watches",
 		.data		= &inotify_max_user_watches,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
-		.proc_handler	= &proc_dointvec_minmax,
-		.strategy	= &sysctl_intvec,
+		.proc_handler	= proc_dointvec_minmax,
 		.extra1		= &zero,
 	},
 	{
-		.ctl_name	= INOTIFY_MAX_QUEUED_EVENTS,
 		.procname	= "max_queued_events",
 		.data		= &inotify_max_queued_events,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
-		.proc_handler	= &proc_dointvec_minmax,
-		.strategy	= &sysctl_intvec,
+		.proc_handler	= proc_dointvec_minmax,
 		.extra1		= &zero
 	},
-	{ .ctl_name = 0 }
+	{ }
 };
 #endif /* CONFIG_SYSCTL */
 
-static inline void get_inotify_dev(struct inotify_device *dev)
+static inline __u32 inotify_arg_to_mask(u32 arg)
 {
-	atomic_inc(&dev->count);
+	__u32 mask;
+
+	/*
+	 * everything should accept their own ignored, cares about children,
+	 * and should receive events when the inode is unmounted
+	 */
+	mask = (FS_IN_IGNORED | FS_EVENT_ON_CHILD | FS_UNMOUNT);
+
+	/* mask off the flags used to open the fd */
+	mask |= (arg & (IN_ALL_EVENTS | IN_ONESHOT | IN_EXCL_UNLINK));
+
+	return mask;
 }
 
-static inline void put_inotify_dev(struct inotify_device *dev)
+static inline u32 inotify_mask_to_arg(__u32 mask)
 {
-	if (atomic_dec_and_test(&dev->count)) {
-		atomic_dec(&dev->user->inotify_devs);
-		free_uid(dev->user);
-		kfree(dev);
-	}
+	return mask & (IN_ALL_EVENTS | IN_ISDIR | IN_UNMOUNT | IN_IGNORED |
+		       IN_Q_OVERFLOW);
 }
 
-/*
- * free_inotify_user_watch - cleans up the watch and its references
- */
-static void free_inotify_user_watch(struct inotify_watch *w)
+/* intofiy userspace file descriptor functions */
+static unsigned int inotify_poll(struct file *file, poll_table *wait)
 {
-	struct inotify_user_watch *watch;
-	struct inotify_device *dev;
+	struct fsnotify_group *group = file->private_data;
+	int ret = 0;
+
+	poll_wait(file, &group->notification_waitq, wait);
+	mutex_lock(&group->notification_mutex);
+	if (!fsnotify_notify_queue_is_empty(group))
+		ret = POLLIN | POLLRDNORM;
+	mutex_unlock(&group->notification_mutex);
+
+	return ret;
+}
 
-	watch = container_of(w, struct inotify_user_watch, wdata);
-	dev = watch->dev;
+static int round_event_name_len(struct fsnotify_event *fsn_event)
+{
+	struct inotify_event_info *event;
 
-	atomic_dec(&dev->user->inotify_watches);
-	put_inotify_dev(dev);
-	kmem_cache_free(watch_cachep, watch);
+	event = INOTIFY_E(fsn_event);
+	if (!event->name_len)
+		return 0;
+	return roundup(event->name_len + 1, sizeof(struct inotify_event));
 }
 
 /*
- * kernel_event - create a new kernel event with the given parameters
+ * Get an inotify_kernel_event if one exists and is small
+ * enough to fit in "count". Return an error pointer if
+ * not large enough.
  *
- * This function can sleep.
+ * Called with the group->notification_mutex held.
  */
-static struct inotify_kernel_event * kernel_event(s32 wd, u32 mask, u32 cookie,
-						  const char *name)
+static struct fsnotify_event *get_one_event(struct fsnotify_group *group,
+					    size_t count)
 {
-	struct inotify_kernel_event *kevent;
+	size_t event_size = sizeof(struct inotify_event);
+	struct fsnotify_event *event;
 
-	kevent = kmem_cache_alloc(event_cachep, GFP_NOFS);
-	if (unlikely(!kevent))
+	if (fsnotify_notify_queue_is_empty(group))
 		return NULL;
 
-	/* we hand this out to user-space, so zero it just in case */
-	memset(&kevent->event, 0, sizeof(struct inotify_event));
-
-	kevent->event.wd = wd;
-	kevent->event.mask = mask;
-	kevent->event.cookie = cookie;
-
-	INIT_LIST_HEAD(&kevent->list);
-
-	if (name) {
-		size_t len, rem, event_size = sizeof(struct inotify_event);
-
-		/*
-		 * We need to pad the filename so as to properly align an
-		 * array of inotify_event structures.  Because the structure is
-		 * small and the common case is a small filename, we just round
-		 * up to the next multiple of the structure's sizeof.  This is
-		 * simple and safe for all architectures.
-		 */
-		len = strlen(name) + 1;
-		rem = event_size - len;
-		if (len > event_size) {
-			rem = event_size - (len % event_size);
-			if (len % event_size == 0)
-				rem = 0;
-		}
+	event = fsnotify_peek_notify_event(group);
 
-		kevent->name = kmalloc(len + rem, GFP_KERNEL);
-		if (unlikely(!kevent->name)) {
-			kmem_cache_free(event_cachep, kevent);
-			return NULL;
-		}
-		memcpy(kevent->name, name, len);
-		if (rem)
-			memset(kevent->name + len, 0, rem);
-		kevent->event.len = len + rem;
-	} else {
-		kevent->event.len = 0;
-		kevent->name = NULL;
-	}
+	pr_debug("%s: group=%p event=%p\n", __func__, group, event);
 
-	return kevent;
-}
+	event_size += round_event_name_len(event);
+	if (event_size > count)
+		return ERR_PTR(-EINVAL);
 
-/*
- * inotify_dev_get_event - return the next event in the given dev's queue
- *
- * Caller must hold dev->ev_mutex.
- */
-static inline struct inotify_kernel_event *
-inotify_dev_get_event(struct inotify_device *dev)
-{
-	return list_entry(dev->events.next, struct inotify_kernel_event, list);
-}
+	/* held the notification_mutex the whole time, so this is the
+	 * same event we peeked above */
+	fsnotify_remove_notify_event(group);
 
-/*
- * inotify_dev_get_last_event - return the last event in the given dev's queue
- *
- * Caller must hold dev->ev_mutex.
- */
-static inline struct inotify_kernel_event *
-inotify_dev_get_last_event(struct inotify_device *dev)
-{
-	if (list_empty(&dev->events))
-		return NULL;
-	return list_entry(dev->events.prev, struct inotify_kernel_event, list);
+	return event;
 }
 
 /*
- * inotify_dev_queue_event - event handler registered with core inotify, adds
- * a new event to the given device
+ * Copy an event to user space, returning how much we copied.
  *
- * Can sleep (calls kernel_event()).
+ * We already checked that the event size is smaller than the
+ * buffer we had in "get_one_event()" above.
  */
-static void inotify_dev_queue_event(struct inotify_watch *w, u32 wd, u32 mask,
-				    u32 cookie, const char *name,
-				    struct inode *ignored)
+static ssize_t copy_event_to_user(struct fsnotify_group *group,
+				  struct fsnotify_event *fsn_event,
+				  char __user *buf)
 {
-	struct inotify_user_watch *watch;
-	struct inotify_device *dev;
-	struct inotify_kernel_event *kevent, *last;
+	struct inotify_event inotify_event;
+	struct inotify_event_info *event;
+	size_t event_size = sizeof(struct inotify_event);
+	size_t name_len;
+	size_t pad_name_len;
 
-	watch = container_of(w, struct inotify_user_watch, wdata);
-	dev = watch->dev;
+	pr_debug("%s: group=%p event=%p\n", __func__, group, fsn_event);
 
-	mutex_lock(&dev->ev_mutex);
+	event = INOTIFY_E(fsn_event);
+	name_len = event->name_len;
+	/*
+	 * round up name length so it is a multiple of event_size
+	 * plus an extra byte for the terminating '\0'.
+	 */
+	pad_name_len = round_event_name_len(fsn_event);
+	inotify_event.len = pad_name_len;
+	inotify_event.mask = inotify_mask_to_arg(fsn_event->mask);
+	inotify_event.wd = event->wd;
+	inotify_event.cookie = event->sync_cookie;
+
+	/* send the main event */
+	if (copy_to_user(buf, &inotify_event, event_size))
+		return -EFAULT;
+
+	buf += event_size;
 
-	/* we can safely put the watch as we don't reference it while
-	 * generating the event
+	/*
+	 * fsnotify only stores the pathname, so here we have to send the pathname
+	 * and then pad that pathname out to a multiple of sizeof(inotify_event)
+	 * with zeros.
 	 */
-	if (mask & IN_IGNORED || w->mask & IN_ONESHOT)
-		put_inotify_watch(w); /* final put */
-
-	/* coalescing: drop this event if it is a dupe of the previous */
-	last = inotify_dev_get_last_event(dev);
-	if (last && last->event.mask == mask && last->event.wd == wd &&
-			last->event.cookie == cookie) {
-		const char *lastname = last->name;
-
-		if (!name && !lastname)
-			goto out;
-		if (name && lastname && !strcmp(lastname, name))
-			goto out;
+	if (pad_name_len) {
+		/* copy the path name */
+		if (copy_to_user(buf, event->name, name_len))
+			return -EFAULT;
+		buf += name_len;
+
+		/* fill userspace with 0's */
+		if (clear_user(buf, pad_name_len - name_len))
+			return -EFAULT;
+		event_size += pad_name_len;
 	}
 
-	/* the queue overflowed and we already sent the Q_OVERFLOW event */
-	if (unlikely(dev->event_count > dev->max_events))
-		goto out;
+	return event_size;
+}
 
-	/* if the queue overflows, we need to notify user space */
-	if (unlikely(dev->event_count == dev->max_events))
-		kevent = kernel_event(-1, IN_Q_OVERFLOW, cookie, NULL);
-	else
-		kevent = kernel_event(wd, mask, cookie, name);
+static ssize_t inotify_read(struct file *file, char __user *buf,
+			    size_t count, loff_t *pos)
+{
+	struct fsnotify_group *group;
+	struct fsnotify_event *kevent;
+	char __user *start;
+	int ret;
+	DEFINE_WAIT(wait);
 
-	if (unlikely(!kevent))
-		goto out;
+	start = buf;
+	group = file->private_data;
 
-	/* queue the event and wake up anyone waiting */
-	dev->event_count++;
-	dev->queue_size += sizeof(struct inotify_event) + kevent->event.len;
-	list_add_tail(&kevent->list, &dev->events);
-	wake_up_interruptible(&dev->wq);
-	kill_fasync(&dev->fa, SIGIO, POLL_IN);
+	while (1) {
+		prepare_to_wait(&group->notification_waitq, &wait, TASK_INTERRUPTIBLE);
 
-out:
-	mutex_unlock(&dev->ev_mutex);
-}
+		mutex_lock(&group->notification_mutex);
+		kevent = get_one_event(group, count);
+		mutex_unlock(&group->notification_mutex);
 
-/*
- * remove_kevent - cleans up the given kevent
- *
- * Caller must hold dev->ev_mutex.
- */
-static void remove_kevent(struct inotify_device *dev,
-			  struct inotify_kernel_event *kevent)
-{
-	list_del(&kevent->list);
+		pr_debug("%s: group=%p kevent=%p\n", __func__, group, kevent);
+
+		if (kevent) {
+			ret = PTR_ERR(kevent);
+			if (IS_ERR(kevent))
+				break;
+			ret = copy_event_to_user(group, kevent, buf);
+			fsnotify_destroy_event(group, kevent);
+			if (ret < 0)
+				break;
+			buf += ret;
+			count -= ret;
+			continue;
+		}
+
+		ret = -EAGAIN;
+		if (file->f_flags & O_NONBLOCK)
+			break;
+		ret = -ERESTARTSYS;
+		if (signal_pending(current))
+			break;
 
-	dev->event_count--;
-	dev->queue_size -= sizeof(struct inotify_event) + kevent->event.len;
+		if (start != buf)
+			break;
+
+		schedule();
+	}
+
+	finish_wait(&group->notification_waitq, &wait);
+	if (start != buf && ret != -EFAULT)
+		ret = buf - start;
+	return ret;
 }
 
-/*
- * free_kevent - frees the given kevent.
- */
-static void free_kevent(struct inotify_kernel_event *kevent)
+static int inotify_release(struct inode *ignored, struct file *file)
 {
-	kfree(kevent->name);
-	kmem_cache_free(event_cachep, kevent);
+	struct fsnotify_group *group = file->private_data;
+
+	pr_debug("%s: group=%p\n", __func__, group);
+
+	/* free this group, matching get was inotify_init->fsnotify_obtain_group */
+	fsnotify_destroy_group(group);
+
+	return 0;
 }
 
-/*
- * inotify_dev_event_dequeue - destroy an event on the given device
- *
- * Caller must hold dev->ev_mutex.
- */
-static void inotify_dev_event_dequeue(struct inotify_device *dev)
+static long inotify_ioctl(struct file *file, unsigned int cmd,
+			  unsigned long arg)
 {
-	if (!list_empty(&dev->events)) {
-		struct inotify_kernel_event *kevent;
-		kevent = inotify_dev_get_event(dev);
-		remove_kevent(dev, kevent);
-		free_kevent(kevent);
+	struct fsnotify_group *group;
+	struct fsnotify_event *fsn_event;
+	void __user *p;
+	int ret = -ENOTTY;
+	size_t send_len = 0;
+
+	group = file->private_data;
+	p = (void __user *) arg;
+
+	pr_debug("%s: group=%p cmd=%u\n", __func__, group, cmd);
+
+	switch (cmd) {
+	case FIONREAD:
+		mutex_lock(&group->notification_mutex);
+		list_for_each_entry(fsn_event, &group->notification_list,
+				    list) {
+			send_len += sizeof(struct inotify_event);
+			send_len += round_event_name_len(fsn_event);
+		}
+		mutex_unlock(&group->notification_mutex);
+		ret = put_user(send_len, (int __user *) p);
+		break;
 	}
+
+	return ret;
 }
 
+static const struct file_operations inotify_fops = {
+	.show_fdinfo	= inotify_show_fdinfo,
+	.poll		= inotify_poll,
+	.read		= inotify_read,
+	.fasync		= fsnotify_fasync,
+	.release	= inotify_release,
+	.unlocked_ioctl	= inotify_ioctl,
+	.compat_ioctl	= inotify_ioctl,
+	.llseek		= noop_llseek,
+};
+
+
 /*
  * find_inode - resolve a user-given path to a specific inode
  */
-static int find_inode(const char __user *dirname, struct path *path,
-		      unsigned flags)
+static int inotify_find_inode(const char __user *dirname, struct path *path, unsigned flags)
 {
 	int error;
 
@@ -378,224 +344,333 @@ static int find_inode(const char __user *dirname, struct path *path,
 	return error;
 }
 
-/*
- * create_watch - creates a watch on the given device.
- *
- * Callers must hold dev->up_mutex.
- */
-static int create_watch(struct inotify_device *dev, struct inode *inode,
-			u32 mask)
+static int inotify_add_to_idr(struct idr *idr, spinlock_t *idr_lock,
+			      struct inotify_inode_mark *i_mark)
 {
-	struct inotify_user_watch *watch;
 	int ret;
 
-	if (atomic_read(&dev->user->inotify_watches) >=
-			inotify_max_user_watches)
-		return -ENOSPC;
+	idr_preload(GFP_KERNEL);
+	spin_lock(idr_lock);
 
-	watch = kmem_cache_alloc(watch_cachep, GFP_KERNEL);
-	if (unlikely(!watch))
-		return -ENOMEM;
+	ret = idr_alloc_cyclic(idr, i_mark, 1, 0, GFP_NOWAIT);
+	if (ret >= 0) {
+		/* we added the mark to the idr, take a reference */
+		i_mark->wd = ret;
+		fsnotify_get_mark(&i_mark->fsn_mark);
+	}
 
-	/* save a reference to device and bump the count to make it official */
-	get_inotify_dev(dev);
-	watch->dev = dev;
+	spin_unlock(idr_lock);
+	idr_preload_end();
+	return ret < 0 ? ret : 0;
+}
 
-	atomic_inc(&dev->user->inotify_watches);
+static struct inotify_inode_mark *inotify_idr_find_locked(struct fsnotify_group *group,
+								int wd)
+{
+	struct idr *idr = &group->inotify_data.idr;
+	spinlock_t *idr_lock = &group->inotify_data.idr_lock;
+	struct inotify_inode_mark *i_mark;
 
-	inotify_init_watch(&watch->wdata);
-	ret = inotify_add_watch(dev->ih, &watch->wdata, inode, mask);
-	if (ret < 0)
-		free_inotify_user_watch(&watch->wdata);
+	assert_spin_locked(idr_lock);
 
-	return ret;
+	i_mark = idr_find(idr, wd);
+	if (i_mark) {
+		struct fsnotify_mark *fsn_mark = &i_mark->fsn_mark;
+
+		fsnotify_get_mark(fsn_mark);
+		/* One ref for being in the idr, one ref we just took */
+		BUG_ON(atomic_read(&fsn_mark->refcnt) < 2);
+	}
+
+	return i_mark;
 }
 
-/* Device Interface */
+static struct inotify_inode_mark *inotify_idr_find(struct fsnotify_group *group,
+							 int wd)
+{
+	struct inotify_inode_mark *i_mark;
+	spinlock_t *idr_lock = &group->inotify_data.idr_lock;
 
-static unsigned int inotify_poll(struct file *file, poll_table *wait)
+	spin_lock(idr_lock);
+	i_mark = inotify_idr_find_locked(group, wd);
+	spin_unlock(idr_lock);
+
+	return i_mark;
+}
+
+static void do_inotify_remove_from_idr(struct fsnotify_group *group,
+				       struct inotify_inode_mark *i_mark)
 {
-	struct inotify_device *dev = file->private_data;
-	int ret = 0;
+	struct idr *idr = &group->inotify_data.idr;
+	spinlock_t *idr_lock = &group->inotify_data.idr_lock;
+	int wd = i_mark->wd;
 
-	poll_wait(file, &dev->wq, wait);
-	mutex_lock(&dev->ev_mutex);
-	if (!list_empty(&dev->events))
-		ret = POLLIN | POLLRDNORM;
-	mutex_unlock(&dev->ev_mutex);
+	assert_spin_locked(idr_lock);
 
-	return ret;
+	idr_remove(idr, wd);
+
+	/* removed from the idr, drop that ref */
+	fsnotify_put_mark(&i_mark->fsn_mark);
 }
 
 /*
- * Get an inotify_kernel_event if one exists and is small
- * enough to fit in "count". Return an error pointer if
- * not large enough.
- *
- * Called with the device ev_mutex held.
+ * Remove the mark from the idr (if present) and drop the reference
+ * on the mark because it was in the idr.
  */
-static struct inotify_kernel_event *get_one_event(struct inotify_device *dev,
-						  size_t count)
+static void inotify_remove_from_idr(struct fsnotify_group *group,
+				    struct inotify_inode_mark *i_mark)
 {
-	size_t event_size = sizeof(struct inotify_event);
-	struct inotify_kernel_event *kevent;
+	spinlock_t *idr_lock = &group->inotify_data.idr_lock;
+	struct inotify_inode_mark *found_i_mark = NULL;
+	int wd;
 
-	if (list_empty(&dev->events))
-		return NULL;
+	spin_lock(idr_lock);
+	wd = i_mark->wd;
 
-	kevent = inotify_dev_get_event(dev);
-	if (kevent->name)
-		event_size += kevent->event.len;
+	/*
+	 * does this i_mark think it is in the idr?  we shouldn't get called
+	 * if it wasn't....
+	 */
+	if (wd == -1) {
+		WARN_ONCE(1, "%s: i_mark=%p i_mark->wd=%d i_mark->group=%p"
+			" i_mark->inode=%p\n", __func__, i_mark, i_mark->wd,
+			i_mark->fsn_mark.group, i_mark->fsn_mark.i.inode);
+		goto out;
+	}
 
-	if (event_size > count)
-		return ERR_PTR(-EINVAL);
+	/* Lets look in the idr to see if we find it */
+	found_i_mark = inotify_idr_find_locked(group, wd);
+	if (unlikely(!found_i_mark)) {
+		WARN_ONCE(1, "%s: i_mark=%p i_mark->wd=%d i_mark->group=%p"
+			" i_mark->inode=%p\n", __func__, i_mark, i_mark->wd,
+			i_mark->fsn_mark.group, i_mark->fsn_mark.i.inode);
+		goto out;
+	}
 
-	remove_kevent(dev, kevent);
-	return kevent;
+	/*
+	 * We found an mark in the idr at the right wd, but it's
+	 * not the mark we were told to remove.  eparis seriously
+	 * fucked up somewhere.
+	 */
+	if (unlikely(found_i_mark != i_mark)) {
+		WARN_ONCE(1, "%s: i_mark=%p i_mark->wd=%d i_mark->group=%p "
+			"mark->inode=%p found_i_mark=%p found_i_mark->wd=%d "
+			"found_i_mark->group=%p found_i_mark->inode=%p\n",
+			__func__, i_mark, i_mark->wd, i_mark->fsn_mark.group,
+			i_mark->fsn_mark.i.inode, found_i_mark, found_i_mark->wd,
+			found_i_mark->fsn_mark.group,
+			found_i_mark->fsn_mark.i.inode);
+		goto out;
+	}
+
+	/*
+	 * One ref for being in the idr
+	 * one ref held by the caller trying to kill us
+	 * one ref grabbed by inotify_idr_find
+	 */
+	if (unlikely(atomic_read(&i_mark->fsn_mark.refcnt) < 3)) {
+		printk(KERN_ERR "%s: i_mark=%p i_mark->wd=%d i_mark->group=%p"
+			" i_mark->inode=%p\n", __func__, i_mark, i_mark->wd,
+			i_mark->fsn_mark.group, i_mark->fsn_mark.i.inode);
+		/* we can't really recover with bad ref cnting.. */
+		BUG();
+	}
+
+	do_inotify_remove_from_idr(group, i_mark);
+out:
+	/* match the ref taken by inotify_idr_find_locked() */
+	if (found_i_mark)
+		fsnotify_put_mark(&found_i_mark->fsn_mark);
+	i_mark->wd = -1;
+	spin_unlock(idr_lock);
 }
 
 /*
- * Copy an event to user space, returning how much we copied.
- *
- * We already checked that the event size is smaller than the
- * buffer we had in "get_one_event()" above.
+ * Send IN_IGNORED for this wd, remove this wd from the idr.
  */
-static ssize_t copy_event_to_user(struct inotify_kernel_event *kevent,
-				  char __user *buf)
+void inotify_ignored_and_remove_idr(struct fsnotify_mark *fsn_mark,
+				    struct fsnotify_group *group)
 {
-	size_t event_size = sizeof(struct inotify_event);
+	struct inotify_inode_mark *i_mark;
 
-	if (copy_to_user(buf, &kevent->event, event_size))
-		return -EFAULT;
+	/* Queue ignore event for the watch */
+	inotify_handle_event(group, NULL, fsn_mark, NULL, FS_IN_IGNORED,
+			     NULL, FSNOTIFY_EVENT_NONE, NULL, 0);
 
-	if (kevent->name) {
-		buf += event_size;
+	i_mark = container_of(fsn_mark, struct inotify_inode_mark, fsn_mark);
+	/* remove this mark from the idr */
+	inotify_remove_from_idr(group, i_mark);
 
-		if (copy_to_user(buf, kevent->name, kevent->event.len))
-			return -EFAULT;
+	atomic_dec(&group->inotify_data.user->inotify_watches);
+}
 
-		event_size += kevent->event.len;
-	}
-	return event_size;
+/* ding dong the mark is dead */
+static void inotify_free_mark(struct fsnotify_mark *fsn_mark)
+{
+	struct inotify_inode_mark *i_mark;
+
+	i_mark = container_of(fsn_mark, struct inotify_inode_mark, fsn_mark);
+
+	kmem_cache_free(inotify_inode_mark_cachep, i_mark);
 }
 
-static ssize_t inotify_read(struct file *file, char __user *buf,
-			    size_t count, loff_t *pos)
+static int inotify_update_existing_watch(struct fsnotify_group *group,
+					 struct inode *inode,
+					 u32 arg)
 {
-	struct inotify_device *dev;
-	char __user *start;
+	struct fsnotify_mark *fsn_mark;
+	struct inotify_inode_mark *i_mark;
+	__u32 old_mask, new_mask;
+	__u32 mask;
+	int add = (arg & IN_MASK_ADD);
 	int ret;
-	DEFINE_WAIT(wait);
 
-	start = buf;
-	dev = file->private_data;
+	mask = inotify_arg_to_mask(arg);
 
-	while (1) {
-		struct inotify_kernel_event *kevent;
+	fsn_mark = fsnotify_find_inode_mark(group, inode);
+	if (!fsn_mark)
+		return -ENOENT;
 
-		prepare_to_wait(&dev->wq, &wait, TASK_INTERRUPTIBLE);
+	i_mark = container_of(fsn_mark, struct inotify_inode_mark, fsn_mark);
 
-		mutex_lock(&dev->ev_mutex);
-		kevent = get_one_event(dev, count);
-		mutex_unlock(&dev->ev_mutex);
+	spin_lock(&fsn_mark->lock);
 
-		if (kevent) {
-			ret = PTR_ERR(kevent);
-			if (IS_ERR(kevent))
-				break;
-			ret = copy_event_to_user(kevent, buf);
-			free_kevent(kevent);
-			if (ret < 0)
-				break;
-			buf += ret;
-			count -= ret;
-			continue;
-		}
+	old_mask = fsn_mark->mask;
+	if (add)
+		fsnotify_set_mark_mask_locked(fsn_mark, (fsn_mark->mask | mask));
+	else
+		fsnotify_set_mark_mask_locked(fsn_mark, mask);
+	new_mask = fsn_mark->mask;
 
-		ret = -EAGAIN;
-		if (file->f_flags & O_NONBLOCK)
-			break;
-		ret = -EINTR;
-		if (signal_pending(current))
-			break;
+	spin_unlock(&fsn_mark->lock);
 
-		if (start != buf)
-			break;
+	if (old_mask != new_mask) {
+		/* more bits in old than in new? */
+		int dropped = (old_mask & ~new_mask);
+		/* more bits in this fsn_mark than the inode's mask? */
+		int do_inode = (new_mask & ~inode->i_fsnotify_mask);
+
+		/* update the inode with this new fsn_mark */
+		if (dropped || do_inode)
+			fsnotify_recalc_inode_mask(inode);
 
-		schedule();
 	}
 
-	finish_wait(&dev->wq, &wait);
-	if (start != buf && ret != -EFAULT)
-		ret = buf - start;
+	/* return the wd */
+	ret = i_mark->wd;
+
+	/* match the get from fsnotify_find_mark() */
+	fsnotify_put_mark(fsn_mark);
+
 	return ret;
 }
 
-static int inotify_fasync(int fd, struct file *file, int on)
+static int inotify_new_watch(struct fsnotify_group *group,
+			     struct inode *inode,
+			     u32 arg)
 {
-	struct inotify_device *dev = file->private_data;
+	struct inotify_inode_mark *tmp_i_mark;
+	__u32 mask;
+	int ret;
+	struct idr *idr = &group->inotify_data.idr;
+	spinlock_t *idr_lock = &group->inotify_data.idr_lock;
 
-	return fasync_helper(fd, file, on, &dev->fa) >= 0 ? 0 : -EIO;
-}
+	mask = inotify_arg_to_mask(arg);
 
-static int inotify_release(struct inode *ignored, struct file *file)
-{
-	struct inotify_device *dev = file->private_data;
+	tmp_i_mark = kmem_cache_alloc(inotify_inode_mark_cachep, GFP_KERNEL);
+	if (unlikely(!tmp_i_mark))
+		return -ENOMEM;
 
-	inotify_destroy(dev->ih);
+	fsnotify_init_mark(&tmp_i_mark->fsn_mark, inotify_free_mark);
+	tmp_i_mark->fsn_mark.mask = mask;
+	tmp_i_mark->wd = -1;
+
+	ret = -ENOSPC;
+	if (atomic_read(&group->inotify_data.user->inotify_watches) >= inotify_max_user_watches)
+		goto out_err;
+
+	ret = inotify_add_to_idr(idr, idr_lock, tmp_i_mark);
+	if (ret)
+		goto out_err;
+
+	/* we are on the idr, now get on the inode */
+	ret = fsnotify_add_mark_locked(&tmp_i_mark->fsn_mark, group, inode,
+				       NULL, 0);
+	if (ret) {
+		/* we failed to get on the inode, get off the idr */
+		inotify_remove_from_idr(group, tmp_i_mark);
+		goto out_err;
+	}
 
-	/* destroy all of the events on this device */
-	mutex_lock(&dev->ev_mutex);
-	while (!list_empty(&dev->events))
-		inotify_dev_event_dequeue(dev);
-	mutex_unlock(&dev->ev_mutex);
+	/* increment the number of watches the user has */
+	atomic_inc(&group->inotify_data.user->inotify_watches);
 
-	/* free this device: the put matching the get in inotify_init() */
-	put_inotify_dev(dev);
+	/* return the watch descriptor for this new mark */
+	ret = tmp_i_mark->wd;
 
-	return 0;
+out_err:
+	/* match the ref from fsnotify_init_mark() */
+	fsnotify_put_mark(&tmp_i_mark->fsn_mark);
+
+	return ret;
 }
 
-static long inotify_ioctl(struct file *file, unsigned int cmd,
-			  unsigned long arg)
+static int inotify_update_watch(struct fsnotify_group *group, struct inode *inode, u32 arg)
 {
-	struct inotify_device *dev;
-	void __user *p;
-	int ret = -ENOTTY;
-
-	dev = file->private_data;
-	p = (void __user *) arg;
+	int ret = 0;
 
-	switch (cmd) {
-	case FIONREAD:
-		ret = put_user(dev->queue_size, (int __user *) p);
-		break;
-	}
+	mutex_lock(&group->mark_mutex);
+	/* try to update and existing watch with the new arg */
+	ret = inotify_update_existing_watch(group, inode, arg);
+	/* no mark present, try to add a new one */
+	if (ret == -ENOENT)
+		ret = inotify_new_watch(group, inode, arg);
+	mutex_unlock(&group->mark_mutex);
 
 	return ret;
 }
 
-static const struct file_operations inotify_fops = {
-	.poll           = inotify_poll,
-	.read           = inotify_read,
-	.fasync         = inotify_fasync,
-	.release        = inotify_release,
-	.unlocked_ioctl = inotify_ioctl,
-	.compat_ioctl	= inotify_ioctl,
-};
+static struct fsnotify_group *inotify_new_group(unsigned int max_events)
+{
+	struct fsnotify_group *group;
+	struct inotify_event_info *oevent;
+
+	group = fsnotify_alloc_group(&inotify_fsnotify_ops);
+	if (IS_ERR(group))
+		return group;
+
+	oevent = kmalloc(sizeof(struct inotify_event_info), GFP_KERNEL);
+	if (unlikely(!oevent)) {
+		fsnotify_destroy_group(group);
+		return ERR_PTR(-ENOMEM);
+	}
+	group->overflow_event = &oevent->fse;
+	fsnotify_init_event(group->overflow_event, NULL, FS_Q_OVERFLOW);
+	oevent->wd = -1;
+	oevent->sync_cookie = 0;
+	oevent->name_len = 0;
+
+	group->max_events = max_events;
+
+	spin_lock_init(&group->inotify_data.idr_lock);
+	idr_init(&group->inotify_data.idr);
+	group->inotify_data.user = get_current_user();
+
+	if (atomic_inc_return(&group->inotify_data.user->inotify_devs) >
+	    inotify_max_user_instances) {
+		fsnotify_destroy_group(group);
+		return ERR_PTR(-EMFILE);
+	}
+
+	return group;
+}
 
-static const struct inotify_operations inotify_user_ops = {
-	.handle_event	= inotify_dev_queue_event,
-	.destroy_watch	= free_inotify_user_watch,
-};
 
+/* inotify syscalls */
 SYSCALL_DEFINE1(inotify_init1, int, flags)
 {
-	struct inotify_device *dev;
-	struct inotify_handle *ih;
-	struct user_struct *user;
-	struct file *filp;
-	int fd, ret;
+	struct fsnotify_group *group;
+	int ret;
 
 	/* Check the IN_* constants for consistency.  */
 	BUILD_BUG_ON(IN_CLOEXEC != O_CLOEXEC);
@@ -604,67 +679,16 @@ SYSCALL_DEFINE1(inotify_init1, int, flags)
 	if (flags & ~(IN_CLOEXEC | IN_NONBLOCK))
 		return -EINVAL;
 
-	fd = get_unused_fd_flags(flags & O_CLOEXEC);
-	if (fd < 0)
-		return fd;
-
-	filp = get_empty_filp();
-	if (!filp) {
-		ret = -ENFILE;
-		goto out_put_fd;
-	}
+	/* fsnotify_obtain_group took a reference to group, we put this when we kill the file in the end */
+	group = inotify_new_group(inotify_max_queued_events);
+	if (IS_ERR(group))
+		return PTR_ERR(group);
 
-	user = get_current_user();
-	if (unlikely(atomic_read(&user->inotify_devs) >=
-			inotify_max_user_instances)) {
-		ret = -EMFILE;
-		goto out_free_uid;
-	}
-
-	dev = kmalloc(sizeof(struct inotify_device), GFP_KERNEL);
-	if (unlikely(!dev)) {
-		ret = -ENOMEM;
-		goto out_free_uid;
-	}
+	ret = anon_inode_getfd("inotify", &inotify_fops, group,
+				  O_RDONLY | flags);
+	if (ret < 0)
+		fsnotify_destroy_group(group);
 
-	ih = inotify_init(&inotify_user_ops);
-	if (IS_ERR(ih)) {
-		ret = PTR_ERR(ih);
-		goto out_free_dev;
-	}
-	dev->ih = ih;
-	dev->fa = NULL;
-
-	filp->f_op = &inotify_fops;
-	filp->f_path.mnt = mntget(inotify_mnt);
-	filp->f_path.dentry = dget(inotify_mnt->mnt_root);
-	filp->f_mapping = filp->f_path.dentry->d_inode->i_mapping;
-	filp->f_mode = FMODE_READ;
-	filp->f_flags = O_RDONLY | (flags & O_NONBLOCK);
-	filp->private_data = dev;
-
-	INIT_LIST_HEAD(&dev->events);
-	init_waitqueue_head(&dev->wq);
-	mutex_init(&dev->ev_mutex);
-	mutex_init(&dev->up_mutex);
-	dev->event_count = 0;
-	dev->queue_size = 0;
-	dev->max_events = inotify_max_queued_events;
-	dev->user = user;
-	atomic_set(&dev->count, 0);
-
-	get_inotify_dev(dev);
-	atomic_inc(&user->inotify_devs);
-	fd_install(fd, filp);
-
-	return fd;
-out_free_dev:
-	kfree(dev);
-out_free_uid:
-	free_uid(user);
-	put_filp(filp);
-out_put_fd:
-	put_unused_fd(fd);
 	return ret;
 }
 
@@ -676,19 +700,23 @@ SYSCALL_DEFINE0(inotify_init)
 SYSCALL_DEFINE3(inotify_add_watch, int, fd, const char __user *, pathname,
 		u32, mask)
 {
+	struct fsnotify_group *group;
 	struct inode *inode;
-	struct inotify_device *dev;
 	struct path path;
-	struct file *filp;
-	int ret, fput_needed;
+	struct fd f;
+	int ret;
 	unsigned flags = 0;
 
-	filp = fget_light(fd, &fput_needed);
-	if (unlikely(!filp))
+	/* don't allow invalid bits: we don't want flags set */
+	if (unlikely(!(mask & ALL_INOTIFY_BITS)))
+		return -EINVAL;
+
+	f = fdget(fd);
+	if (unlikely(!f.file))
 		return -EBADF;
 
 	/* verify that this is indeed an inotify instance */
-	if (unlikely(filp->f_op != &inotify_fops)) {
+	if (unlikely(f.file->f_op != &inotify_fops)) {
 		ret = -EINVAL;
 		goto fput_and_out;
 	}
@@ -698,95 +726,91 @@ SYSCALL_DEFINE3(inotify_add_watch, int, fd, const char __user *, pathname,
 	if (mask & IN_ONLYDIR)
 		flags |= LOOKUP_DIRECTORY;
 
-	ret = find_inode(pathname, &path, flags);
-	if (unlikely(ret))
+	ret = inotify_find_inode(pathname, &path, flags);
+	if (ret)
 		goto fput_and_out;
 
-	/* inode held in place by reference to path; dev by fget on fd */
+	/* inode held in place by reference to path; group by fget on fd */
 	inode = path.dentry->d_inode;
-	dev = filp->private_data;
-
-	mutex_lock(&dev->up_mutex);
-	ret = inotify_find_update_watch(dev->ih, inode, mask);
-	if (ret == -ENOENT)
-		ret = create_watch(dev, inode, mask);
-	mutex_unlock(&dev->up_mutex);
+	group = f.file->private_data;
 
+	/* create/update an inode mark */
+	ret = inotify_update_watch(group, inode, mask);
 	path_put(&path);
 fput_and_out:
-	fput_light(filp, fput_needed);
+	fdput(f);
 	return ret;
 }
 
 SYSCALL_DEFINE2(inotify_rm_watch, int, fd, __s32, wd)
 {
-	struct file *filp;
-	struct inotify_device *dev;
-	int ret, fput_needed;
+	struct fsnotify_group *group;
+	struct inotify_inode_mark *i_mark;
+	struct fd f;
+	int ret = 0;
 
-	filp = fget_light(fd, &fput_needed);
-	if (unlikely(!filp))
+	f = fdget(fd);
+	if (unlikely(!f.file))
 		return -EBADF;
 
 	/* verify that this is indeed an inotify instance */
-	if (unlikely(filp->f_op != &inotify_fops)) {
-		ret = -EINVAL;
+	ret = -EINVAL;
+	if (unlikely(f.file->f_op != &inotify_fops))
 		goto out;
-	}
 
-	dev = filp->private_data;
+	group = f.file->private_data;
 
-	/* we free our watch data when we get IN_IGNORED */
-	ret = inotify_rm_wd(dev->ih, wd);
+	ret = -EINVAL;
+	i_mark = inotify_idr_find(group, wd);
+	if (unlikely(!i_mark))
+		goto out;
+
+	ret = 0;
+
+	fsnotify_destroy_mark(&i_mark->fsn_mark, group);
+
+	/* match ref taken by inotify_idr_find */
+	fsnotify_put_mark(&i_mark->fsn_mark);
 
 out:
-	fput_light(filp, fput_needed);
+	fdput(f);
 	return ret;
 }
 
-static int
-inotify_get_sb(struct file_system_type *fs_type, int flags,
-	       const char *dev_name, void *data, struct vfsmount *mnt)
-{
-	return get_sb_pseudo(fs_type, "inotify", NULL,
-			INOTIFYFS_SUPER_MAGIC, mnt);
-}
-
-static struct file_system_type inotify_fs_type = {
-    .name           = "inotifyfs",
-    .get_sb         = inotify_get_sb,
-    .kill_sb        = kill_anon_super,
-};
-
 /*
- * inotify_user_setup - Our initialization function.  Note that we cannnot return
+ * inotify_user_setup - Our initialization function.  Note that we cannot return
  * error because we have compiled-in VFS hooks.  So an (unlikely) failure here
  * must result in panic().
  */
 static int __init inotify_user_setup(void)
 {
-	int ret;
-
-	ret = register_filesystem(&inotify_fs_type);
-	if (unlikely(ret))
-		panic("inotify: register_filesystem returned %d!\n", ret);
-
-	inotify_mnt = kern_mount(&inotify_fs_type);
-	if (IS_ERR(inotify_mnt))
-		panic("inotify: kern_mount ret %ld!\n", PTR_ERR(inotify_mnt));
+	BUILD_BUG_ON(IN_ACCESS != FS_ACCESS);
+	BUILD_BUG_ON(IN_MODIFY != FS_MODIFY);
+	BUILD_BUG_ON(IN_ATTRIB != FS_ATTRIB);
+	BUILD_BUG_ON(IN_CLOSE_WRITE != FS_CLOSE_WRITE);
+	BUILD_BUG_ON(IN_CLOSE_NOWRITE != FS_CLOSE_NOWRITE);
+	BUILD_BUG_ON(IN_OPEN != FS_OPEN);
+	BUILD_BUG_ON(IN_MOVED_FROM != FS_MOVED_FROM);
+	BUILD_BUG_ON(IN_MOVED_TO != FS_MOVED_TO);
+	BUILD_BUG_ON(IN_CREATE != FS_CREATE);
+	BUILD_BUG_ON(IN_DELETE != FS_DELETE);
+	BUILD_BUG_ON(IN_DELETE_SELF != FS_DELETE_SELF);
+	BUILD_BUG_ON(IN_MOVE_SELF != FS_MOVE_SELF);
+	BUILD_BUG_ON(IN_UNMOUNT != FS_UNMOUNT);
+	BUILD_BUG_ON(IN_Q_OVERFLOW != FS_Q_OVERFLOW);
+	BUILD_BUG_ON(IN_IGNORED != FS_IN_IGNORED);
+	BUILD_BUG_ON(IN_EXCL_UNLINK != FS_EXCL_UNLINK);
+	BUILD_BUG_ON(IN_ISDIR != FS_ISDIR);
+	BUILD_BUG_ON(IN_ONESHOT != FS_IN_ONESHOT);
+
+	BUG_ON(hweight32(ALL_INOTIFY_BITS) != 21);
+
+	inotify_inode_mark_cachep = KMEM_CACHE(inotify_inode_mark, SLAB_PANIC);
 
 	inotify_max_queued_events = 16384;
 	inotify_max_user_instances = 128;
 	inotify_max_user_watches = 8192;
 
-	watch_cachep = kmem_cache_create("inotify_watch_cache",
-					 sizeof(struct inotify_user_watch),
-					 0, SLAB_PANIC, NULL);
-	event_cachep = kmem_cache_create("inotify_event_cache",
-					 sizeof(struct inotify_kernel_event),
-					 0, SLAB_PANIC, NULL);
-
 	return 0;
 }
-
 module_init(inotify_user_setup);
diff --git a/fs/notify/mark.c b/fs/notify/mark.c
new file mode 100644
index 00000000000..d90deaa08e7
--- /dev/null
+++ b/fs/notify/mark.c
@@ -0,0 +1,375 @@
+/*
+ *  Copyright (C) 2008 Red Hat, Inc., Eric Paris <eparis@redhat.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2, or (at your option)
+ *  any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; see the file COPYING.  If not, write to
+ *  the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+/*
+ * fsnotify inode mark locking/lifetime/and refcnting
+ *
+ * REFCNT:
+ * The group->recnt and mark->refcnt tell how many "things" in the kernel
+ * currently are referencing the objects. Both kind of objects typically will
+ * live inside the kernel with a refcnt of 2, one for its creation and one for
+ * the reference a group and a mark hold to each other.
+ * If you are holding the appropriate locks, you can take a reference and the
+ * object itself is guaranteed to survive until the reference is dropped.
+ *
+ * LOCKING:
+ * There are 3 locks involved with fsnotify inode marks and they MUST be taken
+ * in order as follows:
+ *
+ * group->mark_mutex
+ * mark->lock
+ * inode->i_lock
+ *
+ * group->mark_mutex protects the marks_list anchored inside a given group and
+ * each mark is hooked via the g_list.  It also protects the groups private
+ * data (i.e group limits).
+
+ * mark->lock protects the marks attributes like its masks and flags.
+ * Furthermore it protects the access to a reference of the group that the mark
+ * is assigned to as well as the access to a reference of the inode/vfsmount
+ * that is being watched by the mark.
+ *
+ * inode->i_lock protects the i_fsnotify_marks list anchored inside a
+ * given inode and each mark is hooked via the i_list. (and sorta the
+ * free_i_list)
+ *
+ *
+ * LIFETIME:
+ * Inode marks survive between when they are added to an inode and when their
+ * refcnt==0.
+ *
+ * The inode mark can be cleared for a number of different reasons including:
+ * - The inode is unlinked for the last time.  (fsnotify_inode_remove)
+ * - The inode is being evicted from cache. (fsnotify_inode_delete)
+ * - The fs the inode is on is unmounted.  (fsnotify_inode_delete/fsnotify_unmount_inodes)
+ * - Something explicitly requests that it be removed.  (fsnotify_destroy_mark)
+ * - The fsnotify_group associated with the mark is going away and all such marks
+ *   need to be cleaned up. (fsnotify_clear_marks_by_group)
+ *
+ * Worst case we are given an inode and need to clean up all the marks on that
+ * inode.  We take i_lock and walk the i_fsnotify_marks safely.  For each
+ * mark on the list we take a reference (so the mark can't disappear under us).
+ * We remove that mark form the inode's list of marks and we add this mark to a
+ * private list anchored on the stack using i_free_list; we walk i_free_list
+ * and before we destroy the mark we make sure that we dont race with a
+ * concurrent destroy_group by getting a ref to the marks group and taking the
+ * groups mutex.
+
+ * Very similarly for freeing by group, except we use free_g_list.
+ *
+ * This has the very interesting property of being able to run concurrently with
+ * any (or all) other directions.
+ */
+
+#include <linux/fs.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/kthread.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <linux/srcu.h>
+
+#include <linux/atomic.h>
+
+#include <linux/fsnotify_backend.h>
+#include "fsnotify.h"
+
+struct srcu_struct fsnotify_mark_srcu;
+static DEFINE_SPINLOCK(destroy_lock);
+static LIST_HEAD(destroy_list);
+static DECLARE_WAIT_QUEUE_HEAD(destroy_waitq);
+
+void fsnotify_get_mark(struct fsnotify_mark *mark)
+{
+	atomic_inc(&mark->refcnt);
+}
+
+void fsnotify_put_mark(struct fsnotify_mark *mark)
+{
+	if (atomic_dec_and_test(&mark->refcnt)) {
+		if (mark->group)
+			fsnotify_put_group(mark->group);
+		mark->free_mark(mark);
+	}
+}
+
+/*
+ * Any time a mark is getting freed we end up here.
+ * The caller had better be holding a reference to this mark so we don't actually
+ * do the final put under the mark->lock
+ */
+void fsnotify_destroy_mark_locked(struct fsnotify_mark *mark,
+				  struct fsnotify_group *group)
+{
+	struct inode *inode = NULL;
+
+	BUG_ON(!mutex_is_locked(&group->mark_mutex));
+
+	spin_lock(&mark->lock);
+
+	/* something else already called this function on this mark */
+	if (!(mark->flags & FSNOTIFY_MARK_FLAG_ALIVE)) {
+		spin_unlock(&mark->lock);
+		return;
+	}
+
+	mark->flags &= ~FSNOTIFY_MARK_FLAG_ALIVE;
+
+	if (mark->flags & FSNOTIFY_MARK_FLAG_INODE) {
+		inode = mark->i.inode;
+		fsnotify_destroy_inode_mark(mark);
+	} else if (mark->flags & FSNOTIFY_MARK_FLAG_VFSMOUNT)
+		fsnotify_destroy_vfsmount_mark(mark);
+	else
+		BUG();
+
+	list_del_init(&mark->g_list);
+
+	spin_unlock(&mark->lock);
+
+	if (inode && (mark->flags & FSNOTIFY_MARK_FLAG_OBJECT_PINNED))
+		iput(inode);
+	/* release lock temporarily */
+	mutex_unlock(&group->mark_mutex);
+
+	spin_lock(&destroy_lock);
+	list_add(&mark->destroy_list, &destroy_list);
+	spin_unlock(&destroy_lock);
+	wake_up(&destroy_waitq);
+	/*
+	 * We don't necessarily have a ref on mark from caller so the above destroy
+	 * may have actually freed it, unless this group provides a 'freeing_mark'
+	 * function which must be holding a reference.
+	 */
+
+	/*
+	 * Some groups like to know that marks are being freed.  This is a
+	 * callback to the group function to let it know that this mark
+	 * is being freed.
+	 */
+	if (group->ops->freeing_mark)
+		group->ops->freeing_mark(mark, group);
+
+	/*
+	 * __fsnotify_update_child_dentry_flags(inode);
+	 *
+	 * I really want to call that, but we can't, we have no idea if the inode
+	 * still exists the second we drop the mark->lock.
+	 *
+	 * The next time an event arrive to this inode from one of it's children
+	 * __fsnotify_parent will see that the inode doesn't care about it's
+	 * children and will update all of these flags then.  So really this
+	 * is just a lazy update (and could be a perf win...)
+	 */
+
+	atomic_dec(&group->num_marks);
+
+	mutex_lock_nested(&group->mark_mutex, SINGLE_DEPTH_NESTING);
+}
+
+void fsnotify_destroy_mark(struct fsnotify_mark *mark,
+			   struct fsnotify_group *group)
+{
+	mutex_lock_nested(&group->mark_mutex, SINGLE_DEPTH_NESTING);
+	fsnotify_destroy_mark_locked(mark, group);
+	mutex_unlock(&group->mark_mutex);
+}
+
+void fsnotify_set_mark_mask_locked(struct fsnotify_mark *mark, __u32 mask)
+{
+	assert_spin_locked(&mark->lock);
+
+	mark->mask = mask;
+
+	if (mark->flags & FSNOTIFY_MARK_FLAG_INODE)
+		fsnotify_set_inode_mark_mask_locked(mark, mask);
+}
+
+void fsnotify_set_mark_ignored_mask_locked(struct fsnotify_mark *mark, __u32 mask)
+{
+	assert_spin_locked(&mark->lock);
+
+	mark->ignored_mask = mask;
+}
+
+/*
+ * Attach an initialized mark to a given group and fs object.
+ * These marks may be used for the fsnotify backend to determine which
+ * event types should be delivered to which group.
+ */
+int fsnotify_add_mark_locked(struct fsnotify_mark *mark,
+			     struct fsnotify_group *group, struct inode *inode,
+			     struct vfsmount *mnt, int allow_dups)
+{
+	int ret = 0;
+
+	BUG_ON(inode && mnt);
+	BUG_ON(!inode && !mnt);
+	BUG_ON(!mutex_is_locked(&group->mark_mutex));
+
+	/*
+	 * LOCKING ORDER!!!!
+	 * group->mark_mutex
+	 * mark->lock
+	 * inode->i_lock
+	 */
+	spin_lock(&mark->lock);
+	mark->flags |= FSNOTIFY_MARK_FLAG_ALIVE;
+
+	fsnotify_get_group(group);
+	mark->group = group;
+	list_add(&mark->g_list, &group->marks_list);
+	atomic_inc(&group->num_marks);
+	fsnotify_get_mark(mark); /* for i_list and g_list */
+
+	if (inode) {
+		ret = fsnotify_add_inode_mark(mark, group, inode, allow_dups);
+		if (ret)
+			goto err;
+	} else if (mnt) {
+		ret = fsnotify_add_vfsmount_mark(mark, group, mnt, allow_dups);
+		if (ret)
+			goto err;
+	} else {
+		BUG();
+	}
+
+	/* this will pin the object if appropriate */
+	fsnotify_set_mark_mask_locked(mark, mark->mask);
+	spin_unlock(&mark->lock);
+
+	if (inode)
+		__fsnotify_update_child_dentry_flags(inode);
+
+	return ret;
+err:
+	mark->flags &= ~FSNOTIFY_MARK_FLAG_ALIVE;
+	list_del_init(&mark->g_list);
+	fsnotify_put_group(group);
+	mark->group = NULL;
+	atomic_dec(&group->num_marks);
+
+	spin_unlock(&mark->lock);
+
+	spin_lock(&destroy_lock);
+	list_add(&mark->destroy_list, &destroy_list);
+	spin_unlock(&destroy_lock);
+	wake_up(&destroy_waitq);
+
+	return ret;
+}
+
+int fsnotify_add_mark(struct fsnotify_mark *mark, struct fsnotify_group *group,
+		      struct inode *inode, struct vfsmount *mnt, int allow_dups)
+{
+	int ret;
+	mutex_lock(&group->mark_mutex);
+	ret = fsnotify_add_mark_locked(mark, group, inode, mnt, allow_dups);
+	mutex_unlock(&group->mark_mutex);
+	return ret;
+}
+
+/*
+ * clear any marks in a group in which mark->flags & flags is true
+ */
+void fsnotify_clear_marks_by_group_flags(struct fsnotify_group *group,
+					 unsigned int flags)
+{
+	struct fsnotify_mark *lmark, *mark;
+
+	mutex_lock_nested(&group->mark_mutex, SINGLE_DEPTH_NESTING);
+	list_for_each_entry_safe(mark, lmark, &group->marks_list, g_list) {
+		if (mark->flags & flags) {
+			fsnotify_get_mark(mark);
+			fsnotify_destroy_mark_locked(mark, group);
+			fsnotify_put_mark(mark);
+		}
+	}
+	mutex_unlock(&group->mark_mutex);
+}
+
+/*
+ * Given a group, destroy all of the marks associated with that group.
+ */
+void fsnotify_clear_marks_by_group(struct fsnotify_group *group)
+{
+	fsnotify_clear_marks_by_group_flags(group, (unsigned int)-1);
+}
+
+void fsnotify_duplicate_mark(struct fsnotify_mark *new, struct fsnotify_mark *old)
+{
+	assert_spin_locked(&old->lock);
+	new->i.inode = old->i.inode;
+	new->m.mnt = old->m.mnt;
+	if (old->group)
+		fsnotify_get_group(old->group);
+	new->group = old->group;
+	new->mask = old->mask;
+	new->free_mark = old->free_mark;
+}
+
+/*
+ * Nothing fancy, just initialize lists and locks and counters.
+ */
+void fsnotify_init_mark(struct fsnotify_mark *mark,
+			void (*free_mark)(struct fsnotify_mark *mark))
+{
+	memset(mark, 0, sizeof(*mark));
+	spin_lock_init(&mark->lock);
+	atomic_set(&mark->refcnt, 1);
+	mark->free_mark = free_mark;
+}
+
+static int fsnotify_mark_destroy(void *ignored)
+{
+	struct fsnotify_mark *mark, *next;
+	struct list_head private_destroy_list;
+
+	for (;;) {
+		spin_lock(&destroy_lock);
+		/* exchange the list head */
+		list_replace_init(&destroy_list, &private_destroy_list);
+		spin_unlock(&destroy_lock);
+
+		synchronize_srcu(&fsnotify_mark_srcu);
+
+		list_for_each_entry_safe(mark, next, &private_destroy_list, destroy_list) {
+			list_del_init(&mark->destroy_list);
+			fsnotify_put_mark(mark);
+		}
+
+		wait_event_interruptible(destroy_waitq, !list_empty(&destroy_list));
+	}
+
+	return 0;
+}
+
+static int __init fsnotify_mark_init(void)
+{
+	struct task_struct *thread;
+
+	thread = kthread_run(fsnotify_mark_destroy, NULL,
+			     "fsnotify_mark");
+	if (IS_ERR(thread))
+		panic("unable to start fsnotify mark destruction thread.");
+
+	return 0;
+}
+device_initcall(fsnotify_mark_init);
diff --git a/fs/notify/notification.c b/fs/notify/notification.c
new file mode 100644
index 00000000000..1e58402171a
--- /dev/null
+++ b/fs/notify/notification.c
@@ -0,0 +1,196 @@
+/*
+ *  Copyright (C) 2008 Red Hat, Inc., Eric Paris <eparis@redhat.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2, or (at your option)
+ *  any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; see the file COPYING.  If not, write to
+ *  the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+/*
+ * Basic idea behind the notification queue: An fsnotify group (like inotify)
+ * sends the userspace notification about events asynchronously some time after
+ * the event happened.  When inotify gets an event it will need to add that
+ * event to the group notify queue.  Since a single event might need to be on
+ * multiple group's notification queues we can't add the event directly to each
+ * queue and instead add a small "event_holder" to each queue.  This event_holder
+ * has a pointer back to the original event.  Since the majority of events are
+ * going to end up on one, and only one, notification queue we embed one
+ * event_holder into each event.  This means we have a single allocation instead
+ * of always needing two.  If the embedded event_holder is already in use by
+ * another group a new event_holder (from fsnotify_event_holder_cachep) will be
+ * allocated and used.
+ */
+
+#include <linux/fs.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/list.h>
+#include <linux/module.h>
+#include <linux/mount.h>
+#include <linux/mutex.h>
+#include <linux/namei.h>
+#include <linux/path.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+
+#include <linux/atomic.h>
+
+#include <linux/fsnotify_backend.h>
+#include "fsnotify.h"
+
+static atomic_t fsnotify_sync_cookie = ATOMIC_INIT(0);
+
+/**
+ * fsnotify_get_cookie - return a unique cookie for use in synchronizing events.
+ * Called from fsnotify_move, which is inlined into filesystem modules.
+ */
+u32 fsnotify_get_cookie(void)
+{
+	return atomic_inc_return(&fsnotify_sync_cookie);
+}
+EXPORT_SYMBOL_GPL(fsnotify_get_cookie);
+
+/* return true if the notify queue is empty, false otherwise */
+bool fsnotify_notify_queue_is_empty(struct fsnotify_group *group)
+{
+	BUG_ON(!mutex_is_locked(&group->notification_mutex));
+	return list_empty(&group->notification_list) ? true : false;
+}
+
+void fsnotify_destroy_event(struct fsnotify_group *group,
+			    struct fsnotify_event *event)
+{
+	/* Overflow events are per-group and we don't want to free them */
+	if (!event || event->mask == FS_Q_OVERFLOW)
+		return;
+
+	group->ops->free_event(event);
+}
+
+/*
+ * Add an event to the group notification queue.  The group can later pull this
+ * event off the queue to deal with.  The function returns 0 if the event was
+ * added to the queue, 1 if the event was merged with some other queued event,
+ * 2 if the queue of events has overflown.
+ */
+int fsnotify_add_notify_event(struct fsnotify_group *group,
+			      struct fsnotify_event *event,
+			      int (*merge)(struct list_head *,
+					   struct fsnotify_event *))
+{
+	int ret = 0;
+	struct list_head *list = &group->notification_list;
+
+	pr_debug("%s: group=%p event=%p\n", __func__, group, event);
+
+	mutex_lock(&group->notification_mutex);
+
+	if (group->q_len >= group->max_events) {
+		ret = 2;
+		/* Queue overflow event only if it isn't already queued */
+		if (!list_empty(&group->overflow_event->list)) {
+			mutex_unlock(&group->notification_mutex);
+			return ret;
+		}
+		event = group->overflow_event;
+		goto queue;
+	}
+
+	if (!list_empty(list) && merge) {
+		ret = merge(list, event);
+		if (ret) {
+			mutex_unlock(&group->notification_mutex);
+			return ret;
+		}
+	}
+
+queue:
+	group->q_len++;
+	list_add_tail(&event->list, list);
+	mutex_unlock(&group->notification_mutex);
+
+	wake_up(&group->notification_waitq);
+	kill_fasync(&group->fsn_fa, SIGIO, POLL_IN);
+	return ret;
+}
+
+/*
+ * Remove and return the first event from the notification list.  It is the
+ * responsibility of the caller to destroy the obtained event
+ */
+struct fsnotify_event *fsnotify_remove_notify_event(struct fsnotify_group *group)
+{
+	struct fsnotify_event *event;
+
+	BUG_ON(!mutex_is_locked(&group->notification_mutex));
+
+	pr_debug("%s: group=%p\n", __func__, group);
+
+	event = list_first_entry(&group->notification_list,
+				 struct fsnotify_event, list);
+	/*
+	 * We need to init list head for the case of overflow event so that
+	 * check in fsnotify_add_notify_events() works
+	 */
+	list_del_init(&event->list);
+	group->q_len--;
+
+	return event;
+}
+
+/*
+ * This will not remove the event, that must be done with fsnotify_remove_notify_event()
+ */
+struct fsnotify_event *fsnotify_peek_notify_event(struct fsnotify_group *group)
+{
+	BUG_ON(!mutex_is_locked(&group->notification_mutex));
+
+	return list_first_entry(&group->notification_list,
+				struct fsnotify_event, list);
+}
+
+/*
+ * Called when a group is being torn down to clean up any outstanding
+ * event notifications.
+ */
+void fsnotify_flush_notify(struct fsnotify_group *group)
+{
+	struct fsnotify_event *event;
+
+	mutex_lock(&group->notification_mutex);
+	while (!fsnotify_notify_queue_is_empty(group)) {
+		event = fsnotify_remove_notify_event(group);
+		fsnotify_destroy_event(group, event);
+	}
+	mutex_unlock(&group->notification_mutex);
+}
+
+/*
+ * fsnotify_create_event - Allocate a new event which will be sent to each
+ * group's handle_event function if the group was interested in this
+ * particular event.
+ *
+ * @inode the inode which is supposed to receive the event (sometimes a
+ *	parent of the inode to which the event happened.
+ * @mask what actually happened.
+ * @data pointer to the object which was actually affected
+ * @data_type flag indication if the data is a file, path, inode, nothing...
+ * @name the filename, if available
+ */
+void fsnotify_init_event(struct fsnotify_event *event, struct inode *inode,
+			 u32 mask)
+{
+	INIT_LIST_HEAD(&event->list);
+	event->inode = inode;
+	event->mask = mask;
+}
diff --git a/fs/notify/vfsmount_mark.c b/fs/notify/vfsmount_mark.c
new file mode 100644
index 00000000000..68ca5a8704b
--- /dev/null
+++ b/fs/notify/vfsmount_mark.c
@@ -0,0 +1,200 @@
+/*
+ *  Copyright (C) 2008 Red Hat, Inc., Eric Paris <eparis@redhat.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2, or (at your option)
+ *  any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; see the file COPYING.  If not, write to
+ *  the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <linux/fs.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/mount.h>
+#include <linux/mutex.h>
+#include <linux/spinlock.h>
+
+#include <linux/atomic.h>
+
+#include <linux/fsnotify_backend.h>
+#include "fsnotify.h"
+#include "../mount.h"
+
+void fsnotify_clear_marks_by_mount(struct vfsmount *mnt)
+{
+	struct fsnotify_mark *mark, *lmark;
+	struct hlist_node *n;
+	struct mount *m = real_mount(mnt);
+	LIST_HEAD(free_list);
+
+	spin_lock(&mnt->mnt_root->d_lock);
+	hlist_for_each_entry_safe(mark, n, &m->mnt_fsnotify_marks, m.m_list) {
+		list_add(&mark->m.free_m_list, &free_list);
+		hlist_del_init_rcu(&mark->m.m_list);
+		fsnotify_get_mark(mark);
+	}
+	spin_unlock(&mnt->mnt_root->d_lock);
+
+	list_for_each_entry_safe(mark, lmark, &free_list, m.free_m_list) {
+		struct fsnotify_group *group;
+
+		spin_lock(&mark->lock);
+		fsnotify_get_group(mark->group);
+		group = mark->group;
+		spin_unlock(&mark->lock);
+
+		fsnotify_destroy_mark(mark, group);
+		fsnotify_put_mark(mark);
+		fsnotify_put_group(group);
+	}
+}
+
+void fsnotify_clear_vfsmount_marks_by_group(struct fsnotify_group *group)
+{
+	fsnotify_clear_marks_by_group_flags(group, FSNOTIFY_MARK_FLAG_VFSMOUNT);
+}
+
+/*
+ * Recalculate the mask of events relevant to a given vfsmount locked.
+ */
+static void fsnotify_recalc_vfsmount_mask_locked(struct vfsmount *mnt)
+{
+	struct mount *m = real_mount(mnt);
+	struct fsnotify_mark *mark;
+	__u32 new_mask = 0;
+
+	assert_spin_locked(&mnt->mnt_root->d_lock);
+
+	hlist_for_each_entry(mark, &m->mnt_fsnotify_marks, m.m_list)
+		new_mask |= mark->mask;
+	m->mnt_fsnotify_mask = new_mask;
+}
+
+/*
+ * Recalculate the mnt->mnt_fsnotify_mask, or the mask of all FS_* event types
+ * any notifier is interested in hearing for this mount point
+ */
+void fsnotify_recalc_vfsmount_mask(struct vfsmount *mnt)
+{
+	spin_lock(&mnt->mnt_root->d_lock);
+	fsnotify_recalc_vfsmount_mask_locked(mnt);
+	spin_unlock(&mnt->mnt_root->d_lock);
+}
+
+void fsnotify_destroy_vfsmount_mark(struct fsnotify_mark *mark)
+{
+	struct vfsmount *mnt = mark->m.mnt;
+
+	BUG_ON(!mutex_is_locked(&mark->group->mark_mutex));
+	assert_spin_locked(&mark->lock);
+
+	spin_lock(&mnt->mnt_root->d_lock);
+
+	hlist_del_init_rcu(&mark->m.m_list);
+	mark->m.mnt = NULL;
+
+	fsnotify_recalc_vfsmount_mask_locked(mnt);
+
+	spin_unlock(&mnt->mnt_root->d_lock);
+}
+
+static struct fsnotify_mark *fsnotify_find_vfsmount_mark_locked(struct fsnotify_group *group,
+								struct vfsmount *mnt)
+{
+	struct mount *m = real_mount(mnt);
+	struct fsnotify_mark *mark;
+
+	assert_spin_locked(&mnt->mnt_root->d_lock);
+
+	hlist_for_each_entry(mark, &m->mnt_fsnotify_marks, m.m_list) {
+		if (mark->group == group) {
+			fsnotify_get_mark(mark);
+			return mark;
+		}
+	}
+	return NULL;
+}
+
+/*
+ * given a group and vfsmount, find the mark associated with that combination.
+ * if found take a reference to that mark and return it, else return NULL
+ */
+struct fsnotify_mark *fsnotify_find_vfsmount_mark(struct fsnotify_group *group,
+						  struct vfsmount *mnt)
+{
+	struct fsnotify_mark *mark;
+
+	spin_lock(&mnt->mnt_root->d_lock);
+	mark = fsnotify_find_vfsmount_mark_locked(group, mnt);
+	spin_unlock(&mnt->mnt_root->d_lock);
+
+	return mark;
+}
+
+/*
+ * Attach an initialized mark to a given group and vfsmount.
+ * These marks may be used for the fsnotify backend to determine which
+ * event types should be delivered to which groups.
+ */
+int fsnotify_add_vfsmount_mark(struct fsnotify_mark *mark,
+			       struct fsnotify_group *group, struct vfsmount *mnt,
+			       int allow_dups)
+{
+	struct mount *m = real_mount(mnt);
+	struct fsnotify_mark *lmark, *last = NULL;
+	int ret = 0;
+
+	mark->flags |= FSNOTIFY_MARK_FLAG_VFSMOUNT;
+
+	BUG_ON(!mutex_is_locked(&group->mark_mutex));
+	assert_spin_locked(&mark->lock);
+
+	spin_lock(&mnt->mnt_root->d_lock);
+
+	mark->m.mnt = mnt;
+
+	/* is mark the first mark? */
+	if (hlist_empty(&m->mnt_fsnotify_marks)) {
+		hlist_add_head_rcu(&mark->m.m_list, &m->mnt_fsnotify_marks);
+		goto out;
+	}
+
+	/* should mark be in the middle of the current list? */
+	hlist_for_each_entry(lmark, &m->mnt_fsnotify_marks, m.m_list) {
+		last = lmark;
+
+		if ((lmark->group == group) && !allow_dups) {
+			ret = -EEXIST;
+			goto out;
+		}
+
+		if (mark->group->priority < lmark->group->priority)
+			continue;
+
+		if ((mark->group->priority == lmark->group->priority) &&
+		    (mark->group < lmark->group))
+			continue;
+
+		hlist_add_before_rcu(&mark->m.m_list, &lmark->m.m_list);
+		goto out;
+	}
+
+	BUG_ON(last == NULL);
+	/* mark should be the last entry.  last is the current last entry */
+	hlist_add_after_rcu(&last->m.m_list, &mark->m.m_list);
+out:
+	fsnotify_recalc_vfsmount_mask_locked(mnt);
+	spin_unlock(&mnt->mnt_root->d_lock);
+
+	return ret;
+}