inotify, bitches

Signed-off-by: Robert Love <rml@novell.com>

 arch/sparc64/Kconfig       |   13 
 drivers/char/Kconfig       |   13 
 drivers/char/Makefile      |    2 
 drivers/char/inotify.c     | 1053 +++++++++++++++++++++++++++++++++++++++++++++
 drivers/char/misc.c        |   14 
 fs/attr.c                  |   34 -
 fs/compat.c                |   14 
 fs/file_table.c            |    4 
 fs/inode.c                 |    3 
 fs/namei.c                 |   38 -
 fs/open.c                  |    9 
 fs/read_write.c            |   28 -
 fs/super.c                 |    3 
 include/linux/fs.h         |    7 
 include/linux/fsnotify.h   |  235 ++++++++++
 include/linux/inotify.h    |  118 +++++
 include/linux/miscdevice.h |    5 
 include/linux/sched.h      |    2 
 kernel/user.c              |    2 
 19 files changed, 1522 insertions(+), 75 deletions(-)

diff -urN linux-2.6.10/arch/sparc64/Kconfig linux/arch/sparc64/Kconfig
--- linux-2.6.10/arch/sparc64/Kconfig	2004-12-24 16:35:25.000000000 -0500
+++ linux/arch/sparc64/Kconfig	2005-02-01 12:24:26.000000000 -0500
@@ -88,6 +88,19 @@
 	bool
 	default y
 
+config INOTIFY
+	bool "Inotify file change notification support"
+	default y
+	---help---
+	  Say Y here to enable inotify support and the /dev/inotify character
+	  device.  Inotify is a file change notification system and a
+	  replacement for dnotify.  Inotify fixes numerous shortcomings in
+	  dnotify and introduces several new features.  It allows monitoring
+	  of both files and directories via a single open fd.  Multiple file
+	  events are supported.
+
+	  If unsure, say Y.
+
 config SMP
 	bool "Symmetric multi-processing support"
 	---help---
diff -urN linux-2.6.10/drivers/char/inotify.c linux/drivers/char/inotify.c
--- linux-2.6.10/drivers/char/inotify.c	1969-12-31 19:00:00.000000000 -0500
+++ linux/drivers/char/inotify.c	2005-02-09 16:05:07.959265648 -0500
@@ -0,0 +1,1053 @@
+/*
+ * drivers/char/inotify.c - inode-based file event notifications
+ *
+ * Authors:
+ *	John McCutchan	<ttb@tentacle.dhs.org>
+ *	Robert Love	<rml@novell.com>
+ *
+ * Copyright (C) 2005 John McCutchan
+ *
+ * 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/sched.h>
+#include <linux/spinlock.h>
+#include <linux/idr.h>
+#include <linux/slab.h>
+#include <linux/fs.h>
+#include <linux/namei.h>
+#include <linux/poll.h>
+#include <linux/device.h>
+#include <linux/miscdevice.h>
+#include <linux/init.h>
+#include <linux/list.h>
+#include <linux/writeback.h>
+#include <linux/inotify.h>
+
+#include <asm/ioctls.h>
+
+static atomic_t inotify_cookie;
+static kmem_cache_t *watch_cachep;
+static kmem_cache_t *event_cachep;
+static kmem_cache_t *inode_data_cachep;
+
+static int sysfs_attrib_max_user_devices;
+static int sysfs_attrib_max_user_watches;
+static unsigned int sysfs_attrib_max_queued_events;
+
+/*
+ * struct inotify_device - represents an open instance of an inotify device
+ *
+ * For each inotify device, we need to keep track of the events queued on it,
+ * a list of the inodes that we are watching, and so on.
+ *
+ * This structure is protected by 'lock'.  Lock ordering:
+ *
+ * dev->lock (protects dev)
+ *	inode_lock (used to safely walk inode_in_use list)
+ *		inode->i_lock (only needed for getting ref on inode_data)
+ */
+struct inotify_device {
+	wait_queue_head_t 	wait;
+	struct idr		idr;
+	struct list_head 	events;
+	struct list_head 	watches;
+	spinlock_t		lock;
+	unsigned int		queue_size;
+	unsigned int		event_count;
+	unsigned int		max_events;
+	struct user_struct	*user;
+};
+
+struct inotify_watch {
+	s32 			wd;	/* watch descriptor */
+	u32			mask;	/* event mask for this watch */
+	struct inode		*inode;	/* associated inode */
+	struct inotify_device	*dev;	/* associated device */
+	struct list_head	d_list;	/* entry in device's list */
+	struct list_head	i_list; /* entry in inotify_data's list */
+};
+
+/*
+ * A list of these is attached to each instance of the driver.  In read(), this
+ * this list is walked and all events that can fit in the buffer are returned.
+ */
+struct inotify_kernel_event {
+	struct inotify_event	event;
+	struct list_head        list;
+	char			*filename;
+};
+
+static ssize_t show_max_queued_events(struct class_device *class, char *buf)
+{
+	return sprintf(buf, "%d\n", sysfs_attrib_max_queued_events);
+}
+
+static ssize_t store_max_queued_events(struct class_device *class,
+				       const char *buf, size_t count)
+{
+	unsigned int max;
+
+	if (sscanf(buf, "%u", &max) > 0 && max > 0) {
+		sysfs_attrib_max_queued_events = max;
+		return strlen(buf);
+	}
+	return -EINVAL;
+}
+
+static ssize_t show_max_user_devices(struct class_device *class, char *buf)
+{
+	return sprintf(buf, "%d\n", sysfs_attrib_max_user_devices);
+}
+
+static ssize_t store_max_user_devices(struct class_device *class,
+				      const char *buf, size_t count)
+{
+	int max;
+
+	if (sscanf(buf, "%d", &max) > 0 && max > 0) {
+		sysfs_attrib_max_user_devices = max;
+		return strlen(buf);
+	}
+	return -EINVAL;
+}
+
+static ssize_t show_max_user_watches(struct class_device *class, char *buf)
+{
+	return sprintf(buf, "%d\n", sysfs_attrib_max_user_watches);
+}
+
+static ssize_t store_max_user_watches(struct class_device *class,
+				      const char *buf, size_t count)
+{
+	int max;
+
+	if (sscanf(buf, "%d", &max) > 0 && max > 0) {
+		sysfs_attrib_max_user_watches = max;
+		return strlen(buf);
+	}
+	return -EINVAL;
+}
+
+static CLASS_DEVICE_ATTR(max_queued_events, S_IRUGO | S_IWUSR,
+	show_max_queued_events, store_max_queued_events);
+static CLASS_DEVICE_ATTR(max_user_devices, S_IRUGO | S_IWUSR,
+	show_max_user_devices, store_max_user_devices);
+static CLASS_DEVICE_ATTR(max_user_watches, S_IRUGO | S_IWUSR,
+	show_max_user_watches, store_max_user_watches);
+
+static inline void __get_inode_data(struct inotify_inode_data *data)
+{
+	atomic_inc(&data->count);
+}
+
+/*
+ * get_inode_data - pin an inotify_inode_data structure.  Returns the structure
+ * if successful and NULL on failure, which can only occur if inotify_data is
+ * not yet allocated.  The inode must be pinned prior to invocation.
+ */
+static inline struct inotify_inode_data * get_inode_data(struct inode *inode)
+{
+	struct inotify_inode_data *data;
+
+	spin_lock(&inode->i_lock);
+	data = inode->inotify_data;
+	if (data)
+		__get_inode_data(data);
+	spin_unlock(&inode->i_lock);
+
+	return data;
+}
+
+/*
+ * put_inode_data - drop our reference on an inotify_inode_data and the
+ * inode structure in which it lives.  If the reference count on inotify_data
+ * reaches zero, free it.
+ */
+static inline void put_inode_data(struct inode *inode)
+{
+	//spin_lock(&inode->i_lock);
+	if (atomic_dec_and_test(&inode->inotify_data->count)) {
+		kmem_cache_free(inode_data_cachep, inode->inotify_data);
+		inode->inotify_data = NULL;
+	}
+	//spin_unlock(&inode->i_lock);
+}
+
+/*
+ * find_inode - resolve a user-given path to a specific inode and return a nd
+ */
+static int find_inode(const char __user *dirname, struct nameidata *nd)
+{
+	int error;
+
+	error = __user_walk(dirname, LOOKUP_FOLLOW, nd);
+	if (error)
+		return error;
+
+	/* you can only watch an inode if you have read permissions on it */
+	return permission(nd->dentry->d_inode, MAY_READ, NULL);
+}
+
+static struct inotify_kernel_event * kernel_event(s32 wd, u32 mask, u32 cookie,
+						  const char *filename)
+{
+	struct inotify_kernel_event *kevent;
+
+	kevent = kmem_cache_alloc(event_cachep, GFP_ATOMIC);
+	if (!kevent)
+		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 (filename) {
+		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(filename) + 1;
+		rem = event_size - len;
+		if (len > event_size) {
+			rem = event_size - (len % event_size);
+			if (len % event_size == 0)
+				rem = 0;
+		}
+		len += rem;
+
+		kevent->filename = kmalloc(len, GFP_ATOMIC);
+		if (!kevent->filename) {
+			kmem_cache_free(event_cachep, kevent);
+			return NULL;
+		}
+		memset(kevent->filename, 0, len);
+		strncpy(kevent->filename, filename, strlen(filename));
+		kevent->event.len = len;
+	} else {
+		kevent->event.len = 0;
+		kevent->filename = NULL;
+	}
+
+	return kevent;
+}
+
+#define list_to_inotify_kernel_event(pos)	\
+		list_entry((pos), struct inotify_kernel_event, list)
+
+#define inotify_dev_get_event(dev)		\
+		(list_to_inotify_kernel_event(dev->events.next))
+
+/*
+ * inotify_dev_queue_event - add a new event to the given device
+ *
+ * Caller must hold dev->lock.
+ */
+static void inotify_dev_queue_event(struct inotify_device *dev,
+				    struct inotify_watch *watch, u32 mask,
+				    u32 cookie, const char *filename)
+{
+	struct inotify_kernel_event *kevent, *last;
+
+	/* drop this event if it is a dupe of the previous */
+	last = inotify_dev_get_event(dev);
+	if (dev->event_count && last->event.mask == mask &&
+			last->event.wd == watch->wd) {
+		const char *lastname = last->filename;
+
+		if (!filename && !lastname)
+			return;
+		if (filename && lastname && !strcmp(lastname, filename))
+			return;
+	}
+
+	/*
+	 * the queue has already overflowed and we have already sent the
+	 * Q_OVERFLOW event
+	 */
+	if (dev->event_count > dev->max_events)
+		return;
+
+	/* the queue has just overflowed and we need to notify user space */
+	if (dev->event_count == dev->max_events) {
+		kevent = kernel_event(-1, IN_Q_OVERFLOW, cookie, NULL);
+		goto add_event_to_queue;
+	}
+
+	kevent = kernel_event(watch->wd, mask, cookie, filename);
+
+add_event_to_queue:
+	if (!kevent)
+		return;
+
+	/* 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->wait);
+}
+
+static inline int inotify_dev_has_events(struct inotify_device *dev)
+{
+	return !list_empty(&dev->events);
+}
+
+/*
+ * inotify_dev_event_dequeue - destroy an event on the given device
+ *
+ * Caller must hold dev->lock.
+ */
+static void inotify_dev_event_dequeue(struct inotify_device *dev)
+{
+	struct inotify_kernel_event *kevent;
+
+	if (!inotify_dev_has_events(dev))
+		return;
+
+	kevent = inotify_dev_get_event(dev);
+	list_del_init(&kevent->list);
+	if (kevent->filename)
+		kfree(kevent->filename);
+
+	dev->event_count--;
+	dev->queue_size -= sizeof(struct inotify_event) + kevent->event.len;	
+
+	kmem_cache_free(event_cachep, kevent);
+}
+
+/*
+ * inotify_dev_get_wd - returns the next WD for use by the given dev
+ *
+ * This function can sleep.
+ */
+static int inotify_dev_get_wd(struct inotify_device *dev,
+			     struct inotify_watch *watch)
+{
+	int ret;
+
+	if (atomic_read(&dev->user->inotify_watches) >=
+			sysfs_attrib_max_user_watches)
+		return -ENOSPC;
+
+repeat:
+	if (!idr_pre_get(&dev->idr, GFP_KERNEL))
+		return -ENOSPC;
+	spin_lock(&dev->lock);
+	ret = idr_get_new(&dev->idr, watch, &watch->wd);
+	spin_unlock(&dev->lock);
+	if (ret == -EAGAIN) /* more memory is required, try again */
+		goto repeat;
+	else if (ret)       /* the idr is full! */
+		return -ENOSPC;
+
+	atomic_inc(&dev->user->inotify_watches);
+
+	return 0;
+}
+
+/*
+ * inotify_dev_put_wd - release the given WD on the given device
+ *
+ * Caller must hold dev->lock.
+ */
+static int inotify_dev_put_wd(struct inotify_device *dev, s32 wd)
+{
+	if (!dev || wd < 0)
+		return -1;
+
+	atomic_dec(&dev->user->inotify_watches);
+	idr_remove(&dev->idr, wd);
+
+	return 0;
+}
+
+/*
+ * create_watch - creates a watch on the given device.
+ *
+ * Grabs dev->lock, so the caller must not hold it.
+ */
+static struct inotify_watch *create_watch(struct inotify_device *dev,
+					  u32 mask, struct inode *inode)
+{
+	struct inotify_watch *watch;
+
+	watch = kmem_cache_alloc(watch_cachep, GFP_KERNEL);
+	if (!watch)
+		return NULL;
+
+	watch->mask = mask;
+	watch->inode = inode;
+	watch->dev = dev;
+	INIT_LIST_HEAD(&watch->d_list);
+	INIT_LIST_HEAD(&watch->i_list);
+
+	if (inotify_dev_get_wd(dev, watch)) {
+		kmem_cache_free(watch_cachep, watch);
+		return NULL;
+	}
+
+	return watch;
+}
+
+/*
+ * delete_watch - removes the given 'watch' from the given 'dev'
+ *
+ * Caller must hold dev->lock.
+ */
+static void delete_watch(struct inotify_device *dev,
+			 struct inotify_watch *watch)
+{
+	inotify_dev_put_wd(dev, watch->wd);
+	kmem_cache_free(watch_cachep, watch);
+}
+
+/*
+ * inotify_find_dev - find the watch associated with the given inode and dev
+ *
+ * Caller must hold dev->lock.
+ * FIXME: Needs inotify_data->lock too.  Don't need dev->lock, just pin it.
+ */
+static struct inotify_watch *inode_find_dev(struct inode *inode,
+					    struct inotify_device *dev)
+{
+	struct inotify_watch *watch;
+
+	if (!inode->inotify_data)
+		return NULL;
+
+	list_for_each_entry(watch, &inode->inotify_data->watches, i_list) {
+		if (watch->dev == dev)
+			return watch;
+	}
+
+	return NULL;
+}
+
+/*
+ * dev_find_wd - given a (dev,wd) pair, returns the matching inotify_watcher
+ *
+ * Returns the results of looking up (dev,wd) in the idr layer.  NULL is
+ * returned on error.
+ *
+ * The caller must hold dev->lock.
+ */
+static inline struct inotify_watch *dev_find_wd(struct inotify_device *dev,
+						u32 wd)
+{
+	return idr_find(&dev->idr, wd);
+}
+
+static int inotify_dev_is_watching_inode(struct inotify_device *dev,
+					 struct inode *inode)
+{
+	struct inotify_watch *watch;
+
+	list_for_each_entry(watch, &dev->watches, d_list) {
+		if (watch->inode == inode)
+			return 1;
+	}
+
+	return 0;
+}
+
+/*
+ * inotify_dev_add_watcher - add the given watcher to the given device instance
+ *
+ * Caller must hold dev->lock.
+ */
+static int inotify_dev_add_watch(struct inotify_device *dev,
+				 struct inotify_watch *watch)
+{
+	if (!dev || !watch)
+		return -EINVAL;
+
+	list_add(&watch->d_list, &dev->watches);
+	return 0;
+}
+
+/*
+ * inotify_dev_rm_watch - remove the given watch from the given device
+ *
+ * Caller must hold dev->lock because we call inotify_dev_queue_event().
+ */
+static int inotify_dev_rm_watch(struct inotify_device *dev,
+				struct inotify_watch *watch)
+{
+	if (!watch)
+		return -EINVAL;
+
+	inotify_dev_queue_event(dev, watch, IN_IGNORED, 0, NULL);
+	list_del_init(&watch->d_list);
+
+	return 0;
+}
+
+/*
+ * inode_add_watch - add a watch to the given inode
+ *
+ * Callers must hold dev->lock, because we call inode_find_dev().
+ */
+static int inode_add_watch(struct inode *inode, struct inotify_watch *watch)
+{
+	int ret;
+
+	if (!inode || !watch)
+		return -EINVAL;
+
+	spin_lock(&inode->i_lock);
+	if (!inode->inotify_data) {
+		/* inotify_data is not attached to the inode, so add it */
+		inode->inotify_data = kmem_cache_alloc(inode_data_cachep,
+						       GFP_ATOMIC);
+		if (!inode->inotify_data) {
+			ret = -ENOMEM;
+			goto out_lock;
+		}
+
+		atomic_set(&inode->inotify_data->count, 0);
+		INIT_LIST_HEAD(&inode->inotify_data->watches);
+		spin_lock_init(&inode->inotify_data->lock);
+	} else if (inode_find_dev(inode, watch->dev)) {
+		/* a watch is already associated with this (inode,dev) pair */
+		ret = -EINVAL;
+		goto out_lock;
+	}
+	__get_inode_data(inode->inotify_data);
+	spin_unlock(&inode->i_lock);
+
+	list_add(&watch->i_list, &inode->inotify_data->watches);
+
+	return 0;
+out_lock:
+	spin_unlock(&inode->i_lock);
+	return ret;
+}
+
+static int inode_rm_watch(struct inode *inode,
+			  struct inotify_watch *watch)
+{
+	if (!inode || !watch || !inode->inotify_data)
+		return -EINVAL;
+
+	list_del_init(&watch->i_list);
+
+	/* clean up inode->inotify_data */
+	put_inode_data(inode);	
+
+	return 0;
+}
+
+/* Kernel API */
+
+/*
+ * inotify_inode_queue_event - queue an event with the given mask, cookie, and
+ * filename to any watches associated with the given inode.
+ *
+ * inode must be pinned prior to calling.
+ */
+void inotify_inode_queue_event(struct inode *inode, u32 mask, u32 cookie,
+			       const char *name)
+{
+	struct inotify_watch *watch;
+
+	if (!inode->inotify_data)
+		return;
+
+	list_for_each_entry(watch, &inode->inotify_data->watches, i_list) {
+		if (watch->mask & mask) {
+			struct inotify_device *dev = watch->dev;
+			spin_lock(&dev->lock);
+			inotify_dev_queue_event(dev, watch, mask, cookie, name);
+			spin_unlock(&dev->lock);
+		}
+	}
+}
+EXPORT_SYMBOL_GPL(inotify_inode_queue_event);
+
+void inotify_dentry_parent_queue_event(struct dentry *dentry, u32 mask,
+				       u32 cookie, const char *filename)
+{
+	struct dentry *parent;
+	struct inode *inode;
+
+	spin_lock(&dentry->d_lock);
+	parent = dentry->d_parent;
+	inode = parent->d_inode;
+	if (inode->inotify_data) {
+		dget(parent);
+		spin_unlock(&dentry->d_lock);
+		inotify_inode_queue_event(inode, mask, cookie, filename);
+		dput(parent);
+	} else
+		spin_unlock(&dentry->d_lock);
+}
+EXPORT_SYMBOL_GPL(inotify_dentry_parent_queue_event);
+
+u32 inotify_get_cookie(void)
+{
+	atomic_inc(&inotify_cookie);
+	return atomic_read(&inotify_cookie);
+}
+EXPORT_SYMBOL_GPL(inotify_get_cookie);
+
+/*
+ * Caller must hold dev->lock.
+ */
+static void __remove_watch(struct inotify_watch *watch,
+			   struct inotify_device *dev)
+{
+	struct inode *inode;
+
+	inode = watch->inode;
+
+	inode_rm_watch(inode, watch);
+	inotify_dev_rm_watch(dev, watch);
+	delete_watch(dev, watch);
+
+	iput(inode);
+}
+
+/*
+ * destroy_watch - remove a watch from both the device and the inode.
+ *
+ * watch->inode must be pinned.  We drop a reference before returning.  Grabs
+ * dev->lock.
+ */
+static void remove_watch(struct inotify_watch *watch)
+{
+	struct inotify_device *dev = watch->dev;
+
+	spin_lock(&dev->lock);
+	__remove_watch(watch, dev);
+	spin_unlock(&dev->lock);
+}
+
+void inotify_super_block_umount(struct super_block *sb)
+{
+	struct inode *inode;
+
+	spin_lock(&inode_lock);
+
+	/*
+	 * We hold the inode_lock, so the inodes are not going anywhere, and
+	 * we grab a reference on inotify_data before walking its list of
+	 * watches.
+	 */
+	list_for_each_entry(inode, &inode_in_use, i_list) {
+		struct inotify_inode_data *inode_data;
+		struct inotify_watch *watch;
+
+		if (inode->i_sb != sb)
+			continue;
+
+		inode_data = get_inode_data(inode);
+		if (!inode_data)
+			continue;
+
+		list_for_each_entry(watch, &inode_data->watches, i_list) {
+			struct inotify_device *dev = watch->dev;
+			spin_lock(&dev->lock);
+			inotify_dev_queue_event(dev, watch, IN_UNMOUNT, 0,
+						NULL);
+			__remove_watch(watch, dev);
+			spin_unlock(&dev->lock);
+		}
+		put_inode_data(inode);
+	}
+
+	spin_unlock(&inode_lock);
+}
+EXPORT_SYMBOL_GPL(inotify_super_block_umount);
+
+/*
+ * inotify_inode_is_dead - an inode has been deleted, cleanup any watches
+ */
+void inotify_inode_is_dead(struct inode *inode)
+{
+	struct inotify_watch *watch, *next;
+	struct inotify_inode_data *data;
+
+	data = get_inode_data(inode);
+	if (!data)
+		return;
+	list_for_each_entry_safe(watch, next, &data->watches, i_list)
+		remove_watch(watch);
+	put_inode_data(inode);
+}
+EXPORT_SYMBOL_GPL(inotify_inode_is_dead);
+
+/* The driver interface is implemented below */
+
+static unsigned int inotify_poll(struct file *file, poll_table *wait)
+{
+        struct inotify_device *dev;
+
+        dev = file->private_data;
+
+        poll_wait(file, &dev->wait, wait);
+
+        if (inotify_dev_has_events(dev))
+                return POLLIN | POLLRDNORM;
+
+        return 0;
+}
+
+static ssize_t inotify_read(struct file *file, char __user *buf,
+			    size_t count, loff_t *pos)
+{
+	size_t event_size;
+	struct inotify_device *dev;
+	char __user *start;
+	DECLARE_WAITQUEUE(wait, current);
+
+	start = buf;
+	dev = file->private_data;
+
+	/* We only hand out full inotify events */
+	event_size = sizeof(struct inotify_event);
+	if (count < event_size)
+		return 0;
+
+	while (1) {
+		int has_events;
+
+		spin_lock(&dev->lock);
+		has_events = inotify_dev_has_events(dev);
+		spin_unlock(&dev->lock);
+		if (has_events)
+			break;
+
+		if (file->f_flags & O_NONBLOCK)
+			return -EAGAIN;
+
+		if (signal_pending(current))
+			return -EINTR;
+
+		add_wait_queue(&dev->wait, &wait);
+		set_current_state(TASK_INTERRUPTIBLE);
+
+		schedule();
+
+		set_current_state(TASK_RUNNING);		
+		remove_wait_queue(&dev->wait, &wait);
+	}
+
+	while (count >= event_size) {
+		struct inotify_kernel_event *kevent;
+
+		spin_lock(&dev->lock);
+		if (!inotify_dev_has_events(dev)) {
+			spin_unlock(&dev->lock);
+			break;
+		}
+		kevent = inotify_dev_get_event(dev);
+		spin_unlock(&dev->lock);		
+
+		/* We can't send this event, not enough space in the buffer */
+		if (event_size + kevent->event.len > count)
+			break;
+
+		/* Copy the entire event except the string to user space */
+		if (copy_to_user(buf, &kevent->event, event_size)) 
+			return -EFAULT;
+
+		buf += event_size;
+		count -= event_size;
+
+		/* Copy the filename to user space */
+		if (kevent->filename) {
+			if (copy_to_user(buf, kevent->filename,
+					 kevent->event.len))
+				return -EFAULT;
+			buf += kevent->event.len;
+			count -= kevent->event.len;
+		}
+
+		spin_lock(&dev->lock);
+		inotify_dev_event_dequeue(dev);
+		spin_unlock(&dev->lock);
+	}
+
+	return buf - start;
+}
+
+static int inotify_open(struct inode *inode, struct file *file)
+{
+	struct inotify_device *dev;
+	struct user_struct *user;
+	int ret;
+
+	user = get_uid(current->user);
+
+	if (atomic_read(&user->inotify_devs) >= sysfs_attrib_max_user_devices) {
+		ret = -EMFILE;
+		goto out_err;
+	}
+
+	dev = kmalloc(sizeof(struct inotify_device), GFP_KERNEL);
+	if (!dev) {
+		ret = -ENOMEM;
+		goto out_err;
+	}
+
+	atomic_inc(&current->user->inotify_devs);	
+
+	idr_init(&dev->idr);
+
+	INIT_LIST_HEAD(&dev->events);
+	INIT_LIST_HEAD(&dev->watches);
+	init_waitqueue_head(&dev->wait);
+
+	dev->event_count = 0;
+	dev->queue_size = 0;
+	dev->max_events = sysfs_attrib_max_queued_events;
+	dev->user = user;
+	spin_lock_init(&dev->lock);
+
+	file->private_data = dev;
+
+	return 0;
+out_err:
+	free_uid(current->user);
+	return ret;
+}
+
+/*
+ * inotify_release_all_watches - destroy all watches on a given device
+ *
+ * FIXME: We need a lock on the watch here.
+ */
+static void inotify_release_all_watches(struct inotify_device *dev)
+{
+	struct inotify_watch *watch, *next;
+
+	list_for_each_entry_safe(watch, next, &dev->watches, d_list)
+		remove_watch(watch);
+}
+
+/*
+ * inotify_release_all_events - destroy all of the events on a given device
+ */
+static void inotify_release_all_events(struct inotify_device *dev)
+{
+	spin_lock(&dev->lock);
+	while (inotify_dev_has_events(dev))
+		inotify_dev_event_dequeue(dev);
+	spin_unlock(&dev->lock);
+}
+
+static int inotify_release(struct inode *inode, struct file *file)
+{
+	struct inotify_device *dev;
+
+	dev = file->private_data;
+
+	inotify_release_all_watches(dev);
+	inotify_release_all_events(dev);
+
+	atomic_dec(&dev->user->inotify_devs);
+	free_uid(dev->user);
+
+	kfree(dev);
+
+	return 0;
+}
+
+static int inotify_add_watch(struct inotify_device *dev,
+			     struct inotify_watch_request *request)
+{
+	struct inode *inode;
+	struct inotify_watch *watch;
+	struct nameidata nd;
+	int ret;
+
+	ret = find_inode((const char __user*) request->name, &nd);
+	if (ret)
+		return ret;
+
+	/* held in place by references in nd */
+	inode = nd.dentry->d_inode;
+
+	spin_lock(&dev->lock);
+
+	/*
+	 * This handles the case of re-adding a directory we are already
+	 * watching, we just update the mask and return 0
+	 */
+	if (inotify_dev_is_watching_inode(dev, inode)) {
+		struct inotify_watch *owatch;	/* the old watch */
+
+		owatch = inode_find_dev(inode, dev);
+		owatch->mask = request->mask;
+		spin_unlock(&dev->lock);
+		path_release(&nd);
+
+		return owatch->wd;
+	}
+
+	spin_unlock(&dev->lock);
+
+	watch = create_watch(dev, request->mask, inode);
+	if (!watch) {
+		path_release(&nd);
+		return -ENOSPC;
+	}
+
+	spin_lock(&dev->lock);
+
+	/* We can't add anymore watches to this device */
+	if (inotify_dev_add_watch(dev, watch)) {
+		delete_watch(dev, watch);
+		spin_unlock(&dev->lock);
+		path_release(&nd);
+		return -EINVAL;
+	}
+
+	ret = inode_add_watch(inode, watch);
+	if (ret < 0) {
+		list_del_init(&watch->d_list);
+		delete_watch(dev, watch);
+		spin_unlock(&dev->lock);
+		path_release(&nd);
+		return ret;
+	}
+
+	spin_unlock(&dev->lock);
+
+	/*
+	 * Demote the reference to nameidata to a reference to the inode held
+	 * by the watch.
+	 */
+	spin_lock(&inode_lock);
+	__iget(inode);
+	spin_unlock(&inode_lock);
+	path_release(&nd);
+
+	return watch->wd;
+}
+
+static int inotify_ignore(struct inotify_device *dev, s32 wd)
+{
+	struct inotify_watch *watch;
+	int ret = 0;
+
+	spin_lock(&dev->lock);
+	watch = dev_find_wd(dev, wd);
+	spin_unlock(&dev->lock);
+	if (!watch) {
+		ret = -EINVAL;
+		goto out;
+	}
+	__remove_watch(watch, dev);
+
+out:
+	spin_unlock(&dev->lock);
+	return ret;
+}
+
+/*
+ * inotify_ioctl() - our device file's ioctl method
+ *
+ * The VFS serializes all of our calls via the BKL and we rely on that.  We
+ * could, alternatively, grab dev->lock.  Right now lower levels grab that
+ * where needed.
+ */
+static int inotify_ioctl(struct inode *ip, struct file *fp,
+			 unsigned int cmd, unsigned long arg)
+{
+	struct inotify_device *dev;
+	struct inotify_watch_request request;
+	void __user *p;
+	s32 wd;
+
+	dev = fp->private_data;
+	p = (void __user *) arg;
+
+	switch (cmd) {
+	case INOTIFY_WATCH:
+		if (copy_from_user(&request, p, sizeof (request)))
+			return -EFAULT;
+		return inotify_add_watch(dev, &request);
+	case INOTIFY_IGNORE:
+		if (copy_from_user(&wd, p, sizeof (wd)))
+			return -EFAULT;
+		return inotify_ignore(dev, wd);
+	case FIONREAD:
+		return put_user(dev->queue_size, (int __user *) p);
+	default:
+		return -ENOTTY;
+	}
+}
+
+static struct file_operations inotify_fops = {
+	.owner		= THIS_MODULE,
+	.poll		= inotify_poll,
+	.read		= inotify_read,
+	.open		= inotify_open,
+	.release	= inotify_release,
+	.ioctl		= inotify_ioctl,
+};
+
+static struct miscdevice inotify_device = {
+	.minor  = MISC_DYNAMIC_MINOR,
+	.name	= "inotify",
+	.fops	= &inotify_fops,
+};
+
+static int __init inotify_init(void)
+{
+	struct class_device *class;
+	int ret;
+
+	ret = misc_register(&inotify_device);
+	if (ret)
+		return ret;
+
+	sysfs_attrib_max_queued_events = 512;
+	sysfs_attrib_max_user_devices = 64;
+	sysfs_attrib_max_user_watches = 16384;
+
+	class = inotify_device.class;
+	class_device_create_file(class, &class_device_attr_max_queued_events);
+	class_device_create_file(class, &class_device_attr_max_user_devices);
+	class_device_create_file(class, &class_device_attr_max_user_watches);
+
+	atomic_set(&inotify_cookie, 0);
+
+	watch_cachep = kmem_cache_create("inotify_watch_cache",
+			sizeof(struct inotify_watch), 0, SLAB_PANIC,
+			NULL, NULL);
+
+	event_cachep = kmem_cache_create("inotify_event_cache",
+			sizeof(struct inotify_kernel_event), 0,
+			SLAB_PANIC, NULL, NULL);
+
+	inode_data_cachep = kmem_cache_create("inotify_inode_data_cache",
+			sizeof(struct inotify_inode_data), 0, SLAB_PANIC,
+			NULL, NULL);
+
+	printk(KERN_INFO "inotify device minor=%d\n", inotify_device.minor);
+
+	return 0;
+}
+
+module_init(inotify_init);
diff -urN linux-2.6.10/drivers/char/Kconfig linux/drivers/char/Kconfig
--- linux-2.6.10/drivers/char/Kconfig	2004-12-24 16:33:49.000000000 -0500
+++ linux/drivers/char/Kconfig	2005-01-18 16:11:08.000000000 -0500
@@ -62,6 +62,19 @@
 	depends on VT && !S390 && !USERMODE
 	default y
 
+config INOTIFY
+	bool "Inotify file change notification support"
+	default y
+	---help---
+	  Say Y here to enable inotify support and the /dev/inotify character
+	  device.  Inotify is a file change notification system and a
+	  replacement for dnotify.  Inotify fixes numerous shortcomings in
+	  dnotify and introduces several new features.  It allows monitoring
+	  of both files and directories via a single open fd.  Multiple file
+	  events are supported.
+	  
+	  If unsure, say Y.
+
 config SERIAL_NONSTANDARD
 	bool "Non-standard serial port support"
 	---help---
diff -urN linux-2.6.10/drivers/char/Makefile linux/drivers/char/Makefile
--- linux-2.6.10/drivers/char/Makefile	2004-12-24 16:35:29.000000000 -0500
+++ linux/drivers/char/Makefile	2005-01-18 16:11:08.000000000 -0500
@@ -9,6 +9,8 @@
 
 obj-y	 += mem.o random.o tty_io.o n_tty.o tty_ioctl.o
 
+
+obj-$(CONFIG_INOTIFY)           += inotify.o
 obj-$(CONFIG_LEGACY_PTYS)	+= pty.o
 obj-$(CONFIG_UNIX98_PTYS)	+= pty.o
 obj-y				+= misc.o
diff -urN linux-2.6.10/drivers/char/misc.c linux/drivers/char/misc.c
--- linux-2.6.10/drivers/char/misc.c	2004-12-24 16:35:28.000000000 -0500
+++ linux/drivers/char/misc.c	2005-01-18 16:11:08.000000000 -0500
@@ -207,10 +207,9 @@
 int misc_register(struct miscdevice * misc)
 {
 	struct miscdevice *c;
-	struct class_device *class;
 	dev_t dev;
 	int err;
-	
+
 	down(&misc_sem);
 	list_for_each_entry(c, &misc_list, list) {
 		if (c->minor == misc->minor) {
@@ -224,8 +223,7 @@
 		while (--i >= 0)
 			if ( (misc_minors[i>>3] & (1 << (i&7))) == 0)
 				break;
-		if (i<0)
-		{
+		if (i<0) {
 			up(&misc_sem);
 			return -EBUSY;
 		}
@@ -240,10 +238,10 @@
 	}
 	dev = MKDEV(MISC_MAJOR, misc->minor);
 
-	class = class_simple_device_add(misc_class, dev,
-					misc->dev, misc->name);
-	if (IS_ERR(class)) {
-		err = PTR_ERR(class);
+	misc->class = class_simple_device_add(misc_class, dev,
+					      misc->dev, misc->name);
+	if (IS_ERR(misc->class)) {
+		err = PTR_ERR(misc->class);
 		goto out;
 	}
 
diff -urN linux-2.6.10/fs/attr.c linux/fs/attr.c
--- linux-2.6.10/fs/attr.c	2004-12-24 16:34:00.000000000 -0500
+++ linux/fs/attr.c	2005-01-31 15:52:37.000000000 -0500
@@ -10,7 +10,7 @@
 #include <linux/mm.h>
 #include <linux/string.h>
 #include <linux/smp_lock.h>
-#include <linux/dnotify.h>
+#include <linux/fsnotify.h>
 #include <linux/fcntl.h>
 #include <linux/quotaops.h>
 #include <linux/security.h>
@@ -103,31 +103,8 @@
 out:
 	return error;
 }
-
 EXPORT_SYMBOL(inode_setattr);
 
-int setattr_mask(unsigned int ia_valid)
-{
-	unsigned long dn_mask = 0;
-
-	if (ia_valid & ATTR_UID)
-		dn_mask |= DN_ATTRIB;
-	if (ia_valid & ATTR_GID)
-		dn_mask |= DN_ATTRIB;
-	if (ia_valid & ATTR_SIZE)
-		dn_mask |= DN_MODIFY;
-	/* both times implies a utime(s) call */
-	if ((ia_valid & (ATTR_ATIME|ATTR_MTIME)) == (ATTR_ATIME|ATTR_MTIME))
-		dn_mask |= DN_ATTRIB;
-	else if (ia_valid & ATTR_ATIME)
-		dn_mask |= DN_ACCESS;
-	else if (ia_valid & ATTR_MTIME)
-		dn_mask |= DN_MODIFY;
-	if (ia_valid & ATTR_MODE)
-		dn_mask |= DN_ATTRIB;
-	return dn_mask;
-}
-
 int notify_change(struct dentry * dentry, struct iattr * attr)
 {
 	struct inode *inode = dentry->d_inode;
@@ -183,11 +160,10 @@
 				error = inode_setattr(inode, attr);
 		}
 	}
-	if (!error) {
-		unsigned long dn_mask = setattr_mask(ia_valid);
-		if (dn_mask)
-			dnotify_parent(dentry, dn_mask);
-	}
+
+	if (!error)
+		fsnotify_change(dentry, ia_valid);
+
 	return error;
 }
 
diff -urN linux-2.6.10/fs/compat.c linux/fs/compat.c
--- linux-2.6.10/fs/compat.c	2004-12-24 16:34:44.000000000 -0500
+++ linux/fs/compat.c	2005-02-04 12:07:47.000000000 -0500
@@ -35,7 +35,7 @@
 #include <linux/ctype.h>
 #include <linux/module.h>
 #include <linux/dirent.h>
-#include <linux/dnotify.h>
+#include <linux/fsnotify.h>
 #include <linux/highuid.h>
 #include <linux/sunrpc/svc.h>
 #include <linux/nfsd/nfsd.h>
@@ -1192,9 +1192,15 @@
 out:
 	if (iov != iovstack)
 		kfree(iov);
-	if ((ret + (type == READ)) > 0)
-		dnotify_parent(file->f_dentry,
-				(type == READ) ? DN_ACCESS : DN_MODIFY);
+	if ((ret + (type == READ)) > 0) {
+		struct dentry *dentry = file->f_dentry;
+		if (type == READ)
+			fsnotify_access(dentry, dentry->d_inode,
+					dentry->d_name.name);
+		else
+			fsnotify_modify(dentry, dentry->d_inode,
+					dentry->d_name.name);
+	}
 	return ret;
 }
 
diff -urN linux-2.6.10/fs/file_table.c linux/fs/file_table.c
--- linux-2.6.10/fs/file_table.c	2004-12-24 16:33:50.000000000 -0500
+++ linux/fs/file_table.c	2005-01-31 15:46:49.000000000 -0500
@@ -16,6 +16,7 @@
 #include <linux/eventpoll.h>
 #include <linux/mount.h>
 #include <linux/cdev.h>
+#include <linux/fsnotify.h>
 
 /* sysctl tunables... */
 struct files_stat_struct files_stat = {
@@ -121,6 +122,9 @@
 	struct vfsmount *mnt = file->f_vfsmnt;
 	struct inode *inode = dentry->d_inode;
 
+
+	fsnotify_close(dentry, inode, file->f_mode, dentry->d_name.name);
+
 	might_sleep();
 	/*
 	 * The function eventpoll_release() should be the first called
diff -urN linux-2.6.10/fs/inode.c linux/fs/inode.c
--- linux-2.6.10/fs/inode.c	2004-12-24 16:35:40.000000000 -0500
+++ linux/fs/inode.c	2005-01-18 16:11:08.000000000 -0500
@@ -130,6 +130,9 @@
 #ifdef CONFIG_QUOTA
 		memset(&inode->i_dquot, 0, sizeof(inode->i_dquot));
 #endif
+#ifdef CONFIG_INOTIFY
+		inode->inotify_data = NULL;
+#endif
 		inode->i_pipe = NULL;
 		inode->i_bdev = NULL;
 		inode->i_cdev = NULL;
diff -urN linux-2.6.10/fs/namei.c linux/fs/namei.c
--- linux-2.6.10/fs/namei.c	2004-12-24 16:34:30.000000000 -0500
+++ linux/fs/namei.c	2005-01-31 17:24:21.000000000 -0500
@@ -21,7 +21,7 @@
 #include <linux/namei.h>
 #include <linux/quotaops.h>
 #include <linux/pagemap.h>
-#include <linux/dnotify.h>
+#include <linux/fsnotify.h>
 #include <linux/smp_lock.h>
 #include <linux/personality.h>
 #include <linux/security.h>
@@ -1241,7 +1241,7 @@
 	DQUOT_INIT(dir);
 	error = dir->i_op->create(dir, dentry, mode, nd);
 	if (!error) {
-		inode_dir_notify(dir, DN_CREATE);
+		fsnotify_create(dir, dentry->d_name.name);
 		security_inode_post_create(dir, dentry, mode);
 	}
 	return error;
@@ -1555,7 +1555,7 @@
 	DQUOT_INIT(dir);
 	error = dir->i_op->mknod(dir, dentry, mode, dev);
 	if (!error) {
-		inode_dir_notify(dir, DN_CREATE);
+		fsnotify_create(dir, dentry->d_name.name);
 		security_inode_post_mknod(dir, dentry, mode, dev);
 	}
 	return error;
@@ -1628,7 +1628,7 @@
 	DQUOT_INIT(dir);
 	error = dir->i_op->mkdir(dir, dentry, mode);
 	if (!error) {
-		inode_dir_notify(dir, DN_CREATE);
+		fsnotify_mkdir(dir, dentry->d_name.name);
 		security_inode_post_mkdir(dir,dentry, mode);
 	}
 	return error;
@@ -1722,10 +1722,8 @@
 		}
 	}
 	up(&dentry->d_inode->i_sem);
-	if (!error) {
-		inode_dir_notify(dir, DN_DELETE);
-		d_delete(dentry);
-	}
+	if (!error)
+		fsnotify_rmdir(dentry, dentry->d_inode, dir);
 	dput(dentry);
 
 	return error;
@@ -1795,10 +1793,9 @@
 	up(&dentry->d_inode->i_sem);
 
 	/* We don't d_delete() NFS sillyrenamed files--they still exist. */
-	if (!error && !(dentry->d_flags & DCACHE_NFSFS_RENAMED)) {
-		d_delete(dentry);
-		inode_dir_notify(dir, DN_DELETE);
-	}
+	if (!error && !(dentry->d_flags & DCACHE_NFSFS_RENAMED))
+		fsnotify_unlink(dentry->d_inode, dir, dentry);
+
 	return error;
 }
 
@@ -1872,7 +1869,7 @@
 	DQUOT_INIT(dir);
 	error = dir->i_op->symlink(dir, dentry, oldname);
 	if (!error) {
-		inode_dir_notify(dir, DN_CREATE);
+		fsnotify_create(dir, dentry->d_name.name);
 		security_inode_post_symlink(dir, dentry, oldname);
 	}
 	return error;
@@ -1945,7 +1942,7 @@
 	error = dir->i_op->link(old_dentry, dir, new_dentry);
 	up(&old_dentry->d_inode->i_sem);
 	if (!error) {
-		inode_dir_notify(dir, DN_CREATE);
+		fsnotify_create(dir, new_dentry->d_name.name);
 		security_inode_post_link(old_dentry, dir, new_dentry);
 	}
 	return error;
@@ -2109,6 +2106,7 @@
 {
 	int error;
 	int is_dir = S_ISDIR(old_dentry->d_inode->i_mode);
+	char *old_name;
 
 	if (old_dentry->d_inode == new_dentry->d_inode)
  		return 0;
@@ -2130,18 +2128,18 @@
 	DQUOT_INIT(old_dir);
 	DQUOT_INIT(new_dir);
 
+	old_name = fsnotify_oldname_init(old_dentry);
+
 	if (is_dir)
 		error = vfs_rename_dir(old_dir,old_dentry,new_dir,new_dentry);
 	else
 		error = vfs_rename_other(old_dir,old_dentry,new_dir,new_dentry);
 	if (!error) {
-		if (old_dir == new_dir)
-			inode_dir_notify(old_dir, DN_RENAME);
-		else {
-			inode_dir_notify(old_dir, DN_DELETE);
-			inode_dir_notify(new_dir, DN_CREATE);
-		}
+		const char *new_name = old_dentry->d_name.name;
+		fsnotify_move(old_dir, new_dir, old_name, new_name);
 	}
+	fsnotify_oldname_free(old_name);
+
 	return error;
 }
 
diff -urN linux-2.6.10/fs/open.c linux/fs/open.c
--- linux-2.6.10/fs/open.c	2004-12-24 16:33:50.000000000 -0500
+++ linux/fs/open.c	2005-02-02 11:26:06.000000000 -0500
@@ -10,7 +10,7 @@
 #include <linux/file.h>
 #include <linux/smp_lock.h>
 #include <linux/quotaops.h>
-#include <linux/dnotify.h>
+#include <linux/fsnotify.h>
 #include <linux/module.h>
 #include <linux/slab.h>
 #include <linux/tty.h>
@@ -953,9 +953,14 @@
 		fd = get_unused_fd();
 		if (fd >= 0) {
 			struct file *f = filp_open(tmp, flags, mode);
+			struct dentry *dentry;
+
 			error = PTR_ERR(f);
 			if (IS_ERR(f))
 				goto out_error;
+			dentry = f->f_dentry;
+			fsnotify_open(dentry, dentry->d_inode,
+				      dentry->d_name.name);
 			fd_install(fd, f);
 		}
 out:
@@ -1007,7 +1012,7 @@
 			retval = err;
 	}
 
-	dnotify_flush(filp, id);
+	fsnotify_flush(filp, id);
 	locks_remove_posix(filp, id);
 	fput(filp);
 	return retval;
diff -urN linux-2.6.10/fs/read_write.c linux/fs/read_write.c
--- linux-2.6.10/fs/read_write.c	2004-12-24 16:35:00.000000000 -0500
+++ linux/fs/read_write.c	2005-01-31 16:35:05.000000000 -0500
@@ -10,7 +10,7 @@
 #include <linux/file.h>
 #include <linux/uio.h>
 #include <linux/smp_lock.h>
-#include <linux/dnotify.h>
+#include <linux/fsnotify.h>
 #include <linux/security.h>
 #include <linux/module.h>
 #include <linux/syscalls.h>
@@ -216,8 +216,11 @@
 				ret = file->f_op->read(file, buf, count, pos);
 			else
 				ret = do_sync_read(file, buf, count, pos);
-			if (ret > 0)
-				dnotify_parent(file->f_dentry, DN_ACCESS);
+			if (ret > 0) {
+				struct dentry *dentry = file->f_dentry;
+				fsnotify_access(dentry, inode,
+						dentry->d_name.name);
+			}
 		}
 	}
 
@@ -260,8 +263,11 @@
 				ret = file->f_op->write(file, buf, count, pos);
 			else
 				ret = do_sync_write(file, buf, count, pos);
-			if (ret > 0)
-				dnotify_parent(file->f_dentry, DN_MODIFY);
+			if (ret > 0) {
+				struct dentry *dentry = file->f_dentry;
+				fsnotify_modify(dentry, inode,
+						dentry->d_name.name);
+			}
 		}
 	}
 
@@ -493,9 +499,15 @@
 out:
 	if (iov != iovstack)
 		kfree(iov);
-	if ((ret + (type == READ)) > 0)
-		dnotify_parent(file->f_dentry,
-				(type == READ) ? DN_ACCESS : DN_MODIFY);
+	if ((ret + (type == READ)) > 0) {
+		struct dentry *dentry = file->f_dentry;
+		struct inode *inode = dentry->d_inode;
+
+		if (type == READ)
+			fsnotify_access(dentry, inode, dentry->d_name.name);
+		else
+			fsnotify_modify(dentry, inode, dentry->d_name.name);
+	}
 	return ret;
 }
 
diff -urN linux-2.6.10/fs/super.c linux/fs/super.c
--- linux-2.6.10/fs/super.c	2004-12-24 16:34:33.000000000 -0500
+++ linux/fs/super.c	2005-01-31 14:53:38.000000000 -0500
@@ -37,9 +37,9 @@
 #include <linux/writeback.h>		/* for the emergency remount stuff */
 #include <linux/idr.h>
 #include <linux/kobject.h>
+#include <linux/fsnotify.h>
 #include <asm/uaccess.h>
 
-
 void get_filesystem(struct file_system_type *fs);
 void put_filesystem(struct file_system_type *fs);
 struct file_system_type *get_fs_type(const char *name);
@@ -227,6 +227,7 @@
 
 	if (root) {
 		sb->s_root = NULL;
+		fsnotify_sb_umount(sb);
 		shrink_dcache_parent(root);
 		shrink_dcache_anon(&sb->s_anon);
 		dput(root);
diff -urN linux-2.6.10/include/linux/fs.h linux/include/linux/fs.h
--- linux-2.6.10/include/linux/fs.h	2004-12-24 16:34:27.000000000 -0500
+++ linux/include/linux/fs.h	2005-01-18 16:11:08.000000000 -0500
@@ -27,6 +27,7 @@
 struct kstatfs;
 struct vm_area_struct;
 struct vfsmount;
+struct inotify_inode_data;
 
 /*
  * It's silly to have NR_OPEN bigger than NR_FILE, but you can change
@@ -473,6 +474,10 @@
 	struct dnotify_struct	*i_dnotify; /* for directory notifications */
 #endif
 
+#ifdef CONFIG_INOTIFY
+	struct inotify_inode_data *inotify_data;
+#endif
+
 	unsigned long		i_state;
 	unsigned long		dirtied_when;	/* jiffies of first dirtying */
 
@@ -1353,7 +1358,7 @@
 extern int do_remount_sb(struct super_block *sb, int flags,
 			 void *data, int force);
 extern sector_t bmap(struct inode *, sector_t);
-extern int setattr_mask(unsigned int);
+extern void setattr_mask(unsigned int, int *, u32 *);
 extern int notify_change(struct dentry *, struct iattr *);
 extern int permission(struct inode *, int, struct nameidata *);
 extern int generic_permission(struct inode *, int,
diff -urN linux-2.6.10/include/linux/fsnotify.h linux/include/linux/fsnotify.h
--- linux-2.6.10/include/linux/fsnotify.h	1969-12-31 19:00:00.000000000 -0500
+++ linux/include/linux/fsnotify.h	2005-02-04 12:09:48.000000000 -0500
@@ -0,0 +1,235 @@
+#ifndef _LINUX_FS_NOTIFY_H
+#define _LINUX_FS_NOTIFY_H
+
+/*
+ * include/linux/fs_notify.h - generic hooks for filesystem notification, to
+ * reduce in-source duplication from both dnotify and inotify.
+ *
+ * We don't compile any of this away in some complicated menagerie of ifdefs.
+ * Instead, we rely on the code inside to optimize away as needed.
+ *
+ * (C) Copyright 2005 Robert Love
+ */
+
+#ifdef __KERNEL__
+
+#include <linux/dnotify.h>
+#include <linux/inotify.h>
+
+/*
+ * fsnotify_move - file old_name at old_dir was moved to new_name at new_dir
+ */
+static inline void fsnotify_move(struct inode *old_dir, struct inode *new_dir,
+				 const char *old_name, const char *new_name)
+{
+	u32 cookie;
+
+	if (old_dir == new_dir)
+		inode_dir_notify(old_dir, DN_RENAME);
+	else {
+		inode_dir_notify(old_dir, DN_DELETE);
+		inode_dir_notify(new_dir, DN_CREATE);
+	}
+
+	cookie = inotify_get_cookie();
+
+	inotify_inode_queue_event(old_dir, IN_MOVED_FROM, cookie, old_name);
+	inotify_inode_queue_event(new_dir, IN_MOVED_TO, cookie, new_name);
+}
+
+/*
+ * fsnotify_unlink - file was unlinked
+ */
+static inline void fsnotify_unlink(struct inode *inode, struct inode *dir,
+				   struct dentry *dentry)
+{
+	inode_dir_notify(dir, DN_DELETE);
+	inotify_inode_queue_event(dir, IN_DELETE_FILE, 0, dentry->d_name.name);
+	inotify_inode_queue_event(inode, IN_DELETE_SELF, 0, NULL);
+
+	inotify_inode_is_dead(inode);
+	d_delete(dentry);
+}
+
+/*
+ * fsnotify_rmdir - directory was removed
+ */
+static inline void fsnotify_rmdir(struct dentry *dentry, struct inode *inode,
+				  struct inode *dir)
+{
+	inode_dir_notify(dir, DN_DELETE);
+	inotify_inode_queue_event(dir, IN_DELETE_SUBDIR,0,dentry->d_name.name);
+	inotify_inode_queue_event(inode, IN_DELETE_SELF, 0, NULL);
+
+	inotify_inode_is_dead(inode);
+	d_delete(dentry);
+}
+
+/*
+ * fsnotify_create - filename was linked in
+ */
+static inline void fsnotify_create(struct inode *inode, const char *filename)
+{
+	inode_dir_notify(inode, DN_CREATE);
+	inotify_inode_queue_event(inode, IN_CREATE_FILE, 0, filename);
+}
+
+/*
+ * fsnotify_mkdir - directory 'name' was created
+ */
+static inline void fsnotify_mkdir(struct inode *inode, const char *name)
+{
+	inode_dir_notify(inode, DN_CREATE);
+	inotify_inode_queue_event(inode, IN_CREATE_SUBDIR, 0, name);
+}
+
+/*
+ * fsnotify_access - file was read
+ */
+static inline void fsnotify_access(struct dentry *dentry, struct inode *inode,
+				   const char *filename)
+{
+	dnotify_parent(dentry, DN_ACCESS);
+	inotify_dentry_parent_queue_event(dentry, IN_ACCESS, 0,
+					  dentry->d_name.name);
+	inotify_inode_queue_event(inode, IN_ACCESS, 0, NULL);
+}
+
+/*
+ * fsnotify_modify - file was modified
+ */
+static inline void fsnotify_modify(struct dentry *dentry, struct inode *inode,
+				   const char *filename)
+{
+	dnotify_parent(dentry, DN_MODIFY);
+	inotify_dentry_parent_queue_event(dentry, IN_MODIFY, 0, filename);
+	inotify_inode_queue_event(inode, IN_MODIFY, 0, NULL);
+}
+
+/*
+ * fsnotify_open - file was opened
+ */
+static inline void fsnotify_open(struct dentry *dentry, struct inode *inode,
+				 const char *filename)
+{
+	inotify_inode_queue_event(inode, IN_OPEN, 0, NULL);
+	inotify_dentry_parent_queue_event(dentry, IN_OPEN, 0, filename);
+}
+
+/*
+ * fsnotify_close - file was closed
+ */
+static inline void fsnotify_close(struct dentry *dentry, struct inode *inode,
+				  mode_t mode, const char *filename)
+{
+	u32 mask;
+
+	mask = (mode & FMODE_WRITE) ? IN_CLOSE_WRITE : IN_CLOSE_NOWRITE;
+	inotify_dentry_parent_queue_event(dentry, mask, 0, filename);
+	inotify_inode_queue_event(inode, mask, 0, NULL);
+}
+
+/*
+ * fsnotify_change - notify_change event.  file was modified and/or metadata
+ * was changed.
+ */
+static inline void fsnotify_change(struct dentry *dentry, unsigned int ia_valid)
+{
+	int dn_mask = 0;
+	u32 in_mask = 0;
+
+	if (ia_valid & ATTR_UID) {
+		in_mask |= IN_ATTRIB;
+		dn_mask |= DN_ATTRIB;
+	}
+	if (ia_valid & ATTR_GID) {
+		in_mask |= IN_ATTRIB;
+		dn_mask |= DN_ATTRIB;
+	}
+	if (ia_valid & ATTR_SIZE) {
+		in_mask |= IN_MODIFY;
+		dn_mask |= DN_MODIFY;
+	}
+	/* both times implies a utime(s) call */
+	if ((ia_valid & (ATTR_ATIME | ATTR_MTIME)) == (ATTR_ATIME | ATTR_MTIME))
+	{
+		in_mask |= IN_ATTRIB;
+		dn_mask |= DN_ATTRIB;
+	} else if (ia_valid & ATTR_ATIME) {
+		in_mask |= IN_ACCESS;
+		dn_mask |= DN_ACCESS;
+	} else if (ia_valid & ATTR_MTIME) {
+		in_mask |= IN_MODIFY;
+		dn_mask |= DN_MODIFY;
+	}
+	if (ia_valid & ATTR_MODE) {
+		in_mask |= IN_ATTRIB;
+		dn_mask |= DN_ATTRIB;
+	}
+
+	if (dn_mask)
+		dnotify_parent(dentry, dn_mask);
+	if (in_mask) {
+		inotify_inode_queue_event(dentry->d_inode, in_mask, 0, NULL);
+		inotify_dentry_parent_queue_event(dentry, in_mask, 0,
+						  dentry->d_name.name);
+	}
+}
+
+/*
+ * fsnotify_sb_umount - filesystem unmount
+ */
+static inline void fsnotify_sb_umount(struct super_block *sb)
+{
+	inotify_super_block_umount(sb);
+}
+
+/*
+ * fsnotify_flush - flush time!
+ */
+static inline void fsnotify_flush(struct file *filp, fl_owner_t id)
+{
+	dnotify_flush(filp, id);
+}
+
+#ifdef CONFIG_INOTIFY	/* inotify helpers */
+
+/*
+ * fsnotify_oldname_init - save off the old filename before we change it
+ *
+ * this could be kstrdup if only we could add that to lib/string.c
+ */
+static inline char *fsnotify_oldname_init(struct dentry *old_dentry)
+{
+	char *old_name;
+
+	old_name = kmalloc(strlen(old_dentry->d_name.name) + 1, GFP_KERNEL);
+	if (old_name)
+		strcpy(old_name, old_dentry->d_name.name);
+	return old_name;
+}
+
+/*
+ * fsnotify_oldname_free - free the name we got from fsnotify_oldname_init
+ */
+static inline void fsnotify_oldname_free(const char *old_name)
+{
+	kfree(old_name);
+}
+
+#else	/* CONFIG_INOTIFY */
+
+static inline char *fsnotify_oldname_init(struct dentry *old_dentry)
+{
+	return NULL;
+}
+
+static inline void fsnotify_oldname_free(const char *old_name)
+{
+}
+
+#endif	/* ! CONFIG_INOTIFY */
+
+#endif	/* __KERNEL__ */
+
+#endif	/* _LINUX_FS_NOTIFY_H */
diff -urN linux-2.6.10/include/linux/inotify.h linux/include/linux/inotify.h
--- linux-2.6.10/include/linux/inotify.h	1969-12-31 19:00:00.000000000 -0500
+++ linux/include/linux/inotify.h	2005-02-09 16:02:58.291978072 -0500
@@ -0,0 +1,118 @@
+/*
+ * Inode based directory notification for Linux
+ *
+ * Copyright (C) 2005 John McCutchan
+ */
+
+#ifndef _LINUX_INOTIFY_H
+#define _LINUX_INOTIFY_H
+
+#include <linux/types.h>
+#include <linux/limits.h>
+
+/*
+ * struct inotify_event - structure read from the inotify device for each event
+ *
+ * When you are watching a directory, you will receive the filename for events
+ * such as IN_CREATE, IN_DELETE, IN_OPEN, IN_CLOSE, ..., relative to the wd.
+ */
+struct inotify_event {
+	__s32 wd;	/* watch descriptor */
+	__u32 mask;	/* watch mask */
+	__u32 cookie;	/* cookie used for synchronizing two events */
+	size_t len;	/* length (including nulls) of name */
+	char name[0];	/* stub for possible name */
+};
+
+/*
+ * struct inotify_watch_request - represents a watch request
+ *
+ * Pass to the inotify device via the INOTIFY_WATCH ioctl
+ */
+struct inotify_watch_request {
+	char *name;		/* directory name */
+	__u32 mask;		/* event mask */
+};
+
+/* the following are legal, implemented events */
+#define IN_ACCESS		0x00000001	/* File was accessed */
+#define IN_MODIFY		0x00000002	/* File was modified */
+#define IN_ATTRIB		0x00000004	/* File changed attributes */
+#define IN_CLOSE_WRITE		0x00000008	/* Writtable file was closed */
+#define IN_CLOSE_NOWRITE	0x00000010	/* Unwrittable file closed */
+#define IN_OPEN			0x00000020	/* File was opened */
+#define IN_MOVED_FROM		0x00000040	/* File was moved from X */
+#define IN_MOVED_TO		0x00000080	/* File was moved to Y */
+#define IN_DELETE_SUBDIR	0x00000100	/* Subdir was deleted */ 
+#define IN_DELETE_FILE		0x00000200	/* Subfile was deleted */
+#define IN_CREATE_SUBDIR	0x00000400	/* Subdir was created */
+#define IN_CREATE_FILE		0x00000800	/* Subfile was created */
+#define IN_DELETE_SELF		0x00001000	/* Self was deleted */
+#define IN_UNMOUNT		0x00002000	/* Backing fs was unmounted */
+#define IN_Q_OVERFLOW		0x00004000	/* Event queued overflowed */
+#define IN_IGNORED		0x00008000	/* File was ignored */
+
+/* special flags */
+#define IN_ALL_EVENTS		0xffffffff	/* All the events */
+#define IN_CLOSE		(IN_CLOSE_WRITE | IN_CLOSE_NOWRITE)
+
+#define INOTIFY_IOCTL_MAGIC	'Q'
+#define INOTIFY_IOCTL_MAXNR	2
+
+#define INOTIFY_WATCH  		_IOR(INOTIFY_IOCTL_MAGIC, 1, struct inotify_watch_request)
+#define INOTIFY_IGNORE 		_IOR(INOTIFY_IOCTL_MAGIC, 2, int)
+
+#ifdef __KERNEL__
+
+#include <linux/dcache.h>
+#include <linux/fs.h>
+#include <linux/config.h>
+
+struct inotify_inode_data {
+	struct list_head watches;	/* list of watches on this inode */
+	spinlock_t lock;		/* lock protecting the struct */
+	atomic_t count;			/* ref count */
+};
+
+#ifdef CONFIG_INOTIFY
+
+extern void inotify_inode_queue_event(struct inode *, __u32, __u32,
+				      const char *);
+extern void inotify_dentry_parent_queue_event(struct dentry *, __u32, __u32,
+					      const char *);
+extern void inotify_super_block_umount(struct super_block *);
+extern void inotify_inode_is_dead(struct inode *);
+extern __u32 inotify_get_cookie(void);
+
+#else
+
+static inline void inotify_inode_queue_event(struct inode *inode,
+					     __u32 mask, __u32 cookie,
+					     const char *filename)
+{
+}
+
+static inline void inotify_dentry_parent_queue_event(struct dentry *dentry,
+						     __u32 mask, __u32 cookie,
+						     const char *filename)
+{
+}
+
+static inline void inotify_super_block_umount(struct super_block *sb)
+{
+}
+
+static inline void inotify_inode_is_dead(struct inode *inode)
+{
+}
+
+static inline __u32 inotify_get_cookie(void)
+{
+	return 0;
+}
+
+#endif	/* CONFIG_INOTIFY */
+
+#endif	/* __KERNEL __ */
+
+#endif	/* _LINUX_INOTIFY_H */
diff -urN linux-2.6.10/include/linux/miscdevice.h linux/include/linux/miscdevice.h
--- linux-2.6.10/include/linux/miscdevice.h	2004-12-24 16:34:58.000000000 -0500
+++ linux/include/linux/miscdevice.h	2005-01-18 16:11:08.000000000 -0500
@@ -2,6 +2,7 @@
 #define _LINUX_MISCDEVICE_H
 #include <linux/module.h>
 #include <linux/major.h>
+#include <linux/device.h>
 
 #define PSMOUSE_MINOR  1
 #define MS_BUSMOUSE_MINOR 2
@@ -32,13 +33,13 @@
 
 struct device;
 
-struct miscdevice 
-{
+struct miscdevice  {
 	int minor;
 	const char *name;
 	struct file_operations *fops;
 	struct list_head list;
 	struct device *dev;
+	struct class_device *class;
 	char devfs_name[64];
 };
 
diff -urN linux-2.6.10/include/linux/sched.h linux/include/linux/sched.h
--- linux-2.6.10/include/linux/sched.h	2004-12-24 16:33:59.000000000 -0500
+++ linux/include/linux/sched.h	2005-01-18 16:11:08.000000000 -0500
@@ -353,6 +353,8 @@
 	atomic_t processes;	/* How many processes does this user have? */
 	atomic_t files;		/* How many open files does this user have? */
 	atomic_t sigpending;	/* How many pending signals does this user have? */
+	atomic_t inotify_watches;	/* How many inotify watches does this user have? */
+	atomic_t inotify_devs;	/* How many inotify devs does this user have opened? */
 	/* protected by mq_lock	*/
 	unsigned long mq_bytes;	/* How many bytes can be allocated to mqueue? */
 	unsigned long locked_shm; /* How many pages of mlocked shm ? */
diff -urN linux-2.6.10/kernel/user.c linux/kernel/user.c
--- linux-2.6.10/kernel/user.c	2004-12-24 16:34:31.000000000 -0500
+++ linux/kernel/user.c	2005-01-18 16:11:08.000000000 -0500
@@ -119,6 +119,8 @@
 		atomic_set(&new->processes, 0);
 		atomic_set(&new->files, 0);
 		atomic_set(&new->sigpending, 0);
+		atomic_set(&new->inotify_watches, 0);
+		atomic_set(&new->inotify_devs, 0);
 
 		new->mq_bytes = 0;
 		new->locked_shm = 0;