inotify 0.14

@@ -0,0 +1,932 @@
+/*
+ * Inode based directory notifications for Linux.
+ *
+ * Copyright (C) 2004 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.
+ */
+
+/* TODO: 
+ * unmount events don't get sent if filesystem is mounted in two places
+ * dynamically allocate event filename
+ */
+
+#include <linux/bitmap.h>
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/spinlock.h>
+#include <linux/slab.h>
+#include <linux/fs.h>
+#include <linux/namei.h>
+#include <linux/poll.h>
+#include <linux/miscdevice.h>
+#include <linux/init.h>
+#include <linux/list.h>
+#include <linux/writeback.h>
+#include <linux/inotify.h>
+
+#define MAX_INOTIFY_DEVS	  8	/* max open inotify devices */
+#define MAX_INOTIFY_DEV_WATCHES   8192	/* max total watches */
+#define MAX_INOTIFY_QUEUED_EVENTS 256	/* max events queued on the dev */
+
+static atomic_t watch_count;
+static atomic_t inotify_cookie;
+static kmem_cache_t *watch_cachep;
+static kmem_cache_t *event_cachep;
+static kmem_cache_t *inode_data_cachep;
+
+/*
+ * 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.
+ *
+ * 'bitmask' holds one bit for each possible watch descriptor: a set bit
+ * implies that the given WD is valid, unset implies it is not.
+ *
+ * This structure is protected by 'lock'.  Lock ordering:
+ *
+ * inode->i_lock
+ *	dev->lock
+ *		dev->wait->lock
+ *
+ * FIXME: Look at replacing i_lock with i_sem.
+ */
+struct inotify_device {
+	DECLARE_BITMAP(bitmask, MAX_INOTIFY_DEV_WATCHES);
+	wait_queue_head_t 	wait;
+	struct list_head 	events;
+	struct list_head 	watches;
+	spinlock_t		lock;
+	unsigned int		event_count;
+	unsigned int		nr_watches;
+};
+
+struct inotify_watch {
+	s32 			wd;	/* watch descriptor */
+	u32			mask;
+	struct inode *		inode;
+	struct inotify_device *	dev;
+	struct list_head	d_list;	/* device list */
+	struct list_head	i_list; /* inode list */
+	struct list_head	u_list; /* unmount list */
+};
+#define inotify_watch_d_list(pos) list_entry((pos), struct inotify_watch, d_list)
+#define inotify_watch_i_list(pos) list_entry((pos), struct inotify_watch, i_list)
+#define inotify_watch_u_list(pos) list_entry((pos), struct inotify_watch, u_list)
+
+/*
+ * A list of these is attached to each instance of the driver
+ * when the drivers read() gets called, this list is walked and
+ * all events that can fit in the buffer get delivered
+ */
+struct inotify_kernel_event {
+        struct list_head        list;
+	struct inotify_event	event;
+};
+
+/*
+ * find_inode - resolve a user-given path to a specific inode and iget() it
+ */
+static struct inode * find_inode(const char __user *dirname)
+{
+	struct inode *inode;
+	struct nameidata nd;
+	int error;
+
+	error = __user_walk(dirname, LOOKUP_FOLLOW, &nd);
+	if (error) {
+		inode = ERR_PTR(error);
+		goto out;
+	}
+
+	inode = nd.dentry->d_inode;
+
+	/* you can only watch an inode if you have read permissions on it */
+	error = vfs_permission(inode, MAY_READ);
+	if (error) {
+		inode = ERR_PTR(error);
+		goto release_and_out;
+	}
+
+	__iget(inode);
+release_and_out:
+	path_release(&nd);
+out:
+	return inode;
+}
+
+static inline void unref_inode(struct inode *inode)
+{
+	iput(inode);
+}
+
+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)
+		goto out;
+
+	/* we hand this out to user-space, so zero it out just in case */
+	memset(kevent, 0, sizeof(struct inotify_kernel_event));
+
+	kevent->event.wd = wd;
+	kevent->event.mask = mask;
+	kevent->event.cookie = cookie;
+	INIT_LIST_HEAD(&kevent->list);
+
+	if (filename) {
+		strncpy(kevent->event.filename, filename,
+			INOTIFY_FILENAME_MAX);
+		kevent->event.filename[INOTIFY_FILENAME_MAX-1] = '\0';
+	} else
+		kevent->event.filename[0] = '\0';
+
+out:
+	return kevent;
+}
+
+void delete_kernel_event(struct inotify_kernel_event *kevent)
+{
+	if (!kevent)
+		return;
+	kmem_cache_free(event_cachep, 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))
+#define inotify_dev_has_events(dev)	(!list_empty(&dev->events))
+
+/* Does this events mask get sent to the watch ? */
+#define event_and(event_mask,watches_mask) 	((event_mask == IN_UNMOUNT) || \
+						(event_mask == IN_IGNORED) || \
+						(event_mask & watches_mask))
+
+/*
+ * 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;
+
+	/*
+	 * Check if the new event is a duplicate of the last event queued.
+	 */
+	last = inotify_dev_get_event(dev);
+	if (dev->event_count && last->event.mask == mask &&
+			last->event.wd == watch->wd) {
+		/* Check if the filenames match */
+		if (!filename && last->event.filename[0] == '\0')
+			return;
+		if (filename && !strcmp(last->event.filename, filename))
+			return;
+	}
+
+	/*
+	 * the queue has already overflowed and we have already sent the
+	 * Q_OVERFLOW event
+	 */
+	if (dev->event_count > MAX_INOTIFY_QUEUED_EVENTS)
+		return;
+
+	/* the queue has just overflowed and we need to notify user space */
+	if (dev->event_count == MAX_INOTIFY_QUEUED_EVENTS) {
+		dev->event_count++;
+		kevent = kernel_event(-1, IN_Q_OVERFLOW, cookie, NULL);
+		goto add_event_to_queue;
+	}
+
+	if (!event_and(mask, watch->inode->inotify_data->watch_mask) ||
+			!event_and(mask, watch->mask))
+		return;
+
+	dev->event_count++;
+	kevent = kernel_event(watch->wd, mask, cookie, filename);
+
+add_event_to_queue:
+	if (!kevent) {
+		dev->event_count--;
+		return;
+	}
+
+	/* queue the event and wake up anyone waiting */
+	list_add_tail(&kevent->list, &dev->events);
+	wake_up_interruptible(&dev->wait);
+}
+
+/*
+ * 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(&kevent->list);
+	dev->event_count--;
+	delete_kernel_event(kevent);
+
+}
+
+/*
+ * inotify_dev_get_wd - returns the next WD for use by the given dev
+ *
+ * Caller must hold dev->lock before calling.
+ */
+static int inotify_dev_get_wd(struct inotify_device *dev)
+{
+	s32 wd;
+
+	if (!dev || dev->nr_watches == MAX_INOTIFY_DEV_WATCHES)
+		return -1;
+
+	dev->nr_watches++;
+	wd = find_first_zero_bit(dev->bitmask, MAX_INOTIFY_DEV_WATCHES);
+	set_bit(wd, dev->bitmask);
+
+	return wd;
+}
+
+/*
+ * 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;
+
+	dev->nr_watches--;
+	clear_bit(wd, dev->bitmask);
+
+	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);
+	INIT_LIST_HEAD(&watch->u_list);
+
+	spin_lock(&dev->lock);
+	watch->wd = inotify_dev_get_wd(dev);
+	spin_unlock(&dev->lock);
+
+	if (watch->wd < 0) {
+		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.
+ */
+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;
+}
+
+static struct inotify_watch *dev_find_wd(struct inotify_device *dev, s32 wd)
+{
+	struct inotify_watch *watch;
+
+	list_for_each_entry(watch, &dev->watches, d_list) {
+		if (watch->wd == wd)
+			return watch;
+	}
+
+	return NULL;
+}
+
+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;
+
+	if (dev_find_wd (dev, watch->wd))
+		return -EINVAL;
+
+	if (dev->nr_watches == MAX_INOTIFY_DEV_WATCHES)
+		return -ENOSPC;
+
+	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(&watch->d_list);
+
+	return 0;
+}
+
+void inode_update_watch_mask(struct inode *inode)
+{
+	struct inotify_watch *watch;
+	u32 new_mask;
+
+	if (!inode->inotify_data)
+		return;
+
+	new_mask = 0;
+	list_for_each_entry(watch, &inode->inotify_data->watches, i_list)
+		new_mask |= watch->mask;
+
+	inode->inotify_data->watch_mask = new_mask;
+}
+
+/*
+ * 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)
+{
+	if (!inode || !watch || inode_find_dev(inode, watch->dev))
+		return -EINVAL;
+
+	/*
+	 * This inode doesn't have an inotify_data structure attached to it
+	 */
+	if (!inode->inotify_data) {
+		inode->inotify_data = kmem_cache_alloc(inode_data_cachep,
+						       GFP_ATOMIC);
+		INIT_LIST_HEAD(&inode->inotify_data->watches);
+		inode->inotify_data->watch_mask = 0;
+		inode->inotify_data->watch_count = 0;
+	}
+	list_add(&watch->i_list, &inode->inotify_data->watches);
+	inode->inotify_data->watch_count++;
+	inode_update_watch_mask(inode);
+
+	return 0;
+}
+
+static int inode_rm_watch(struct inode *inode,
+			  struct inotify_watch *watch)
+{
+	if (!inode || !watch || !inode->inotify_data)
+		return -EINVAL;
+
+	list_del(&watch->i_list);
+	inode->inotify_data->watch_count--;
+
+	if (!inode->inotify_data->watch_count) {
+		kmem_cache_free(inode_data_cachep, inode->inotify_data);
+		inode->inotify_data = NULL;
+	}
+
+	inode_update_watch_mask(inode);
+
+	return 0;
+}
+
+/* Kernel API */
+
+void inotify_inode_queue_event(struct inode *inode, u32 mask, u32 cookie,
+			       const char *filename)
+{
+	struct inotify_watch *watch;
+
+	if (!inode->inotify_data)
+		return;
+
+	spin_lock(&inode->i_lock);
+
+	list_for_each_entry(watch, &inode->inotify_data->watches, i_list) {
+		spin_lock(&watch->dev->lock);
+		inotify_dev_queue_event(watch->dev, watch, mask, cookie,
+					filename);
+		spin_unlock(&watch->dev->lock);
+	}
+
+	spin_unlock(&inode->i_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;
+
+	parent = dget_parent(dentry);
+	inotify_inode_queue_event(parent->d_inode, mask, cookie, filename);
+	dput(parent);
+}
+EXPORT_SYMBOL_GPL(inotify_dentry_parent_queue_event);
+
+u32 inotify_get_cookie()
+{
+	atomic_inc(&inotify_cookie);
+
+	return atomic_read(&inotify_cookie);
+}
+EXPORT_SYMBOL_GPL(inotify_get_cookie);
+
+static void ignore_helper(struct inotify_watch *watch, int event)
+{
+	struct inotify_device *dev;
+	struct inode *inode;
+
+	inode = watch->inode;
+	dev = watch->dev;
+
+	spin_lock(&inode->i_lock);
+	spin_lock(&dev->lock);
+
+	if (event)
+		inotify_dev_queue_event(dev, watch, event, 0, NULL);
+
+	inode_rm_watch(inode, watch);
+	inotify_dev_rm_watch(watch->dev, watch);
+	list_del(&watch->u_list);
+
+	delete_watch(dev, watch);
+	spin_unlock(&dev->lock);
+	spin_unlock(&inode->i_lock);
+
+	unref_inode(inode);
+}
+
+static void process_umount_list(struct list_head *umount)
+{
+	struct inotify_watch *watch, *next;
+
+	list_for_each_entry_safe(watch, next, umount, u_list)
+		ignore_helper(watch, IN_UNMOUNT);
+}
+
+/*
+ * build_umount_list - build a list of watches affected by an unmount.
+ *
+ * Caller must hold inode_lock.
+ */
+static void build_umount_list(struct list_head *head, struct super_block *sb,
+			      struct list_head *umount)
+{
+	struct inode *inode;
+
+	list_for_each_entry(inode, head, i_list) {
+		struct inotify_watch *watch;
+
+		if (inode->i_sb != sb)
+			continue;
+
+		if (!inode->inotify_data)
+			continue;
+
+		spin_lock(&inode->i_lock);
+
+		list_for_each_entry(watch, &inode->inotify_data->watches,
+				    i_list)
+			list_add(&watch->u_list, umount);
+
+		spin_unlock(&inode->i_lock);
+	}
+}
+
+void inotify_super_block_umount(struct super_block *sb)
+{
+	struct list_head umount;
+
+	INIT_LIST_HEAD(&umount);
+
+	spin_lock(&inode_lock);
+	build_umount_list(&inode_in_use, sb, &umount);
+	spin_unlock(&inode_lock);
+
+	process_umount_list(&umount);
+}
+EXPORT_SYMBOL_GPL(inotify_super_block_umount);
+
+/*
+ * inotify_inode_is_dead - an inode has been deleted, cleanup any watches
+ *
+ * FIXME: Callers need to always hold inode->i_lock.
+ */
+void inotify_inode_is_dead(struct inode *inode)
+{
+	struct inotify_watch *watch, *next;
+	struct inotify_inode_data *data;
+
+	data = inode->inotify_data;
+	if (!data)
+		return;
+
+	list_for_each_entry_safe(watch, next, &data->watches, i_list)
+		ignore_helper(watch, 0);
+}
+EXPORT_SYMBOL_GPL(inotify_inode_is_dead);
+
+/*
+ * setattr_mask_inotify - return the desired event mask based on the given
+ * attribute bitmask.
+ */
+u32 setattr_mask_inotify(unsigned int ia_valid)
+{
+	u32 mask = 0;
+
+	if (ia_valid & ATTR_UID)
+		mask |= IN_ATTRIB;
+	if (ia_valid & ATTR_GID)
+		mask |= IN_ATTRIB;
+	if (ia_valid & ATTR_SIZE)
+		mask |= IN_MODIFY;
+	/* both times implies a utime(s) call */
+	if ((ia_valid & (ATTR_ATIME|ATTR_MTIME)) == (ATTR_ATIME|ATTR_MTIME))
+		mask |= IN_ATTRIB;
+	else if (ia_valid & ATTR_ATIME)
+		mask |= IN_ACCESS;
+	else if (ia_valid & ATTR_MTIME)
+		mask |= IN_MODIFY;
+	if (ia_valid & ATTR_MODE)
+		mask |= IN_ATTRIB;
+	return mask;
+}
+EXPORT_SYMBOL(setattr_mask_inotify);
+
+/* 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 *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 -EINVAL;
+
+	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 -ERESTARTSYS;
+
+		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);
+		if (copy_to_user(buf, &kevent->event, event_size))
+			return -EFAULT;
+
+		spin_lock(&dev->lock);
+		inotify_dev_event_dequeue(dev);
+		spin_unlock(&dev->lock);
+		count -= event_size;
+		buf += event_size;
+	}
+
+	return buf - start;
+}
+
+static int inotify_open(struct inode *inode, struct file *file)
+{
+	struct inotify_device *dev;
+
+	if (atomic_read(&watch_count) == MAX_INOTIFY_DEVS)
+		return -ENODEV;
+
+	atomic_inc(&watch_count);
+
+	dev = kmalloc(sizeof(struct inotify_device), GFP_KERNEL);
+	if (!dev)
+		return -ENOMEM;
+
+	bitmap_zero(dev->bitmask, MAX_INOTIFY_DEV_WATCHES);
+
+	INIT_LIST_HEAD(&dev->events);
+	INIT_LIST_HEAD(&dev->watches);
+	init_waitqueue_head(&dev->wait);
+
+	dev->event_count = 0;
+	dev->nr_watches = 0;
+	dev->lock = SPIN_LOCK_UNLOCKED;
+
+	file->private_data = dev;
+
+	return 0;
+}
+
+/*
+ * inotify_release_all_watches - destroy all watches on a given device
+ *
+ * FIXME: Do we want a lock 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)
+		ignore_helper(watch, 0);
+}
+
+/*
+ * 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);
+	kfree(dev);
+
+	atomic_dec(&watch_count);
+	return 0;
+}
+
+static int inotify_watch(struct inotify_device *dev,
+			 struct inotify_watch_request *request)
+{
+	struct inode *inode;
+	struct inotify_watch *watch;
+
+	inode = find_inode(request->dirname);
+	if (IS_ERR(inode))
+		return PTR_ERR(inode);
+
+	spin_lock(&inode->i_lock);
+	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;
+		inode_update_watch_mask(inode);
+		spin_unlock(&dev->lock);
+		spin_unlock(&inode->i_lock);		
+		unref_inode(inode);
+
+		return owatch->wd;
+	}
+
+	spin_unlock(&dev->lock);
+	spin_unlock(&inode->i_lock);	
+
+	watch = create_watch(dev, request->mask, inode);
+	if (!watch) {
+		unref_inode(inode);
+		return -ENOSPC;
+	}
+
+	spin_lock(&inode->i_lock);
+	spin_lock(&dev->lock);
+
+	/* We can't add anymore watches to this device */
+	if (inotify_dev_add_watch(dev, watch) == -ENOSPC) {
+		delete_watch(dev, watch);
+		spin_unlock(&dev->lock);
+		spin_unlock(&inode->i_lock);		
+		unref_inode(inode);
+		return -ENOSPC;
+	}
+
+	inode_add_watch(inode, watch);
+
+	spin_unlock(&dev->lock);
+	spin_unlock(&inode->i_lock);
+
+	return watch->wd;
+}
+
+static int inotify_ignore(struct inotify_device *dev, s32 wd)
+{
+	struct inotify_watch *watch;
+
+	watch = dev_find_wd(dev, wd);
+	if (!watch)
+		return -EINVAL;
+	ignore_helper(watch, 0);
+
+	return 0;
+}
+
+/*
+ * 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_watch(dev, &request);
+	case INOTIFY_IGNORE:
+		if (copy_from_user(&wd, p, sizeof (wd)))
+			return -EFAULT;
+		return inotify_ignore(dev, wd);
+	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,
+};
+
+struct miscdevice inotify_device = {
+	.minor  = MISC_DYNAMIC_MINOR,
+	.name	= "inotify",
+	.fops	= &inotify_fops,
+};
+
+static int __init inotify_init(void)
+{
+	int ret;
+
+	ret = misc_register(&inotify_device);
+	if (ret)
+		return ret;
+
+	atomic_set(&watch_count, 0);
+	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 init: minor=%d\n", inotify_device.minor);
+
+	return 0;
+}
+
+module_init(inotify_init);

154 lines not on this page · raw message

@@ -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

@@ -61,7 +62,7 @@
 };
 extern struct inodes_stat_t inodes_stat;
 
-extern int leases_enable, dir_notify_enable, lease_break_time;
+extern int leases_enable, lease_break_time;
 
 #define NR_FILE  8192	/* this can well be larger on a larger system */
 #define NR_RESERVED_FILES 10 /* reserved for root */

@@ -455,8 +456,14 @@
 	struct cdev		*i_cdev;
 	int			i_cindex;
 
+#ifdef CONFIG_DNOTIFY
 	unsigned long		i_dnotify_mask; /* Directory notify events */
 	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 */

@@ -1315,7 +1322,7 @@
 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 vfs_permission(struct inode *, int);

@@ -36,6 +36,7 @@
 #include <linux/writeback.h>		/* for the emergency remount stuff */
 #include <linux/idr.h>
 #include <asm/uaccess.h>
+#include <linux/inotify.h>
 
 
 void get_filesystem(struct file_system_type *fs);

@@ -204,6 +205,7 @@
 
 	if (root) {
 		sb->s_root = NULL;
+		inotify_super_block_umount(sb);
 		shrink_dcache_parent(root);
 		shrink_dcache_anon(&sb->s_anon);
 		dput(root);

@@ -11,6 +11,7 @@
 #include <linux/uio.h>
 #include <linux/smp_lock.h>
 #include <linux/dnotify.h>
+#include <linux/inotify.h>
 #include <linux/security.h>
 #include <linux/module.h>
 

@@ -216,8 +217,14 @@
 				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;
+				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);
+			}
 		}
 	}
 

@@ -260,8 +267,14 @@
 				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;
+				dnotify_parent(dentry, DN_MODIFY);
+				inotify_dentry_parent_queue_event(dentry,
+						IN_MODIFY, 0, dentry->d_name.name);
+				inotify_inode_queue_event(inode, IN_MODIFY, 0,
+						NULL);
+			}
 		}
 	}
 

@@ -493,9 +506,15 @@
 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;
+		dnotify_parent(dentry, (type == READ) ? DN_ACCESS : DN_MODIFY);
+		inotify_dentry_parent_queue_event(dentry,
+				(type == READ) ? IN_ACCESS : IN_MODIFY, 0,
+				dentry->d_name.name);
+		inotify_inode_queue_event (dentry->d_inode,
+				(type == READ) ? IN_ACCESS : IN_MODIFY, 0, NULL);
+	}
 	return ret;
 }
 

@@ -11,6 +11,7 @@
 #include <linux/smp_lock.h>
 #include <linux/quotaops.h>
 #include <linux/dnotify.h>
+#include <linux/inotify.h>
 #include <linux/module.h>
 #include <linux/slab.h>
 #include <linux/tty.h>

@@ -955,6 +956,10 @@
 			error = PTR_ERR(f);
 			if (IS_ERR(f))
 				goto out_error;
+			inotify_inode_queue_event(f->f_dentry->d_inode,
+					IN_OPEN, 0, NULL);
+			inotify_dentry_parent_queue_event(f->f_dentry, IN_OPEN,
+					0, f->f_dentry->d_name.name);
 			fd_install(fd, f);
 		}
 out:

@@ -22,6 +22,7 @@
 #include <linux/quotaops.h>
 #include <linux/pagemap.h>
 #include <linux/dnotify.h>
+#include <linux/inotify.h>
 #include <linux/smp_lock.h>
 #include <linux/personality.h>
 #include <linux/security.h>

@@ -1221,6 +1222,8 @@
 	error = dir->i_op->create(dir, dentry, mode, nd);
 	if (!error) {
 		inode_dir_notify(dir, DN_CREATE);
+		inotify_inode_queue_event(dir, IN_CREATE_FILE,
+				0, dentry->d_name.name);
 		security_inode_post_create(dir, dentry, mode);
 	}
 	return error;

@@ -1535,6 +1538,8 @@
 	error = dir->i_op->mknod(dir, dentry, mode, dev);
 	if (!error) {
 		inode_dir_notify(dir, DN_CREATE);
+		inotify_inode_queue_event(dir, IN_CREATE_FILE, 0,
+				dentry->d_name.name);
 		security_inode_post_mknod(dir, dentry, mode, dev);
 	}
 	return error;

@@ -1608,6 +1613,8 @@
 	error = dir->i_op->mkdir(dir, dentry, mode);
 	if (!error) {
 		inode_dir_notify(dir, DN_CREATE);
+		inotify_inode_queue_event(dir, IN_CREATE_SUBDIR, 0,
+				dentry->d_name.name);
 		security_inode_post_mkdir(dir,dentry, mode);
 	}
 	return error;

@@ -1703,6 +1710,11 @@
 	up(&dentry->d_inode->i_sem);
 	if (!error) {
 		inode_dir_notify(dir, DN_DELETE);
+		inotify_inode_queue_event(dir, IN_DELETE_SUBDIR, 0,
+				dentry->d_name.name);
+		inotify_inode_queue_event(dentry->d_inode, IN_DELETE_SELF, 0,
+				NULL);
+		inotify_inode_is_dead (dentry->d_inode);
 		d_delete(dentry);
 	}
 	dput(dentry);

@@ -1775,8 +1787,13 @@
 
 	/* 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);
+		inotify_inode_queue_event(dir, IN_DELETE_FILE, 0,
+				dentry->d_name.name);
+		inotify_inode_queue_event(dentry->d_inode, IN_DELETE_SELF, 0,
+				NULL);
+		inotify_inode_is_dead (dentry->d_inode);
+		d_delete(dentry);
 	}
 	return error;
 }

@@ -1853,6 +1870,8 @@
 	error = dir->i_op->symlink(dir, dentry, oldname);
 	if (!error) {
 		inode_dir_notify(dir, DN_CREATE);
+		inotify_inode_queue_event(dir, IN_CREATE_FILE, 0,
+				dentry->d_name.name);
 		security_inode_post_symlink(dir, dentry, oldname);
 	}
 	return error;

@@ -1926,6 +1945,8 @@
 	up(&old_dentry->d_inode->i_sem);
 	if (!error) {
 		inode_dir_notify(dir, DN_CREATE);
+		inotify_inode_queue_event(dir, IN_CREATE_FILE, 0, 
+					new_dentry->d_name.name);
 		security_inode_post_link(old_dentry, dir, new_dentry);
 	}
 	return error;

@@ -2089,6 +2110,8 @@
 {
 	int error;
 	int is_dir = S_ISDIR(old_dentry->d_inode->i_mode);
+	char *old_name;
+	u32 cookie;
 
 	if (old_dentry->d_inode == new_dentry->d_inode)
  		return 0;

@@ -2110,6 +2133,8 @@
 	DQUOT_INIT(old_dir);
 	DQUOT_INIT(new_dir);
 
+	old_name = inotify_oldname_init(old_dentry);
+
 	if (is_dir)
 		error = vfs_rename_dir(old_dir,old_dentry,new_dir,new_dentry);
 	else

@@ -2121,7 +2146,15 @@
 			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_dentry->d_name.name);
 	}
+	inotify_oldname_free(old_name);
+
 	return error;
 }
 

@@ -114,6 +114,9 @@
 	if (inode) {
 		struct address_space * const mapping = &inode->i_data;
 
+#ifdef CONFIG_INOTIFY
+		inode->inotify_data = NULL;
+#endif
 		inode->i_sb = sb;
 		inode->i_blkbits = sb->s_blocksize_bits;
 		inode->i_flags = 0;

97 lines not on this page · raw message

@@ -16,6 +16,7 @@
 #include <linux/eventpoll.h>
 #include <linux/mount.h>
 #include <linux/cdev.h>
+#include <linux/inotify.h>
 
 /* sysctl tunables... */
 struct files_stat_struct files_stat = {

@@ -168,6 +169,12 @@
 	struct dentry *dentry = file->f_dentry;
 	struct vfsmount *mnt = file->f_vfsmnt;
 	struct inode *inode = dentry->d_inode;
+	u32 mask;
+
+
+	mask = (file->f_mode & FMODE_WRITE) ? IN_CLOSE_WRITE : IN_CLOSE_NOWRITE;
+	inotify_dentry_parent_queue_event(dentry, mask, 0, dentry->d_name.name);
+	inotify_inode_queue_event(inode, mask, 0, NULL);
 
 	/*
 	 * The function eventpoll_release() should be the first called

@@ -13,6 +13,7 @@
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  * General Public License for more details.
  */
+
 #include <linux/fs.h>
 #include <linux/module.h>
 #include <linux/sched.h>

@@ -21,8 +22,6 @@
 #include <linux/spinlock.h>
 #include <linux/slab.h>
 
-int dir_notify_enable = 1;
-
 static kmem_cache_t *dn_cache;
 
 static void redo_inode_mask(struct inode *inode)

@@ -72,8 +71,6 @@
 		dnotify_flush(filp, id);
 		return 0;
 	}
-	if (!dir_notify_enable)
-		return -EINVAL;
 	inode = filp->f_dentry->d_inode;
 	if (!S_ISDIR(inode->i_mode))
 		return -ENOTDIR;

@@ -146,6 +143,29 @@
 
 EXPORT_SYMBOL(__inode_dir_notify);
 
+int setattr_mask_dnotify(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;
+}
+EXPORT_SYMBOL(setattr_mask_dnotify);
+
 /*
  * This is hopelessly wrong, but unfixable without API changes.  At
  * least it doesn't oops the kernel...

@@ -157,9 +177,6 @@
 {
 	struct dentry *parent;
 
-	if (!dir_notify_enable)
-		return;
-
 	spin_lock(&dentry->d_lock);
 	parent = dentry->d_parent;
 	if (parent->d_inode->i_dnotify_mask & event) {

70 lines not on this page · raw message

@@ -868,14 +868,6 @@
 		.proc_handler	= &proc_dointvec,
 	},
 	{
-		.ctl_name	= FS_DIR_NOTIFY,
-		.procname	= "dir-notify-enable",
-		.data		= &dir_notify_enable,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec,
-	},
-	{
 		.ctl_name	= FS_LEASE_TIME,
 		.procname	= "lease-break-time",
 		.data		= &lease_break_time,