1 files changed, 290 insertions, 241 deletions

diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c
index 063224812b7..3fdc8a3e113 100644
--- a/fs/notify/fanotify/fanotify_user.c
+++ b/fs/notify/fanotify/fanotify_user.c
@@ -13,23 +13,36 @@
 #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;
-static struct kmem_cache *fanotify_response_event_cache __read_mostly;
-
-struct fanotify_response_event {
-	struct list_head list;
-	__s32 fd;
-	struct fsnotify_event *event;
-};
+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
@@ -56,11 +69,11 @@ static struct fsnotify_event *get_one_event(struct fsnotify_group *group,
 	return fsnotify_remove_notify_event(group);
 }
 
-static int create_fd(struct fsnotify_group *group, struct fsnotify_event *event)
+static int create_fd(struct fsnotify_group *group,
+		     struct fanotify_event_info *event,
+		     struct file **file)
 {
 	int client_fd;
-	struct dentry *dentry;
-	struct vfsmount *mnt;
 	struct file *new_file;
 
 	pr_debug("%s: group=%p event=%p\n", __func__, group, event);
@@ -69,22 +82,14 @@ static int create_fd(struct fsnotify_group *group, struct fsnotify_event *event)
 	if (client_fd < 0)
 		return client_fd;
 
-	if (event->data_type != FSNOTIFY_EVENT_PATH) {
-		WARN_ON(1);
-		put_unused_fd(client_fd);
-		return -EINVAL;
-	}
-
 	/*
 	 * we need a new file handle for the userspace program so it can read even if it was
 	 * originally opened O_WRONLY.
 	 */
-	dentry = dget(event->path.dentry);
-	mnt = mntget(event->path.mnt);
 	/* 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 (dentry && mnt)
-		new_file = dentry_open(dentry, mnt,
+	if (event->path.dentry && event->path.mnt)
+		new_file = dentry_open(&event->path,
 				       group->fanotify_data.f_flags | FMODE_NONOTIFY,
 				       current_cred());
 	else
@@ -100,62 +105,77 @@ static int create_fd(struct fsnotify_group *group, struct fsnotify_event *event)
 		put_unused_fd(client_fd);
 		client_fd = PTR_ERR(new_file);
 	} else {
-		fd_install(client_fd, new_file);
+		*file = new_file;
 	}
 
 	return client_fd;
 }
 
-static ssize_t fill_event_metadata(struct fsnotify_group *group,
-				   struct fanotify_event_metadata *metadata,
-				   struct fsnotify_event *event)
+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, event);
+		 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->mask = event->mask & FAN_ALL_OUTGOING_EVENTS;
+	metadata->reserved = 0;
+	metadata->mask = fsn_event->mask & FAN_ALL_OUTGOING_EVENTS;
 	metadata->pid = pid_vnr(event->tgid);
-	metadata->fd = create_fd(group, event);
+	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 metadata->fd;
+	return ret;
 }
 
 #ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
-static struct fanotify_response_event *dequeue_re(struct fsnotify_group *group,
-						  __s32 fd)
+static struct fanotify_perm_event_info *dequeue_event(
+				struct fsnotify_group *group, int fd)
 {
-	struct fanotify_response_event *re, *return_re = NULL;
+	struct fanotify_perm_event_info *event, *return_e = NULL;
 
-	mutex_lock(&group->fanotify_data.access_mutex);
-	list_for_each_entry(re, &group->fanotify_data.access_list, list) {
-		if (re->fd != fd)
+	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(&re->list);
-		return_re = re;
+		list_del_init(&event->fae.fse.list);
+		return_e = event;
 		break;
 	}
-	mutex_unlock(&group->fanotify_data.access_mutex);
+	spin_unlock(&group->fanotify_data.access_lock);
 
-	pr_debug("%s: found return_re=%p\n", __func__, return_re);
+	pr_debug("%s: found return_re=%p\n", __func__, return_e);
 
-	return return_re;
+	return return_e;
 }
 
 static int process_access_response(struct fsnotify_group *group,
 				   struct fanotify_response *response_struct)
 {
-	struct fanotify_response_event *re;
-	__s32 fd = response_struct->fd;
-	__u32 response = response_struct->response;
+	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 responce or we will clean it up after the
+	 * userspace can send a valid response or we will clean it up after the
 	 * timeout
 	 */
 	switch (response) {
@@ -169,83 +189,15 @@ static int process_access_response(struct fsnotify_group *group,
 	if (fd < 0)
 		return -EINVAL;
 
-	re = dequeue_re(group, fd);
-	if (!re)
+	event = dequeue_event(group, fd);
+	if (!event)
 		return -ENOENT;
 
-	re->event->response = response;
-
+	event->response = response;
 	wake_up(&group->fanotify_data.access_waitq);
 
-	kmem_cache_free(fanotify_response_event_cache, re);
-
 	return 0;
 }
-
-static int prepare_for_access_response(struct fsnotify_group *group,
-				       struct fsnotify_event *event,
-				       __s32 fd)
-{
-	struct fanotify_response_event *re;
-
-	if (!(event->mask & FAN_ALL_PERM_EVENTS))
-		return 0;
-
-	re = kmem_cache_alloc(fanotify_response_event_cache, GFP_KERNEL);
-	if (!re)
-		return -ENOMEM;
-
-	re->event = event;
-	re->fd = fd;
-
-	mutex_lock(&group->fanotify_data.access_mutex);
-
-	if (group->fanotify_data.bypass_perm) {
-		mutex_unlock(&group->fanotify_data.access_mutex);
-		kmem_cache_free(fanotify_response_event_cache, re);
-		event->response = FAN_ALLOW;
-		return 0;
-	}
-		
-	list_add_tail(&re->list, &group->fanotify_data.access_list);
-	mutex_unlock(&group->fanotify_data.access_mutex);
-
-	return 0;
-}
-
-static void remove_access_response(struct fsnotify_group *group,
-				   struct fsnotify_event *event,
-				   __s32 fd)
-{
-	struct fanotify_response_event *re;
-
-	if (!(event->mask & FAN_ALL_PERM_EVENTS))
-		return;
-
-	re = dequeue_re(group, fd);
-	if (!re)
-		return;
-
-	BUG_ON(re->event != event);
-
-	kmem_cache_free(fanotify_response_event_cache, re);
-
-	return;
-}
-#else
-static int prepare_for_access_response(struct fsnotify_group *group,
-				       struct fsnotify_event *event,
-				       __s32 fd)
-{
-	return 0;
-}
-
-static void remove_access_response(struct fsnotify_group *group,
-				   struct fsnotify_event *event,
-				   __s32 fd)
-{
-	return;
-}
 #endif
 
 static ssize_t copy_event_to_user(struct fsnotify_group *group,
@@ -253,28 +205,35 @@ static ssize_t copy_event_to_user(struct fsnotify_group *group,
 				  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);
 
-	fd = fill_event_metadata(group, &fanotify_event_metadata, event);
-	if (fd < 0)
-		return fd;
+	ret = fill_event_metadata(group, &fanotify_event_metadata, event, &f);
+	if (ret < 0)
+		return ret;
 
-	ret = prepare_for_access_response(group, event, fd);
-	if (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;
 
-	ret = -EFAULT;
-	if (copy_to_user(buf, &fanotify_event_metadata, FAN_EVENT_METADATA_LEN))
-		goto out_kill_access_response;
+#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
+	if (event->mask & FAN_ALL_PERM_EVENTS)
+		FANOTIFY_PE(event)->fd = fd;
+#endif
 
-	return FAN_EVENT_METADATA_LEN;
+	if (fd != FAN_NOFD)
+		fd_install(fd, f);
+	return fanotify_event_metadata.event_len;
 
-out_kill_access_response:
-	remove_access_response(group, event, fd);
 out_close_fd:
-	sys_close(fd);
+	if (fd != FAN_NOFD) {
+		put_unused_fd(fd);
+		fput(f);
+	}
 	return ret;
 }
 
@@ -314,30 +273,50 @@ static ssize_t fanotify_read(struct file *file, char __user *buf,
 		kevent = get_one_event(group, count);
 		mutex_unlock(&group->notification_mutex);
 
-		if (kevent) {
+		if (IS_ERR(kevent)) {
 			ret = PTR_ERR(kevent);
-			if (IS_ERR(kevent))
+			break;
+		}
+
+		if (!kevent) {
+			ret = -EAGAIN;
+			if (file->f_flags & O_NONBLOCK)
 				break;
-			ret = copy_event_to_user(group, kevent, buf);
-			fsnotify_put_event(kevent);
-			if (ret < 0)
+
+			ret = -ERESTARTSYS;
+			if (signal_pending(current))
+				break;
+
+			if (start != buf)
 				break;
-			buf += ret;
-			count -= ret;
+			schedule();
 			continue;
 		}
 
-		ret = -EAGAIN;
-		if (file->f_flags & O_NONBLOCK)
-			break;
-		ret = -ERESTARTSYS;
-		if (signal_pending(current))
-			break;
-
-		if (start != buf)
-			break;
-
-		schedule();
+		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);
@@ -378,27 +357,27 @@ static int fanotify_release(struct inode *ignored, struct file *file)
 	struct fsnotify_group *group = file->private_data;
 
 #ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
-	struct fanotify_response_event *re, *lre;
+	struct fanotify_perm_event_info *event, *next;
 
-	mutex_lock(&group->fanotify_data.access_mutex);
+	spin_lock(&group->fanotify_data.access_lock);
 
-	group->fanotify_data.bypass_perm = true;
+	atomic_inc(&group->fanotify_data.bypass_perm);
 
-	list_for_each_entry_safe(re, lre, &group->fanotify_data.access_list, list) {
-		pr_debug("%s: found group=%p re=%p event=%p\n", __func__, group,
-			 re, re->event);
+	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(&re->list);
-		re->event->response = FAN_ALLOW;
-
-		kmem_cache_free(fanotify_response_event_cache, re);
+		list_del_init(&event->fae.fse.list);
+		event->response = FAN_ALLOW;
 	}
-	mutex_unlock(&group->fanotify_data.access_mutex);
+	spin_unlock(&group->fanotify_data.access_lock);
 
 	wake_up(&group->fanotify_data.access_waitq);
 #endif
+
 	/* matches the fanotify_init->fsnotify_alloc_group */
-	fsnotify_put_group(group);
+	fsnotify_destroy_group(group);
 
 	return 0;
 }
@@ -406,7 +385,7 @@ static int fanotify_release(struct inode *ignored, struct file *file)
 static long fanotify_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 {
 	struct fsnotify_group *group;
-	struct fsnotify_event_holder *holder;
+	struct fsnotify_event *fsn_event;
 	void __user *p;
 	int ret = -ENOTTY;
 	size_t send_len = 0;
@@ -418,7 +397,7 @@ static long fanotify_ioctl(struct file *file, unsigned int cmd, unsigned long ar
 	switch (cmd) {
 	case FIONREAD:
 		mutex_lock(&group->notification_mutex);
-		list_for_each_entry(holder, &group->notification_list, event_list)
+		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);
@@ -429,6 +408,7 @@ static long fanotify_ioctl(struct file *file, unsigned int cmd, unsigned long ar
 }
 
 static const struct file_operations fanotify_fops = {
+	.show_fdinfo	= fanotify_show_fdinfo,
 	.poll		= fanotify_poll,
 	.read		= fanotify_read,
 	.write		= fanotify_write,
@@ -453,24 +433,22 @@ static int fanotify_find_path(int dfd, const char __user *filename,
 		 dfd, filename, flags);
 
 	if (filename == NULL) {
-		struct file *file;
-		int fput_needed;
+		struct fd f = fdget(dfd);
 
 		ret = -EBADF;
-		file = fget_light(dfd, &fput_needed);
-		if (!file)
+		if (!f.file)
 			goto out;
 
 		ret = -ENOTDIR;
 		if ((flags & FAN_MARK_ONLYDIR) &&
-		    !(S_ISDIR(file->f_path.dentry->d_inode->i_mode))) {
-			fput_light(file, fput_needed);
+		    !(S_ISDIR(file_inode(f.file)->i_mode))) {
+			fdput(f);
 			goto out;
 		}
 
-		*path = file->f_path;
+		*path = f.file->f_path;
 		path_get(path);
-		fput_light(file, fput_needed);
+		fdput(f);
 	} else {
 		unsigned int lookup_flags = 0;
 
@@ -494,7 +472,8 @@ out:
 
 static __u32 fanotify_mark_remove_from_mask(struct fsnotify_mark *fsn_mark,
 					    __u32 mask,
-					    unsigned int flags)
+					    unsigned int flags,
+					    int *destroy)
 {
 	__u32 oldmask;
 
@@ -508,8 +487,7 @@ static __u32 fanotify_mark_remove_from_mask(struct fsnotify_mark *fsn_mark,
 	}
 	spin_unlock(&fsn_mark->lock);
 
-	if (!(oldmask & ~mask))
-		fsnotify_destroy_mark(fsn_mark);
+	*destroy = !(oldmask & ~mask);
 
 	return mask & oldmask;
 }
@@ -520,14 +498,23 @@ static int fanotify_remove_vfsmount_mark(struct fsnotify_group *group,
 {
 	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)
+	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);
 
-	removed = fanotify_mark_remove_from_mask(fsn_mark, mask, flags);
 	fsnotify_put_mark(fsn_mark);
-	if (removed & mnt->mnt_fsnotify_mask)
+	if (removed & real_mount(mnt)->mnt_fsnotify_mask)
 		fsnotify_recalc_vfsmount_mask(mnt);
 
 	return 0;
@@ -539,12 +526,21 @@ static int fanotify_remove_inode_mark(struct fsnotify_group *group,
 {
 	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)
+	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);
 
-	removed = fanotify_mark_remove_from_mask(fsn_mark, mask, flags);
 	/* matches the fsnotify_find_inode_mark() */
 	fsnotify_put_mark(fsn_mark);
 	if (removed & inode->i_fsnotify_mask)
@@ -580,6 +576,31 @@ static __u32 fanotify_mark_add_to_mask(struct fsnotify_mark *fsn_mark,
 	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)
@@ -587,29 +608,22 @@ static int fanotify_add_vfsmount_mark(struct fsnotify_group *group,
 	struct fsnotify_mark *fsn_mark;
 	__u32 added;
 
+	mutex_lock(&group->mark_mutex);
 	fsn_mark = fsnotify_find_vfsmount_mark(group, mnt);
 	if (!fsn_mark) {
-		int ret;
-
-		if (atomic_read(&group->num_marks) > group->fanotify_data.max_marks)
-			return -ENOSPC;
-
-		fsn_mark = kmem_cache_alloc(fanotify_mark_cache, GFP_KERNEL);
-		if (!fsn_mark)
-			return -ENOMEM;
-
-		fsnotify_init_mark(fsn_mark, fanotify_free_mark);
-		ret = fsnotify_add_mark(fsn_mark, group, NULL, mnt, 0);
-		if (ret) {
-			fanotify_free_mark(fsn_mark);
-			return ret;
+		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);
-	fsnotify_put_mark(fsn_mark);
-	if (added & ~mnt->mnt_fsnotify_mask)
+	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;
 }
 
@@ -632,28 +646,22 @@ static int fanotify_add_inode_mark(struct fsnotify_group *group,
 	    (atomic_read(&inode->i_writecount) > 0))
 		return 0;
 
+	mutex_lock(&group->mark_mutex);
 	fsn_mark = fsnotify_find_inode_mark(group, inode);
 	if (!fsn_mark) {
-		int ret;
-
-		if (atomic_read(&group->num_marks) > group->fanotify_data.max_marks)
-			return -ENOSPC;
-
-		fsn_mark = kmem_cache_alloc(fanotify_mark_cache, GFP_KERNEL);
-		if (!fsn_mark)
-			return -ENOMEM;
-
-		fsnotify_init_mark(fsn_mark, fanotify_free_mark);
-		ret = fsnotify_add_mark(fsn_mark, group, inode, NULL, 0);
-		if (ret) {
-			fanotify_free_mark(fsn_mark);
-			return ret;
+		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);
-	fsnotify_put_mark(fsn_mark);
+	mutex_unlock(&group->mark_mutex);
+
 	if (added & ~inode->i_fsnotify_mask)
 		fsnotify_recalc_inode_mask(inode);
+
+	fsnotify_put_mark(fsn_mark);
 	return 0;
 }
 
@@ -663,6 +671,7 @@ 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);
@@ -673,6 +682,18 @@ SYSCALL_DEFINE2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags)
 	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);
@@ -687,17 +708,29 @@ SYSCALL_DEFINE2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags)
 
 	/* fsnotify_alloc_group takes a ref.  Dropped in fanotify_release */
 	group = fsnotify_alloc_group(&fanotify_fsnotify_ops);
-	if (IS_ERR(group))
+	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
-	mutex_init(&group->fanotify_data.access_mutex);
+	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:
@@ -711,13 +744,13 @@ SYSCALL_DEFINE2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags)
 		break;
 	default:
 		fd = -EINVAL;
-		goto out_put_group;
+		goto out_destroy_group;
 	}
 
 	if (flags & FAN_UNLIMITED_QUEUE) {
 		fd = -EPERM;
 		if (!capable(CAP_SYS_ADMIN))
-			goto out_put_group;
+			goto out_destroy_group;
 		group->max_events = UINT_MAX;
 	} else {
 		group->max_events = FANOTIFY_DEFAULT_MAX_EVENTS;
@@ -726,7 +759,7 @@ SYSCALL_DEFINE2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags)
 	if (flags & FAN_UNLIMITED_MARKS) {
 		fd = -EPERM;
 		if (!capable(CAP_SYS_ADMIN))
-			goto out_put_group;
+			goto out_destroy_group;
 		group->fanotify_data.max_marks = UINT_MAX;
 	} else {
 		group->fanotify_data.max_marks = FANOTIFY_DEFAULT_MAX_MARKS;
@@ -734,25 +767,25 @@ SYSCALL_DEFINE2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags)
 
 	fd = anon_inode_getfd("[fanotify]", &fanotify_fops, group, f_flags);
 	if (fd < 0)
-		goto out_put_group;
+		goto out_destroy_group;
 
 	return fd;
 
-out_put_group:
-	fsnotify_put_group(group);
+out_destroy_group:
+	fsnotify_destroy_group(group);
 	return fd;
 }
 
-SYSCALL_DEFINE(fanotify_mark)(int fanotify_fd, unsigned int flags,
-			      __u64 mask, int dfd,
-			      const char  __user * pathname)
+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 file *filp;
+	struct fd f;
 	struct path path;
-	int ret, fput_needed;
+	int ret;
 
 	pr_debug("%s: fanotify_fd=%d flags=%x dfd=%d pathname=%p mask=%llx\n",
 		 __func__, fanotify_fd, flags, dfd, pathname, mask);
@@ -764,9 +797,14 @@ SYSCALL_DEFINE(fanotify_mark)(int fanotify_fd, unsigned int flags,
 	if (flags & ~FAN_ALL_MARK_FLAGS)
 		return -EINVAL;
 	switch (flags & (FAN_MARK_ADD | FAN_MARK_REMOVE | FAN_MARK_FLUSH)) {
-	case FAN_MARK_ADD:
+	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;
@@ -784,15 +822,15 @@ SYSCALL_DEFINE(fanotify_mark)(int fanotify_fd, unsigned int flags,
 #endif
 		return -EINVAL;
 
-	filp = fget_light(fanotify_fd, &fput_needed);
-	if (unlikely(!filp))
+	f = fdget(fanotify_fd);
+	if (unlikely(!f.file))
 		return -EBADF;
 
 	/* verify that this is indeed an fanotify instance */
 	ret = -EINVAL;
-	if (unlikely(filp->f_op != &fanotify_fops))
+	if (unlikely(f.file->f_op != &fanotify_fops))
 		goto fput_and_out;
-	group = filp->private_data;
+	group = f.file->private_data;
 
 	/*
 	 * group->priority == FS_PRIO_0 == FAN_CLASS_NOTIF.  These are not
@@ -803,6 +841,15 @@ SYSCALL_DEFINE(fanotify_mark)(int fanotify_fd, unsigned int flags,
 	    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;
@@ -814,7 +861,7 @@ SYSCALL_DEFINE(fanotify_mark)(int fanotify_fd, unsigned int flags,
 		mnt = path.mnt;
 
 	/* create/update an inode mark */
-	switch (flags & (FAN_MARK_ADD | FAN_MARK_REMOVE | FAN_MARK_FLUSH)) {
+	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);
@@ -827,43 +874,45 @@ SYSCALL_DEFINE(fanotify_mark)(int fanotify_fd, unsigned int flags,
 		else
 			ret = fanotify_remove_inode_mark(group, inode, mask, flags);
 		break;
-	case FAN_MARK_FLUSH:
-		if (flags & FAN_MARK_MOUNT)
-			fsnotify_clear_vfsmount_marks_by_group(group);
-		else
-			fsnotify_clear_inode_marks_by_group(group);
-		break;
 	default:
 		ret = -EINVAL;
 	}
 
 	path_put(&path);
 fput_and_out:
-	fput_light(filp, fput_needed);
+	fdput(f);
 	return ret;
 }
 
-#ifdef CONFIG_HAVE_SYSCALL_WRAPPERS
-asmlinkage long SyS_fanotify_mark(long fanotify_fd, long flags, __u64 mask,
-				  long dfd, long pathname)
+#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 SYSC_fanotify_mark((int) fanotify_fd, (unsigned int) flags,
-				  mask, (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);
 }
-SYSCALL_ALIAS(sys_fanotify_mark, SyS_fanotify_mark);
 #endif
 
 /*
- * fanotify_user_setup - Our initialization function.  Note that we cannnot return
+ * 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_response_event_cache = KMEM_CACHE(fanotify_response_event,
-						   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;
 }