3 files changed, 393 insertions, 287 deletions

diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c
index 0c2f9122b26..ee9cb3795c2 100644
--- a/fs/notify/fanotify/fanotify.c
+++ b/fs/notify/fanotify/fanotify.c
@@ -9,91 +9,59 @@
 #include <linux/types.h>
 #include <linux/wait.h>
 
-static bool should_merge(struct fsnotify_event *old, struct fsnotify_event *new)
+#include "fanotify.h"
+
+static bool should_merge(struct fsnotify_event *old_fsn,
+			 struct fsnotify_event *new_fsn)
 {
-	pr_debug("%s: old=%p new=%p\n", __func__, old, new);
+	struct fanotify_event_info *old, *new;
 
-	if (old->to_tell == new->to_tell &&
-	    old->data_type == new->data_type &&
-	    old->tgid == new->tgid) {
-		switch (old->data_type) {
-		case (FSNOTIFY_EVENT_PATH):
-#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
-			/* dont merge two permission events */
-			if ((old->mask & FAN_ALL_PERM_EVENTS) &&
-			    (new->mask & FAN_ALL_PERM_EVENTS))
-				return false;
-#endif
-			if ((old->path.mnt == new->path.mnt) &&
-			    (old->path.dentry == new->path.dentry))
-				return true;
-			break;
-		case (FSNOTIFY_EVENT_NONE):
-			return true;
-		default:
-			BUG();
-		};
-	}
+	pr_debug("%s: old=%p new=%p\n", __func__, old_fsn, new_fsn);
+	old = FANOTIFY_E(old_fsn);
+	new = FANOTIFY_E(new_fsn);
+
+	if (old_fsn->inode == new_fsn->inode && old->tgid == new->tgid &&
+	    old->path.mnt == new->path.mnt &&
+	    old->path.dentry == new->path.dentry)
+		return true;
 	return false;
 }
 
 /* and the list better be locked by something too! */
-static struct fsnotify_event *fanotify_merge(struct list_head *list,
-					     struct fsnotify_event *event)
+static int fanotify_merge(struct list_head *list, struct fsnotify_event *event)
 {
-	struct fsnotify_event_holder *test_holder;
-	struct fsnotify_event *test_event = NULL;
-	struct fsnotify_event *new_event;
+	struct fsnotify_event *test_event;
+	bool do_merge = false;
 
 	pr_debug("%s: list=%p event=%p\n", __func__, list, event);
 
+#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
+	/*
+	 * Don't merge a permission event with any other event so that we know
+	 * the event structure we have created in fanotify_handle_event() is the
+	 * one we should check for permission response.
+	 */
+	if (event->mask & FAN_ALL_PERM_EVENTS)
+		return 0;
+#endif
 
-	list_for_each_entry_reverse(test_holder, list, event_list) {
-		if (should_merge(test_holder->event, event)) {
-			test_event = test_holder->event;
+	list_for_each_entry_reverse(test_event, list, list) {
+		if (should_merge(test_event, event)) {
+			do_merge = true;
 			break;
 		}
 	}
 
-	if (!test_event)
-		return NULL;
-
-	fsnotify_get_event(test_event);
-
-	/* if they are exactly the same we are done */
-	if (test_event->mask == event->mask)
-		return test_event;
-
-	/*
-	 * if the refcnt == 2 this is the only queue
-	 * for this event and so we can update the mask
-	 * in place.
-	 */
-	if (atomic_read(&test_event->refcnt) == 2) {
-		test_event->mask |= event->mask;
-		return test_event;
-	}
-
-	new_event = fsnotify_clone_event(test_event);
-
-	/* done with test_event */
-	fsnotify_put_event(test_event);
-
-	/* couldn't allocate memory, merge was not possible */
-	if (unlikely(!new_event))
-		return ERR_PTR(-ENOMEM);
-
-	/* build new event and replace it on the list */
-	new_event->mask = (test_event->mask | event->mask);
-	fsnotify_replace_event(test_holder, new_event);
+	if (!do_merge)
+		return 0;
 
-	/* we hold a reference on new_event from clone_event */
-	return new_event;
+	test_event->mask |= event->mask;
+	return 1;
 }
 
 #ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
-static int fanotify_get_response_from_access(struct fsnotify_group *group,
-					     struct fsnotify_event *event)
+static int fanotify_get_response(struct fsnotify_group *group,
+				 struct fanotify_perm_event_info *event)
 {
 	int ret;
 
@@ -106,7 +74,6 @@ static int fanotify_get_response_from_access(struct fsnotify_group *group,
 		return 0;
 
 	/* userspace responded, convert to something usable */
-	spin_lock(&event->lock);
 	switch (event->response) {
 	case FAN_ALLOW:
 		ret = 0;
@@ -116,7 +83,6 @@ static int fanotify_get_response_from_access(struct fsnotify_group *group,
 		ret = -EPERM;
 	}
 	event->response = 0;
-	spin_unlock(&event->lock);
 
 	pr_debug("%s: group=%p event=%p about to return ret=%d\n", __func__,
 		 group, event, ret);
@@ -125,58 +91,17 @@ static int fanotify_get_response_from_access(struct fsnotify_group *group,
 }
 #endif
 
-static int fanotify_handle_event(struct fsnotify_group *group,
-				 struct fsnotify_mark *inode_mark,
-				 struct fsnotify_mark *fanotify_mark,
-				 struct fsnotify_event *event)
-{
-	int ret = 0;
-	struct fsnotify_event *notify_event = NULL;
-
-	BUILD_BUG_ON(FAN_ACCESS != FS_ACCESS);
-	BUILD_BUG_ON(FAN_MODIFY != FS_MODIFY);
-	BUILD_BUG_ON(FAN_CLOSE_NOWRITE != FS_CLOSE_NOWRITE);
-	BUILD_BUG_ON(FAN_CLOSE_WRITE != FS_CLOSE_WRITE);
-	BUILD_BUG_ON(FAN_OPEN != FS_OPEN);
-	BUILD_BUG_ON(FAN_EVENT_ON_CHILD != FS_EVENT_ON_CHILD);
-	BUILD_BUG_ON(FAN_Q_OVERFLOW != FS_Q_OVERFLOW);
-	BUILD_BUG_ON(FAN_OPEN_PERM != FS_OPEN_PERM);
-	BUILD_BUG_ON(FAN_ACCESS_PERM != FS_ACCESS_PERM);
-	BUILD_BUG_ON(FAN_ONDIR != FS_ISDIR);
-
-	pr_debug("%s: group=%p event=%p\n", __func__, group, event);
-
-	notify_event = fsnotify_add_notify_event(group, event, NULL, fanotify_merge);
-	if (IS_ERR(notify_event))
-		return PTR_ERR(notify_event);
-
-#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
-	if (event->mask & FAN_ALL_PERM_EVENTS) {
-		/* if we merged we need to wait on the new event */
-		if (notify_event)
-			event = notify_event;
-		ret = fanotify_get_response_from_access(group, event);
-	}
-#endif
-
-	if (notify_event)
-		fsnotify_put_event(notify_event);
-
-	return ret;
-}
-
-static bool fanotify_should_send_event(struct fsnotify_group *group,
-				       struct inode *to_tell,
-				       struct fsnotify_mark *inode_mark,
+static bool fanotify_should_send_event(struct fsnotify_mark *inode_mark,
 				       struct fsnotify_mark *vfsmnt_mark,
-				       __u32 event_mask, void *data, int data_type)
+				       u32 event_mask,
+				       void *data, int data_type)
 {
 	__u32 marks_mask, marks_ignored_mask;
 	struct path *path = data;
 
-	pr_debug("%s: group=%p to_tell=%p inode_mark=%p vfsmnt_mark=%p "
-		 "mask=%x data=%p data_type=%d\n", __func__, group, to_tell,
-		 inode_mark, vfsmnt_mark, event_mask, data, data_type);
+	pr_debug("%s: inode_mark=%p vfsmnt_mark=%p mask=%x data=%p"
+		 " data_type=%d\n", __func__, inode_mark, vfsmnt_mark,
+		 event_mask, data, data_type);
 
 	/* if we don't have enough info to send an event to userspace say no */
 	if (data_type != FSNOTIFY_EVENT_PATH)
@@ -217,6 +142,93 @@ static bool fanotify_should_send_event(struct fsnotify_group *group,
 	return false;
 }
 
+struct fanotify_event_info *fanotify_alloc_event(struct inode *inode, u32 mask,
+						 struct path *path)
+{
+	struct fanotify_event_info *event;
+
+#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
+	if (mask & FAN_ALL_PERM_EVENTS) {
+		struct fanotify_perm_event_info *pevent;
+
+		pevent = kmem_cache_alloc(fanotify_perm_event_cachep,
+					  GFP_KERNEL);
+		if (!pevent)
+			return NULL;
+		event = &pevent->fae;
+		pevent->response = 0;
+		goto init;
+	}
+#endif
+	event = kmem_cache_alloc(fanotify_event_cachep, GFP_KERNEL);
+	if (!event)
+		return NULL;
+init: __maybe_unused
+	fsnotify_init_event(&event->fse, inode, mask);
+	event->tgid = get_pid(task_tgid(current));
+	if (path) {
+		event->path = *path;
+		path_get(&event->path);
+	} else {
+		event->path.mnt = NULL;
+		event->path.dentry = NULL;
+	}
+	return event;
+}
+
+static int fanotify_handle_event(struct fsnotify_group *group,
+				 struct inode *inode,
+				 struct fsnotify_mark *inode_mark,
+				 struct fsnotify_mark *fanotify_mark,
+				 u32 mask, void *data, int data_type,
+				 const unsigned char *file_name, u32 cookie)
+{
+	int ret = 0;
+	struct fanotify_event_info *event;
+	struct fsnotify_event *fsn_event;
+
+	BUILD_BUG_ON(FAN_ACCESS != FS_ACCESS);
+	BUILD_BUG_ON(FAN_MODIFY != FS_MODIFY);
+	BUILD_BUG_ON(FAN_CLOSE_NOWRITE != FS_CLOSE_NOWRITE);
+	BUILD_BUG_ON(FAN_CLOSE_WRITE != FS_CLOSE_WRITE);
+	BUILD_BUG_ON(FAN_OPEN != FS_OPEN);
+	BUILD_BUG_ON(FAN_EVENT_ON_CHILD != FS_EVENT_ON_CHILD);
+	BUILD_BUG_ON(FAN_Q_OVERFLOW != FS_Q_OVERFLOW);
+	BUILD_BUG_ON(FAN_OPEN_PERM != FS_OPEN_PERM);
+	BUILD_BUG_ON(FAN_ACCESS_PERM != FS_ACCESS_PERM);
+	BUILD_BUG_ON(FAN_ONDIR != FS_ISDIR);
+
+	if (!fanotify_should_send_event(inode_mark, fanotify_mark, mask, data,
+					data_type))
+		return 0;
+
+	pr_debug("%s: group=%p inode=%p mask=%x\n", __func__, group, inode,
+		 mask);
+
+	event = fanotify_alloc_event(inode, mask, data);
+	if (unlikely(!event))
+		return -ENOMEM;
+
+	fsn_event = &event->fse;
+	ret = fsnotify_add_notify_event(group, fsn_event, fanotify_merge);
+	if (ret) {
+		/* Permission events shouldn't be merged */
+		BUG_ON(ret == 1 && mask & FAN_ALL_PERM_EVENTS);
+		/* Our event wasn't used in the end. Free it. */
+		fsnotify_destroy_event(group, fsn_event);
+
+		return 0;
+	}
+
+#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
+	if (mask & FAN_ALL_PERM_EVENTS) {
+		ret = fanotify_get_response(group, FANOTIFY_PE(fsn_event));
+		fsnotify_destroy_event(group, fsn_event);
+	}
+#endif
+	return ret;
+}
+
 static void fanotify_free_group_priv(struct fsnotify_group *group)
 {
 	struct user_struct *user;
@@ -226,10 +238,25 @@ static void fanotify_free_group_priv(struct fsnotify_group *group)
 	free_uid(user);
 }
 
+static void fanotify_free_event(struct fsnotify_event *fsn_event)
+{
+	struct fanotify_event_info *event;
+
+	event = FANOTIFY_E(fsn_event);
+	path_put(&event->path);
+	put_pid(event->tgid);
+#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
+	if (fsn_event->mask & FAN_ALL_PERM_EVENTS) {
+		kmem_cache_free(fanotify_perm_event_cachep,
+				FANOTIFY_PE(fsn_event));
+		return;
+	}
+#endif
+	kmem_cache_free(fanotify_event_cachep, event);
+}
+
 const struct fsnotify_ops fanotify_fsnotify_ops = {
 	.handle_event = fanotify_handle_event,
-	.should_send_event = fanotify_should_send_event,
 	.free_group_priv = fanotify_free_group_priv,
-	.free_event_priv = NULL,
-	.freeing_mark = NULL,
+	.free_event = fanotify_free_event,
 };
diff --git a/fs/notify/fanotify/fanotify.h b/fs/notify/fanotify/fanotify.h
new file mode 100644
index 00000000000..2a5fb14115d
--- /dev/null
+++ b/fs/notify/fanotify/fanotify.h
@@ -0,0 +1,50 @@
+#include <linux/fsnotify_backend.h>
+#include <linux/path.h>
+#include <linux/slab.h>
+
+extern struct kmem_cache *fanotify_event_cachep;
+extern struct kmem_cache *fanotify_perm_event_cachep;
+
+/*
+ * Structure for normal fanotify events. It gets allocated in
+ * fanotify_handle_event() and freed when the information is retrieved by
+ * userspace
+ */
+struct fanotify_event_info {
+	struct fsnotify_event fse;
+	/*
+	 * We hold ref to this path so it may be dereferenced at any point
+	 * during this object's lifetime
+	 */
+	struct path path;
+	struct pid *tgid;
+};
+
+#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
+/*
+ * Structure for permission fanotify events. It gets allocated and freed in
+ * fanotify_handle_event() since we wait there for user response. When the
+ * information is retrieved by userspace the structure is moved from
+ * group->notification_list to group->fanotify_data.access_list to wait for
+ * user response.
+ */
+struct fanotify_perm_event_info {
+	struct fanotify_event_info fae;
+	int response;	/* userspace answer to question */
+	int fd;		/* fd we passed to userspace for this event */
+};
+
+static inline struct fanotify_perm_event_info *
+FANOTIFY_PE(struct fsnotify_event *fse)
+{
+	return container_of(fse, struct fanotify_perm_event_info, fae.fse);
+}
+#endif
+
+static inline struct fanotify_event_info *FANOTIFY_E(struct fsnotify_event *fse)
+{
+	return container_of(fse, struct fanotify_event_info, fse);
+}
+
+struct fanotify_event_info *fanotify_alloc_event(struct inode *inode, u32 mask,
+						 struct path *path);
diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c
index 6c80083a984..3fdc8a3e113 100644
--- a/fs/notify/fanotify/fanotify_user.c
+++ b/fs/notify/fanotify/fanotify_user.c
@@ -19,21 +19,30 @@
 
 #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
@@ -61,8 +70,8 @@ static struct fsnotify_event *get_one_event(struct fsnotify_group *group,
 }
 
 static int create_fd(struct fsnotify_group *group,
-			struct fsnotify_event *event,
-			struct file **file)
+		     struct fanotify_event_info *event,
+		     struct file **file)
 {
 	int client_fd;
 	struct file *new_file;
@@ -73,12 +82,6 @@ static int create_fd(struct fsnotify_group *group,
 	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.
@@ -109,22 +112,25 @@ static int create_fd(struct fsnotify_group *group,
 }
 
 static int fill_event_metadata(struct fsnotify_group *group,
-				   struct fanotify_event_metadata *metadata,
-				   struct fsnotify_event *event,
-				   struct file **file)
+			       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);
-	if (unlikely(event->mask & FAN_Q_OVERFLOW))
+	if (unlikely(fsn_event->mask & FAN_Q_OVERFLOW))
 		metadata->fd = FAN_NOFD;
 	else {
 		metadata->fd = create_fd(group, event, file);
@@ -136,33 +142,34 @@ static int fill_event_metadata(struct fsnotify_group *group,
 }
 
 #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);
@@ -182,58 +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 (atomic_read(&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;
-}
-
-#else
-static int prepare_for_access_response(struct fsnotify_group *group,
-				       struct fsnotify_event *event,
-				       __s32 fd)
-{
 	return 0;
 }
-
 #endif
 
 static ssize_t copy_event_to_user(struct fsnotify_group *group,
@@ -248,7 +212,7 @@ static ssize_t copy_event_to_user(struct fsnotify_group *group,
 
 	ret = fill_event_metadata(group, &fanotify_event_metadata, event, &f);
 	if (ret < 0)
-		goto out;
+		return ret;
 
 	fd = fanotify_event_metadata.fd;
 	ret = -EFAULT;
@@ -256,9 +220,10 @@ static ssize_t copy_event_to_user(struct fsnotify_group *group,
 			 fanotify_event_metadata.event_len))
 		goto out_close_fd;
 
-	ret = prepare_for_access_response(group, event, fd);
-	if (ret)
-		goto out_close_fd;
+#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
+	if (event->mask & FAN_ALL_PERM_EVENTS)
+		FANOTIFY_PE(event)->fd = fd;
+#endif
 
 	if (fd != FAN_NOFD)
 		fd_install(fd, f);
@@ -269,13 +234,6 @@ out_close_fd:
 		put_unused_fd(fd);
 		fput(f);
 	}
-out:
-#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
-	if (event->mask & FAN_ALL_PERM_EVENTS) {
-		event->response = FAN_DENY;
-		wake_up(&group->fanotify_data.access_waitq);
-	}
-#endif
 	return ret;
 }
 
@@ -315,30 +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);
@@ -379,29 +357,25 @@ 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);
 
 	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
 
-	if (file->f_flags & FASYNC)
-		fsnotify_fasync(-1, file, 0);
-
 	/* matches the fanotify_init->fsnotify_alloc_group */
 	fsnotify_destroy_group(group);
 
@@ -411,7 +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;
@@ -423,7 +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);
@@ -526,14 +500,18 @@ static int fanotify_remove_vfsmount_mark(struct fsnotify_group *group,
 	__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(fsn_mark, group);
+		fsnotify_destroy_mark_locked(fsn_mark, group);
+	mutex_unlock(&group->mark_mutex);
 
 	fsnotify_put_mark(fsn_mark);
 	if (removed & real_mount(mnt)->mnt_fsnotify_mask)
@@ -550,14 +528,19 @@ static int fanotify_remove_inode_mark(struct fsnotify_group *group,
 	__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(fsn_mark, group);
+		fsnotify_destroy_mark_locked(fsn_mark, group);
+	mutex_unlock(&group->mark_mutex);
+
 	/* matches the fsnotify_find_inode_mark() */
 	fsnotify_put_mark(fsn_mark);
 	if (removed & inode->i_fsnotify_mask)
@@ -593,35 +576,55 @@ 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)
 {
 	struct fsnotify_mark *fsn_mark;
 	__u32 added;
-	int ret = 0;
 
+	mutex_lock(&group->mark_mutex);
 	fsn_mark = fsnotify_find_vfsmount_mark(group, mnt);
 	if (!fsn_mark) {
-		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)
-			goto err;
+		fsn_mark = fanotify_add_new_mark(group, NULL, mnt);
+		if (IS_ERR(fsn_mark)) {
+			mutex_unlock(&group->mark_mutex);
+			return PTR_ERR(fsn_mark);
+		}
 	}
 	added = fanotify_mark_add_to_mask(fsn_mark, mask, flags);
+	mutex_unlock(&group->mark_mutex);
 
 	if (added & ~real_mount(mnt)->mnt_fsnotify_mask)
 		fsnotify_recalc_vfsmount_mask(mnt);
-err:
+
 	fsnotify_put_mark(fsn_mark);
-	return ret;
+	return 0;
 }
 
 static int fanotify_add_inode_mark(struct fsnotify_group *group,
@@ -630,7 +633,6 @@ static int fanotify_add_inode_mark(struct fsnotify_group *group,
 {
 	struct fsnotify_mark *fsn_mark;
 	__u32 added;
-	int ret = 0;
 
 	pr_debug("%s: group=%p inode=%p\n", __func__, group, inode);
 
@@ -644,27 +646,23 @@ 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) {
-		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)
-			goto err;
+		fsn_mark = fanotify_add_new_mark(group, inode, NULL);
+		if (IS_ERR(fsn_mark)) {
+			mutex_unlock(&group->mark_mutex);
+			return PTR_ERR(fsn_mark);
+		}
 	}
 	added = fanotify_mark_add_to_mask(fsn_mark, mask, flags);
+	mutex_unlock(&group->mark_mutex);
 
 	if (added & ~inode->i_fsnotify_mask)
 		fsnotify_recalc_inode_mask(inode);
-err:
+
 	fsnotify_put_mark(fsn_mark);
-	return ret;
+	return 0;
 }
 
 /* fanotify syscalls */
@@ -673,6 +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);
@@ -683,6 +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);
@@ -705,9 +716,18 @@ SYSCALL_DEFINE2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags)
 	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);
@@ -781,7 +801,10 @@ SYSCALL_DEFINE5(fanotify_mark, int, fanotify_fd, unsigned int, flags,
 	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;
@@ -818,6 +841,15 @@ SYSCALL_DEFINE5(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;
@@ -829,7 +861,7 @@ SYSCALL_DEFINE5(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);
@@ -842,12 +874,6 @@ SYSCALL_DEFINE5(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;
 	}
@@ -866,9 +892,9 @@ COMPAT_SYSCALL_DEFINE6(fanotify_mark,
 {
 	return sys_fanotify_mark(fanotify_fd, flags,
 #ifdef __BIG_ENDIAN
-				((__u64)mask1 << 32) | mask0,
-#else
 				((__u64)mask0 << 32) | mask1,
+#else
+				((__u64)mask1 << 32) | mask0,
 #endif
 				 dfd, pathname);
 }
@@ -882,8 +908,11 @@ COMPAT_SYSCALL_DEFINE6(fanotify_mark,
 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;
 }