5 files changed, 191 insertions, 154 deletions

diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c
index dc638f786d5..ee9cb3795c2 100644
--- a/fs/notify/fanotify/fanotify.c
+++ b/fs/notify/fanotify/fanotify.c
@@ -60,8 +60,8 @@ static int fanotify_merge(struct list_head *list, struct fsnotify_event *event)
 }
 
 #ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
-static int fanotify_get_response_from_access(struct fsnotify_group *group,
-					     struct fanotify_event_info *event)
+static int fanotify_get_response(struct fsnotify_group *group,
+				 struct fanotify_perm_event_info *event)
 {
 	int ret;
 
@@ -142,6 +142,40 @@ static bool fanotify_should_send_event(struct fsnotify_mark *inode_mark,
 	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,
@@ -171,25 +205,11 @@ static int fanotify_handle_event(struct fsnotify_group *group,
 	pr_debug("%s: group=%p inode=%p mask=%x\n", __func__, group, inode,
 		 mask);
 
-	event = kmem_cache_alloc(fanotify_event_cachep, GFP_KERNEL);
+	event = fanotify_alloc_event(inode, mask, data);
 	if (unlikely(!event))
 		return -ENOMEM;
 
 	fsn_event = &event->fse;
-	fsnotify_init_event(fsn_event, inode, mask);
-	event->tgid = get_pid(task_tgid(current));
-	if (data_type == FSNOTIFY_EVENT_PATH) {
-		struct path *path = data;
-		event->path = *path;
-		path_get(&event->path);
-	} else {
-		event->path.mnt = NULL;
-		event->path.dentry = NULL;
-	}
-#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
-	event->response = 0;
-#endif
-
 	ret = fsnotify_add_notify_event(group, fsn_event, fanotify_merge);
 	if (ret) {
 		/* Permission events shouldn't be merged */
@@ -202,7 +222,7 @@ static int fanotify_handle_event(struct fsnotify_group *group,
 
 #ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
 	if (mask & FAN_ALL_PERM_EVENTS) {
-		ret = fanotify_get_response_from_access(group, event);
+		ret = fanotify_get_response(group, FANOTIFY_PE(fsn_event));
 		fsnotify_destroy_event(group, fsn_event);
 	}
 #endif
@@ -225,6 +245,13 @@ static void fanotify_free_event(struct fsnotify_event *fsn_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);
 }
 
diff --git a/fs/notify/fanotify/fanotify.h b/fs/notify/fanotify/fanotify.h
index 32a2f034fb9..2a5fb14115d 100644
--- a/fs/notify/fanotify/fanotify.h
+++ b/fs/notify/fanotify/fanotify.h
@@ -3,13 +3,12 @@
 #include <linux/slab.h>
 
 extern struct kmem_cache *fanotify_event_cachep;
+extern struct kmem_cache *fanotify_perm_event_cachep;
 
 /*
- * Lifetime of the structure differs for normal and permission events. In both
- * cases the structure is allocated in fanotify_handle_event(). For normal
- * events the structure is freed immediately after reporting it to userspace.
- * For permission events we free it only after we receive response from
- * userspace.
+ * 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;
@@ -19,12 +18,33 @@ struct fanotify_event_info {
 	 */
 	struct path path;
 	struct pid *tgid;
+};
+
 #ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
-	u32 response;	/* userspace answer to question */
-#endif
+/*
+ * 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 287a22c0414..3fdc8a3e113 100644
--- a/fs/notify/fanotify/fanotify_user.c
+++ b/fs/notify/fanotify/fanotify_user.c
@@ -25,17 +25,24 @@
 #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 kmem_cache *fanotify_event_cachep __read_mostly;
-
-struct fanotify_response_event {
-	struct list_head list;
-	__s32 fd;
-	struct fanotify_event_info *event;
-};
+struct kmem_cache *fanotify_perm_event_cachep __read_mostly;
 
 /*
  * Get an fsnotify notification event if one exists and is small
@@ -135,33 +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);
@@ -181,58 +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 = FANOTIFY_E(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);
-		FANOTIFY_E(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,
@@ -247,7 +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;
@@ -255,9 +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);
@@ -268,13 +234,6 @@ out_close_fd:
 		put_unused_fd(fd);
 		fput(f);
 	}
-out:
-#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
-	if (event->mask & FAN_ALL_PERM_EVENTS) {
-		FANOTIFY_E(event)->response = FAN_DENY;
-		wake_up(&group->fanotify_data.access_waitq);
-	}
-#endif
 	return ret;
 }
 
@@ -314,35 +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);
-			/*
-			 * Permission events get destroyed after we
-			 * receive response
-			 */
-			if (!(kevent->mask & FAN_ALL_PERM_EVENTS))
-				fsnotify_destroy_event(group, kevent);
-			if (ret < 0)
+
+			ret = -ERESTARTSYS;
+			if (signal_pending(current))
 				break;
-			buf += ret;
-			count -= ret;
+
+			if (start != buf)
+				break;
+			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);
@@ -383,22 +357,21 @@ 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_del_init(&re->list);
-		re->event->response = FAN_ALLOW;
+	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);
 
-		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
@@ -709,6 +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);
@@ -731,21 +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 = kmem_cache_alloc(fanotify_event_cachep, GFP_KERNEL);
+	oevent = fanotify_alloc_event(NULL, FS_Q_OVERFLOW, NULL);
 	if (unlikely(!oevent)) {
 		fd = -ENOMEM;
 		goto out_destroy_group;
 	}
 	group->overflow_event = &oevent->fse;
-	fsnotify_init_event(group->overflow_event, NULL, FS_Q_OVERFLOW);
-	oevent->tgid = get_pid(task_tgid(current));
-	oevent->path.mnt = NULL;
-	oevent->path.dentry = NULL;
 
+	if (force_o_largefile())
+		event_f_flags |= O_LARGEFILE;
 	group->fanotify_data.f_flags = event_f_flags;
 #ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS
-	oevent->response = 0;
-	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);
@@ -819,7 +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;
@@ -856,6 +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;
@@ -867,7 +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);
@@ -880,12 +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;
 	}
@@ -920,9 +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;
 }
diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c
index 78a2ca3966c..cc423a30a0c 100644
--- a/fs/notify/inotify/inotify_user.c
+++ b/fs/notify/inotify/inotify_user.c
@@ -57,7 +57,7 @@ static struct kmem_cache *inotify_inode_mark_cachep __read_mostly;
 
 static int zero;
 
-ctl_table inotify_table[] = {
+struct ctl_table inotify_table[] = {
 	{
 		.procname	= "max_user_instances",
 		.data		= &inotify_max_user_instances,
diff --git a/fs/notify/mark.c b/fs/notify/mark.c
index 923fe4a5f50..d90deaa08e7 100644
--- a/fs/notify/mark.c
+++ b/fs/notify/mark.c
@@ -340,7 +340,7 @@ void fsnotify_init_mark(struct fsnotify_mark *mark,
 static int fsnotify_mark_destroy(void *ignored)
 {
 	struct fsnotify_mark *mark, *next;
-	LIST_HEAD(private_destroy_list);
+	struct list_head private_destroy_list;
 
 	for (;;) {
 		spin_lock(&destroy_lock);