6.6.18-rc1 review

@@ -2,8 +2,9 @@
 /*
  *  event_inode.c - part of tracefs, a pseudo file system for activating tracing
  *
- *  Copyright (C) 2020-23 VMware Inc, author: Steven Rostedt (VMware) <[email protected]>
+ *  Copyright (C) 2020-23 VMware Inc, author: Steven Rostedt <[email protected]>
  *  Copyright (C) 2020-23 VMware Inc, author: Ajay Kaher <[email protected]>
+ *  Copyright (C) 2023 Google, author: Steven Rostedt <[email protected]>
  *
  *  eventfs is used to dynamically create inodes and dentries based on the
  *  meta data provided by the tracing system.

@@ -23,46 +24,6 @@
 #include <linux/delay.h>
 #include "internal.h"
 
-struct eventfs_inode {
-	struct list_head	e_top_files;
-};
-
-/*
- * struct eventfs_file - hold the properties of the eventfs files and
- *                       directories.
- * @name:	the name of the file or directory to create
- * @d_parent:   holds parent's dentry
- * @dentry:     once accessed holds dentry
- * @list:	file or directory to be added to parent directory
- * @ei:		list of files and directories within directory
- * @fop:	file_operations for file or directory
- * @iop:	inode_operations for file or directory
- * @data:	something that the caller will want to get to later on
- * @mode:	the permission that the file or directory should have
- */
-struct eventfs_file {
-	const char			*name;
-	struct dentry			*d_parent;
-	struct dentry			*dentry;
-	struct list_head		list;
-	struct eventfs_inode		*ei;
-	const struct file_operations	*fop;
-	const struct inode_operations	*iop;
-	/*
-	 * Union - used for deletion
-	 * @del_list:	list of eventfs_file to delete
-	 * @rcu:	eventfs_file to delete in RCU
-	 * @is_freed:	node is freed if one of the above is set
-	 */
-	union {
-		struct list_head	del_list;
-		struct rcu_head		rcu;
-		unsigned long		is_freed;
-	};
-	void				*data;
-	umode_t				mode;
-};
-
 static DEFINE_MUTEX(eventfs_mutex);
 DEFINE_STATIC_SRCU(eventfs_srcu);
 

@@ -93,16 +54,9 @@ static const struct file_operations even
  * @data: something that the caller will want to get to later on.
  * @fop: struct file_operations that should be used for this file.
  *
- * This is the basic "create a file" function for tracefs.  It allows for a
- * wide range of flexibility in creating a file.
- *
- * This function will return a pointer to a dentry if it succeeds.  This
- * pointer must be passed to the tracefs_remove() function when the file is
- * to be removed (no automatic cleanup happens if your module is unloaded,
- * you are responsible here.)  If an error occurs, %NULL will be returned.
- *
- * If tracefs is not enabled in the kernel, the value -%ENODEV will be
- * returned.
+ * This function creates a dentry that represents a file in the eventsfs_inode
+ * directory. The inode.i_private pointer will point to @data in the open()
+ * call.
  */
 static struct dentry *create_file(const char *name, umode_t mode,
 				  struct dentry *parent, void *data,

@@ -118,6 +72,7 @@ static struct dentry *create_file(const
 	if (WARN_ON_ONCE(!S_ISREG(mode)))
 		return NULL;
 
+	WARN_ON_ONCE(!parent);
 	dentry = eventfs_start_creating(name, parent);
 
 	if (IS_ERR(dentry))

@@ -142,20 +97,11 @@ static struct dentry *create_file(const
  * create_dir - create a dir in the tracefs filesystem
  * @name: the name of the file to create.
  * @parent: parent dentry for this file.
- * @data: something that the caller will want to get to later on.
- *
- * This is the basic "create a dir" function for eventfs.  It allows for a
- * wide range of flexibility in creating a dir.
- *
- * This function will return a pointer to a dentry if it succeeds.  This
- * pointer must be passed to the tracefs_remove() function when the file is
- * to be removed (no automatic cleanup happens if your module is unloaded,
- * you are responsible here.)  If an error occurs, %NULL will be returned.
  *
- * If tracefs is not enabled in the kernel, the value -%ENODEV will be
- * returned.
+ * This function will create a dentry for a directory represented by
+ * a eventfs_inode.
  */
-static struct dentry *create_dir(const char *name, struct dentry *parent, void *data)
+static struct dentry *create_dir(const char *name, struct dentry *parent)
 {
 	struct tracefs_inode *ti;
 	struct dentry *dentry;

@@ -172,7 +118,6 @@ static struct dentry *create_dir(const c
 	inode->i_mode = S_IFDIR | S_IRWXU | S_IRUGO | S_IXUGO;
 	inode->i_op = &eventfs_root_dir_inode_operations;
 	inode->i_fop = &eventfs_file_operations;
-	inode->i_private = data;
 
 	ti = get_tracefs(inode);
 	ti->flags |= TRACEFS_EVENT_INODE;

@@ -185,18 +130,18 @@ static struct dentry *create_dir(const c
 }
 
 /**
- * eventfs_set_ef_status_free - set the ef->status to free
+ * eventfs_set_ei_status_free - remove the dentry reference from an eventfs_inode
  * @ti: the tracefs_inode of the dentry
- * @dentry: dentry who's status to be freed
+ * @dentry: dentry which has the reference to remove.
  *
- * eventfs_set_ef_status_free will be called if no more
- * references remain
+ * Remove the association between a dentry from an eventfs_inode.
  */
-void eventfs_set_ef_status_free(struct tracefs_inode *ti, struct dentry *dentry)
+void eventfs_set_ei_status_free(struct tracefs_inode *ti, struct dentry *dentry)
 {
 	struct tracefs_inode *ti_parent;
+	struct eventfs_inode *ei_child, *tmp;
 	struct eventfs_inode *ei;
-	struct eventfs_file *ef, *tmp;
+	int i;
 
 	/* The top level events directory may be freed by this */
 	if (unlikely(ti->flags & TRACEFS_EVENT_TOP_INODE)) {

@@ -207,9 +152,9 @@ void eventfs_set_ef_status_free(struct t
 		ei = ti->private;
 
 		/* Record all the top level files */
-		list_for_each_entry_srcu(ef, &ei->e_top_files, list,
+		list_for_each_entry_srcu(ei_child, &ei->children, list,
 					 lockdep_is_held(&eventfs_mutex)) {
-			list_add_tail(&ef->del_list, &ef_del_list);
+			list_add_tail(&ei_child->del_list, &ef_del_list);
 		}
 
 		/* Nothing should access this, but just in case! */

@@ -218,11 +163,13 @@ void eventfs_set_ef_status_free(struct t
 		mutex_unlock(&eventfs_mutex);
 
 		/* Now safely free the top level files and their children */
-		list_for_each_entry_safe(ef, tmp, &ef_del_list, del_list) {
-			list_del(&ef->del_list);
-			eventfs_remove(ef);
+		list_for_each_entry_safe(ei_child, tmp, &ef_del_list, del_list) {
+			list_del(&ei_child->del_list);
+			eventfs_remove_dir(ei_child);
 		}
 
+		kfree_const(ei->name);
+		kfree(ei->d_children);
 		kfree(ei);
 		return;
 	}

@@ -233,68 +180,162 @@ void eventfs_set_ef_status_free(struct t
 	if (!ti_parent || !(ti_parent->flags & TRACEFS_EVENT_INODE))
 		goto out;
 
-	ef = dentry->d_fsdata;
-	if (!ef)
+	ei = dentry->d_fsdata;
+	if (!ei)
 		goto out;
 
 	/*
-	 * If ef was freed, then the LSB bit is set for d_fsdata.
+	 * If ei was freed, then the LSB bit is set for d_fsdata.
 	 * But this should not happen, as it should still have a
 	 * ref count that prevents it. Warn in case it does.
 	 */
-	if (WARN_ON_ONCE((unsigned long)ef & 1))
+	if (WARN_ON_ONCE((unsigned long)ei & 1))
 		goto out;
 
+	/* This could belong to one of the files of the ei */
+	if (ei->dentry != dentry) {
+		for (i = 0; i < ei->nr_entries; i++) {
+			if (ei->d_children[i] == dentry)
+				break;
+		}
+		if (WARN_ON_ONCE(i == ei->nr_entries))
+			goto out;
+		ei->d_children[i] = NULL;
+	} else {
+		ei->dentry = NULL;
+	}
+
 	dentry->d_fsdata = NULL;
-	ef->dentry = NULL;
-out:
+ out:
 	mutex_unlock(&eventfs_mutex);
 }
 
 /**
+ * create_file_dentry - create a dentry for a file of an eventfs_inode
+ * @ei: the eventfs_inode that the file will be created under
+ * @e_dentry: a pointer to the d_children[] of the @ei
+ * @parent: The parent dentry of the created file.
+ * @name: The name of the file to create
+ * @mode: The mode of the file.
+ * @data: The data to use to set the inode of the file with on open()
+ * @fops: The fops of the file to be created.
+ * @lookup: If called by the lookup routine, in which case, dput() the created dentry.
+ *
+ * Create a dentry for a file of an eventfs_inode @ei and place it into the
+ * address located at @e_dentry. If the @e_dentry already has a dentry, then
+ * just do a dget() on it and return. Otherwise create the dentry and attach it.
+ */
+static struct dentry *
+create_file_dentry(struct eventfs_inode *ei, struct dentry **e_dentry,
+		   struct dentry *parent, const char *name, umode_t mode, void *data,
+		   const struct file_operations *fops, bool lookup)
+{
+	struct dentry *dentry;
+	bool invalidate = false;
+
+	mutex_lock(&eventfs_mutex);
+	/* If the e_dentry already has a dentry, use it */
+	if (*e_dentry) {
+		/* lookup does not need to up the ref count */
+		if (!lookup)
+			dget(*e_dentry);
+		mutex_unlock(&eventfs_mutex);
+		return *e_dentry;
+	}
+	mutex_unlock(&eventfs_mutex);
+
+	/* The lookup already has the parent->d_inode locked */
+	if (!lookup)
+		inode_lock(parent->d_inode);
+
+	dentry = create_file(name, mode, parent, data, fops);
+
+	if (!lookup)
+		inode_unlock(parent->d_inode);
+
+	mutex_lock(&eventfs_mutex);
+
+	if (IS_ERR_OR_NULL(dentry)) {
+		/*
+		 * When the mutex was released, something else could have
+		 * created the dentry for this e_dentry. In which case
+		 * use that one.
+		 *
+		 * Note, with the mutex held, the e_dentry cannot have content
+		 * and the ei->is_freed be true at the same time.
+		 */
+		WARN_ON_ONCE(ei->is_freed);
+		dentry = *e_dentry;
+		/* The lookup does not need to up the dentry refcount */
+		if (dentry && !lookup)
+			dget(dentry);
+		mutex_unlock(&eventfs_mutex);
+		return dentry;
+	}
+
+	if (!*e_dentry && !ei->is_freed) {
+		*e_dentry = dentry;
+		dentry->d_fsdata = ei;
+	} else {
+		/*
+		 * Should never happen unless we get here due to being freed.
+		 * Otherwise it means two dentries exist with the same name.
+		 */
+		WARN_ON_ONCE(!ei->is_freed);
+		invalidate = true;
+	}
+	mutex_unlock(&eventfs_mutex);
+
+	if (invalidate)
+		d_invalidate(dentry);
+
+	if (lookup || invalidate)
+		dput(dentry);
+
+	return invalidate ? NULL : dentry;
+}
+
+/**
  * eventfs_post_create_dir - post create dir routine
- * @ef: eventfs_file of recently created dir
+ * @ei: eventfs_inode of recently created dir
  *
  * Map the meta-data of files within an eventfs dir to their parent dentry
  */
-static void eventfs_post_create_dir(struct eventfs_file *ef)
+static void eventfs_post_create_dir(struct eventfs_inode *ei)
 {
-	struct eventfs_file *ef_child;
+	struct eventfs_inode *ei_child;
 	struct tracefs_inode *ti;
 
 	/* srcu lock already held */
 	/* fill parent-child relation */
-	list_for_each_entry_srcu(ef_child, &ef->ei->e_top_files, list,
+	list_for_each_entry_srcu(ei_child, &ei->children, list,
 				 srcu_read_lock_held(&eventfs_srcu)) {
-		ef_child->d_parent = ef->dentry;
+		ei_child->d_parent = ei->dentry;
 	}
 
-	ti = get_tracefs(ef->dentry->d_inode);
-	ti->private = ef->ei;
+	ti = get_tracefs(ei->dentry->d_inode);
+	ti->private = ei;
 }
 
 /**
- * create_dentry - helper function to create dentry
- * @ef: eventfs_file of file or directory to create
- * @parent: parent dentry
- * @lookup: true if called from lookup routine
+ * create_dir_dentry - Create a directory dentry for the eventfs_inode
+ * @ei: The eventfs_inode to create the directory for
+ * @parent: The dentry of the parent of this directory
+ * @lookup: True if this is called by the lookup code
  *
- * Used to create a dentry for file/dir, executes post dentry creation routine
+ * This creates and attaches a directory dentry to the eventfs_inode @ei.
  */
 static struct dentry *
-create_dentry(struct eventfs_file *ef, struct dentry *parent, bool lookup)
+create_dir_dentry(struct eventfs_inode *ei, struct dentry *parent, bool lookup)
 {
 	bool invalidate = false;
-	struct dentry *dentry;
+	struct dentry *dentry = NULL;
 
 	mutex_lock(&eventfs_mutex);
-	if (ef->is_freed) {
-		mutex_unlock(&eventfs_mutex);
-		return NULL;
-	}
-	if (ef->dentry) {
-		dentry = ef->dentry;
-		/* On dir open, up the ref count */
+	if (ei->dentry) {
+		/* If the dentry already has a dentry, use it */
+		dentry = ei->dentry;
+		/* lookup does not need to up the ref count */
 		if (!lookup)
 			dget(dentry);
 		mutex_unlock(&eventfs_mutex);

@@ -302,42 +343,44 @@ create_dentry(struct eventfs_file *ef, s
 	}
 	mutex_unlock(&eventfs_mutex);
 
+	/* The lookup already has the parent->d_inode locked */
 	if (!lookup)
 		inode_lock(parent->d_inode);
 
-	if (ef->ei)
-		dentry = create_dir(ef->name, parent, ef->data);
-	else
-		dentry = create_file(ef->name, ef->mode, parent,
-				     ef->data, ef->fop);
+	dentry = create_dir(ei->name, parent);
 
 	if (!lookup)
 		inode_unlock(parent->d_inode);
 
 	mutex_lock(&eventfs_mutex);
-	if (IS_ERR_OR_NULL(dentry)) {
-		/* If the ef was already updated get it */
-		dentry = ef->dentry;
+
+	if (IS_ERR_OR_NULL(dentry) && !ei->is_freed) {
+		/*
+		 * When the mutex was released, something else could have
+		 * created the dentry for this e_dentry. In which case
+		 * use that one.
+		 *
+		 * Note, with the mutex held, the e_dentry cannot have content
+		 * and the ei->is_freed be true at the same time.
+		 */
+		dentry = ei->dentry;
 		if (dentry && !lookup)
 			dget(dentry);
 		mutex_unlock(&eventfs_mutex);
 		return dentry;
 	}
 
-	if (!ef->dentry && !ef->is_freed) {
-		ef->dentry = dentry;
-		if (ef->ei)
-			eventfs_post_create_dir(ef);
-		dentry->d_fsdata = ef;
+	if (!ei->dentry && !ei->is_freed) {
+		ei->dentry = dentry;
+		eventfs_post_create_dir(ei);
+		dentry->d_fsdata = ei;
 	} else {
-		/* A race here, should try again (unless freed) */
-		invalidate = true;
-
 		/*
 		 * Should never happen unless we get here due to being freed.
 		 * Otherwise it means two dentries exist with the same name.
 		 */
-		WARN_ON_ONCE(!ef->is_freed);
+		WARN_ON_ONCE(!ei->is_freed);
+		invalidate = true;
 	}
 	mutex_unlock(&eventfs_mutex);
 	if (invalidate)

@@ -349,50 +392,85 @@ create_dentry(struct eventfs_file *ef, s
 	return invalidate ? NULL : dentry;
 }
 
-static bool match_event_file(struct eventfs_file *ef, const char *name)
-{
-	bool ret;
-
-	mutex_lock(&eventfs_mutex);
-	ret = !ef->is_freed && strcmp(ef->name, name) == 0;
-	mutex_unlock(&eventfs_mutex);
-
-	return ret;
-}
-
 /**
  * eventfs_root_lookup - lookup routine to create file/dir
  * @dir: in which a lookup is being done
  * @dentry: file/dir dentry
- * @flags: to pass as flags parameter to simple lookup
+ * @flags: Just passed to simple_lookup()
  *
- * Used to create a dynamic file/dir within @dir. Use the eventfs_inode
- * list of meta data to find the information needed to create the file/dir.
+ * Used to create dynamic file/dir with-in @dir, search with-in @ei
+ * list, if @dentry found go ahead and create the file/dir
  */
+
 static struct dentry *eventfs_root_lookup(struct inode *dir,
 					  struct dentry *dentry,
 					  unsigned int flags)
 {
+	const struct file_operations *fops;
+	const struct eventfs_entry *entry;
+	struct eventfs_inode *ei_child;
 	struct tracefs_inode *ti;
 	struct eventfs_inode *ei;
-	struct eventfs_file *ef;
+	struct dentry *ei_dentry = NULL;
 	struct dentry *ret = NULL;
+	const char *name = dentry->d_name.name;
+	bool created = false;
+	umode_t mode;
+	void *data;
 	int idx;
+	int i;
+	int r;
 
 	ti = get_tracefs(dir);
 	if (!(ti->flags & TRACEFS_EVENT_INODE))
 		return NULL;
 
-	ei = ti->private;
+	/* Grab srcu to prevent the ei from going away */
 	idx = srcu_read_lock(&eventfs_srcu);
-	list_for_each_entry_srcu(ef, &ei->e_top_files, list,
+
+	/*
+	 * Grab the eventfs_mutex to consistent value from ti->private.
+	 * This s
+	 */
+	mutex_lock(&eventfs_mutex);
+	ei = READ_ONCE(ti->private);
+	if (ei)
+		ei_dentry = READ_ONCE(ei->dentry);
+	mutex_unlock(&eventfs_mutex);
+
+	if (!ei || !ei_dentry)
+		goto out;
+
+	data = ei->data;
+
+	list_for_each_entry_srcu(ei_child, &ei->children, list,
 				 srcu_read_lock_held(&eventfs_srcu)) {
-		if (!match_event_file(ef, dentry->d_name.name))
+		if (strcmp(ei_child->name, name) != 0)
 			continue;
 		ret = simple_lookup(dir, dentry, flags);
-		create_dentry(ef, ef->d_parent, true);
+		create_dir_dentry(ei_child, ei_dentry, true);
+		created = true;
 		break;
 	}
+
+	if (created)
+		goto out;
+
+	for (i = 0; i < ei->nr_entries; i++) {
+		entry = &ei->entries[i];
+		if (strcmp(name, entry->name) == 0) {
+			void *cdata = data;
+			r = entry->callback(name, &mode, &cdata, &fops);
+			if (r <= 0)
+				continue;
+			ret = simple_lookup(dir, dentry, flags);
+			create_file_dentry(ei, &ei->d_children[i],
+					   ei_dentry, name, mode, cdata,
+					   fops, true);
+			break;
+		}
+	}
+ out:
 	srcu_read_unlock(&eventfs_srcu, idx);
 	return ret;
 }

@@ -432,29 +510,48 @@ static int eventfs_release(struct inode
 	return dcache_dir_close(inode, file);
 }
 
+static int add_dentries(struct dentry ***dentries, struct dentry *d, int cnt)
+{
+	struct dentry **tmp;
+
+	tmp = krealloc(*dentries, sizeof(d) * (cnt + 2), GFP_KERNEL);
+	if (!tmp)
+		return -1;
+	tmp[cnt] = d;
+	tmp[cnt + 1] = NULL;
+	*dentries = tmp;
+	return 0;
+}
+
 /**
  * dcache_dir_open_wrapper - eventfs open wrapper
  * @inode: not used
- * @file: dir to be opened (to create its child)
+ * @file: dir to be opened (to create it's children)
  *
- * Used to dynamically create the file/dir within @file. @file is really a
- * directory and all the files/dirs of the children within @file will be
- * created. If any of the files/dirs have already been created, their
- * reference count will be incremented.
+ * Used to dynamic create file/dir with-in @file, all the
+ * file/dir will be created. If already created then references
+ * will be increased
  */
 static int dcache_dir_open_wrapper(struct inode *inode, struct file *file)
 {
+	const struct file_operations *fops;
+	const struct eventfs_entry *entry;
+	struct eventfs_inode *ei_child;
 	struct tracefs_inode *ti;
 	struct eventfs_inode *ei;
-	struct eventfs_file *ef;
 	struct dentry_list *dlist;
 	struct dentry **dentries = NULL;
-	struct dentry *dentry = file_dentry(file);
+	struct dentry *parent = file_dentry(file);
 	struct dentry *d;
 	struct inode *f_inode = file_inode(file);
+	const char *name = parent->d_name.name;
+	umode_t mode;
+	void *data;
 	int cnt = 0;
 	int idx;
 	int ret;
+	int i;
+	int r;
 
 	ti = get_tracefs(f_inode);
 	if (!(ti->flags & TRACEFS_EVENT_INODE))

@@ -463,25 +560,51 @@ static int dcache_dir_open_wrapper(struc
 	if (WARN_ON_ONCE(file->private_data))
 		return -EINVAL;
 
+	idx = srcu_read_lock(&eventfs_srcu);
+
+	mutex_lock(&eventfs_mutex);
+	ei = READ_ONCE(ti->private);
+	mutex_unlock(&eventfs_mutex);
+
+	if (!ei) {
+		srcu_read_unlock(&eventfs_srcu, idx);
+		return -EINVAL;
+	}
+
+
+	data = ei->data;
+
 	dlist = kmalloc(sizeof(*dlist), GFP_KERNEL);
-	if (!dlist)
+	if (!dlist) {
+		srcu_read_unlock(&eventfs_srcu, idx);
 		return -ENOMEM;
+	}
 
-	ei = ti->private;
-	idx = srcu_read_lock(&eventfs_srcu);
-	list_for_each_entry_srcu(ef, &ei->e_top_files, list,
+	list_for_each_entry_srcu(ei_child, &ei->children, list,
 				 srcu_read_lock_held(&eventfs_srcu)) {
-		d = create_dentry(ef, dentry, false);
+		d = create_dir_dentry(ei_child, parent, false);
 		if (d) {
-			struct dentry **tmp;
+			ret = add_dentries(&dentries, d, cnt);
+			if (ret < 0)
+				break;
+			cnt++;
+		}
+	}
 
-			tmp = krealloc(dentries, sizeof(d) * (cnt + 2), GFP_KERNEL);
-			if (!tmp)
+	for (i = 0; i < ei->nr_entries; i++) {
+		void *cdata = data;
+		entry = &ei->entries[i];
+		name = entry->name;
+		r = entry->callback(name, &mode, &cdata, &fops);
+		if (r <= 0)
+			continue;
+		d = create_file_dentry(ei, &ei->d_children[i],
+				       parent, name, mode, cdata, fops, false);
+		if (d) {
+			ret = add_dentries(&dentries, d, cnt);
+			if (ret < 0)
 				break;
-			tmp[cnt] = d;
-			tmp[cnt + 1] = NULL;
 			cnt++;
-			dentries = tmp;
 		}
 	}
 	srcu_read_unlock(&eventfs_srcu, idx);

@@ -514,63 +637,90 @@ static int dcache_readdir_wrapper(struct
 }
 
 /**
- * eventfs_prepare_ef - helper function to prepare eventfs_file
- * @name: the name of the file/directory to create.
- * @mode: the permission that the file should have.
- * @fop: struct file_operations that should be used for this file/directory.
- * @iop: struct inode_operations that should be used for this file/directory.
- * @data: something that the caller will want to get to later on. The
- *        inode.i_private pointer will point to this value on the open() call.
- *
- * This function allocates and fills the eventfs_file structure.
- */
-static struct eventfs_file *eventfs_prepare_ef(const char *name, umode_t mode,
-					const struct file_operations *fop,
-					const struct inode_operations *iop,
-					void *data)
+ * eventfs_create_dir - Create the eventfs_inode for this directory
+ * @name: The name of the directory to create.
+ * @parent: The eventfs_inode of the parent directory.
+ * @entries: A list of entries that represent the files under this directory
+ * @size: The number of @entries
+ * @data: The default data to pass to the files (an entry may override it).
+ *
+ * This function creates the descriptor to represent a directory in the
+ * eventfs. This descriptor is an eventfs_inode, and it is returned to be
+ * used to create other children underneath.
+ *
+ * The @entries is an array of eventfs_entry structures which has:
+ *	const char		 *name
+ *	eventfs_callback	callback;
+ *
+ * The name is the name of the file, and the callback is a pointer to a function
+ * that will be called when the file is reference (either by lookup or by
+ * reading a directory). The callback is of the prototype:
+ *
+ *    int callback(const char *name, umode_t *mode, void **data,
+ *		   const struct file_operations **fops);
+ *
+ * When a file needs to be created, this callback will be called with
+ *   name = the name of the file being created (so that the same callback
+ *          may be used for multiple files).
+ *   mode = a place to set the file's mode
+ *   data = A pointer to @data, and the callback may replace it, which will
+ *         cause the file created to pass the new data to the open() call.
+ *   fops = the fops to use for the created file.
+ */
+struct eventfs_inode *eventfs_create_dir(const char *name, struct eventfs_inode *parent,
+					 const struct eventfs_entry *entries,
+					 int size, void *data)
 {
-	struct eventfs_file *ef;
+	struct eventfs_inode *ei;
+
+	if (!parent)
+		return ERR_PTR(-EINVAL);
 
-	ef = kzalloc(sizeof(*ef), GFP_KERNEL);
-	if (!ef)
+	ei = kzalloc(sizeof(*ei), GFP_KERNEL);
+	if (!ei)
 		return ERR_PTR(-ENOMEM);
 
-	ef->name = kstrdup(name, GFP_KERNEL);
-	if (!ef->name) {
-		kfree(ef);
+	ei->name = kstrdup_const(name, GFP_KERNEL);
+	if (!ei->name) {
+		kfree(ei);
 		return ERR_PTR(-ENOMEM);
 	}
 
-	if (S_ISDIR(mode)) {
-		ef->ei = kzalloc(sizeof(*ef->ei), GFP_KERNEL);
-		if (!ef->ei) {
-			kfree(ef->name);
-			kfree(ef);
+	if (size) {
+		ei->d_children = kzalloc(sizeof(*ei->d_children) * size, GFP_KERNEL);
+		if (!ei->d_children) {
+			kfree_const(ei->name);
+			kfree(ei);
 			return ERR_PTR(-ENOMEM);
 		}
-		INIT_LIST_HEAD(&ef->ei->e_top_files);
-	} else {
-		ef->ei = NULL;
 	}
 
-	ef->iop = iop;
-	ef->fop = fop;
-	ef->mode = mode;
-	ef->data = data;
-	return ef;
+	ei->entries = entries;
+	ei->nr_entries = size;
+	ei->data = data;
+	INIT_LIST_HEAD(&ei->children);
+
+	mutex_lock(&eventfs_mutex);
+	list_add_tail(&ei->list, &parent->children);
+	ei->d_parent = parent->dentry;
+	mutex_unlock(&eventfs_mutex);
+
+	return ei;
 }
 
 /**
- * eventfs_create_events_dir - create the trace event structure
- * @name: the name of the directory to create.
- * @parent: parent dentry for this file.  This should be a directory dentry
- *          if set.  If this parameter is NULL, then the directory will be
- *          created in the root of the tracefs filesystem.
+ * eventfs_create_events_dir - create the top level events directory
+ * @name: The name of the top level directory to create.
+ * @parent: Parent dentry for this file in the tracefs directory.
+ * @entries: A list of entries that represent the files under this directory
+ * @size: The number of @entries
+ * @data: The default data to pass to the files (an entry may override it).
  *
  * This function creates the top of the trace event directory.
  */
-struct dentry *eventfs_create_events_dir(const char *name,
-					 struct dentry *parent)
+struct eventfs_inode *eventfs_create_events_dir(const char *name, struct dentry *parent,
+						const struct eventfs_entry *entries,
+						int size, void *data)
 {
 	struct dentry *dentry = tracefs_start_creating(name, parent);
 	struct eventfs_inode *ei;

@@ -581,19 +731,32 @@ struct dentry *eventfs_create_events_dir
 		return NULL;
 
 	if (IS_ERR(dentry))
-		return dentry;
+		return (struct eventfs_inode *)dentry;
 
 	ei = kzalloc(sizeof(*ei), GFP_KERNEL);
 	if (!ei)
-		return ERR_PTR(-ENOMEM);
+		goto fail;
+
 	inode = tracefs_get_inode(dentry->d_sb);
-	if (unlikely(!inode)) {
-		kfree(ei);
-		tracefs_failed_creating(dentry);
-		return ERR_PTR(-ENOMEM);
-	}
+	if (unlikely(!inode))
+		goto fail;
+
+	if (size) {
+		ei->d_children = kzalloc(sizeof(*ei->d_children) * size, GFP_KERNEL);
+		if (!ei->d_children)
+			goto fail;
+	}
+
+	ei->dentry = dentry;
+	ei->entries = entries;
+	ei->nr_entries = size;
+	ei->data = data;
+	ei->name = kstrdup_const(name, GFP_KERNEL);
+	if (!ei->name)
+		goto fail;
 
-	INIT_LIST_HEAD(&ei->e_top_files);
+	INIT_LIST_HEAD(&ei->children);
+	INIT_LIST_HEAD(&ei->list);
 
 	ti = get_tracefs(inode);
 	ti->flags |= TRACEFS_EVENT_INODE | TRACEFS_EVENT_TOP_INODE;

@@ -608,193 +771,41 @@ struct dentry *eventfs_create_events_dir
 	d_instantiate(dentry, inode);
 	inc_nlink(dentry->d_parent->d_inode);
 	fsnotify_mkdir(dentry->d_parent->d_inode, dentry);
-	return tracefs_end_creating(dentry);
-}
+	tracefs_end_creating(dentry);
 
-/**
- * eventfs_add_subsystem_dir - add eventfs subsystem_dir to list to create later
- * @name: the name of the file to create.
- * @parent: parent dentry for this dir.
- *
- * This function adds eventfs subsystem dir to list.
- * And all these dirs are created on the fly when they are looked up,
- * and the dentry and inodes will be removed when they are done.
- */
-struct eventfs_file *eventfs_add_subsystem_dir(const char *name,
-					       struct dentry *parent)
-{
-	struct tracefs_inode *ti_parent;
-	struct eventfs_inode *ei_parent;
-	struct eventfs_file *ef;
+	/* Will call dput when the directory is removed */
+	dget(dentry);
 
-	if (security_locked_down(LOCKDOWN_TRACEFS))
-		return NULL;
-
-	if (!parent)
-		return ERR_PTR(-EINVAL);
+	return ei;
 
-	ti_parent = get_tracefs(parent->d_inode);
-	ei_parent = ti_parent->private;
-
-	ef = eventfs_prepare_ef(name, S_IFDIR, NULL, NULL, NULL);
-	if (IS_ERR(ef))
-		return ef;
-
-	mutex_lock(&eventfs_mutex);
-	list_add_tail(&ef->list, &ei_parent->e_top_files);
-	ef->d_parent = parent;
-	mutex_unlock(&eventfs_mutex);
-	return ef;
+ fail:
+	kfree(ei->d_children);
+	kfree(ei);
+	tracefs_failed_creating(dentry);
+	return ERR_PTR(-ENOMEM);
 }
 
-/**
- * eventfs_add_dir - add eventfs dir to list to create later
- * @name: the name of the file to create.
- * @ef_parent: parent eventfs_file for this dir.
- *
- * This function adds eventfs dir to list.
- * And all these dirs are created on the fly when they are looked up,
- * and the dentry and inodes will be removed when they are done.
- */
-struct eventfs_file *eventfs_add_dir(const char *name,
-				     struct eventfs_file *ef_parent)
+static void free_ei(struct rcu_head *head)
 {
-	struct eventfs_file *ef;
+	struct eventfs_inode *ei = container_of(head, struct eventfs_inode, rcu);
 
-	if (security_locked_down(LOCKDOWN_TRACEFS))
-		return NULL;
-
-	if (!ef_parent)
-		return ERR_PTR(-EINVAL);
-
-	ef = eventfs_prepare_ef(name, S_IFDIR, NULL, NULL, NULL);
-	if (IS_ERR(ef))
-		return ef;
-
-	mutex_lock(&eventfs_mutex);
-	list_add_tail(&ef->list, &ef_parent->ei->e_top_files);
-	ef->d_parent = ef_parent->dentry;
-	mutex_unlock(&eventfs_mutex);
-	return ef;
-}
-
-/**
- * eventfs_add_events_file - add the data needed to create a file for later reference
- * @name: the name of the file to create.
- * @mode: the permission that the file should have.
- * @parent: parent dentry for this file.
- * @data: something that the caller will want to get to later on.
- * @fop: struct file_operations that should be used for this file.
- *
- * This function is used to add the information needed to create a
- * dentry/inode within the top level events directory. The file created
- * will have the @mode permissions. The @data will be used to fill the
- * inode.i_private when the open() call is done. The dentry and inodes are
- * all created when they are referenced, and removed when they are no
- * longer referenced.
- */
-int eventfs_add_events_file(const char *name, umode_t mode,
-			 struct dentry *parent, void *data,
-			 const struct file_operations *fop)
-{
-	struct tracefs_inode *ti;
-	struct eventfs_inode *ei;
-	struct eventfs_file *ef;
-
-	if (security_locked_down(LOCKDOWN_TRACEFS))
-		return -ENODEV;
-
-	if (!parent)
-		return -EINVAL;
-
-	if (!(mode & S_IFMT))
-		mode |= S_IFREG;
-
-	if (!parent->d_inode)
-		return -EINVAL;
-
-	ti = get_tracefs(parent->d_inode);
-	if (!(ti->flags & TRACEFS_EVENT_INODE))
-		return -EINVAL;
-
-	ei = ti->private;
-	ef = eventfs_prepare_ef(name, mode, fop, NULL, data);
-
-	if (IS_ERR(ef))
-		return -ENOMEM;
-
-	mutex_lock(&eventfs_mutex);
-	list_add_tail(&ef->list, &ei->e_top_files);
-	ef->d_parent = parent;
-	mutex_unlock(&eventfs_mutex);
-	return 0;
-}
-
-/**
- * eventfs_add_file - add eventfs file to list to create later
- * @name: the name of the file to create.
- * @mode: the permission that the file should have.
- * @ef_parent: parent eventfs_file for this file.
- * @data: something that the caller will want to get to later on.
- * @fop: struct file_operations that should be used for this file.
- *
- * This function is used to add the information needed to create a
- * file within a subdirectory of the events directory. The file created
- * will have the @mode permissions. The @data will be used to fill the
- * inode.i_private when the open() call is done. The dentry and inodes are
- * all created when they are referenced, and removed when they are no
- * longer referenced.
- */
-int eventfs_add_file(const char *name, umode_t mode,
-		     struct eventfs_file *ef_parent,
-		     void *data,
-		     const struct file_operations *fop)
-{
-	struct eventfs_file *ef;
-
-	if (security_locked_down(LOCKDOWN_TRACEFS))
-		return -ENODEV;
-
-	if (!ef_parent)
-		return -EINVAL;
-
-	if (!(mode & S_IFMT))
-		mode |= S_IFREG;
-
-	ef = eventfs_prepare_ef(name, mode, fop, NULL, data);
-	if (IS_ERR(ef))
-		return -ENOMEM;
-
-	mutex_lock(&eventfs_mutex);
-	list_add_tail(&ef->list, &ef_parent->ei->e_top_files);
-	ef->d_parent = ef_parent->dentry;
-	mutex_unlock(&eventfs_mutex);
-	return 0;
-}
-
-static void free_ef(struct rcu_head *head)
-{
-	struct eventfs_file *ef = container_of(head, struct eventfs_file, rcu);
-
-	kfree(ef->name);
-	kfree(ef->ei);
-	kfree(ef);
+	kfree_const(ei->name);
+	kfree(ei->d_children);
+	kfree(ei);
 }
 
 /**
  * eventfs_remove_rec - remove eventfs dir or file from list
- * @ef: eventfs_file to be removed.
- * @head: to create list of eventfs_file to be deleted
- * @level: to check recursion depth
+ * @ei: eventfs_inode to be removed.
  *
- * The helper function eventfs_remove_rec() is used to clean up and free the
- * associated data from eventfs for both of the added functions.
+ * This function recursively remove eventfs_inode which
+ * contains info of file or dir.
  */
-static void eventfs_remove_rec(struct eventfs_file *ef, struct list_head *head, int level)
+static void eventfs_remove_rec(struct eventfs_inode *ei, struct list_head *head, int level)
 {
-	struct eventfs_file *ef_child;
+	struct eventfs_inode *ei_child;
 
-	if (!ef)
+	if (!ei)
 		return;
 	/*
 	 * Check recursion depth. It should never be greater than 3:

@@ -806,62 +817,68 @@ static void eventfs_remove_rec(struct ev
 	if (WARN_ON_ONCE(level > 3))
 		return;
 
-	if (ef->ei) {
-		/* search for nested folders or files */
-		list_for_each_entry_srcu(ef_child, &ef->ei->e_top_files, list,
-					 lockdep_is_held(&eventfs_mutex)) {
-			eventfs_remove_rec(ef_child, head, level + 1);
-		}
+	/* search for nested folders or files */
+	list_for_each_entry_srcu(ei_child, &ei->children, list,
+				 lockdep_is_held(&eventfs_mutex)) {
+		eventfs_remove_rec(ei_child, head, level + 1);
 	}
 
-	list_del_rcu(&ef->list);
-	list_add_tail(&ef->del_list, head);
+	list_del_rcu(&ei->list);
+	list_add_tail(&ei->del_list, head);
 }
 
+static void unhook_dentry(struct dentry **dentry, struct dentry **list)
+{
+	if (*dentry) {
+		unsigned long ptr = (unsigned long)*list;
+
+		/* Keep the dentry from being freed yet */
+		dget(*dentry);
+
+		/*
+		 * Paranoid: The dget() above should prevent the dentry
+		 * from being freed and calling eventfs_set_ei_status_free().
+		 * But just in case, set the link list LSB pointer to 1
+		 * and have eventfs_set_ei_status_free() check that to
+		 * make sure that if it does happen, it will not think
+		 * the d_fsdata is an eventfs_inode.
+		 *
+		 * For this to work, no eventfs_inode should be allocated
+		 * on a odd space, as the ef should always be allocated
+		 * to be at least word aligned. Check for that too.
+		 */
+		WARN_ON_ONCE(ptr & 1);
+
+		(*dentry)->d_fsdata = (void *)(ptr | 1);
+		*list = *dentry;
+		*dentry = NULL;
+	}
+}
 /**
  * eventfs_remove - remove eventfs dir or file from list
- * @ef: eventfs_file to be removed.
+ * @ei: eventfs_inode to be removed.
  *
  * This function acquire the eventfs_mutex lock and call eventfs_remove_rec()
  */
-void eventfs_remove(struct eventfs_file *ef)
+void eventfs_remove_dir(struct eventfs_inode *ei)
 {
-	struct eventfs_file *tmp;
-	LIST_HEAD(ef_del_list);
+	struct eventfs_inode *tmp;
+	LIST_HEAD(ei_del_list);
 	struct dentry *dentry_list = NULL;
 	struct dentry *dentry;
+	int i;
 
-	if (!ef)
+	if (!ei)
 		return;
 
 	mutex_lock(&eventfs_mutex);
-	eventfs_remove_rec(ef, &ef_del_list, 0);
-	list_for_each_entry_safe(ef, tmp, &ef_del_list, del_list) {
-		if (ef->dentry) {
-			unsigned long ptr = (unsigned long)dentry_list;
-
-			/* Keep the dentry from being freed yet */
-			dget(ef->dentry);
-
-			/*
-			 * Paranoid: The dget() above should prevent the dentry
-			 * from being freed and calling eventfs_set_ef_status_free().
-			 * But just in case, set the link list LSB pointer to 1
-			 * and have eventfs_set_ef_status_free() check that to
-			 * make sure that if it does happen, it will not think
-			 * the d_fsdata is an event_file.
-			 *
-			 * For this to work, no event_file should be allocated
-			 * on a odd space, as the ef should always be allocated
-			 * to be at least word aligned. Check for that too.
-			 */
-			WARN_ON_ONCE(ptr & 1);
-
-			ef->dentry->d_fsdata = (void *)(ptr | 1);
-			dentry_list = ef->dentry;
-			ef->dentry = NULL;
-		}
-		call_srcu(&eventfs_srcu, &ef->rcu, free_ef);
+	eventfs_remove_rec(ei, &ei_del_list, 0);
+
+	list_for_each_entry_safe(ei, tmp, &ei_del_list, del_list) {
+		for (i = 0; i < ei->nr_entries; i++)
+			unhook_dentry(&ei->d_children[i], &dentry_list);
+		unhook_dentry(&ei->dentry, &dentry_list);
+		call_srcu(&eventfs_srcu, &ei->rcu, free_ei);
 	}
 	mutex_unlock(&eventfs_mutex);
 

@@ -876,8 +893,8 @@ void eventfs_remove(struct eventfs_file
 		mutex_lock(&eventfs_mutex);
 		/* dentry should now have at least a single reference */
 		WARN_ONCE((int)d_count(dentry) < 1,
-			  "dentry %p less than one reference (%d) after invalidate\n",
-			  dentry, d_count(dentry));
+			  "dentry %px (%s) less than one reference (%d) after invalidate\n",
+			  dentry, dentry->d_name.name, d_count(dentry));
 		mutex_unlock(&eventfs_mutex);
 		dput(dentry);
 	}

@@ -385,7 +385,7 @@ static void tracefs_dentry_iput(struct d
 
 	ti = get_tracefs(inode);
 	if (ti && ti->flags & TRACEFS_EVENT_INODE)
-		eventfs_set_ef_status_free(ti, dentry);
+		eventfs_set_ei_status_free(ti, dentry);
 	iput(inode);
 }
 

@@ -13,6 +13,41 @@ struct tracefs_inode {
 	struct inode            vfs_inode;
 };
 
+/*
+ * struct eventfs_inode - hold the properties of the eventfs directories.
+ * @list:	link list into the parent directory
+ * @entries:	the array of entries representing the files in the directory
+ * @name:	the name of the directory to create
+ * @children:	link list into the child eventfs_inode
+ * @dentry:     the dentry of the directory
+ * @d_parent:   pointer to the parent's dentry
+ * @d_children: The array of dentries to represent the files when created
+ * @data:	The private data to pass to the callbacks
+ * @nr_entries: The number of items in @entries
+ */
+struct eventfs_inode {
+	struct list_head		list;
+	const struct eventfs_entry	*entries;
+	const char			*name;
+	struct list_head		children;
+	struct dentry			*dentry;
+	struct dentry			*d_parent;
+	struct dentry			**d_children;
+	void				*data;
+	/*
+	 * Union - used for deletion
+	 * @del_list:	list of eventfs_inode to delete
+	 * @rcu:	eventfs_indoe to delete in RCU
+	 * @is_freed:	node is freed if one of the above is set
+	 */
+	union {
+		struct list_head	del_list;
+		struct rcu_head		rcu;
+		unsigned long		is_freed;
+	};
+	int				nr_entries;
+};
+
 static inline struct tracefs_inode *get_tracefs(const struct inode *inode)
 {
 	return container_of(inode, struct tracefs_inode, vfs_inode);

@@ -25,6 +60,6 @@ struct inode *tracefs_get_inode(struct s
 struct dentry *eventfs_start_creating(const char *name, struct dentry *parent);
 struct dentry *eventfs_failed_creating(struct dentry *dentry);
 struct dentry *eventfs_end_creating(struct dentry *dentry);
-void eventfs_set_ef_status_free(struct tracefs_inode *ti, struct dentry *dentry);
+void eventfs_set_ei_status_free(struct tracefs_inode *ti, struct dentry *dentry);
 
 #endif /* _TRACEFS_INTERNAL_H */

@@ -652,7 +652,7 @@ struct trace_event_file {
 	struct list_head		list;
 	struct trace_event_call		*event_call;
 	struct event_filter __rcu	*filter;
-	struct eventfs_file             *ef;
+	struct eventfs_inode		*ei;
 	struct trace_array		*tr;
 	struct trace_subsystem_dir	*system;
 	struct list_head		triggers;

@@ -23,26 +23,25 @@ struct file_operations;
 
 struct eventfs_file;
 
-struct dentry *eventfs_create_events_dir(const char *name,
-					 struct dentry *parent);
+typedef int (*eventfs_callback)(const char *name, umode_t *mode, void **data,
+				const struct file_operations **fops);
 
-struct eventfs_file *eventfs_add_subsystem_dir(const char *name,
-					       struct dentry *parent);
+struct eventfs_entry {
+	const char			*name;
+	eventfs_callback		callback;
+};
 
-struct eventfs_file *eventfs_add_dir(const char *name,
-				     struct eventfs_file *ef_parent);
+struct eventfs_inode;
 
-int eventfs_add_file(const char *name, umode_t mode,
-		     struct eventfs_file *ef_parent, void *data,
-		     const struct file_operations *fops);
+struct eventfs_inode *eventfs_create_events_dir(const char *name, struct dentry *parent,
+						const struct eventfs_entry *entries,
+						int size, void *data);
 
-int eventfs_add_events_file(const char *name, umode_t mode,
-			 struct dentry *parent, void *data,
-			 const struct file_operations *fops);
+struct eventfs_inode *eventfs_create_dir(const char *name, struct eventfs_inode *parent,
+					 const struct eventfs_entry *entries,
+					 int size, void *data);
 
-void eventfs_remove(struct eventfs_file *ef);
-
-void eventfs_remove_events_dir(struct dentry *dentry);
+void eventfs_remove_dir(struct eventfs_inode *ei);
 
 struct dentry *tracefs_create_file(const char *name, umode_t mode,
 				   struct dentry *parent, void *data,

@@ -9760,7 +9760,6 @@ static __init void create_trace_instance
 static void
 init_tracer_tracefs(struct trace_array *tr, struct dentry *d_tracer)
 {
-	struct trace_event_file *file;
 	int cpu;
 
 	trace_create_file("available_tracers", TRACE_MODE_READ, d_tracer,

@@ -9793,11 +9792,7 @@ init_tracer_tracefs(struct trace_array *
 	trace_create_file("trace_marker", 0220, d_tracer,
 			  tr, &tracing_mark_fops);
 
-	file = __find_event_file(tr, "ftrace", "print");
-	if (file && file->ef)
-		eventfs_add_file("trigger", TRACE_MODE_WRITE, file->ef,
-				  file, &event_trigger_fops);
-	tr->trace_marker_file = file;
+	tr->trace_marker_file = __find_event_file(tr, "ftrace", "print");
 
 	trace_create_file("trace_marker_raw", 0220, d_tracer,
 			  tr, &tracing_mark_raw_fops);

@@ -381,7 +381,7 @@ struct trace_array {
 	struct dentry		*dir;
 	struct dentry		*options;
 	struct dentry		*percpu_dir;
-	struct dentry		*event_dir;
+	struct eventfs_inode	*event_dir;
 	struct trace_options	*topts;
 	struct list_head	systems;
 	struct list_head	events;

@@ -1345,7 +1345,7 @@ struct trace_subsystem_dir {
 	struct list_head		list;
 	struct event_subsystem		*subsystem;
 	struct trace_array		*tr;
-	struct eventfs_file             *ef;
+	struct eventfs_inode		*ei;
 	int				ref_count;
 	int				nr_events;
 };

424 lines not on this page · raw message

@@ -901,22 +901,17 @@ void eventfs_remove_dir(struct eventfs_i
 }
 
 /**
- * eventfs_remove_events_dir - remove eventfs dir or file from list
- * @dentry: events's dentry to be removed.
+ * eventfs_remove_events_dir - remove the top level eventfs directory
+ * @ei: the event_inode returned by eventfs_create_events_dir().
  *
- * This function remove events main directory
+ * This function removes the events main directory
  */
-void eventfs_remove_events_dir(struct dentry *dentry)
+void eventfs_remove_events_dir(struct eventfs_inode *ei)
 {
-	struct tracefs_inode *ti;
+	struct dentry *dentry = ei->dentry;
 
-	if (!dentry || !dentry->d_inode)
-		return;
+	eventfs_remove_dir(ei);
 
-	ti = get_tracefs(dentry->d_inode);
-	if (!ti || !(ti->flags & TRACEFS_EVENT_INODE))
-		return;
-
-	d_invalidate(dentry);
+	/* Matches the dget() from eventfs_create_events_dir() */
 	dput(dentry);
 }

@@ -41,6 +41,7 @@ struct eventfs_inode *eventfs_create_dir
 					 const struct eventfs_entry *entries,
 					 int size, void *data);
 
+void eventfs_remove_events_dir(struct eventfs_inode *ei);
 void eventfs_remove_dir(struct eventfs_inode *ei);
 
 struct dentry *tracefs_create_file(const char *name, umode_t mode,

@@ -3893,7 +3893,7 @@ int event_trace_del_tracer(struct trace_
 
 	down_write(&trace_event_sem);
 	__trace_remove_event_dirs(tr);
-	eventfs_remove_dir(tr->event_dir);
+	eventfs_remove_events_dir(tr->event_dir);
 	up_write(&trace_event_sem);
 
 	tr->event_dir = NULL;

@@ -731,7 +731,7 @@ struct eventfs_inode *eventfs_create_eve
 		return NULL;
 
 	if (IS_ERR(dentry))
-		return (struct eventfs_inode *)dentry;
+		return ERR_CAST(dentry);
 
 	ei = kzalloc(sizeof(*ei), GFP_KERNEL);
 	if (!ei)

@@ -735,7 +735,7 @@ struct eventfs_inode *eventfs_create_eve
 
 	ei = kzalloc(sizeof(*ei), GFP_KERNEL);
 	if (!ei)
-		goto fail;
+		goto fail_ei;
 
 	inode = tracefs_get_inode(dentry->d_sb);
 	if (unlikely(!inode))

@@ -781,6 +781,7 @@ struct eventfs_inode *eventfs_create_eve
  fail:
 	kfree(ei->d_children);
 	kfree(ei);
+ fail_ei:
 	tracefs_failed_creating(dentry);
 	return ERR_PTR(-ENOMEM);
 }

@@ -856,7 +856,7 @@ static void unhook_dentry(struct dentry
 	}
 }
 /**
- * eventfs_remove - remove eventfs dir or file from list
+ * eventfs_remove_dir - remove eventfs dir or file from list
  * @ei: eventfs_inode to be removed.
  *
  * This function acquire the eventfs_mutex lock and call eventfs_remove_rec()

@@ -264,8 +264,9 @@ create_file_dentry(struct eventfs_inode
 		 * Note, with the mutex held, the e_dentry cannot have content
 		 * and the ei->is_freed be true at the same time.
 		 */
-		WARN_ON_ONCE(ei->is_freed);
 		dentry = *e_dentry;
+		if (WARN_ON_ONCE(dentry && ei->is_freed))
+			dentry = NULL;
 		/* The lookup does not need to up the dentry refcount */
 		if (dentry && !lookup)
 			dget(dentry);

@@ -37,7 +37,7 @@ struct eventfs_inode {
 	/*
 	 * Union - used for deletion
 	 * @del_list:	list of eventfs_inode to delete
-	 * @rcu:	eventfs_indoe to delete in RCU
+	 * @rcu:	eventfs_inode to delete in RCU
 	 * @is_freed:	node is freed if one of the above is set
 	 */
 	union {

@@ -774,9 +774,6 @@ struct eventfs_inode *eventfs_create_eve
 	fsnotify_mkdir(dentry->d_parent->d_inode, dentry);
 	tracefs_end_creating(dentry);
 
-	/* Will call dput when the directory is removed */
-	dget(dentry);
-
 	return ei;
 
  fail:

@@ -796,9 +796,11 @@ static void free_ei(struct rcu_head *hea
 /**
  * eventfs_remove_rec - remove eventfs dir or file from list
  * @ei: eventfs_inode to be removed.
+ * @head: the list head to place the deleted @ei and children
+ * @level: prevent recursion from going more than 3 levels deep.
  *
- * This function recursively remove eventfs_inode which
- * contains info of file or dir.
+ * This function recursively removes eventfs_inodes which
+ * contains info of files and/or directories.
  */
 static void eventfs_remove_rec(struct eventfs_inode *ei, struct list_head *head, int level)
 {

@@ -824,6 +824,8 @@ static void eventfs_remove_rec(struct ev
 		eventfs_remove_rec(ei_child, head, level + 1);
 	}
 
+	ei->is_freed = 1;
+
 	list_del_rcu(&ei->list);
 	list_add_tail(&ei->del_list, head);
 }

@@ -23,6 +23,7 @@ struct tracefs_inode {
  * @d_parent:   pointer to the parent's dentry
  * @d_children: The array of dentries to represent the files when created
  * @data:	The private data to pass to the callbacks
+ * @is_freed:	Flag set if the eventfs is on its way to be freed
  * @nr_entries: The number of items in @entries
  */
 struct eventfs_inode {

@@ -38,14 +39,13 @@ struct eventfs_inode {
 	 * Union - used for deletion
 	 * @del_list:	list of eventfs_inode to delete
 	 * @rcu:	eventfs_inode to delete in RCU
-	 * @is_freed:	node is freed if one of the above is set
 	 */
 	union {
 		struct list_head	del_list;
 		struct rcu_head		rcu;
-		unsigned long		is_freed;
 	};
-	int				nr_entries;
+	unsigned int			is_freed:1;
+	unsigned int			nr_entries:31;
 };
 
 static inline struct tracefs_inode *get_tracefs(const struct inode *inode)

@@ -129,6 +129,13 @@ static struct dentry *create_dir(const c
 	return eventfs_end_creating(dentry);
 }
 
+static void free_ei(struct eventfs_inode *ei)
+{
+	kfree_const(ei->name);
+	kfree(ei->d_children);
+	kfree(ei);
+}
+
 /**
  * eventfs_set_ei_status_free - remove the dentry reference from an eventfs_inode
  * @ti: the tracefs_inode of the dentry

@@ -168,9 +175,7 @@ void eventfs_set_ei_status_free(struct t
 			eventfs_remove_dir(ei_child);
 		}
 
-		kfree_const(ei->name);
-		kfree(ei->d_children);
-		kfree(ei);
+		free_ei(ei);
 		return;
 	}
 

@@ -784,13 +789,11 @@ struct eventfs_inode *eventfs_create_eve
 	return ERR_PTR(-ENOMEM);
 }
 
-static void free_ei(struct rcu_head *head)
+static void free_rcu_ei(struct rcu_head *head)
 {
 	struct eventfs_inode *ei = container_of(head, struct eventfs_inode, rcu);
 
-	kfree_const(ei->name);
-	kfree(ei->d_children);
-	kfree(ei);
+	free_ei(ei);
 }
 
 /**

@@ -881,7 +884,7 @@ void eventfs_remove_dir(struct eventfs_i
 		for (i = 0; i < ei->nr_entries; i++)
 			unhook_dentry(&ei->d_children[i], &dentry_list);
 		unhook_dentry(&ei->dentry, &dentry_list);
-		call_srcu(&eventfs_srcu, &ei->rcu, free_ei);
+		call_srcu(&eventfs_srcu, &ei->rcu, free_rcu_ei);
 	}
 	mutex_unlock(&eventfs_mutex);
 

@@ -24,7 +24,20 @@
 #include <linux/delay.h>
 #include "internal.h"
 
+/*
+ * eventfs_mutex protects the eventfs_inode (ei) dentry. Any access
+ * to the ei->dentry must be done under this mutex and after checking
+ * if ei->is_freed is not set. The ei->dentry is released under the
+ * mutex at the same time ei->is_freed is set. If ei->is_freed is set
+ * then the ei->dentry is invalid.
+ */
 static DEFINE_MUTEX(eventfs_mutex);
+
+/*
+ * The eventfs_inode (ei) itself is protected by SRCU. It is released from
+ * its parent's list and will have is_freed set (under eventfs_mutex).
+ * After the SRCU grace period is over, the ei may be freed.
+ */
 DEFINE_STATIC_SRCU(eventfs_srcu);
 
 static struct dentry *eventfs_root_lookup(struct inode *dir,

@@ -239,6 +252,10 @@ create_file_dentry(struct eventfs_inode
 	bool invalidate = false;
 
 	mutex_lock(&eventfs_mutex);
+	if (ei->is_freed) {
+		mutex_unlock(&eventfs_mutex);
+		return NULL;
+	}
 	/* If the e_dentry already has a dentry, use it */
 	if (*e_dentry) {
 		/* lookup does not need to up the ref count */

@@ -312,6 +329,8 @@ static void eventfs_post_create_dir(stru
 	struct eventfs_inode *ei_child;
 	struct tracefs_inode *ti;
 
+	lockdep_assert_held(&eventfs_mutex);
+
 	/* srcu lock already held */
 	/* fill parent-child relation */
 	list_for_each_entry_srcu(ei_child, &ei->children, list,

@@ -325,6 +344,7 @@ static void eventfs_post_create_dir(stru
 
 /**
  * create_dir_dentry - Create a directory dentry for the eventfs_inode
+ * @pei: The eventfs_inode parent of ei.
  * @ei: The eventfs_inode to create the directory for
  * @parent: The dentry of the parent of this directory
  * @lookup: True if this is called by the lookup code

@@ -332,12 +352,17 @@ static void eventfs_post_create_dir(stru
  * This creates and attaches a directory dentry to the eventfs_inode @ei.
  */
 static struct dentry *
-create_dir_dentry(struct eventfs_inode *ei, struct dentry *parent, bool lookup)
+create_dir_dentry(struct eventfs_inode *pei, struct eventfs_inode *ei,
+		  struct dentry *parent, bool lookup)
 {
 	bool invalidate = false;
 	struct dentry *dentry = NULL;
 
 	mutex_lock(&eventfs_mutex);
+	if (pei->is_freed || ei->is_freed) {
+		mutex_unlock(&eventfs_mutex);
+		return NULL;
+	}
 	if (ei->dentry) {
 		/* If the dentry already has a dentry, use it */
 		dentry = ei->dentry;

@@ -440,7 +465,7 @@ static struct dentry *eventfs_root_looku
 	 */
 	mutex_lock(&eventfs_mutex);
 	ei = READ_ONCE(ti->private);
-	if (ei)
+	if (ei && !ei->is_freed)
 		ei_dentry = READ_ONCE(ei->dentry);
 	mutex_unlock(&eventfs_mutex);
 

@@ -454,7 +479,7 @@ static struct dentry *eventfs_root_looku
 		if (strcmp(ei_child->name, name) != 0)
 			continue;
 		ret = simple_lookup(dir, dentry, flags);
-		create_dir_dentry(ei_child, ei_dentry, true);
+		create_dir_dentry(ei, ei_child, ei_dentry, true);
 		created = true;
 		break;
 	}

@@ -588,7 +613,7 @@ static int dcache_dir_open_wrapper(struc
 
 	list_for_each_entry_srcu(ei_child, &ei->children, list,
 				 srcu_read_lock_held(&eventfs_srcu)) {
-		d = create_dir_dentry(ei_child, parent, false);
+		d = create_dir_dentry(ei, ei_child, parent, false);
 		if (d) {
 			ret = add_dentries(&dentries, d, cnt);
 			if (ret < 0)

@@ -705,12 +730,20 @@ struct eventfs_inode *eventfs_create_dir
 	ei->nr_entries = size;
 	ei->data = data;
 	INIT_LIST_HEAD(&ei->children);
+	INIT_LIST_HEAD(&ei->list);
 
 	mutex_lock(&eventfs_mutex);
-	list_add_tail(&ei->list, &parent->children);
-	ei->d_parent = parent->dentry;
+	if (!parent->is_freed) {
+		list_add_tail(&ei->list, &parent->children);
+		ei->d_parent = parent->dentry;
+	}
 	mutex_unlock(&eventfs_mutex);
 
+	/* Was the parent freed? */
+	if (list_empty(&ei->list)) {
+		free_ei(ei);
+		ei = NULL;
+	}
 	return ei;
 }
 

@@ -24,6 +24,7 @@ struct tracefs_inode {
  * @d_children: The array of dentries to represent the files when created
  * @data:	The private data to pass to the callbacks
  * @is_freed:	Flag set if the eventfs is on its way to be freed
+ *                Note if is_freed is set, then dentry is corrupted.
  * @nr_entries: The number of items in @entries
  */
 struct eventfs_inode {

@@ -31,7 +32,7 @@ struct eventfs_inode {
 	const struct eventfs_entry	*entries;
 	const char			*name;
 	struct list_head		children;
-	struct dentry			*dentry;
+	struct dentry			*dentry; /* Check is_freed to access */
 	struct dentry			*d_parent;
 	struct dentry			**d_children;
 	void				*data;

@@ -40,6 +40,15 @@ static DEFINE_MUTEX(eventfs_mutex);
  */
 DEFINE_STATIC_SRCU(eventfs_srcu);
 
+/* Mode is unsigned short, use the upper bits for flags */
+enum {
+	EVENTFS_SAVE_MODE	= BIT(16),
+	EVENTFS_SAVE_UID	= BIT(17),
+	EVENTFS_SAVE_GID	= BIT(18),
+};
+
+#define EVENTFS_MODE_MASK	(EVENTFS_SAVE_MODE - 1)
+
 static struct dentry *eventfs_root_lookup(struct inode *dir,
 					  struct dentry *dentry,
 					  unsigned int flags);

@@ -47,8 +56,89 @@ static int dcache_dir_open_wrapper(struc
 static int dcache_readdir_wrapper(struct file *file, struct dir_context *ctx);
 static int eventfs_release(struct inode *inode, struct file *file);
 
+static void update_attr(struct eventfs_attr *attr, struct iattr *iattr)
+{
+	unsigned int ia_valid = iattr->ia_valid;
+
+	if (ia_valid & ATTR_MODE) {
+		attr->mode = (attr->mode & ~EVENTFS_MODE_MASK) |
+			(iattr->ia_mode & EVENTFS_MODE_MASK) |
+			EVENTFS_SAVE_MODE;
+	}
+	if (ia_valid & ATTR_UID) {
+		attr->mode |= EVENTFS_SAVE_UID;
+		attr->uid = iattr->ia_uid;
+	}
+	if (ia_valid & ATTR_GID) {
+		attr->mode |= EVENTFS_SAVE_GID;
+		attr->gid = iattr->ia_gid;
+	}
+}
+
+static int eventfs_set_attr(struct mnt_idmap *idmap, struct dentry *dentry,
+			    struct iattr *iattr)
+{
+	const struct eventfs_entry *entry;
+	struct eventfs_inode *ei;
+	const char *name;
+	int ret;
+
+	mutex_lock(&eventfs_mutex);
+	ei = dentry->d_fsdata;
+	/* The LSB is set when the eventfs_inode is being freed */
+	if (((unsigned long)ei & 1UL) || ei->is_freed) {
+		/* Do not allow changes if the event is about to be removed. */
+		mutex_unlock(&eventfs_mutex);
+		return -ENODEV;
+	}
+
+	/* Preallocate the children mode array if necessary */
+	if (!(dentry->d_inode->i_mode & S_IFDIR)) {
+		if (!ei->entry_attrs) {
+			ei->entry_attrs = kzalloc(sizeof(*ei->entry_attrs) * ei->nr_entries,
+						  GFP_KERNEL);
+			if (!ei->entry_attrs) {
+				ret = -ENOMEM;
+				goto out;
+			}
+		}
+	}
+
+	ret = simple_setattr(idmap, dentry, iattr);
+	if (ret < 0)
+		goto out;
+
+	/*
+	 * If this is a dir, then update the ei cache, only the file
+	 * mode is saved in the ei->m_children, and the ownership is
+	 * determined by the parent directory.
+	 */
+	if (dentry->d_inode->i_mode & S_IFDIR) {
+		update_attr(&ei->attr, iattr);
+
+	} else {
+		name = dentry->d_name.name;
+
+		for (int i = 0; i < ei->nr_entries; i++) {
+			entry = &ei->entries[i];
+			if (strcmp(name, entry->name) == 0) {
+				update_attr(&ei->entry_attrs[i], iattr);
+				break;
+			}
+		}
+	}
+ out:
+	mutex_unlock(&eventfs_mutex);
+	return ret;
+}
+
 static const struct inode_operations eventfs_root_dir_inode_operations = {
 	.lookup		= eventfs_root_lookup,
+	.setattr	= eventfs_set_attr,
+};
+
+static const struct inode_operations eventfs_file_inode_operations = {
+	.setattr	= eventfs_set_attr,
 };
 
 static const struct file_operations eventfs_file_operations = {

@@ -59,10 +149,30 @@ static const struct file_operations even
 	.release	= eventfs_release,
 };
 
+static void update_inode_attr(struct inode *inode, struct eventfs_attr *attr, umode_t mode)
+{
+	if (!attr) {
+		inode->i_mode = mode;
+		return;
+	}
+
+	if (attr->mode & EVENTFS_SAVE_MODE)
+		inode->i_mode = attr->mode & EVENTFS_MODE_MASK;
+	else
+		inode->i_mode = mode;
+
+	if (attr->mode & EVENTFS_SAVE_UID)
+		inode->i_uid = attr->uid;
+
+	if (attr->mode & EVENTFS_SAVE_GID)
+		inode->i_gid = attr->gid;
+}
+
 /**
  * create_file - create a file in the tracefs filesystem
  * @name: the name of the file to create.
  * @mode: the permission that the file should have.
+ * @attr: saved attributes changed by user
  * @parent: parent dentry for this file.
  * @data: something that the caller will want to get to later on.
  * @fop: struct file_operations that should be used for this file.

@@ -72,6 +182,7 @@ static const struct file_operations even
  * call.
  */
 static struct dentry *create_file(const char *name, umode_t mode,
+				  struct eventfs_attr *attr,
 				  struct dentry *parent, void *data,
 				  const struct file_operations *fop)
 {

@@ -95,7 +206,10 @@ static struct dentry *create_file(const
 	if (unlikely(!inode))
 		return eventfs_failed_creating(dentry);
 
-	inode->i_mode = mode;
+	/* If the user updated the directory's attributes, use them */
+	update_inode_attr(inode, attr, mode);
+
+	inode->i_op = &eventfs_file_inode_operations;
 	inode->i_fop = fop;
 	inode->i_private = data;
 

@@ -108,19 +222,19 @@ static struct dentry *create_file(const
 
 /**
  * create_dir - create a dir in the tracefs filesystem
- * @name: the name of the file to create.
+ * @ei: the eventfs_inode that represents the directory to create
  * @parent: parent dentry for this file.
  *
  * This function will create a dentry for a directory represented by
  * a eventfs_inode.
  */
-static struct dentry *create_dir(const char *name, struct dentry *parent)
+static struct dentry *create_dir(struct eventfs_inode *ei, struct dentry *parent)
 {
 	struct tracefs_inode *ti;
 	struct dentry *dentry;
 	struct inode *inode;
 
-	dentry = eventfs_start_creating(name, parent);
+	dentry = eventfs_start_creating(ei->name, parent);
 	if (IS_ERR(dentry))
 		return dentry;
 

@@ -128,7 +242,9 @@ static struct dentry *create_dir(const c
 	if (unlikely(!inode))
 		return eventfs_failed_creating(dentry);
 
-	inode->i_mode = S_IFDIR | S_IRWXU | S_IRUGO | S_IXUGO;
+	/* If the user updated the directory's attributes, use them */
+	update_inode_attr(inode, &ei->attr, S_IFDIR | S_IRWXU | S_IRUGO | S_IXUGO);
+
 	inode->i_op = &eventfs_root_dir_inode_operations;
 	inode->i_fop = &eventfs_file_operations;
 

@@ -146,6 +262,7 @@ static void free_ei(struct eventfs_inode
 {
 	kfree_const(ei->name);
 	kfree(ei->d_children);
+	kfree(ei->entry_attrs);
 	kfree(ei);
 }
 

@@ -231,7 +348,7 @@ void eventfs_set_ei_status_free(struct t
 /**
  * create_file_dentry - create a dentry for a file of an eventfs_inode
  * @ei: the eventfs_inode that the file will be created under
- * @e_dentry: a pointer to the d_children[] of the @ei
+ * @idx: the index into the d_children[] of the @ei
  * @parent: The parent dentry of the created file.
  * @name: The name of the file to create
  * @mode: The mode of the file.

@@ -244,10 +361,12 @@ void eventfs_set_ei_status_free(struct t
  * just do a dget() on it and return. Otherwise create the dentry and attach it.
  */
 static struct dentry *
-create_file_dentry(struct eventfs_inode *ei, struct dentry **e_dentry,
+create_file_dentry(struct eventfs_inode *ei, int idx,
 		   struct dentry *parent, const char *name, umode_t mode, void *data,
 		   const struct file_operations *fops, bool lookup)
 {
+	struct eventfs_attr *attr = NULL;
+	struct dentry **e_dentry = &ei->d_children[idx];
 	struct dentry *dentry;
 	bool invalidate = false;
 

@@ -264,13 +383,18 @@ create_file_dentry(struct eventfs_inode
 		mutex_unlock(&eventfs_mutex);
 		return *e_dentry;
 	}
+
+	/* ei->entry_attrs are protected by SRCU */
+	if (ei->entry_attrs)
+		attr = &ei->entry_attrs[idx];
+
 	mutex_unlock(&eventfs_mutex);
 
 	/* The lookup already has the parent->d_inode locked */
 	if (!lookup)
 		inode_lock(parent->d_inode);
 
-	dentry = create_file(name, mode, parent, data, fops);
+	dentry = create_file(name, mode, attr, parent, data, fops);
 
 	if (!lookup)
 		inode_unlock(parent->d_inode);

@@ -378,7 +502,7 @@ create_dir_dentry(struct eventfs_inode *
 	if (!lookup)
 		inode_lock(parent->d_inode);
 
-	dentry = create_dir(ei->name, parent);
+	dentry = create_dir(ei, parent);
 
 	if (!lookup)
 		inode_unlock(parent->d_inode);

@@ -495,8 +619,7 @@ static struct dentry *eventfs_root_looku
 			if (r <= 0)
 				continue;
 			ret = simple_lookup(dir, dentry, flags);
-			create_file_dentry(ei, &ei->d_children[i],
-					   ei_dentry, name, mode, cdata,
+			create_file_dentry(ei, i, ei_dentry, name, mode, cdata,
 					   fops, true);
 			break;
 		}

@@ -629,8 +752,7 @@ static int dcache_dir_open_wrapper(struc
 		r = entry->callback(name, &mode, &cdata, &fops);
 		if (r <= 0)
 			continue;
-		d = create_file_dentry(ei, &ei->d_children[i],
-				       parent, name, mode, cdata, fops, false);
+		d = create_file_dentry(ei, i, parent, name, mode, cdata, fops, false);
 		if (d) {
 			ret = add_dentries(&dentries, d, cnt);
 			if (ret < 0)

@@ -14,6 +14,18 @@ struct tracefs_inode {
 };
 
 /*
+ * struct eventfs_attr - cache the mode and ownership of a eventfs entry
+ * @mode:	saved mode plus flags of what is saved
+ * @uid:	saved uid if changed
+ * @gid:	saved gid if changed
+ */
+struct eventfs_attr {
+	int				mode;
+	kuid_t				uid;
+	kgid_t				gid;
+};
+
+/*
  * struct eventfs_inode - hold the properties of the eventfs directories.
  * @list:	link list into the parent directory
  * @entries:	the array of entries representing the files in the directory

@@ -22,6 +34,8 @@ struct tracefs_inode {
  * @dentry:     the dentry of the directory
  * @d_parent:   pointer to the parent's dentry
  * @d_children: The array of dentries to represent the files when created
+ * @entry_attrs: Saved mode and ownership of the @d_children
+ * @attr:	Saved mode and ownership of eventfs_inode itself
  * @data:	The private data to pass to the callbacks
  * @is_freed:	Flag set if the eventfs is on its way to be freed
  *                Note if is_freed is set, then dentry is corrupted.

@@ -35,6 +49,8 @@ struct eventfs_inode {
 	struct dentry			*dentry; /* Check is_freed to access */
 	struct dentry			*d_parent;
 	struct dentry			**d_children;
+	struct eventfs_attr		*entry_attrs;
+	struct eventfs_attr		attr;
 	void				*data;
 	/*
 	 * Union - used for deletion

@@ -615,7 +615,13 @@ static struct dentry *eventfs_root_looku
 		entry = &ei->entries[i];
 		if (strcmp(name, entry->name) == 0) {
 			void *cdata = data;
-			r = entry->callback(name, &mode, &cdata, &fops);
+			mutex_lock(&eventfs_mutex);
+			/* If ei->is_freed, then the event itself may be too */
+			if (!ei->is_freed)
+				r = entry->callback(name, &mode, &cdata, &fops);
+			else
+				r = -1;
+			mutex_unlock(&eventfs_mutex);
 			if (r <= 0)
 				continue;
 			ret = simple_lookup(dir, dentry, flags);

@@ -749,7 +755,13 @@ static int dcache_dir_open_wrapper(struc
 		void *cdata = data;
 		entry = &ei->entries[i];
 		name = entry->name;
-		r = entry->callback(name, &mode, &cdata, &fops);
+		mutex_lock(&eventfs_mutex);
+		/* If ei->is_freed, then the event itself may be too */
+		if (!ei->is_freed)
+			r = entry->callback(name, &mode, &cdata, &fops);
+		else
+			r = -1;
+		mutex_unlock(&eventfs_mutex);
 		if (r <= 0)
 			continue;
 		d = create_file_dentry(ei, i, parent, name, mode, cdata, fops, false);

@@ -819,6 +831,10 @@ static int dcache_readdir_wrapper(struct
  *   data = A pointer to @data, and the callback may replace it, which will
  *         cause the file created to pass the new data to the open() call.
  *   fops = the fops to use for the created file.
+ *
+ * NB. @callback is called while holding internal locks of the eventfs
+ *     system. The callback must not call any code that might also call into
+ *     the tracefs or eventfs system or it will risk creating a deadlock.
  */
 struct eventfs_inode *eventfs_create_dir(const char *name, struct eventfs_inode *parent,
 					 const struct eventfs_entry *entries,

@@ -878,6 +894,8 @@ struct eventfs_inode *eventfs_create_dir
  * @data: The default data to pass to the files (an entry may override it).
  *
  * This function creates the top of the trace event directory.
+ *
+ * See eventfs_create_dir() for use of @entries.
  */
 struct eventfs_inode *eventfs_create_events_dir(const char *name, struct dentry *parent,
 						const struct eventfs_entry *entries,

52 lines not on this page · raw message

@@ -85,8 +85,7 @@ static int eventfs_set_attr(struct mnt_i
 
 	mutex_lock(&eventfs_mutex);
 	ei = dentry->d_fsdata;
-	/* The LSB is set when the eventfs_inode is being freed */
-	if (((unsigned long)ei & 1UL) || ei->is_freed) {
+	if (ei->is_freed) {
 		/* Do not allow changes if the event is about to be removed. */
 		mutex_unlock(&eventfs_mutex);
 		return -ENODEV;

@@ -276,35 +275,17 @@ static void free_ei(struct eventfs_inode
 void eventfs_set_ei_status_free(struct tracefs_inode *ti, struct dentry *dentry)
 {
 	struct tracefs_inode *ti_parent;
-	struct eventfs_inode *ei_child, *tmp;
 	struct eventfs_inode *ei;
 	int i;
 
 	/* The top level events directory may be freed by this */
 	if (unlikely(ti->flags & TRACEFS_EVENT_TOP_INODE)) {
-		LIST_HEAD(ef_del_list);
-
 		mutex_lock(&eventfs_mutex);
-
 		ei = ti->private;
-
-		/* Record all the top level files */
-		list_for_each_entry_srcu(ei_child, &ei->children, list,
-					 lockdep_is_held(&eventfs_mutex)) {
-			list_add_tail(&ei_child->del_list, &ef_del_list);
-		}
-
 		/* Nothing should access this, but just in case! */
 		ti->private = NULL;
-
 		mutex_unlock(&eventfs_mutex);
 
-		/* Now safely free the top level files and their children */
-		list_for_each_entry_safe(ei_child, tmp, &ef_del_list, del_list) {
-			list_del(&ei_child->del_list);
-			eventfs_remove_dir(ei_child);
-		}
-
 		free_ei(ei);
 		return;
 	}

@@ -319,14 +300,6 @@ void eventfs_set_ei_status_free(struct t
 	if (!ei)
 		goto out;
 
-	/*
-	 * If ei was freed, then the LSB bit is set for d_fsdata.
-	 * But this should not happen, as it should still have a
-	 * ref count that prevents it. Warn in case it does.
-	 */
-	if (WARN_ON_ONCE((unsigned long)ei & 1))
-		goto out;
-
 	/* This could belong to one of the files of the ei */
 	if (ei->dentry != dentry) {
 		for (i = 0; i < ei->nr_entries; i++) {

@@ -336,6 +309,8 @@ void eventfs_set_ei_status_free(struct t
 		if (WARN_ON_ONCE(i == ei->nr_entries))
 			goto out;
 		ei->d_children[i] = NULL;
+	} else if (ei->is_freed) {
+		free_ei(ei);
 	} else {
 		ei->dentry = NULL;
 	}

@@ -962,13 +937,65 @@ struct eventfs_inode *eventfs_create_eve
 	return ERR_PTR(-ENOMEM);
 }
 
+static LLIST_HEAD(free_list);
+
+static void eventfs_workfn(struct work_struct *work)
+{
+        struct eventfs_inode *ei, *tmp;
+        struct llist_node *llnode;
+
+	llnode = llist_del_all(&free_list);
+        llist_for_each_entry_safe(ei, tmp, llnode, llist) {
+		/* This dput() matches the dget() from unhook_dentry() */
+		for (int i = 0; i < ei->nr_entries; i++) {
+			if (ei->d_children[i])
+				dput(ei->d_children[i]);
+		}
+		/* This should only get here if it had a dentry */
+		if (!WARN_ON_ONCE(!ei->dentry))
+			dput(ei->dentry);
+        }
+}
+
+static DECLARE_WORK(eventfs_work, eventfs_workfn);
+
 static void free_rcu_ei(struct rcu_head *head)
 {
 	struct eventfs_inode *ei = container_of(head, struct eventfs_inode, rcu);
 
+	if (ei->dentry) {
+		/* Do not free the ei until all references of dentry are gone */
+		if (llist_add(&ei->llist, &free_list))
+			queue_work(system_unbound_wq, &eventfs_work);
+		return;
+	}
+
+	/* If the ei doesn't have a dentry, neither should its children */
+	for (int i = 0; i < ei->nr_entries; i++) {
+		WARN_ON_ONCE(ei->d_children[i]);
+	}
+
 	free_ei(ei);
 }
 
+static void unhook_dentry(struct dentry *dentry)
+{
+	if (!dentry)
+		return;
+
+	/* Keep the dentry from being freed yet (see eventfs_workfn()) */
+	dget(dentry);
+
+	dentry->d_fsdata = NULL;
+	d_invalidate(dentry);
+	mutex_lock(&eventfs_mutex);
+	/* dentry should now have at least a single reference */
+	WARN_ONCE((int)d_count(dentry) < 1,
+		  "dentry %px (%s) less than one reference (%d) after invalidate\n",
+		  dentry, dentry->d_name.name, d_count(dentry));
+	mutex_unlock(&eventfs_mutex);
+}
+
 /**
  * eventfs_remove_rec - remove eventfs dir or file from list
  * @ei: eventfs_inode to be removed.

@@ -1006,33 +1033,6 @@ static void eventfs_remove_rec(struct ev
 	list_add_tail(&ei->del_list, head);
 }
 
-static void unhook_dentry(struct dentry **dentry, struct dentry **list)
-{
-	if (*dentry) {
-		unsigned long ptr = (unsigned long)*list;
-
-		/* Keep the dentry from being freed yet */
-		dget(*dentry);
-
-		/*
-		 * Paranoid: The dget() above should prevent the dentry
-		 * from being freed and calling eventfs_set_ei_status_free().
-		 * But just in case, set the link list LSB pointer to 1
-		 * and have eventfs_set_ei_status_free() check that to
-		 * make sure that if it does happen, it will not think
-		 * the d_fsdata is an eventfs_inode.
-		 *
-		 * For this to work, no eventfs_inode should be allocated
-		 * on a odd space, as the ef should always be allocated
-		 * to be at least word aligned. Check for that too.
-		 */
-		WARN_ON_ONCE(ptr & 1);
-
-		(*dentry)->d_fsdata = (void *)(ptr | 1);
-		*list = *dentry;
-		*dentry = NULL;
-	}
-}
 /**
  * eventfs_remove_dir - remove eventfs dir or file from list
  * @ei: eventfs_inode to be removed.

@@ -1043,40 +1043,28 @@ void eventfs_remove_dir(struct eventfs_i
 {
 	struct eventfs_inode *tmp;
 	LIST_HEAD(ei_del_list);
-	struct dentry *dentry_list = NULL;
-	struct dentry *dentry;
-	int i;
 
 	if (!ei)
 		return;
 
+	/*
+	 * Move the deleted eventfs_inodes onto the ei_del_list
+	 * which will also set the is_freed value. Note, this has to be
+	 * done under the eventfs_mutex, but the deletions of
+	 * the dentries must be done outside the eventfs_mutex.
+	 * Hence moving them to this temporary list.
+	 */
 	mutex_lock(&eventfs_mutex);
 	eventfs_remove_rec(ei, &ei_del_list, 0);
+	mutex_unlock(&eventfs_mutex);
 
 	list_for_each_entry_safe(ei, tmp, &ei_del_list, del_list) {
-		for (i = 0; i < ei->nr_entries; i++)
-			unhook_dentry(&ei->d_children[i], &dentry_list);
-		unhook_dentry(&ei->dentry, &dentry_list);
+		for (int i = 0; i < ei->nr_entries; i++)
+			unhook_dentry(ei->d_children[i]);
+		unhook_dentry(ei->dentry);
+		list_del(&ei->del_list);
 		call_srcu(&eventfs_srcu, &ei->rcu, free_rcu_ei);
 	}
-	mutex_unlock(&eventfs_mutex);
-
-	while (dentry_list) {
-		unsigned long ptr;
-
-		dentry = dentry_list;
-		ptr = (unsigned long)dentry->d_fsdata & ~1UL;
-		dentry_list = (struct dentry *)ptr;
-		dentry->d_fsdata = NULL;
-		d_invalidate(dentry);
-		mutex_lock(&eventfs_mutex);
-		/* dentry should now have at least a single reference */
-		WARN_ONCE((int)d_count(dentry) < 1,
-			  "dentry %px (%s) less than one reference (%d) after invalidate\n",
-			  dentry, dentry->d_name.name, d_count(dentry));
-		mutex_unlock(&eventfs_mutex);
-		dput(dentry);
-	}
 }
 
 /**

@@ -54,10 +54,12 @@ struct eventfs_inode {
 	void				*data;
 	/*
 	 * Union - used for deletion
+	 * @llist:	for calling dput() if needed after RCU
 	 * @del_list:	list of eventfs_inode to delete
 	 * @rcu:	eventfs_inode to delete in RCU
 	 */
 	union {
+		struct llist_node	llist;
 		struct list_head	del_list;
 		struct rcu_head		rcu;
 	};

@@ -274,28 +274,11 @@ static void free_ei(struct eventfs_inode
  */
 void eventfs_set_ei_status_free(struct tracefs_inode *ti, struct dentry *dentry)
 {
-	struct tracefs_inode *ti_parent;
 	struct eventfs_inode *ei;
 	int i;
 
-	/* The top level events directory may be freed by this */
-	if (unlikely(ti->flags & TRACEFS_EVENT_TOP_INODE)) {
-		mutex_lock(&eventfs_mutex);
-		ei = ti->private;
-		/* Nothing should access this, but just in case! */
-		ti->private = NULL;
-		mutex_unlock(&eventfs_mutex);
-
-		free_ei(ei);
-		return;
-	}
-
 	mutex_lock(&eventfs_mutex);
 
-	ti_parent = get_tracefs(dentry->d_parent->d_inode);
-	if (!ti_parent || !(ti_parent->flags & TRACEFS_EVENT_INODE))
-		goto out;
-
 	ei = dentry->d_fsdata;
 	if (!ei)
 		goto out;

@@ -920,6 +903,8 @@ struct eventfs_inode *eventfs_create_eve
 	inode->i_op = &eventfs_root_dir_inode_operations;
 	inode->i_fop = &eventfs_file_operations;
 
+	dentry->d_fsdata = ei;
+
 	/* directory inodes start off with i_nlink == 2 (for "." entry) */
 	inc_nlink(inode);
 	d_instantiate(dentry, inode);

@@ -967,30 +967,29 @@ static void unhook_dentry(struct dentry
 {
 	if (!dentry)
 		return;
-
-	/* Keep the dentry from being freed yet (see eventfs_workfn()) */
+	/*
+	 * Need to add a reference to the dentry that is expected by
+	 * simple_recursive_removal(), which will include a dput().
+	 */
 	dget(dentry);
 
-	dentry->d_fsdata = NULL;
-	d_invalidate(dentry);
-	mutex_lock(&eventfs_mutex);
-	/* dentry should now have at least a single reference */
-	WARN_ONCE((int)d_count(dentry) < 1,
-		  "dentry %px (%s) less than one reference (%d) after invalidate\n",
-		  dentry, dentry->d_name.name, d_count(dentry));
-	mutex_unlock(&eventfs_mutex);
+	/*
+	 * Also add a reference for the dput() in eventfs_workfn().
+	 * That is required as that dput() will free the ei after
+	 * the SRCU grace period is over.
+	 */
+	dget(dentry);
 }
 
 /**
  * eventfs_remove_rec - remove eventfs dir or file from list
  * @ei: eventfs_inode to be removed.
- * @head: the list head to place the deleted @ei and children
  * @level: prevent recursion from going more than 3 levels deep.
  *
  * This function recursively removes eventfs_inodes which
  * contains info of files and/or directories.
  */
-static void eventfs_remove_rec(struct eventfs_inode *ei, struct list_head *head, int level)
+static void eventfs_remove_rec(struct eventfs_inode *ei, int level)
 {
 	struct eventfs_inode *ei_child;
 

@@ -1009,13 +1008,26 @@ static void eventfs_remove_rec(struct ev
 	/* search for nested folders or files */
 	list_for_each_entry_srcu(ei_child, &ei->children, list,
 				 lockdep_is_held(&eventfs_mutex)) {
-		eventfs_remove_rec(ei_child, head, level + 1);
+		/* Children only have dentry if parent does */
+		WARN_ON_ONCE(ei_child->dentry && !ei->dentry);
+		eventfs_remove_rec(ei_child, level + 1);
 	}
 
+
 	ei->is_freed = 1;
 
+	for (int i = 0; i < ei->nr_entries; i++) {
+		if (ei->d_children[i]) {
+			/* Children only have dentry if parent does */
+			WARN_ON_ONCE(!ei->dentry);
+			unhook_dentry(ei->d_children[i]);
+		}
+	}
+
+	unhook_dentry(ei->dentry);
+
 	list_del_rcu(&ei->list);
-	list_add_tail(&ei->del_list, head);
+	call_srcu(&eventfs_srcu, &ei->rcu, free_rcu_ei);
 }
 
 /**

@@ -1026,30 +1038,22 @@ static void eventfs_remove_rec(struct ev
  */
 void eventfs_remove_dir(struct eventfs_inode *ei)
 {
-	struct eventfs_inode *tmp;
-	LIST_HEAD(ei_del_list);
+	struct dentry *dentry;
 
 	if (!ei)
 		return;
 
-	/*
-	 * Move the deleted eventfs_inodes onto the ei_del_list
-	 * which will also set the is_freed value. Note, this has to be
-	 * done under the eventfs_mutex, but the deletions of
-	 * the dentries must be done outside the eventfs_mutex.
-	 * Hence moving them to this temporary list.
-	 */
 	mutex_lock(&eventfs_mutex);
-	eventfs_remove_rec(ei, &ei_del_list, 0);
+	dentry = ei->dentry;
+	eventfs_remove_rec(ei, 0);
 	mutex_unlock(&eventfs_mutex);
 
-	list_for_each_entry_safe(ei, tmp, &ei_del_list, del_list) {
-		for (int i = 0; i < ei->nr_entries; i++)
-			unhook_dentry(ei->d_children[i]);
-		unhook_dentry(ei->dentry);
-		list_del(&ei->del_list);
-		call_srcu(&eventfs_srcu, &ei->rcu, free_rcu_ei);
-	}
+	/*
+	 * If any of the ei children has a dentry, then the ei itself
+	 * must have a dentry.
+	 */
+	if (dentry)
+		simple_recursive_removal(dentry, NULL);
 }
 
 /**

@@ -1060,10 +1064,17 @@ void eventfs_remove_dir(struct eventfs_i
  */
 void eventfs_remove_events_dir(struct eventfs_inode *ei)
 {
-	struct dentry *dentry = ei->dentry;
+	struct dentry *dentry;
 
+	dentry = ei->dentry;
 	eventfs_remove_dir(ei);
 
-	/* Matches the dget() from eventfs_create_events_dir() */
+	/*
+	 * Matches the dget() done by tracefs_start_creating()
+	 * in eventfs_create_events_dir() when it the dentry was
+	 * created. In other words, it's a normal dentry that
+	 * sticks around while the other ei->dentry are created
+	 * and destroyed dynamically.
+	 */
 	dput(dentry);
 }

@@ -55,12 +55,10 @@ struct eventfs_inode {
 	/*
 	 * Union - used for deletion
 	 * @llist:	for calling dput() if needed after RCU
-	 * @del_list:	list of eventfs_inode to delete
 	 * @rcu:	eventfs_inode to delete in RCU
 	 */
 	union {
 		struct llist_node	llist;
-		struct list_head	del_list;
 		struct rcu_head		rcu;
 	};
 	unsigned int			is_freed:1;

@@ -27,16 +27,16 @@
 /*
  * eventfs_mutex protects the eventfs_inode (ei) dentry. Any access
  * to the ei->dentry must be done under this mutex and after checking
- * if ei->is_freed is not set. The ei->dentry is released under the
- * mutex at the same time ei->is_freed is set. If ei->is_freed is set
- * then the ei->dentry is invalid.
+ * if ei->is_freed is not set. When ei->is_freed is set, the dentry
+ * is on its way to being freed after the last dput() is made on it.
  */
 static DEFINE_MUTEX(eventfs_mutex);
 
 /*
  * The eventfs_inode (ei) itself is protected by SRCU. It is released from
  * its parent's list and will have is_freed set (under eventfs_mutex).
- * After the SRCU grace period is over, the ei may be freed.
+ * After the SRCU grace period is over and the last dput() is called
+ * the ei is freed.
  */
 DEFINE_STATIC_SRCU(eventfs_srcu);
 

@@ -365,12 +365,14 @@ create_file_dentry(struct eventfs_inode
 		 * created the dentry for this e_dentry. In which case
 		 * use that one.
 		 *
-		 * Note, with the mutex held, the e_dentry cannot have content
-		 * and the ei->is_freed be true at the same time.
+		 * If ei->is_freed is set, the e_dentry is currently on its
+		 * way to being freed, don't return it. If e_dentry is NULL
+		 * it means it was already freed.
 		 */
-		dentry = *e_dentry;
-		if (WARN_ON_ONCE(dentry && ei->is_freed))
+		if (ei->is_freed)
 			dentry = NULL;
+		else
+			dentry = *e_dentry;
 		/* The lookup does not need to up the dentry refcount */
 		if (dentry && !lookup)
 			dget(dentry);

@@ -473,8 +475,8 @@ create_dir_dentry(struct eventfs_inode *
 		 * created the dentry for this e_dentry. In which case
 		 * use that one.
 		 *
-		 * Note, with the mutex held, the e_dentry cannot have content
-		 * and the ei->is_freed be true at the same time.
+		 * If ei->is_freed is set, the e_dentry is currently on its
+		 * way to being freed.
 		 */
 		dentry = ei->dentry;
 		if (dentry && !lookup)

@@ -326,7 +326,6 @@ create_file_dentry(struct eventfs_inode
 	struct eventfs_attr *attr = NULL;
 	struct dentry **e_dentry = &ei->d_children[idx];
 	struct dentry *dentry;
-	bool invalidate = false;
 
 	mutex_lock(&eventfs_mutex);
 	if (ei->is_freed) {

@@ -389,17 +388,14 @@ create_file_dentry(struct eventfs_inode
 		 * Otherwise it means two dentries exist with the same name.
 		 */
 		WARN_ON_ONCE(!ei->is_freed);
-		invalidate = true;
+		dentry = NULL;
 	}
 	mutex_unlock(&eventfs_mutex);
 
-	if (invalidate)
-		d_invalidate(dentry);
-
-	if (lookup || invalidate)
+	if (lookup)
 		dput(dentry);
 
-	return invalidate ? NULL : dentry;
+	return dentry;
 }
 
 /**

@@ -439,7 +435,6 @@ static struct dentry *
 create_dir_dentry(struct eventfs_inode *pei, struct eventfs_inode *ei,
 		  struct dentry *parent, bool lookup)
 {
-	bool invalidate = false;
 	struct dentry *dentry = NULL;
 
 	mutex_lock(&eventfs_mutex);

@@ -495,16 +490,14 @@ create_dir_dentry(struct eventfs_inode *
 		 * Otherwise it means two dentries exist with the same name.
 		 */
 		WARN_ON_ONCE(!ei->is_freed);
-		invalidate = true;
+		dentry = NULL;
 	}
 	mutex_unlock(&eventfs_mutex);
-	if (invalidate)
-		d_invalidate(dentry);
 
-	if (lookup || invalidate)
+	if (lookup)
 		dput(dentry);
 
-	return invalidate ? NULL : dentry;
+	return dentry;
 }
 
 /**

@@ -95,7 +95,7 @@ static int eventfs_set_attr(struct mnt_i
 	if (!(dentry->d_inode->i_mode & S_IFDIR)) {
 		if (!ei->entry_attrs) {
 			ei->entry_attrs = kzalloc(sizeof(*ei->entry_attrs) * ei->nr_entries,
-						  GFP_KERNEL);
+						  GFP_NOFS);
 			if (!ei->entry_attrs) {
 				ret = -ENOMEM;
 				goto out;

@@ -627,7 +627,7 @@ static int add_dentries(struct dentry **
 {
 	struct dentry **tmp;
 
-	tmp = krealloc(*dentries, sizeof(d) * (cnt + 2), GFP_KERNEL);
+	tmp = krealloc(*dentries, sizeof(d) * (cnt + 2), GFP_NOFS);
 	if (!tmp)
 		return -1;
 	tmp[cnt] = d;

@@ -347,15 +347,8 @@ create_file_dentry(struct eventfs_inode
 
 	mutex_unlock(&eventfs_mutex);
 
-	/* The lookup already has the parent->d_inode locked */
-	if (!lookup)
-		inode_lock(parent->d_inode);
-
 	dentry = create_file(name, mode, attr, parent, data, fops);
 
-	if (!lookup)
-		inode_unlock(parent->d_inode);
-
 	mutex_lock(&eventfs_mutex);
 
 	if (IS_ERR_OR_NULL(dentry)) {

@@ -453,15 +446,8 @@ create_dir_dentry(struct eventfs_inode *
 	}
 	mutex_unlock(&eventfs_mutex);
 
-	/* The lookup already has the parent->d_inode locked */
-	if (!lookup)
-		inode_lock(parent->d_inode);
-
 	dentry = create_dir(ei, parent);
 
-	if (!lookup)
-		inode_unlock(parent->d_inode);
-
 	mutex_lock(&eventfs_mutex);
 
 	if (IS_ERR_OR_NULL(dentry) && !ei->is_freed) {

@@ -693,6 +679,7 @@ static int dcache_dir_open_wrapper(struc
 		return -ENOMEM;
 	}
 
+	inode_lock(parent->d_inode);
 	list_for_each_entry_srcu(ei_child, &ei->children, list,
 				 srcu_read_lock_held(&eventfs_srcu)) {
 		d = create_dir_dentry(ei, ei_child, parent, false);

@@ -725,6 +712,7 @@ static int dcache_dir_open_wrapper(struc
 			cnt++;
 		}
 	}
+	inode_unlock(parent->d_inode);
 	srcu_read_unlock(&eventfs_srcu, idx);
 	ret = dcache_dir_open(inode, file);
 

@@ -509,20 +509,15 @@ struct dentry *eventfs_start_creating(co
 	struct dentry *dentry;
 	int error;
 
+	/* Must always have a parent. */
+	if (WARN_ON_ONCE(!parent))
+		return ERR_PTR(-EINVAL);
+
 	error = simple_pin_fs(&trace_fs_type, &tracefs_mount,
 			      &tracefs_mount_count);
 	if (error)
 		return ERR_PTR(error);
 
-	/*
-	 * If the parent is not specified, we create it in the root.
-	 * We need the root dentry to do this, which is in the super
-	 * block. A pointer to that is in the struct vfsmount that we
-	 * have around.
-	 */
-	if (!parent)
-		parent = tracefs_mount->mnt_root;
-
 	if (unlikely(IS_DEADDIR(parent->d_inode)))
 		dentry = ERR_PTR(-ENOENT);
 	else

@@ -327,6 +327,8 @@ create_file_dentry(struct eventfs_inode
 	struct dentry **e_dentry = &ei->d_children[idx];
 	struct dentry *dentry;
 
+	WARN_ON_ONCE(!inode_is_locked(parent->d_inode));
+
 	mutex_lock(&eventfs_mutex);
 	if (ei->is_freed) {
 		mutex_unlock(&eventfs_mutex);

@@ -430,6 +432,8 @@ create_dir_dentry(struct eventfs_inode *
 {
 	struct dentry *dentry = NULL;
 
+	WARN_ON_ONCE(!inode_is_locked(parent->d_inode));
+
 	mutex_lock(&eventfs_mutex);
 	if (pei->is_freed || ei->is_freed) {
 		mutex_unlock(&eventfs_mutex);

@@ -546,6 +546,8 @@ static struct dentry *eventfs_root_looku
 		if (strcmp(ei_child->name, name) != 0)
 			continue;
 		ret = simple_lookup(dir, dentry, flags);
+		if (IS_ERR(ret))
+			goto out;
 		create_dir_dentry(ei, ei_child, ei_dentry, true);
 		created = true;
 		break;

@@ -568,6 +570,8 @@ static struct dentry *eventfs_root_looku
 			if (r <= 0)
 				continue;
 			ret = simple_lookup(dir, dentry, flags);
+			if (IS_ERR(ret))
+				goto out;
 			create_file_dentry(ei, i, ei_dentry, name, mode, cdata,
 					   fops, true);
 			break;

@@ -148,7 +148,8 @@ static const struct file_operations even
 	.release	= eventfs_release,
 };
 
-static void update_inode_attr(struct inode *inode, struct eventfs_attr *attr, umode_t mode)
+static void update_inode_attr(struct dentry *dentry, struct inode *inode,
+			      struct eventfs_attr *attr, umode_t mode)
 {
 	if (!attr) {
 		inode->i_mode = mode;

@@ -162,9 +163,13 @@ static void update_inode_attr(struct ino
 
 	if (attr->mode & EVENTFS_SAVE_UID)
 		inode->i_uid = attr->uid;
+	else
+		inode->i_uid = d_inode(dentry->d_parent)->i_uid;
 
 	if (attr->mode & EVENTFS_SAVE_GID)
 		inode->i_gid = attr->gid;
+	else
+		inode->i_gid = d_inode(dentry->d_parent)->i_gid;
 }
 
 /**

@@ -206,7 +211,7 @@ static struct dentry *create_file(const
 		return eventfs_failed_creating(dentry);
 
 	/* If the user updated the directory's attributes, use them */
-	update_inode_attr(inode, attr, mode);
+	update_inode_attr(dentry, inode, attr, mode);
 
 	inode->i_op = &eventfs_file_inode_operations;
 	inode->i_fop = fop;

@@ -242,7 +247,8 @@ static struct dentry *create_dir(struct
 		return eventfs_failed_creating(dentry);
 
 	/* If the user updated the directory's attributes, use them */
-	update_inode_attr(inode, &ei->attr, S_IFDIR | S_IRWXU | S_IRUGO | S_IXUGO);
+	update_inode_attr(dentry, inode, &ei->attr,
+			  S_IFDIR | S_IRWXU | S_IRUGO | S_IXUGO);
 
 	inode->i_op = &eventfs_root_dir_inode_operations;
 	inode->i_fop = &eventfs_file_operations;