Merge branch 'overlayfs-next' of...

Merge branch 'overlayfs-next' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs into for-next

Documentation/filesystems/overlayfs.txt

@@ -159,6 +159,22 @@ overlay filesystem (though an operation on the name of the file such as
 rename or unlink will of course be noticed and handled).
 
 
+Multiple lower layers
+---------------------
+
+Multiple lower layers can now be given using the the colon (":") as a
+separator character between the directory names.  For example:
+
+  mount -t overlay overlay -olowerdir=/lower1:/lower2:/lower3 /merged
+
+As the example shows, "upperdir=" and "workdir=" may be omitted.  In
+that case the overlay will be read-only.
+
+The specified lower directories will be stacked beginning from the
+rightmost one and going left.  In the above example lower1 will be the
+top, lower2 the middle and lower3 the bottom layer.
+
+
 Non-standard behavior
 ---------------------
 
@@ -196,3 +212,15 @@ Changes to the underlying filesystems while part of a mounted overlay
 filesystem are not allowed.  If the underlying filesystem is changed,
 the behavior of the overlay is undefined, though it will not result in
 a crash or deadlock.
+
+Testsuite
+---------
+
+There's testsuite developed by David Howells at:
+
+  git://git.infradead.org/users/dhowells/unionmount-testsuite.git
+
+Run as root:
+
+  # cd unionmount-testsuite
+  # ./run --ov

fs/overlayfs/copy_up.c

@@ -191,7 +191,6 @@ int ovl_set_attr(struct dentry *upperdentry, struct kstat *stat)
 		ovl_set_timestamps(upperdentry, stat);
 
 	return err;
-
 }
 
 static int ovl_copy_up_locked(struct dentry *workdir, struct dentry *upperdir,
@@ -385,7 +384,7 @@ int ovl_copy_up(struct dentry *dentry)
 		struct kstat stat;
 		enum ovl_path_type type = ovl_path_type(dentry);
 
-		if (type != OVL_PATH_LOWER)
+		if (OVL_TYPE_UPPER(type))
 			break;
 
 		next = dget(dentry);
@@ -394,7 +393,7 @@ int ovl_copy_up(struct dentry *dentry)
 			parent = dget_parent(next);
 
 			type = ovl_path_type(parent);
-			if (type != OVL_PATH_LOWER)
+			if (OVL_TYPE_UPPER(type))
 				break;
 
 			dput(next);

fs/overlayfs/dir.c

@@ -118,14 +118,14 @@ int ovl_create_real(struct inode *dir, struct dentry *newdentry,
 
 static int ovl_set_opaque(struct dentry *upperdentry)
 {
-	return ovl_do_setxattr(upperdentry, ovl_opaque_xattr, "y", 1, 0);
+	return ovl_do_setxattr(upperdentry, OVL_XATTR_OPAQUE, "y", 1, 0);
 }
 
 static void ovl_remove_opaque(struct dentry *upperdentry)
 {
 	int err;
 
-	err = ovl_do_removexattr(upperdentry, ovl_opaque_xattr);
+	err = ovl_do_removexattr(upperdentry, OVL_XATTR_OPAQUE);
 	if (err) {
 		pr_warn("overlayfs: failed to remove opaque from '%s' (%i)\n",
 			upperdentry->d_name.name, err);
@@ -152,7 +152,7 @@ static int ovl_dir_getattr(struct vfsmount *mnt, struct dentry *dentry,
 	 * correct link count.  nlink=1 seems to pacify 'find' and
 	 * other utilities.
 	 */
-	if (type == OVL_PATH_MERGE)
+	if (OVL_TYPE_MERGE(type))
 		stat->nlink = 1;
 
 	return 0;
@@ -506,7 +506,7 @@ static int ovl_remove_and_whiteout(struct dentry *dentry, bool is_dir)
 	struct dentry *opaquedir = NULL;
 	int err;
 
-	if (is_dir) {
+	if (is_dir && OVL_TYPE_MERGE_OR_LOWER(ovl_path_type(dentry))) {
 		opaquedir = ovl_check_empty_and_clear(dentry);
 		err = PTR_ERR(opaquedir);
 		if (IS_ERR(opaquedir))
@@ -630,7 +630,7 @@ static int ovl_do_remove(struct dentry *dentry, bool is_dir)
 		goto out_drop_write;
 
 	type = ovl_path_type(dentry);
-	if (type == OVL_PATH_PURE_UPPER) {
+	if (OVL_TYPE_PURE_UPPER(type)) {
 		err = ovl_remove_upper(dentry, is_dir);
 	} else {
 		const struct cred *old_cred;
@@ -712,7 +712,7 @@ static int ovl_rename2(struct inode *olddir, struct dentry *old,
 	/* Don't copy up directory trees */
 	old_type = ovl_path_type(old);
 	err = -EXDEV;
-	if ((old_type == OVL_PATH_LOWER || old_type == OVL_PATH_MERGE) && is_dir)
+	if (OVL_TYPE_MERGE_OR_LOWER(old_type) && is_dir)
 		goto out;
 
 	if (new->d_inode) {
@@ -725,25 +725,25 @@ static int ovl_rename2(struct inode *olddir, struct dentry *old,
 
 		new_type = ovl_path_type(new);
 		err = -EXDEV;
-		if (!overwrite && (new_type == OVL_PATH_LOWER || new_type == OVL_PATH_MERGE) && new_is_dir)
+		if (!overwrite && OVL_TYPE_MERGE_OR_LOWER(new_type) && new_is_dir)
 			goto out;
 
 		err = 0;
-		if (new_type == OVL_PATH_LOWER && old_type == OVL_PATH_LOWER) {
+		if (!OVL_TYPE_UPPER(new_type) && !OVL_TYPE_UPPER(old_type)) {
 			if (ovl_dentry_lower(old)->d_inode ==
 			    ovl_dentry_lower(new)->d_inode)
 				goto out;
 		}
-		if (new_type != OVL_PATH_LOWER && old_type != OVL_PATH_LOWER) {
+		if (OVL_TYPE_UPPER(new_type) && OVL_TYPE_UPPER(old_type)) {
 			if (ovl_dentry_upper(old)->d_inode ==
 			    ovl_dentry_upper(new)->d_inode)
 				goto out;
 		}
 	} else {
 		if (ovl_dentry_is_opaque(new))
-			new_type = OVL_PATH_UPPER;
+			new_type = __OVL_PATH_UPPER;
 		else
-			new_type = OVL_PATH_PURE_UPPER;
+			new_type = __OVL_PATH_UPPER | __OVL_PATH_PURE;
 	}
 
 	err = ovl_want_write(old);
@@ -763,8 +763,8 @@ static int ovl_rename2(struct inode *olddir, struct dentry *old,
 			goto out_drop_write;
 	}
 
-	old_opaque = old_type != OVL_PATH_PURE_UPPER;
-	new_opaque = new_type != OVL_PATH_PURE_UPPER;
+	old_opaque = !OVL_TYPE_PURE_UPPER(old_type);
+	new_opaque = !OVL_TYPE_PURE_UPPER(new_type);
 
 	if (old_opaque || new_opaque) {
 		err = -ENOMEM;
@@ -787,7 +787,7 @@ static int ovl_rename2(struct inode *olddir, struct dentry *old,
 		old_cred = override_creds(override_cred);
 	}
 
-	if (overwrite && (new_type == OVL_PATH_LOWER || new_type == OVL_PATH_MERGE) && new_is_dir) {
+	if (overwrite && OVL_TYPE_MERGE_OR_LOWER(new_type) && new_is_dir) {
 		opaquedir = ovl_check_empty_and_clear(new);
 		err = PTR_ERR(opaquedir);
 		if (IS_ERR(opaquedir)) {

fs/overlayfs/inode.c

@@ -205,7 +205,7 @@ static int ovl_readlink(struct dentry *dentry, char __user *buf, int bufsiz)
 
 static bool ovl_is_private_xattr(const char *name)
 {
-	return strncmp(name, "trusted.overlay.", 14) == 0;
+	return strncmp(name, OVL_XATTR_PRE_NAME, OVL_XATTR_PRE_LEN) == 0;
 }
 
 int ovl_setxattr(struct dentry *dentry, const char *name,
@@ -238,7 +238,10 @@ int ovl_setxattr(struct dentry *dentry, const char *name,
 static bool ovl_need_xattr_filter(struct dentry *dentry,
 				  enum ovl_path_type type)
 {
-	return type == OVL_PATH_UPPER && S_ISDIR(dentry->d_inode->i_mode);
+	if ((type & (__OVL_PATH_PURE | __OVL_PATH_UPPER)) == __OVL_PATH_UPPER)
+		return S_ISDIR(dentry->d_inode->i_mode);
+	else
+		return false;
 }
 
 ssize_t ovl_getxattr(struct dentry *dentry, const char *name,
@@ -299,7 +302,7 @@ int ovl_removexattr(struct dentry *dentry, const char *name)
 	if (ovl_need_xattr_filter(dentry, type) && ovl_is_private_xattr(name))
 		goto out_drop_write;
 
-	if (type == OVL_PATH_LOWER) {
+	if (!OVL_TYPE_UPPER(type)) {
 		err = vfs_getxattr(realpath.dentry, name, NULL, 0);
 		if (err < 0)
 			goto out_drop_write;
@@ -321,7 +324,7 @@ int ovl_removexattr(struct dentry *dentry, const char *name)
 static bool ovl_open_need_copy_up(int flags, enum ovl_path_type type,
 				  struct dentry *realdentry)
 {
-	if (type != OVL_PATH_LOWER)
+	if (OVL_TYPE_UPPER(type))
 		return false;
 
 	if (special_file(realdentry->d_inode->i_mode))
@@ -430,5 +433,4 @@ struct inode *ovl_new_inode(struct super_block *sb, umode_t mode,
 	}
 
 	return inode;
-
 }

fs/overlayfs/overlayfs.h

@@ -12,13 +12,20 @@
 struct ovl_entry;
 
 enum ovl_path_type {
-	OVL_PATH_PURE_UPPER,
-	OVL_PATH_UPPER,
-	OVL_PATH_MERGE,
-	OVL_PATH_LOWER,
+	__OVL_PATH_PURE		= (1 << 0),
+	__OVL_PATH_UPPER	= (1 << 1),
+	__OVL_PATH_MERGE	= (1 << 2),
 };
 
-extern const char *ovl_opaque_xattr;
+#define OVL_TYPE_UPPER(type)	((type) & __OVL_PATH_UPPER)
+#define OVL_TYPE_MERGE(type)	((type) & __OVL_PATH_MERGE)
+#define OVL_TYPE_PURE_UPPER(type) ((type) & __OVL_PATH_PURE)
+#define OVL_TYPE_MERGE_OR_LOWER(type) \
+	(OVL_TYPE_MERGE(type) || !OVL_TYPE_UPPER(type))
+
+#define OVL_XATTR_PRE_NAME "trusted.overlay."
+#define OVL_XATTR_PRE_LEN  16
+#define OVL_XATTR_OPAQUE   OVL_XATTR_PRE_NAME"opaque"
 
 static inline int ovl_do_rmdir(struct inode *dir, struct dentry *dentry)
 {
@@ -130,6 +137,7 @@ void ovl_dentry_version_inc(struct dentry *dentry);
 void ovl_path_upper(struct dentry *dentry, struct path *path);
 void ovl_path_lower(struct dentry *dentry, struct path *path);
 enum ovl_path_type ovl_path_real(struct dentry *dentry, struct path *path);
+int ovl_path_next(int idx, struct dentry *dentry, struct path *path);
 struct dentry *ovl_dentry_upper(struct dentry *dentry);
 struct dentry *ovl_dentry_lower(struct dentry *dentry);
 struct dentry *ovl_dentry_real(struct dentry *dentry);

fs/overlayfs/readdir.c

@@ -24,7 +24,6 @@ struct ovl_cache_entry {
 	struct list_head l_node;
 	struct rb_node node;
 	bool is_whiteout;
-	bool is_cursor;
 	char name[];
 };
 
@@ -40,6 +39,7 @@ struct ovl_readdir_data {
 	struct rb_root root;
 	struct list_head *list;
 	struct list_head middle;
+	struct dentry *dir;
 	int count;
 	int err;
 };
@@ -48,7 +48,7 @@ struct ovl_dir_file {
 	bool is_real;
 	bool is_upper;
 	struct ovl_dir_cache *cache;
-	struct ovl_cache_entry cursor;
+	struct list_head *cursor;
 	struct file *realfile;
 	struct file *upperfile;
 };
@@ -79,23 +79,49 @@ static struct ovl_cache_entry *ovl_cache_entry_find(struct rb_root *root,
 	return NULL;
 }
 
-static struct ovl_cache_entry *ovl_cache_entry_new(const char *name, int len,
+static struct ovl_cache_entry *ovl_cache_entry_new(struct dentry *dir,
+						   const char *name, int len,
 						   u64 ino, unsigned int d_type)
 {
 	struct ovl_cache_entry *p;
 	size_t size = offsetof(struct ovl_cache_entry, name[len + 1]);
 
 	p = kmalloc(size, GFP_KERNEL);
-	if (p) {
+	if (!p)
+		return NULL;
+
 	memcpy(p->name, name, len);
 	p->name[len] = '\0';
 	p->len = len;
 	p->type = d_type;
 	p->ino = ino;
 	p->is_whiteout = false;
-		p->is_cursor = false;
+
+	if (d_type == DT_CHR) {
+		struct dentry *dentry;
+		const struct cred *old_cred;
+		struct cred *override_cred;
+
+		override_cred = prepare_creds();
+		if (!override_cred) {
+			kfree(p);
+			return NULL;
 		}
 
+		/*
+		 * CAP_DAC_OVERRIDE for lookup
+		 */
+		cap_raise(override_cred->cap_effective, CAP_DAC_OVERRIDE);
+		old_cred = override_creds(override_cred);
+
+		dentry = lookup_one_len(name, dir, len);
+		if (!IS_ERR(dentry)) {
+			p->is_whiteout = ovl_is_whiteout(dentry);
+			dput(dentry);
+		}
+		revert_creds(old_cred);
+		put_cred(override_cred);
+	}
 	return p;
 }
 
@@ -122,7 +148,7 @@ static int ovl_cache_entry_add_rb(struct ovl_readdir_data *rdd,
 			return 0;
 	}
 
-	p = ovl_cache_entry_new(name, len, ino, d_type);
+	p = ovl_cache_entry_new(rdd->dir, name, len, ino, d_type);
 	if (p == NULL)
 		return -ENOMEM;
 
@@ -143,7 +169,7 @@ static int ovl_fill_lower(struct ovl_readdir_data *rdd,
 	if (p) {
 		list_move_tail(&p->l_node, &rdd->middle);
 	} else {
-		p = ovl_cache_entry_new(name, namelen, ino, d_type);
+		p = ovl_cache_entry_new(rdd->dir, name, namelen, ino, d_type);
 		if (p == NULL)
 			rdd->err = -ENOMEM;
 		else
@@ -168,7 +194,6 @@ static void ovl_cache_put(struct ovl_dir_file *od, struct dentry *dentry)
 {
 	struct ovl_dir_cache *cache = od->cache;
 
-	list_del_init(&od->cursor.l_node);
 	WARN_ON(cache->refcount <= 0);
 	cache->refcount--;
 	if (!cache->refcount) {
@@ -204,6 +229,7 @@ static inline int ovl_dir_read(struct path *realpath,
 	if (IS_ERR(realfile))
 		return PTR_ERR(realfile);
 
+	rdd->dir = realpath->dentry;
 	rdd->ctx.pos = 0;
 	do {
 		rdd->count = 0;
@@ -227,108 +253,58 @@ static void ovl_dir_reset(struct file *file)
 	if (cache && ovl_dentry_version_get(dentry) != cache->version) {
 		ovl_cache_put(od, dentry);
 		od->cache = NULL;
+		od->cursor = NULL;
 	}
-	WARN_ON(!od->is_real && type != OVL_PATH_MERGE);
-	if (od->is_real && type == OVL_PATH_MERGE)
+	WARN_ON(!od->is_real && !OVL_TYPE_MERGE(type));
+	if (od->is_real && OVL_TYPE_MERGE(type))
 		od->is_real = false;
 }
 
-static int ovl_dir_mark_whiteouts(struct dentry *dir,
-				  struct ovl_readdir_data *rdd)
-{
-	struct ovl_cache_entry *p;
-	struct dentry *dentry;
-	const struct cred *old_cred;
-	struct cred *override_cred;
-
-	override_cred = prepare_creds();
-	if (!override_cred) {
-		ovl_cache_free(rdd->list);
-		return -ENOMEM;
-	}
-
-	/*
-	 * CAP_DAC_OVERRIDE for lookup
-	 */
-	cap_raise(override_cred->cap_effective, CAP_DAC_OVERRIDE);
-	old_cred = override_creds(override_cred);
-
-	mutex_lock(&dir->d_inode->i_mutex);
-	list_for_each_entry(p, rdd->list, l_node) {
-		if (p->is_cursor)
-			continue;
-
-		if (p->type != DT_CHR)
-			continue;
-
-		dentry = lookup_one_len(p->name, dir, p->len);
-		if (IS_ERR(dentry))
-			continue;
-
-		p->is_whiteout = ovl_is_whiteout(dentry);
-		dput(dentry);
-	}
-	mutex_unlock(&dir->d_inode->i_mutex);
-
-	revert_creds(old_cred);
-	put_cred(override_cred);
-
-	return 0;
-}
-
 static int ovl_dir_read_merged(struct dentry *dentry, struct list_head *list)
 {
 	int err;
-	struct path lowerpath;
-	struct path upperpath;
+	struct path realpath;
 	struct ovl_readdir_data rdd = {
 		.ctx.actor = ovl_fill_merge,
 		.list = list,
 		.root = RB_ROOT,
 		.is_merge = false,
 	};
+	int idx, next;
 
-	ovl_path_lower(dentry, &lowerpath);
-	ovl_path_upper(dentry, &upperpath);
-
-	if (upperpath.dentry) {
-		err = ovl_dir_read(&upperpath, &rdd);
-		if (err)
-			goto out;
+	for (idx = 0; idx != -1; idx = next) {
+		next = ovl_path_next(idx, dentry, &realpath);
 
-		if (lowerpath.dentry) {
-			err = ovl_dir_mark_whiteouts(upperpath.dentry, &rdd);
+		if (next != -1) {
+			err = ovl_dir_read(&realpath, &rdd);
 			if (err)
-				goto out;
-		}
-	}
-	if (lowerpath.dentry) {
+				break;
+		} else {
 			/*
-		 * Insert lowerpath entries before upperpath ones, this allows
-		 * offsets to be reasonably constant
+			 * Insert lowest layer entries before upper ones, this
+			 * allows offsets to be reasonably constant
 			 */
 			list_add(&rdd.middle, rdd.list);
 			rdd.is_merge = true;
-		err = ovl_dir_read(&lowerpath, &rdd);
+			err = ovl_dir_read(&realpath, &rdd);
 			list_del(&rdd.middle);
 		}
-out:
+	}
 	return err;
 }
 
 static void ovl_seek_cursor(struct ovl_dir_file *od, loff_t pos)
 {
-	struct ovl_cache_entry *p;
+	struct list_head *p;
 	loff_t off = 0;
 
-	list_for_each_entry(p, &od->cache->entries, l_node) {
-		if (p->is_cursor)
-			continue;
+	list_for_each(p, &od->cache->entries) {
 		if (off >= pos)
 			break;
 		off++;
 	}
-	list_move_tail(&od->cursor.l_node, &p->l_node);
+	/* Cursor is safe since the cache is stable */
+	od->cursor = p;
 }
 
 static struct ovl_dir_cache *ovl_cache_get(struct dentry *dentry)
@@ -367,6 +343,7 @@ static int ovl_iterate(struct file *file, struct dir_context *ctx)
 {
 	struct ovl_dir_file *od = file->private_data;
 	struct dentry *dentry = file->f_path.dentry;
+	struct ovl_cache_entry *p;
 
 	if (!ctx->pos)
 		ovl_dir_reset(file);
@@ -385,20 +362,14 @@ static int ovl_iterate(struct file *file, struct dir_context *ctx)
 		ovl_seek_cursor(od, ctx->pos);
 	}
 
-	while (od->cursor.l_node.next != &od->cache->entries) {
-		struct ovl_cache_entry *p;
-
-		p = list_entry(od->cursor.l_node.next, struct ovl_cache_entry, l_node);
-		/* Skip cursors */
-		if (!p->is_cursor) {
-			if (!p->is_whiteout) {
+	while (od->cursor != &od->cache->entries) {
+		p = list_entry(od->cursor, struct ovl_cache_entry, l_node);
+		if (!p->is_whiteout)
 			if (!dir_emit(ctx, p->name, p->len, p->ino, p->type))
 				break;
-			}
+		od->cursor = p->l_node.next;
 		ctx->pos++;
 	}
-		list_move(&od->cursor.l_node, &p->l_node);
-	}
 	return 0;
 }
 
@@ -452,7 +423,7 @@ static int ovl_dir_fsync(struct file *file, loff_t start, loff_t end,
 	/*
 	 * Need to check if we started out being a lower dir, but got copied up
 	 */
-	if (!od->is_upper && ovl_path_type(dentry) != OVL_PATH_LOWER) {
+	if (!od->is_upper && OVL_TYPE_UPPER(ovl_path_type(dentry))) {
 		struct inode *inode = file_inode(file);
 
 		realfile = lockless_dereference(od->upperfile);
@@ -516,11 +487,9 @@ static int ovl_dir_open(struct inode *inode, struct file *file)
 		kfree(od);
 		return PTR_ERR(realfile);
 	}
-	INIT_LIST_HEAD(&od->cursor.l_node);
 	od->realfile = realfile;
-	od->is_real = (type != OVL_PATH_MERGE);
-	od->is_upper = (type != OVL_PATH_LOWER);
-	od->cursor.is_cursor = true;
+	od->is_real = !OVL_TYPE_MERGE(type);
+	od->is_upper = OVL_TYPE_UPPER(type);
 	file->private_data = od;
 
 	return 0;