Merge branch 'for-linus-2' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs

Pull more vfs updates from Al Viro:
 "Assorted stuff from this cycle.  The big ones here are multilayer
  overlayfs from Miklos and beginning of sorting ->d_inode accesses out
  from David"

* 'for-linus-2' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (51 commits)
  autofs4 copy_dev_ioctl(): keep the value of ->size we'd used for allocation
  procfs: fix race between symlink removals and traversals
  debugfs: leave freeing a symlink body until inode eviction
  Documentation/filesystems/Locking: ->get_sb() is long gone
  trylock_super(): replacement for grab_super_passive()
  fanotify: Fix up scripted S_ISDIR/S_ISREG/S_ISLNK conversions
  Cachefiles: Fix up scripted S_ISDIR/S_ISREG/S_ISLNK conversions
  VFS: (Scripted) Convert S_ISLNK/DIR/REG(dentry->d_inode) to d_is_*(dentry)
  SELinux: Use d_is_positive() rather than testing dentry->d_inode
  Smack: Use d_is_positive() rather than testing dentry->d_inode
  TOMOYO: Use d_is_dir() rather than d_inode and S_ISDIR()
  Apparmor: Use d_is_positive/negative() rather than testing dentry->d_inode
  Apparmor: mediated_filesystem() should use dentry->d_sb not inode->i_sb
  VFS: Split DCACHE_FILE_TYPE into regular and special types
  VFS: Add a fallthrough flag for marking virtual dentries
  VFS: Add a whiteout dentry type
  VFS: Introduce inode-getting helpers for layered/unioned fs environments
  Infiniband: Fix potential NULL d_inode dereference
  posix_acl: fix reference leaks in posix_acl_create
  autofs4: Wrong format for printing dentry
  ...

fs/jffs2/super.c

@@ -138,7 +138,7 @@ static struct dentry *jffs2_get_parent(struct dentry *child)
 	struct jffs2_inode_info *f;
 	uint32_t pino;
 
-	BUG_ON(!S_ISDIR(child->d_inode->i_mode));
+	BUG_ON(!d_is_dir(child));
 
 	f = JFFS2_INODE_INFO(child->d_inode);
 

fs/libfs.c

@@ -329,7 +329,7 @@ int simple_rename(struct inode *old_dir, struct dentry *old_dentry,
 		struct inode *new_dir, struct dentry *new_dentry)
 {
 	struct inode *inode = old_dentry->d_inode;
-	int they_are_dirs = S_ISDIR(old_dentry->d_inode->i_mode);
+	int they_are_dirs = d_is_dir(old_dentry);
 
 	if (!simple_empty(new_dentry))
 		return -ENOTEMPTY;

fs/namei.c

@@ -2814,7 +2814,7 @@ static int atomic_open(struct nameidata *nd, struct dentry *dentry,
 			} else if (!dentry->d_inode) {
 				goto out;
 			} else if ((open_flag & O_TRUNC) &&
-				   S_ISREG(dentry->d_inode->i_mode)) {
+				   d_is_reg(dentry)) {
 				goto out;
 			}
 			/* will fail later, go on to get the right error */

fs/namespace.c

@@ -1907,8 +1907,8 @@ static int graft_tree(struct mount *mnt, struct mount *p, struct mountpoint *mp)
 	if (mnt->mnt.mnt_sb->s_flags & MS_NOUSER)
 		return -EINVAL;
 
-	if (S_ISDIR(mp->m_dentry->d_inode->i_mode) !=
-	      S_ISDIR(mnt->mnt.mnt_root->d_inode->i_mode))
+	if (d_is_dir(mp->m_dentry) !=
+	      d_is_dir(mnt->mnt.mnt_root))
 		return -ENOTDIR;
 
 	return attach_recursive_mnt(mnt, p, mp, NULL);
@@ -2180,8 +2180,8 @@ static int do_move_mount(struct path *path, const char *old_name)
 	if (!mnt_has_parent(old))
 		goto out1;
 
-	if (S_ISDIR(path->dentry->d_inode->i_mode) !=
-	      S_ISDIR(old_path.dentry->d_inode->i_mode))
+	if (d_is_dir(path->dentry) !=
+	      d_is_dir(old_path.dentry))
 		goto out1;
 	/*
 	 * Don't move a mount residing in a shared parent.
@@ -2271,7 +2271,7 @@ static int do_add_mount(struct mount *newmnt, struct path *path, int mnt_flags)
 		goto unlock;
 
 	err = -EINVAL;
-	if (S_ISLNK(newmnt->mnt.mnt_root->d_inode->i_mode))
+	if (d_is_symlink(newmnt->mnt.mnt_root))
 		goto unlock;
 
 	newmnt->mnt.mnt_flags = mnt_flags;

fs/nfsd/nfs4recover.c

@@ -583,7 +583,7 @@ nfs4_reset_recoverydir(char *recdir)
 	if (status)
 		return status;
 	status = -ENOTDIR;
-	if (S_ISDIR(path.dentry->d_inode->i_mode)) {
+	if (d_is_dir(path.dentry)) {
 		strcpy(user_recovery_dirname, recdir);
 		status = 0;
 	}
@@ -1426,7 +1426,7 @@ nfsd4_client_tracking_init(struct net *net)
 	nn->client_tracking_ops = &nfsd4_legacy_tracking_ops;
 	status = kern_path(nfs4_recoverydir(), LOOKUP_FOLLOW, &path);
 	if (!status) {
-		status = S_ISDIR(path.dentry->d_inode->i_mode);
+		status = d_is_dir(path.dentry);
 		path_put(&path);
 		if (status)
 			goto do_init;

fs/nfsd/nfsfh.c

@@ -114,8 +114,8 @@ static inline __be32 check_pseudo_root(struct svc_rqst *rqstp,
 	 * We're exposing only the directories and symlinks that have to be
 	 * traversed on the way to real exports:
 	 */
-	if (unlikely(!S_ISDIR(dentry->d_inode->i_mode) &&
-		     !S_ISLNK(dentry->d_inode->i_mode)))
+	if (unlikely(!d_is_dir(dentry) &&
+		     !d_is_symlink(dentry)))
 		return nfserr_stale;
 	/*
 	 * A pseudoroot export gives permission to access only one
@@ -259,7 +259,7 @@ static __be32 nfsd_set_fh_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp)
 		goto out;
 	}
 
-	if (S_ISDIR(dentry->d_inode->i_mode) &&
+	if (d_is_dir(dentry) &&
 			(dentry->d_flags & DCACHE_DISCONNECTED)) {
 		printk("nfsd: find_fh_dentry returned a DISCONNECTED directory: %pd2\n",
 				dentry);
@@ -414,7 +414,7 @@ static inline void _fh_update_old(struct dentry *dentry,
 {
 	fh->ofh_ino = ino_t_to_u32(dentry->d_inode->i_ino);
 	fh->ofh_generation = dentry->d_inode->i_generation;
-	if (S_ISDIR(dentry->d_inode->i_mode) ||
+	if (d_is_dir(dentry) ||
 	    (exp->ex_flags & NFSEXP_NOSUBTREECHECK))
 		fh->ofh_dirino = 0;
 }

fs/nfsd/vfs.c

@@ -615,9 +615,9 @@ nfsd_access(struct svc_rqst *rqstp, struct svc_fh *fhp, u32 *access, u32 *suppor
 	export = fhp->fh_export;
 	dentry = fhp->fh_dentry;
 
-	if (S_ISREG(dentry->d_inode->i_mode))
+	if (d_is_reg(dentry))
 		map = nfs3_regaccess;
-	else if (S_ISDIR(dentry->d_inode->i_mode))
+	else if (d_is_dir(dentry))
 		map = nfs3_diraccess;
 	else
 		map = nfs3_anyaccess;
@@ -1402,7 +1402,7 @@ do_nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp,
 
 		switch (createmode) {
 		case NFS3_CREATE_UNCHECKED:
-			if (! S_ISREG(dchild->d_inode->i_mode))
+			if (! d_is_reg(dchild))
 				goto out;
 			else if (truncp) {
 				/* in nfsv4, we need to treat this case a little
@@ -1615,7 +1615,7 @@ nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp,
 	if (err)
 		goto out;
 	err = nfserr_isdir;
-	if (S_ISDIR(tfhp->fh_dentry->d_inode->i_mode))
+	if (d_is_dir(tfhp->fh_dentry))
 		goto out;
 	err = nfserr_perm;
 	if (!len)

fs/notify/fanotify/fanotify.c

@@ -115,8 +115,8 @@ static bool fanotify_should_send_event(struct fsnotify_mark *inode_mark,
 		return false;
 
 	/* sorry, fanotify only gives a damn about files and dirs */
-	if (!S_ISREG(path->dentry->d_inode->i_mode) &&
-	    !S_ISDIR(path->dentry->d_inode->i_mode))
+	if (!d_is_reg(path->dentry) &&
+	    !d_can_lookup(path->dentry))
 		return false;
 
 	if (inode_mark && vfsmnt_mark) {
@@ -139,7 +139,7 @@ static bool fanotify_should_send_event(struct fsnotify_mark *inode_mark,
 		BUG();
 	}
 
-	if (S_ISDIR(path->dentry->d_inode->i_mode) &&
+	if (d_is_dir(path->dentry) &&
 	    !(marks_mask & FS_ISDIR & ~marks_ignored_mask))
 		return false;
 

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

@@ -19,7 +19,7 @@ void ovl_cleanup(struct inode *wdir, struct dentry *wdentry)
 	int err;
 
 	dget(wdentry);
-	if (S_ISDIR(wdentry->d_inode->i_mode))
+	if (d_is_dir(wdentry))
 		err = ovl_do_rmdir(wdir, wdentry);
 	else
 		err = ovl_do_unlink(wdir, wdentry);
@@ -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;
@@ -693,7 +693,7 @@ static int ovl_rename2(struct inode *olddir, struct dentry *old,
 	bool new_create = false;
 	bool cleanup_whiteout = false;
 	bool overwrite = !(flags & RENAME_EXCHANGE);
-	bool is_dir = S_ISDIR(old->d_inode->i_mode);
+	bool is_dir = d_is_dir(old);
 	bool new_is_dir = false;
 	struct dentry *opaquedir = NULL;
 	const struct cred *old_cred = NULL;
@@ -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) {
@@ -720,30 +720,30 @@ static int ovl_rename2(struct inode *olddir, struct dentry *old,
 		if (err)
 			goto out;
 
-		if (S_ISDIR(new->d_inode->i_mode))
+		if (d_is_dir(new))
 			new_is_dir = true;
 
 		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) {
-		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 (!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;
+
+	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);
+	for (idx = 0; idx != -1; idx = next) {
+		next = ovl_path_next(idx, dentry, &realpath);
 
-	if (upperpath.dentry) {
-		err = ovl_dir_read(&upperpath, &rdd);
-		if (err)
-			goto out;
-
-		if (lowerpath.dentry) {
-			err = ovl_dir_mark_whiteouts(upperpath.dentry, &rdd);
+		if (next != -1) {
+			err = ovl_dir_read(&realpath, &rdd);
 			if (err)
-				goto out;
+				break;
+		} else {
+			/*
+			 * 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(&realpath, &rdd);
+			list_del(&rdd.middle);
 		}
 	}
-	if (lowerpath.dentry) {
-		/*
-		 * Insert lowerpath entries before upperpath ones, this allows
-		 * offsets to be reasonably constant
-		 */
-		list_add(&rdd.middle, rdd.list);
-		rdd.is_merge = true;
-		err = ovl_dir_read(&lowerpath, &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);
@@ -385,19 +362,13 @@ 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) {
-				if (!dir_emit(ctx, p->name, p->len, p->ino, p->type))
-					break;
-			}
-			ctx->pos++;
-		}
-		list_move(&od->cursor.l_node, &p->l_node);
+	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++;
 	}
 	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;

fs/posix_acl.c

@@ -564,13 +564,11 @@ posix_acl_create(struct inode *dir, umode_t *mode,
 
 	*acl = posix_acl_clone(p, GFP_NOFS);
 	if (!*acl)
-		return -ENOMEM;
+		goto no_mem;
 
 	ret = posix_acl_create_masq(*acl, mode);
-	if (ret < 0) {
-		posix_acl_release(*acl);
-		return -ENOMEM;
-	}
+	if (ret < 0)
+		goto no_mem_clone;
 
 	if (ret == 0) {
 		posix_acl_release(*acl);
@@ -591,6 +589,12 @@ posix_acl_create(struct inode *dir, umode_t *mode,
 	*default_acl = NULL;
 	*acl = NULL;
 	return 0;
+
+no_mem_clone:
+	posix_acl_release(*acl);
+no_mem:
+	posix_acl_release(p);
+	return -ENOMEM;
 }
 EXPORT_SYMBOL_GPL(posix_acl_create);
 
@@ -772,7 +776,7 @@ posix_acl_xattr_get(struct dentry *dentry, const char *name,
 
 	if (!IS_POSIXACL(dentry->d_inode))
 		return -EOPNOTSUPP;
-	if (S_ISLNK(dentry->d_inode->i_mode))
+	if (d_is_symlink(dentry))
 		return -EOPNOTSUPP;
 
 	acl = get_acl(dentry->d_inode, type);
@@ -832,7 +836,7 @@ posix_acl_xattr_list(struct dentry *dentry, char *list, size_t list_size,
 
 	if (!IS_POSIXACL(dentry->d_inode))
 		return -EOPNOTSUPP;
-	if (S_ISLNK(dentry->d_inode->i_mode))
+	if (d_is_symlink(dentry))
 		return -EOPNOTSUPP;
 
 	if (type == ACL_TYPE_ACCESS)

fs/proc/generic.c

@@ -19,7 +19,6 @@
 #include <linux/mount.h>
 #include <linux/init.h>
 #include <linux/idr.h>
-#include <linux/namei.h>
 #include <linux/bitops.h>
 #include <linux/spinlock.h>
 #include <linux/completion.h>
@@ -223,17 +222,6 @@ void proc_free_inum(unsigned int inum)
 	spin_unlock_irqrestore(&proc_inum_lock, flags);
 }
 
-static void *proc_follow_link(struct dentry *dentry, struct nameidata *nd)
-{
-	nd_set_link(nd, __PDE_DATA(dentry->d_inode));
-	return NULL;
-}
-
-static const struct inode_operations proc_link_inode_operations = {
-	.readlink	= generic_readlink,
-	.follow_link	= proc_follow_link,
-};
-
 /*
  * Don't create negative dentries here, return -ENOENT by hand
  * instead.

fs/proc/inode.c

@@ -23,6 +23,7 @@
 #include <linux/slab.h>
 #include <linux/mount.h>
 #include <linux/magic.h>
+#include <linux/namei.h>
 
 #include <asm/uaccess.h>
 
@@ -393,6 +394,26 @@ static const struct file_operations proc_reg_file_ops_no_compat = {
 };
 #endif
 
+static void *proc_follow_link(struct dentry *dentry, struct nameidata *nd)
+{
+	struct proc_dir_entry *pde = PDE(dentry->d_inode);
+	if (unlikely(!use_pde(pde)))
+		return ERR_PTR(-EINVAL);
+	nd_set_link(nd, pde->data);
+	return pde;
+}
+
+static void proc_put_link(struct dentry *dentry, struct nameidata *nd, void *p)
+{
+	unuse_pde(p);
+}
+
+const struct inode_operations proc_link_inode_operations = {
+	.readlink	= generic_readlink,
+	.follow_link	= proc_follow_link,
+	.put_link	= proc_put_link,
+};
+
 struct inode *proc_get_inode(struct super_block *sb, struct proc_dir_entry *de)
 {
 	struct inode *inode = new_inode_pseudo(sb);

fs/proc/internal.h

@@ -200,6 +200,7 @@ struct pde_opener {
 	int closing;
 	struct completion *c;
 };
+extern const struct inode_operations proc_link_inode_operations;
 
 extern const struct inode_operations proc_pid_link_inode_operations;
 

fs/reiserfs/xattr.c

@@ -266,7 +266,7 @@ static int reiserfs_for_each_xattr(struct inode *inode,
 		for (i = 0; !err && i < buf.count && buf.dentries[i]; i++) {
 			struct dentry *dentry = buf.dentries[i];
 
-			if (!S_ISDIR(dentry->d_inode->i_mode))
+			if (!d_is_dir(dentry))
 				err = action(dentry, data);
 
 			dput(dentry);
@@ -322,7 +322,7 @@ static int delete_one_xattr(struct dentry *dentry, void *data)
 	struct inode *dir = dentry->d_parent->d_inode;
 
 	/* This is the xattr dir, handle specially. */
-	if (S_ISDIR(dentry->d_inode->i_mode))
+	if (d_is_dir(dentry))
 		return xattr_rmdir(dir, dentry);
 
 	return xattr_unlink(dir, dentry);

fs/super.c

@@ -71,7 +71,7 @@ static unsigned long super_cache_scan(struct shrinker *shrink,
 	if (!(sc->gfp_mask & __GFP_FS))
 		return SHRINK_STOP;
 
-	if (!grab_super_passive(sb))
+	if (!trylock_super(sb))
 		return SHRINK_STOP;
 
 	if (sb->s_op->nr_cached_objects)
@@ -105,7 +105,7 @@ static unsigned long super_cache_scan(struct shrinker *shrink,
 		freed += sb->s_op->free_cached_objects(sb, sc);
 	}
 
-	drop_super(sb);
+	up_read(&sb->s_umount);
 	return freed;
 }
 
@@ -118,7 +118,7 @@ static unsigned long super_cache_count(struct shrinker *shrink,
 	sb = container_of(shrink, struct super_block, s_shrink);
 
 	/*
-	 * Don't call grab_super_passive as it is a potential
+	 * Don't call trylock_super as it is a potential
 	 * scalability bottleneck. The counts could get updated
 	 * between super_cache_count and super_cache_scan anyway.
 	 * Call to super_cache_count with shrinker_rwsem held
@@ -348,35 +348,31 @@ static int grab_super(struct super_block *s) __releases(sb_lock)
 }
 
 /*
- *	grab_super_passive - acquire a passive reference
+ *	trylock_super - try to grab ->s_umount shared
  *	@sb: reference we are trying to grab
  *
- *	Tries to acquire a passive reference. This is used in places where we
+ *	Try to prevent fs shutdown.  This is used in places where we
  *	cannot take an active reference but we need to ensure that the
- *	superblock does not go away while we are working on it. It returns
- *	false if a reference was not gained, and returns true with the s_umount
- *	lock held in read mode if a reference is gained. On successful return,
- *	the caller must drop the s_umount lock and the passive reference when
- *	done.
+ *	filesystem is not shut down while we are working on it. It returns
+ *	false if we cannot acquire s_umount or if we lose the race and
+ *	filesystem already got into shutdown, and returns true with the s_umount
+ *	lock held in read mode in case of success. On successful return,
+ *	the caller must drop the s_umount lock when done.
+ *
+ *	Note that unlike get_super() et.al. this one does *not* bump ->s_count.
+ *	The reason why it's safe is that we are OK with doing trylock instead
+ *	of down_read().  There's a couple of places that are OK with that, but
+ *	it's very much not a general-purpose interface.
  */
-bool grab_super_passive(struct super_block *sb)
+bool trylock_super(struct super_block *sb)
 {
-	spin_lock(&sb_lock);
-	if (hlist_unhashed(&sb->s_instances)) {
-		spin_unlock(&sb_lock);
-		return false;
-	}
-
-	sb->s_count++;
-	spin_unlock(&sb_lock);
-
 	if (down_read_trylock(&sb->s_umount)) {
-		if (sb->s_root && (sb->s_flags & MS_BORN))
+		if (!hlist_unhashed(&sb->s_instances) &&
+		    sb->s_root && (sb->s_flags & MS_BORN))
 			return true;
 		up_read(&sb->s_umount);
 	}
 
-	put_super(sb);
 	return false;
 }