Merge tag 'ovl-update-5.7' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs

Pull overlayfs update from Miklos Szeredi:

 - Fix failure to copy-up files from certain NFSv4 mounts

 - Sort out inconsistencies between st_ino and i_ino (used in /proc/locks)

 - Allow consistent (POSIX-y) inode numbering in more cases

 - Allow virtiofs to be used as upper layer

 - Miscellaneous cleanups and fixes

* tag 'ovl-update-5.7' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs:
  ovl: document xino expected behavior
  ovl: enable xino automatically in more cases
  ovl: avoid possible inode number collisions with xino=on
  ovl: use a private non-persistent ino pool
  ovl: fix WARN_ON nlink drop to zero
  ovl: fix a typo in comment
  ovl: replace zero-length array with flexible-array member
  ovl: ovl_obtain_alias(): don't call d_instantiate_anon() for old
  ovl: strict upper fs requirements for remote upper fs
  ovl: check if upper fs supports RENAME_WHITEOUT
  ovl: allow remote upper
  ovl: decide if revalidate needed on a per-dentry basis
  ovl: separate detection of remote upper layer from stacked overlay
  ovl: restructure dentry revalidation
  ovl: ignore failure to copy up unknown xattrs
  ovl: document permission model
  ovl: simplify i_ino initialization
  ovl: factor out helper ovl_get_root()
  ovl: fix out of date comment and unreachable code
  ovl: fix value of i_ino for lower hardlink corner case

Documentation/filesystems/overlayfs.rst

@@ -40,13 +40,46 @@ On 64bit systems, even if all overlay layers are not on the same
 underlying filesystem, the same compliant behavior could be achieved
 with the "xino" feature.  The "xino" feature composes a unique object
 identifier from the real object st_ino and an underlying fsid index.
+
 If all underlying filesystems support NFS file handles and export file
 handles with 32bit inode number encoding (e.g. ext4), overlay filesystem
 will use the high inode number bits for fsid.  Even when the underlying
 filesystem uses 64bit inode numbers, users can still enable the "xino"
 feature with the "-o xino=on" overlay mount option.  That is useful for the
 case of underlying filesystems like xfs and tmpfs, which use 64bit inode
-numbers, but are very unlikely to use the high inode number bit.
+numbers, but are very unlikely to use the high inode number bits.  In case
+the underlying inode number does overflow into the high xino bits, overlay
+filesystem will fall back to the non xino behavior for that inode.
+
+The following table summarizes what can be expected in different overlay
+configurations.
+
+Inode properties
+````````````````
+
++--------------+------------+------------+-----------------+----------------+
+|Configuration | Persistent | Uniform    | st_ino == d_ino | d_ino == i_ino |
+|              | st_ino     | st_dev     |                 | [*]            |
++==============+=====+======+=====+======+========+========+========+=======+
+|              | dir | !dir | dir | !dir |  dir   +  !dir  |  dir   | !dir  |
++--------------+-----+------+-----+------+--------+--------+--------+-------+
+| All layers   |  Y  |  Y   |  Y  |  Y   |  Y     |   Y    |  Y     |  Y    |
+| on same fs   |     |      |     |      |        |        |        |       |
++--------------+-----+------+-----+------+--------+--------+--------+-------+
+| Layers not   |  N  |  Y   |  Y  |  N   |  N     |   Y    |  N     |  Y    |
+| on same fs,  |     |      |     |      |        |        |        |       |
+| xino=off     |     |      |     |      |        |        |        |       |
++--------------+-----+------+-----+------+--------+--------+--------+-------+
+| xino=on/auto |  Y  |  Y   |  Y  |  Y   |  Y     |   Y    |  Y     |  Y    |
+|              |     |      |     |      |        |        |        |       |
++--------------+-----+------+-----+------+--------+--------+--------+-------+
+| xino=on/auto,|  N  |  Y   |  Y  |  N   |  N     |   Y    |  N     |  Y    |
+| ino overflow |     |      |     |      |        |        |        |       |
++--------------+-----+------+-----+------+--------+--------+--------+-------+
+
+[*] nfsd v3 readdirplus verifies d_ino == i_ino. i_ino is exposed via several
+/proc files, such as /proc/locks and /proc/self/fdinfo/<fd> of an inotify
+file descriptor.
 
 
 Upper and Lower
@@ -248,6 +281,50 @@ overlay filesystem (though an operation on the name of the file such as
 rename or unlink will of course be noticed and handled).
 
 
+Permission model
+----------------
+
+Permission checking in the overlay filesystem follows these principles:
+
+ 1) permission check SHOULD return the same result before and after copy up
+
+ 2) task creating the overlay mount MUST NOT gain additional privileges
+
+ 3) non-mounting task MAY gain additional privileges through the overlay,
+ compared to direct access on underlying lower or upper filesystems
+
+This is achieved by performing two permission checks on each access
+
+ a) check if current task is allowed access based on local DAC (owner,
+    group, mode and posix acl), as well as MAC checks
+
+ b) check if mounting task would be allowed real operation on lower or
+    upper layer based on underlying filesystem permissions, again including
+    MAC checks
+
+Check (a) ensures consistency (1) since owner, group, mode and posix acls
+are copied up.  On the other hand it can result in server enforced
+permissions (used by NFS, for example) being ignored (3).
+
+Check (b) ensures that no task gains permissions to underlying layers that
+the mounting task does not have (2).  This also means that it is possible
+to create setups where the consistency rule (1) does not hold; normally,
+however, the mounting task will have sufficient privileges to perform all
+operations.
+
+Another way to demonstrate this model is drawing parallels between
+
+  mount -t overlay overlay -olowerdir=/lower,upperdir=/upper,... /merged
+
+and
+
+  cp -a /lower /upper
+  mount --bind /upper /merged
+
+The resulting access permissions should be the same.  The difference is in
+the time of copy (on-demand vs. up-front).
+
+
 Multiple lower layers
 ---------------------
 
@@ -383,7 +460,8 @@ guarantee that the values of st_ino and st_dev returned by stat(2) and the
 value of d_ino returned by readdir(3) will act like on a normal filesystem.
 E.g. the value of st_dev may be different for two objects in the same
 overlay filesystem and the value of st_ino for directory objects may not be
-persistent and could change even while the overlay filesystem is mounted.
+persistent and could change even while the overlay filesystem is mounted, as
+summarized in the `Inode properties`_ table above.
 
 
 Changes to underlying filesystems

fs/overlayfs/copy_up.c

@@ -36,6 +36,13 @@ static int ovl_ccup_get(char *buf, const struct kernel_param *param)
 module_param_call(check_copy_up, ovl_ccup_set, ovl_ccup_get, NULL, 0644);
 MODULE_PARM_DESC(check_copy_up, "Obsolete; does nothing");
 
+static bool ovl_must_copy_xattr(const char *name)
+{
+	return !strcmp(name, XATTR_POSIX_ACL_ACCESS) ||
+	       !strcmp(name, XATTR_POSIX_ACL_DEFAULT) ||
+	       !strncmp(name, XATTR_SECURITY_PREFIX, XATTR_SECURITY_PREFIX_LEN);
+}
+
 int ovl_copy_xattr(struct dentry *old, struct dentry *new)
 {
 	ssize_t list_size, size, value_size = 0;
@@ -107,8 +114,13 @@ int ovl_copy_xattr(struct dentry *old, struct dentry *new)
 			continue; /* Discard */
 		}
 		error = vfs_setxattr(new, name, value, size, 0);
-		if (error)
-			break;
+		if (error) {
+			if (error != -EOPNOTSUPP || ovl_must_copy_xattr(name))
+				break;
+
+			/* Ignore failure to copy unknown xattrs */
+			error = 0;
+		}
 	}
 	kfree(value);
 out:

fs/overlayfs/dir.c

@@ -42,7 +42,7 @@ int ovl_cleanup(struct inode *wdir, struct dentry *wdentry)
 	return err;
 }
 
-static struct dentry *ovl_lookup_temp(struct dentry *workdir)
+struct dentry *ovl_lookup_temp(struct dentry *workdir)
 {
 	struct dentry *temp;
 	char name[20];
@@ -243,6 +243,9 @@ static int ovl_instantiate(struct dentry *dentry, struct inode *inode,
 
 	ovl_dir_modified(dentry->d_parent, false);
 	ovl_dentry_set_upper_alias(dentry);
+	ovl_dentry_update_reval(dentry, newdentry,
+			DCACHE_OP_REVALIDATE | DCACHE_OP_WEAK_REVALIDATE);
+
 	if (!hardlink) {
 		/*
 		 * ovl_obtain_alias() can be called after ovl_create_real()
@@ -819,6 +822,28 @@ static bool ovl_pure_upper(struct dentry *dentry)
 	       !ovl_test_flag(OVL_WHITEOUTS, d_inode(dentry));
 }
 
+static void ovl_drop_nlink(struct dentry *dentry)
+{
+	struct inode *inode = d_inode(dentry);
+	struct dentry *alias;
+
+	/* Try to find another, hashed alias */
+	spin_lock(&inode->i_lock);
+	hlist_for_each_entry(alias, &inode->i_dentry, d_u.d_alias) {
+		if (alias != dentry && !d_unhashed(alias))
+			break;
+	}
+	spin_unlock(&inode->i_lock);
+
+	/*
+	 * Changes to underlying layers may cause i_nlink to lose sync with
+	 * reality.  In this case prevent the link count from going to zero
+	 * prematurely.
+	 */
+	if (inode->i_nlink > !!alias)
+		drop_nlink(inode);
+}
+
 static int ovl_do_remove(struct dentry *dentry, bool is_dir)
 {
 	int err;
@@ -856,7 +881,7 @@ static int ovl_do_remove(struct dentry *dentry, bool is_dir)
 		if (is_dir)
 			clear_nlink(dentry->d_inode);
 		else
-			drop_nlink(dentry->d_inode);
+			ovl_drop_nlink(dentry);
 	}
 	ovl_nlink_end(dentry);
 
@@ -1201,7 +1226,7 @@ static int ovl_rename(struct inode *olddir, struct dentry *old,
 		if (new_is_dir)
 			clear_nlink(d_inode(new));
 		else
-			drop_nlink(d_inode(new));
+			ovl_drop_nlink(new);
 	}
 
 	ovl_dir_modified(old->d_parent, ovl_type_origin(old) ||

fs/overlayfs/export.c

@@ -308,29 +308,35 @@ static struct dentry *ovl_obtain_alias(struct super_block *sb,
 		ovl_set_flag(OVL_UPPERDATA, inode);
 
 	dentry = d_find_any_alias(inode);
-	if (!dentry) {
-		dentry = d_alloc_anon(inode->i_sb);
-		if (!dentry)
-			goto nomem;
-		oe = ovl_alloc_entry(lower ? 1 : 0);
-		if (!oe)
-			goto nomem;
-
-		if (lower) {
-			oe->lowerstack->dentry = dget(lower);
-			oe->lowerstack->layer = lowerpath->layer;
-		}
-		dentry->d_fsdata = oe;
-		if (upper_alias)
-			ovl_dentry_set_upper_alias(dentry);
+	if (dentry)
+		goto out_iput;
+
+	dentry = d_alloc_anon(inode->i_sb);
+	if (unlikely(!dentry))
+		goto nomem;
+	oe = ovl_alloc_entry(lower ? 1 : 0);
+	if (!oe)
+		goto nomem;
+
+	if (lower) {
+		oe->lowerstack->dentry = dget(lower);
+		oe->lowerstack->layer = lowerpath->layer;
 	}
+	dentry->d_fsdata = oe;
+	if (upper_alias)
+		ovl_dentry_set_upper_alias(dentry);
+
+	ovl_dentry_update_reval(dentry, upper,
+			DCACHE_OP_REVALIDATE | DCACHE_OP_WEAK_REVALIDATE);
 
 	return d_instantiate_anon(dentry, inode);
 
 nomem:
-	iput(inode);
 	dput(dentry);
-	return ERR_PTR(-ENOMEM);
+	dentry = ERR_PTR(-ENOMEM);
+out_iput:
+	iput(inode);
+	return dentry;
 }
 
 /* Get the upper or lower dentry in stach whose on layer @idx */

fs/overlayfs/inode.c

@@ -79,6 +79,7 @@ static int ovl_map_dev_ino(struct dentry *dentry, struct kstat *stat, int fsid)
 {
 	bool samefs = ovl_same_fs(dentry->d_sb);
 	unsigned int xinobits = ovl_xino_bits(dentry->d_sb);
+	unsigned int xinoshift = 64 - xinobits;
 
 	if (samefs) {
 		/*
@@ -89,22 +90,22 @@ static int ovl_map_dev_ino(struct dentry *dentry, struct kstat *stat, int fsid)
 		stat->dev = dentry->d_sb->s_dev;
 		return 0;
 	} else if (xinobits) {
-		unsigned int shift = 64 - xinobits;
 		/*
 		 * All inode numbers of underlying fs should not be using the
 		 * high xinobits, so we use high xinobits to partition the
 		 * overlay st_ino address space. The high bits holds the fsid
-		 * (upper fsid is 0). This way overlay inode numbers are unique
-		 * and all inodes use overlay st_dev. Inode numbers are also
-		 * persistent for a given layer configuration.
+		 * (upper fsid is 0). The lowest xinobit is reserved for mapping
+		 * the non-peresistent inode numbers range in case of overflow.
+		 * This way all overlay inode numbers are unique and use the
+		 * overlay st_dev.
 		 */
-		if (stat->ino >> shift) {
-			pr_warn_ratelimited("inode number too big (%pd2, ino=%llu, xinobits=%d)\n",
-					    dentry, stat->ino, xinobits);
-		} else {
-			stat->ino |= ((u64)fsid) << shift;
+		if (likely(!(stat->ino >> xinoshift))) {
+			stat->ino |= ((u64)fsid) << (xinoshift + 1);
 			stat->dev = dentry->d_sb->s_dev;
 			return 0;
+		} else if (ovl_xino_warn(dentry->d_sb)) {
+			pr_warn_ratelimited("inode number too big (%pd2, ino=%llu, xinobits=%d)\n",
+					    dentry, stat->ino, xinobits);
 		}
 	}
 
@@ -504,7 +505,7 @@ static const struct address_space_operations ovl_aops = {
 
 /*
  * It is possible to stack overlayfs instance on top of another
- * overlayfs instance as lower layer. We need to annonate the
+ * overlayfs instance as lower layer. We need to annotate the
  * stackable i_mutex locks according to stack level of the super
  * block instance. An overlayfs instance can never be in stack
  * depth 0 (there is always a real fs below it).  An overlayfs
@@ -561,27 +562,73 @@ static inline void ovl_lockdep_annotate_inode_mutex_key(struct inode *inode)
 #endif
 }
 
-static void ovl_fill_inode(struct inode *inode, umode_t mode, dev_t rdev,
-			   unsigned long ino, int fsid)
+static void ovl_next_ino(struct inode *inode)
+{
+	struct ovl_fs *ofs = inode->i_sb->s_fs_info;
+
+	inode->i_ino = atomic_long_inc_return(&ofs->last_ino);
+	if (unlikely(!inode->i_ino))
+		inode->i_ino = atomic_long_inc_return(&ofs->last_ino);
+}
+
+static void ovl_map_ino(struct inode *inode, unsigned long ino, int fsid)
 {
 	int xinobits = ovl_xino_bits(inode->i_sb);
+	unsigned int xinoshift = 64 - xinobits;
 
 	/*
 	 * When d_ino is consistent with st_ino (samefs or i_ino has enough
 	 * bits to encode layer), set the same value used for st_ino to i_ino,
 	 * so inode number exposed via /proc/locks and a like will be
 	 * consistent with d_ino and st_ino values. An i_ino value inconsistent
-	 * with d_ino also causes nfsd readdirplus to fail.  When called from
-	 * ovl_new_inode(), ino arg is 0, so i_ino will be updated to real
-	 * upper inode i_ino on ovl_inode_init() or ovl_inode_update().
+	 * with d_ino also causes nfsd readdirplus to fail.
 	 */
-	if (ovl_same_dev(inode->i_sb)) {
-		inode->i_ino = ino;
-		if (xinobits && fsid && !(ino >> (64 - xinobits)))
-			inode->i_ino |= (unsigned long)fsid << (64 - xinobits);
-	} else {
-		inode->i_ino = get_next_ino();
+	inode->i_ino = ino;
+	if (ovl_same_fs(inode->i_sb)) {
+		return;
+	} else if (xinobits && likely(!(ino >> xinoshift))) {
+		inode->i_ino |= (unsigned long)fsid << (xinoshift + 1);
+		return;
+	}
+
+	/*
+	 * For directory inodes on non-samefs with xino disabled or xino
+	 * overflow, we allocate a non-persistent inode number, to be used for
+	 * resolving st_ino collisions in ovl_map_dev_ino().
+	 *
+	 * To avoid ino collision with legitimate xino values from upper
+	 * layer (fsid 0), use the lowest xinobit to map the non
+	 * persistent inode numbers to the unified st_ino address space.
+	 */
+	if (S_ISDIR(inode->i_mode)) {
+		ovl_next_ino(inode);
+		if (xinobits) {
+			inode->i_ino &= ~0UL >> xinobits;
+			inode->i_ino |= 1UL << xinoshift;
+		}
 	}
+}
+
+void ovl_inode_init(struct inode *inode, struct ovl_inode_params *oip,
+		    unsigned long ino, int fsid)
+{
+	struct inode *realinode;
+
+	if (oip->upperdentry)
+		OVL_I(inode)->__upperdentry = oip->upperdentry;
+	if (oip->lowerpath && oip->lowerpath->dentry)
+		OVL_I(inode)->lower = igrab(d_inode(oip->lowerpath->dentry));
+	if (oip->lowerdata)
+		OVL_I(inode)->lowerdata = igrab(d_inode(oip->lowerdata));
+
+	realinode = ovl_inode_real(inode);
+	ovl_copyattr(realinode, inode);
+	ovl_copyflags(realinode, inode);
+	ovl_map_ino(inode, ino, fsid);
+}
+
+static void ovl_fill_inode(struct inode *inode, umode_t mode, dev_t rdev)
+{
 	inode->i_mode = mode;
 	inode->i_flags |= S_NOCMTIME;
 #ifdef CONFIG_FS_POSIX_ACL
@@ -719,7 +766,7 @@ struct inode *ovl_new_inode(struct super_block *sb, umode_t mode, dev_t rdev)
 
 	inode = new_inode(sb);
 	if (inode)
-		ovl_fill_inode(inode, mode, rdev, 0, 0);
+		ovl_fill_inode(inode, mode, rdev);
 
 	return inode;
 }
@@ -891,7 +938,7 @@ struct inode *ovl_get_inode(struct super_block *sb,
 	struct dentry *lowerdentry = lowerpath ? lowerpath->dentry : NULL;
 	bool bylower = ovl_hash_bylower(sb, upperdentry, lowerdentry,
 					oip->index);
-	int fsid = bylower ? oip->lowerpath->layer->fsid : 0;
+	int fsid = bylower ? lowerpath->layer->fsid : 0;
 	bool is_dir, metacopy = false;
 	unsigned long ino = 0;
 	int err = oip->newinode ? -EEXIST : -ENOMEM;
@@ -941,9 +988,11 @@ struct inode *ovl_get_inode(struct super_block *sb,
 			err = -ENOMEM;
 			goto out_err;
 		}
+		ino = realinode->i_ino;
+		fsid = lowerpath->layer->fsid;
 	}
-	ovl_fill_inode(inode, realinode->i_mode, realinode->i_rdev, ino, fsid);
-	ovl_inode_init(inode, upperdentry, lowerdentry, oip->lowerdata);
+	ovl_fill_inode(inode, realinode->i_mode, realinode->i_rdev);
+	ovl_inode_init(inode, oip, ino, fsid);
 
 	if (upperdentry && ovl_is_impuredir(upperdentry))
 		ovl_set_flag(OVL_IMPURE, inode);

fs/overlayfs/namei.c

@@ -845,7 +845,7 @@ struct dentry *ovl_lookup(struct inode *dir, struct dentry *dentry,
 		if (err)
 			goto out;
 
-		if (upperdentry && unlikely(ovl_dentry_remote(upperdentry))) {
+		if (upperdentry && upperdentry->d_flags & DCACHE_OP_REAL) {
 			dput(upperdentry);
 			err = -EREMOTE;
 			goto out;
@@ -1076,6 +1076,9 @@ struct dentry *ovl_lookup(struct inode *dir, struct dentry *dentry,
 			goto out_free_oe;
 	}
 
+	ovl_dentry_update_reval(dentry, upperdentry,
+			DCACHE_OP_REVALIDATE | DCACHE_OP_WEAK_REVALIDATE);
+
 	revert_creds(old_cred);
 	if (origin_path) {
 		dput(origin_path->dentry);

fs/overlayfs/overlayfs.h

@@ -48,6 +48,12 @@ enum ovl_entry_flag {
 	OVL_E_CONNECTED,
 };
 
+enum {
+	OVL_XINO_OFF,
+	OVL_XINO_AUTO,
+	OVL_XINO_ON,
+};
+
 /*
  * The tuple (fh,uuid) is a universal unique identifier for a copy up origin,
  * where:
@@ -87,7 +93,7 @@ struct ovl_fb {
 	u8 flags;	/* OVL_FH_FLAG_* */
 	u8 type;	/* fid_type of fid */
 	uuid_t uuid;	/* uuid of filesystem */
-	u32 fid[0];	/* file identifier should be 32bit aligned in-memory */
+	u32 fid[];	/* file identifier should be 32bit aligned in-memory */
 } __packed;
 
 /* In-memory and on-wire format for overlay file handle */
@@ -230,6 +236,8 @@ bool ovl_index_all(struct super_block *sb);
 bool ovl_verify_lower(struct super_block *sb);
 struct ovl_entry *ovl_alloc_entry(unsigned int numlower);
 bool ovl_dentry_remote(struct dentry *dentry);
+void ovl_dentry_update_reval(struct dentry *dentry, struct dentry *upperdentry,
+			     unsigned int mask);
 bool ovl_dentry_weird(struct dentry *dentry);
 enum ovl_path_type ovl_path_type(struct dentry *dentry);
 void ovl_path_upper(struct dentry *dentry, struct path *path);
@@ -264,8 +272,6 @@ void ovl_set_upperdata(struct inode *inode);
 bool ovl_redirect_dir(struct super_block *sb);
 const char *ovl_dentry_get_redirect(struct dentry *dentry);
 void ovl_dentry_set_redirect(struct dentry *dentry, const char *redirect);
-void ovl_inode_init(struct inode *inode, struct dentry *upperdentry,
-		    struct dentry *lowerdentry, struct dentry *lowerdata);
 void ovl_inode_update(struct inode *inode, struct dentry *upperdentry);
 void ovl_dir_modified(struct dentry *dentry, bool impurity);
 u64 ovl_dentry_version_get(struct dentry *dentry);
@@ -301,6 +307,16 @@ static inline bool ovl_is_impuredir(struct dentry *dentry)
 	return ovl_check_dir_xattr(dentry, OVL_XATTR_IMPURE);
 }
 
+/*
+ * With xino=auto, we do best effort to keep all inodes on same st_dev and
+ * d_ino consistent with st_ino.
+ * With xino=on, we do the same effort but we warn if we failed.
+ */
+static inline bool ovl_xino_warn(struct super_block *sb)
+{
+	return OVL_FS(sb)->config.xino == OVL_XINO_ON;
+}
+
 /* All layers on same fs? */
 static inline bool ovl_same_fs(struct super_block *sb)
 {
@@ -410,6 +426,8 @@ struct ovl_inode_params {
 	char *redirect;
 	struct dentry *lowerdata;
 };
+void ovl_inode_init(struct inode *inode, struct ovl_inode_params *oip,
+		    unsigned long ino, int fsid);
 struct inode *ovl_new_inode(struct super_block *sb, umode_t mode, dev_t rdev);
 struct inode *ovl_lookup_inode(struct super_block *sb, struct dentry *real,
 			       bool is_upper);
@@ -451,6 +469,7 @@ struct ovl_cattr {
 struct dentry *ovl_create_real(struct inode *dir, struct dentry *newdentry,
 			       struct ovl_cattr *attr);
 int ovl_cleanup(struct inode *dir, struct dentry *dentry);
+struct dentry *ovl_lookup_temp(struct dentry *workdir);
 struct dentry *ovl_create_temp(struct dentry *workdir, struct ovl_cattr *attr);
 
 /* file.c */

fs/overlayfs/ovl_entry.h

@@ -75,6 +75,8 @@ struct ovl_fs {
 	struct inode *indexdir_trap;
 	/* -1: disabled, 0: same fs, 1..32: number of unused ino bits */
 	int xino_mode;
+	/* For allocation of non-persistent inode numbers */
+	atomic_long_t last_ino;
 };
 
 static inline struct ovl_fs *OVL_FS(struct super_block *sb)

fs/overlayfs/readdir.c

@@ -438,15 +438,23 @@ static struct ovl_dir_cache *ovl_cache_get(struct dentry *dentry)
 
 /* Map inode number to lower fs unique range */
 static u64 ovl_remap_lower_ino(u64 ino, int xinobits, int fsid,
-			       const char *name, int namelen)
+			       const char *name, int namelen, bool warn)
 {
-	if (ino >> (64 - xinobits)) {
-		pr_warn_ratelimited("d_ino too big (%.*s, ino=%llu, xinobits=%d)\n",
-				    namelen, name, ino, xinobits);
+	unsigned int xinoshift = 64 - xinobits;
+
+	if (unlikely(ino >> xinoshift)) {
+		if (warn) {
+			pr_warn_ratelimited("d_ino too big (%.*s, ino=%llu, xinobits=%d)\n",
+					    namelen, name, ino, xinobits);
+		}
 		return ino;
 	}
 
-	return ino | ((u64)fsid) << (64 - xinobits);
+	/*
+	 * The lowest xinobit is reserved for mapping the non-peresistent inode
+	 * numbers range, but this range is only exposed via st_ino, not here.
+	 */
+	return ino | ((u64)fsid) << (xinoshift + 1);
 }
 
 /*
@@ -515,7 +523,8 @@ static int ovl_cache_update_ino(struct path *path, struct ovl_cache_entry *p)
 	} else if (xinobits && !OVL_TYPE_UPPER(type)) {
 		ino = ovl_remap_lower_ino(ino, xinobits,
 					  ovl_layer_lower(this)->fsid,
-					  p->name, p->len);
+					  p->name, p->len,
+					  ovl_xino_warn(dir->d_sb));
 	}
 
 out:
@@ -645,6 +654,7 @@ struct ovl_readdir_translate {
 	u64 parent_ino;
 	int fsid;
 	int xinobits;
+	bool xinowarn;
 };
 
 static int ovl_fill_real(struct dir_context *ctx, const char *name,
@@ -665,7 +675,7 @@ static int ovl_fill_real(struct dir_context *ctx, const char *name,
 			ino = p->ino;
 	} else if (rdt->xinobits) {
 		ino = ovl_remap_lower_ino(ino, rdt->xinobits, rdt->fsid,
-					  name, namelen);
+					  name, namelen, rdt->xinowarn);
 	}
 
 	return orig_ctx->actor(orig_ctx, name, namelen, offset, ino, d_type);
@@ -696,6 +706,7 @@ static int ovl_iterate_real(struct file *file, struct dir_context *ctx)
 		.ctx.actor = ovl_fill_real,
 		.orig_ctx = ctx,
 		.xinobits = ovl_xino_bits(dir->d_sb),
+		.xinowarn = ovl_xino_warn(dir->d_sb),
 	};
 
 	if (rdt.xinobits && lower_layer)

fs/overlayfs/super.c

@@ -113,53 +113,54 @@ static struct dentry *ovl_d_real(struct dentry *dentry,
 	return dentry;
 }
 
-static int ovl_dentry_revalidate(struct dentry *dentry, unsigned int flags)
+static int ovl_revalidate_real(struct dentry *d, unsigned int flags, bool weak)
 {
-	struct ovl_entry *oe = dentry->d_fsdata;
-	unsigned int i;
 	int ret = 1;
 
-	for (i = 0; i < oe->numlower; i++) {
-		struct dentry *d = oe->lowerstack[i].dentry;
-
-		if (d->d_flags & DCACHE_OP_REVALIDATE) {
-			ret = d->d_op->d_revalidate(d, flags);
-			if (ret < 0)
-				return ret;
-			if (!ret) {
-				if (!(flags & LOOKUP_RCU))
-					d_invalidate(d);
-				return -ESTALE;
-			}
+	if (weak) {
+		if (d->d_flags & DCACHE_OP_WEAK_REVALIDATE)
+			ret =  d->d_op->d_weak_revalidate(d, flags);
+	} else if (d->d_flags & DCACHE_OP_REVALIDATE) {
+		ret = d->d_op->d_revalidate(d, flags);
+		if (!ret) {
+			if (!(flags & LOOKUP_RCU))
+				d_invalidate(d);
+			ret = -ESTALE;
 		}
 	}
-	return 1;
+	return ret;
 }
 
-static int ovl_dentry_weak_revalidate(struct dentry *dentry, unsigned int flags)
+static int ovl_dentry_revalidate_common(struct dentry *dentry,
+					unsigned int flags, bool weak)
 {
 	struct ovl_entry *oe = dentry->d_fsdata;
+	struct dentry *upper;
 	unsigned int i;
 	int ret = 1;
 
-	for (i = 0; i < oe->numlower; i++) {
-		struct dentry *d = oe->lowerstack[i].dentry;
+	upper = ovl_dentry_upper(dentry);
+	if (upper)
+		ret = ovl_revalidate_real(upper, flags, weak);
 
-		if (d->d_flags & DCACHE_OP_WEAK_REVALIDATE) {
-			ret = d->d_op->d_weak_revalidate(d, flags);
-			if (ret <= 0)
-				break;
-		}
+	for (i = 0; ret > 0 && i < oe->numlower; i++) {
+		ret = ovl_revalidate_real(oe->lowerstack[i].dentry, flags,
+					  weak);
 	}
 	return ret;
 }
 
-static const struct dentry_operations ovl_dentry_operations = {
-	.d_release = ovl_dentry_release,
-	.d_real = ovl_d_real,
-};
+static int ovl_dentry_revalidate(struct dentry *dentry, unsigned int flags)
+{
+	return ovl_dentry_revalidate_common(dentry, flags, false);
+}
+
+static int ovl_dentry_weak_revalidate(struct dentry *dentry, unsigned int flags)
+{
+	return ovl_dentry_revalidate_common(dentry, flags, true);
+}
 
-static const struct dentry_operations ovl_reval_dentry_operations = {
+static const struct dentry_operations ovl_dentry_operations = {
 	.d_release = ovl_dentry_release,
 	.d_real = ovl_d_real,
 	.d_revalidate = ovl_dentry_revalidate,
@@ -316,12 +317,6 @@ static const char *ovl_redirect_mode_def(void)
 	return ovl_redirect_dir_def ? "on" : "off";
 }
 
-enum {
-	OVL_XINO_OFF,
-	OVL_XINO_AUTO,
-	OVL_XINO_ON,
-};
-
 static const char * const ovl_xino_str[] = {
 	"off",
 	"auto",
@@ -751,13 +746,12 @@ static int ovl_mount_dir(const char *name, struct path *path)
 		ovl_unescape(tmp);
 		err = ovl_mount_dir_noesc(tmp, path);
 
-		if (!err)
-			if (ovl_dentry_remote(path->dentry)) {
-				pr_err("filesystem on '%s' not supported as upperdir\n",
-				       tmp);
-				path_put_init(path);
-				err = -EINVAL;
-			}
+		if (!err && path->dentry->d_flags & DCACHE_OP_REAL) {
+			pr_err("filesystem on '%s' not supported as upperdir\n",
+			       tmp);
+			path_put_init(path);
+			err = -EINVAL;
+		}
 		kfree(tmp);
 	}
 	return err;
@@ -778,7 +772,7 @@ static int ovl_check_namelen(struct path *path, struct ovl_fs *ofs,
 }
 
 static int ovl_lower_dir(const char *name, struct path *path,
-			 struct ovl_fs *ofs, int *stack_depth, bool *remote)
+			 struct ovl_fs *ofs, int *stack_depth)
 {
 	int fh_type;
 	int err;
@@ -793,9 +787,6 @@ static int ovl_lower_dir(const char *name, struct path *path,
 
 	*stack_depth = max(*stack_depth, path->mnt->mnt_sb->s_stack_depth);
 
-	if (ovl_dentry_remote(path->dentry))
-		*remote = true;
-
 	/*
 	 * The inodes index feature and NFS export need to encode and decode
 	 * file handles, so they require that all layers support them.
@@ -1074,11 +1065,73 @@ static int ovl_get_upper(struct super_block *sb, struct ovl_fs *ofs,
 	return err;
 }
 
+/*
+ * Returns 1 if RENAME_WHITEOUT is supported, 0 if not supported and
+ * negative values if error is encountered.
+ */
+static int ovl_check_rename_whiteout(struct dentry *workdir)
+{
+	struct inode *dir = d_inode(workdir);
+	struct dentry *temp;
+	struct dentry *dest;
+	struct dentry *whiteout;
+	struct name_snapshot name;
+	int err;
+
+	inode_lock_nested(dir, I_MUTEX_PARENT);
+
+	temp = ovl_create_temp(workdir, OVL_CATTR(S_IFREG | 0));
+	err = PTR_ERR(temp);
+	if (IS_ERR(temp))
+		goto out_unlock;
+
+	dest = ovl_lookup_temp(workdir);
+	err = PTR_ERR(dest);
+	if (IS_ERR(dest)) {
+		dput(temp);
+		goto out_unlock;
+	}
+
+	/* Name is inline and stable - using snapshot as a copy helper */
+	take_dentry_name_snapshot(&name, temp);
+	err = ovl_do_rename(dir, temp, dir, dest, RENAME_WHITEOUT);
+	if (err) {
+		if (err == -EINVAL)
+			err = 0;
+		goto cleanup_temp;
+	}
+
+	whiteout = lookup_one_len(name.name.name, workdir, name.name.len);
+	err = PTR_ERR(whiteout);
+	if (IS_ERR(whiteout))
+		goto cleanup_temp;
+
+	err = ovl_is_whiteout(whiteout);
+
+	/* Best effort cleanup of whiteout and temp file */
+	if (err)
+		ovl_cleanup(dir, whiteout);
+	dput(whiteout);
+
+cleanup_temp:
+	ovl_cleanup(dir, temp);
+	release_dentry_name_snapshot(&name);
+	dput(temp);
+	dput(dest);
+
+out_unlock:
+	inode_unlock(dir);
+
+	return err;
+}
+
 static int ovl_make_workdir(struct super_block *sb, struct ovl_fs *ofs,
 			    struct path *workpath)
 {
 	struct vfsmount *mnt = ofs->upper_mnt;
 	struct dentry *temp;
+	bool rename_whiteout;
+	bool d_type;
 	int fh_type;
 	int err;
 
@@ -1104,11 +1157,8 @@ static int ovl_make_workdir(struct super_block *sb, struct ovl_fs *ofs,
 	if (err < 0)
 		goto out;
 
-	/*
-	 * We allowed this configuration and don't want to break users over
-	 * kernel upgrade. So warn instead of erroring out.
-	 */
-	if (!err)
+	d_type = err;
+	if (!d_type)
 		pr_warn("upper fs needs to support d_type.\n");
 
 	/* Check if upper/work fs supports O_TMPFILE */
@@ -1119,6 +1169,16 @@ static int ovl_make_workdir(struct super_block *sb, struct ovl_fs *ofs,
 	else
 		pr_warn("upper fs does not support tmpfile.\n");
 
+
+	/* Check if upper/work fs supports RENAME_WHITEOUT */
+	err = ovl_check_rename_whiteout(ofs->workdir);
+	if (err < 0)
+		goto out;
+
+	rename_whiteout = err;
+	if (!rename_whiteout)
+		pr_warn("upper fs does not support RENAME_WHITEOUT.\n");
+
 	/*
 	 * Check if upper/work fs supports trusted.overlay.* xattr
 	 */
@@ -1133,6 +1193,18 @@ static int ovl_make_workdir(struct super_block *sb, struct ovl_fs *ofs,
 		vfs_removexattr(ofs->workdir, OVL_XATTR_OPAQUE);
 	}
 
+	/*
+	 * We allowed sub-optimal upper fs configuration and don't want to break
+	 * users over kernel upgrade, but we never allowed remote upper fs, so
+	 * we can enforce strict requirements for remote upper fs.
+	 */
+	if (ovl_dentry_remote(ofs->workdir) &&
+	    (!d_type || !rename_whiteout || ofs->noxattr)) {
+		pr_err("upper fs missing required features.\n");
+		err = -EINVAL;
+		goto out;
+	}
+
 	/* Check if upper/work fs supports file handles */
 	fh_type = ovl_can_decode_fh(ofs->workdir->d_sb);
 	if (ofs->config.index && !fh_type) {
@@ -1401,11 +1473,12 @@ static int ovl_get_layers(struct super_block *sb, struct ovl_fs *ofs,
 
 	/*
 	 * When all layers on same fs, overlay can use real inode numbers.
-	 * With mount option "xino=on", mounter declares that there are enough
-	 * free high bits in underlying fs to hold the unique fsid.
+	 * With mount option "xino=<on|auto>", mounter declares that there are
+	 * enough free high bits in underlying fs to hold the unique fsid.
 	 * If overlayfs does encounter underlying inodes using the high xino
 	 * bits reserved for fsid, it emits a warning and uses the original
-	 * inode number.
+	 * inode number or a non persistent inode number allocated from a
+	 * dedicated range.
 	 */
 	if (ofs->numfs - !ofs->upper_mnt == 1) {
 		if (ofs->config.xino == OVL_XINO_ON)
@@ -1413,14 +1486,16 @@ static int ovl_get_layers(struct super_block *sb, struct ovl_fs *ofs,
 		ofs->xino_mode = 0;
 	} else if (ofs->config.xino == OVL_XINO_OFF) {
 		ofs->xino_mode = -1;
-	} else if (ofs->config.xino == OVL_XINO_ON && ofs->xino_mode < 0) {
+	} else if (ofs->xino_mode < 0) {
 		/*
 		 * This is a roundup of number of bits needed for encoding
-		 * fsid, where fsid 0 is reserved for upper fs even with
-		 * lower only overlay.
+		 * fsid, where fsid 0 is reserved for upper fs (even with
+		 * lower only overlay) +1 extra bit is reserved for the non
+		 * persistent inode number range that is used for resolving
+		 * xino lower bits overflow.
 		 */
-		BUILD_BUG_ON(ilog2(OVL_MAX_STACK) > 31);
-		ofs->xino_mode = ilog2(ofs->numfs - 1) + 1;
+		BUILD_BUG_ON(ilog2(OVL_MAX_STACK) > 30);
+		ofs->xino_mode = ilog2(ofs->numfs - 1) + 2;
 	}
 
 	if (ofs->xino_mode > 0) {
@@ -1440,7 +1515,6 @@ static struct ovl_entry *ovl_get_lowerstack(struct super_block *sb,
 	char *lowertmp, *lower;
 	struct path *stack = NULL;
 	unsigned int stacklen, numlower = 0, i;
-	bool remote = false;
 	struct ovl_entry *oe;
 
 	err = -ENOMEM;
@@ -1472,7 +1546,7 @@ static struct ovl_entry *ovl_get_lowerstack(struct super_block *sb,
 	lower = lowertmp;
 	for (numlower = 0; numlower < stacklen; numlower++) {
 		err = ovl_lower_dir(lower, &stack[numlower], ofs,
-				    &sb->s_stack_depth, &remote);
+				    &sb->s_stack_depth);
 		if (err)
 			goto out_err;
 
@@ -1500,11 +1574,6 @@ static struct ovl_entry *ovl_get_lowerstack(struct super_block *sb,
 		oe->lowerstack[i].layer = &ofs->layers[i+1];
 	}
 
-	if (remote)
-		sb->s_d_op = &ovl_reval_dentry_operations;
-	else
-		sb->s_d_op = &ovl_dentry_operations;
-
 out:
 	for (i = 0; i < numlower; i++)
 		path_put(&stack[i]);
@@ -1589,6 +1658,44 @@ static int ovl_check_overlapping_layers(struct super_block *sb,
 	return 0;
 }
 
+static struct dentry *ovl_get_root(struct super_block *sb,
+				   struct dentry *upperdentry,
+				   struct ovl_entry *oe)
+{
+	struct dentry *root;
+	struct ovl_path *lowerpath = &oe->lowerstack[0];
+	unsigned long ino = d_inode(lowerpath->dentry)->i_ino;
+	int fsid = lowerpath->layer->fsid;
+	struct ovl_inode_params oip = {
+		.upperdentry = upperdentry,
+		.lowerpath = lowerpath,
+	};
+
+	root = d_make_root(ovl_new_inode(sb, S_IFDIR, 0));
+	if (!root)
+		return NULL;
+
+	root->d_fsdata = oe;
+
+	if (upperdentry) {
+		/* Root inode uses upper st_ino/i_ino */
+		ino = d_inode(upperdentry)->i_ino;
+		fsid = 0;
+		ovl_dentry_set_upper_alias(root);
+		if (ovl_is_impuredir(upperdentry))
+			ovl_set_flag(OVL_IMPURE, d_inode(root));
+	}
+
+	/* Root is always merge -> can have whiteouts */
+	ovl_set_flag(OVL_WHITEOUTS, d_inode(root));
+	ovl_dentry_set_flag(OVL_E_CONNECTED, root);
+	ovl_set_upperdata(d_inode(root));
+	ovl_inode_init(d_inode(root), &oip, ino, fsid);
+	ovl_dentry_update_reval(root, upperdentry, DCACHE_OP_WEAK_REVALIDATE);
+
+	return root;
+}
+
 static int ovl_fill_super(struct super_block *sb, void *data, int silent)
 {
 	struct path upperpath = { };
@@ -1598,6 +1705,8 @@ static int ovl_fill_super(struct super_block *sb, void *data, int silent)
 	struct cred *cred;
 	int err;
 
+	sb->s_d_op = &ovl_dentry_operations;
+
 	err = -ENOMEM;
 	ofs = kzalloc(sizeof(struct ovl_fs), GFP_KERNEL);
 	if (!ofs)
@@ -1624,6 +1733,7 @@ static int ovl_fill_super(struct super_block *sb, void *data, int silent)
 
 	sb->s_stack_depth = 0;
 	sb->s_maxbytes = MAX_LFS_FILESIZE;
+	atomic_long_set(&ofs->last_ino, 1);
 	/* Assume underlaying fs uses 32bit inodes unless proven otherwise */
 	if (ofs->config.xino != OVL_XINO_OFF) {
 		ofs->xino_mode = BITS_PER_LONG - 32;
@@ -1710,25 +1820,11 @@ static int ovl_fill_super(struct super_block *sb, void *data, int silent)
 	sb->s_flags |= SB_POSIXACL;
 
 	err = -ENOMEM;
-	root_dentry = d_make_root(ovl_new_inode(sb, S_IFDIR, 0));
+	root_dentry = ovl_get_root(sb, upperpath.dentry, oe);
 	if (!root_dentry)
 		goto out_free_oe;
 
-	root_dentry->d_fsdata = oe;
-
 	mntput(upperpath.mnt);
-	if (upperpath.dentry) {
-		ovl_dentry_set_upper_alias(root_dentry);
-		if (ovl_is_impuredir(upperpath.dentry))
-			ovl_set_flag(OVL_IMPURE, d_inode(root_dentry));
-	}
-
-	/* Root is always merge -> can have whiteouts */
-	ovl_set_flag(OVL_WHITEOUTS, d_inode(root_dentry));
-	ovl_dentry_set_flag(OVL_E_CONNECTED, root_dentry);
-	ovl_set_upperdata(d_inode(root_dentry));
-	ovl_inode_init(d_inode(root_dentry), upperpath.dentry,
-		       ovl_dentry_lower(root_dentry), NULL);
 
 	sb->s_root = root_dentry;
 

fs/overlayfs/util.c

@@ -93,8 +93,24 @@ struct ovl_entry *ovl_alloc_entry(unsigned int numlower)
 bool ovl_dentry_remote(struct dentry *dentry)
 {
 	return dentry->d_flags &
-		(DCACHE_OP_REVALIDATE | DCACHE_OP_WEAK_REVALIDATE |
-		 DCACHE_OP_REAL);
+		(DCACHE_OP_REVALIDATE | DCACHE_OP_WEAK_REVALIDATE);
+}
+
+void ovl_dentry_update_reval(struct dentry *dentry, struct dentry *upperdentry,
+			     unsigned int mask)
+{
+	struct ovl_entry *oe = OVL_E(dentry);
+	unsigned int i, flags = 0;
+
+	if (upperdentry)
+		flags |= upperdentry->d_flags;
+	for (i = 0; i < oe->numlower; i++)
+		flags |= oe->lowerstack[i].dentry->d_flags;
+
+	spin_lock(&dentry->d_lock);
+	dentry->d_flags &= ~mask;
+	dentry->d_flags |= flags & mask;
+	spin_unlock(&dentry->d_lock);
 }
 
 bool ovl_dentry_weird(struct dentry *dentry)
@@ -386,24 +402,6 @@ void ovl_dentry_set_redirect(struct dentry *dentry, const char *redirect)
 	oi->redirect = redirect;
 }
 
-void ovl_inode_init(struct inode *inode, struct dentry *upperdentry,
-		    struct dentry *lowerdentry, struct dentry *lowerdata)
-{
-	struct inode *realinode = d_inode(upperdentry ?: lowerdentry);
-
-	if (upperdentry)
-		OVL_I(inode)->__upperdentry = upperdentry;
-	if (lowerdentry)
-		OVL_I(inode)->lower = igrab(d_inode(lowerdentry));
-	if (lowerdata)
-		OVL_I(inode)->lowerdata = igrab(d_inode(lowerdata));
-
-	ovl_copyattr(realinode, inode);
-	ovl_copyflags(realinode, inode);
-	if (!inode->i_ino)
-		inode->i_ino = realinode->i_ino;
-}
-
 void ovl_inode_update(struct inode *inode, struct dentry *upperdentry)
 {
 	struct inode *upperinode = d_inode(upperdentry);
@@ -416,8 +414,6 @@ void ovl_inode_update(struct inode *inode, struct dentry *upperdentry)
 	smp_wmb();
 	OVL_I(inode)->__upperdentry = upperdentry;
 	if (inode_unhashed(inode)) {
-		if (!inode->i_ino)
-			inode->i_ino = upperinode->i_ino;
 		inode->i_private = upperinode;
 		__insert_inode_hash(inode, (unsigned long) upperinode);
 	}