Merge tag 'ovl-fixes-5.3' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs

Pull overlayfs fixes from Miklos Szeredi:
 "Fix a regression in docker introduced by overlayfs changes in 4.19.
  Also fix a couple of miscellaneous bugs"

* tag 'ovl-fixes-5.3' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs:
  ovl: filter of trusted xattr results in audit
  ovl: Fix dereferencing possible ERR_PTR()
  ovl: fix regression caused by overlapping layers detection

Documentation/filesystems/overlayfs.txt

@@ -302,7 +302,7 @@ beneath or above the path of another overlay lower layer path.
 
 Using an upper layer path and/or a workdir path that are already used by
 another overlay mount is not allowed and may fail with EBUSY.  Using
-partially overlapping paths is not allowed but will not fail with EBUSY.
+partially overlapping paths is not allowed and may fail with EBUSY.
 If files are accessed from two overlayfs mounts which share or overlap the
 upper layer and/or workdir path the behavior of the overlay is undefined,
 though it will not result in a crash or deadlock.

fs/overlayfs/export.c

@@ -227,9 +227,8 @@ static int ovl_d_to_fh(struct dentry *dentry, char *buf, int buflen)
 	/* Encode an upper or lower file handle */
 	fh = ovl_encode_real_fh(enc_lower ? ovl_dentry_lower(dentry) :
 				ovl_dentry_upper(dentry), !enc_lower);
-	err = PTR_ERR(fh);
 	if (IS_ERR(fh))
-		goto fail;
+		return PTR_ERR(fh);
 
 	err = -EOVERFLOW;
 	if (fh->len > buflen)

fs/overlayfs/inode.c

@@ -383,7 +383,8 @@ static bool ovl_can_list(const char *s)
 		return true;
 
 	/* Never list trusted.overlay, list other trusted for superuser only */
-	return !ovl_is_private_xattr(s) && capable(CAP_SYS_ADMIN);
+	return !ovl_is_private_xattr(s) &&
+	       ns_capable_noaudit(&init_user_ns, CAP_SYS_ADMIN);
 }
 
 ssize_t ovl_listxattr(struct dentry *dentry, char *list, size_t size)

fs/overlayfs/ovl_entry.h

@@ -66,6 +66,7 @@ struct ovl_fs {
 	bool workdir_locked;
 	/* Traps in ovl inode cache */
 	struct inode *upperdir_trap;
+	struct inode *workbasedir_trap;
 	struct inode *workdir_trap;
 	struct inode *indexdir_trap;
 	/* Inode numbers in all layers do not use the high xino_bits */

fs/overlayfs/super.c

@@ -212,6 +212,7 @@ static void ovl_free_fs(struct ovl_fs *ofs)
 {
 	unsigned i;
 
+	iput(ofs->workbasedir_trap);
 	iput(ofs->indexdir_trap);
 	iput(ofs->workdir_trap);
 	iput(ofs->upperdir_trap);
@@ -1003,6 +1004,25 @@ static int ovl_setup_trap(struct super_block *sb, struct dentry *dir,
 	return 0;
 }
 
+/*
+ * Determine how we treat concurrent use of upperdir/workdir based on the
+ * index feature. This is papering over mount leaks of container runtimes,
+ * for example, an old overlay mount is leaked and now its upperdir is
+ * attempted to be used as a lower layer in a new overlay mount.
+ */
+static int ovl_report_in_use(struct ovl_fs *ofs, const char *name)
+{
+	if (ofs->config.index) {
+		pr_err("overlayfs: %s is in-use as upperdir/workdir of another mount, mount with '-o index=off' to override exclusive upperdir protection.\n",
+		       name);
+		return -EBUSY;
+	} else {
+		pr_warn("overlayfs: %s is in-use as upperdir/workdir of another mount, accessing files from both mounts will result in undefined behavior.\n",
+			name);
+		return 0;
+	}
+}
+
 static int ovl_get_upper(struct super_block *sb, struct ovl_fs *ofs,
 			 struct path *upperpath)
 {
@@ -1040,14 +1060,12 @@ static int ovl_get_upper(struct super_block *sb, struct ovl_fs *ofs,
 	upper_mnt->mnt_flags &= ~(MNT_NOATIME | MNT_NODIRATIME | MNT_RELATIME);
 	ofs->upper_mnt = upper_mnt;
 
-	err = -EBUSY;
 	if (ovl_inuse_trylock(ofs->upper_mnt->mnt_root)) {
 		ofs->upperdir_locked = true;
-	} else if (ofs->config.index) {
-		pr_err("overlayfs: upperdir is in-use by another mount, mount with '-o index=off' to override exclusive upperdir protection.\n");
-		goto out;
 	} else {
-		pr_warn("overlayfs: upperdir is in-use by another mount, accessing files from both mounts will result in undefined behavior.\n");
+		err = ovl_report_in_use(ofs, "upperdir");
+		if (err)
+			goto out;
 	}
 
 	err = 0;
@@ -1157,16 +1175,19 @@ static int ovl_get_workdir(struct super_block *sb, struct ovl_fs *ofs,
 
 	ofs->workbasedir = dget(workpath.dentry);
 
-	err = -EBUSY;
 	if (ovl_inuse_trylock(ofs->workbasedir)) {
 		ofs->workdir_locked = true;
-	} else if (ofs->config.index) {
-		pr_err("overlayfs: workdir is in-use by another mount, mount with '-o index=off' to override exclusive workdir protection.\n");
-		goto out;
 	} else {
-		pr_warn("overlayfs: workdir is in-use by another mount, accessing files from both mounts will result in undefined behavior.\n");
+		err = ovl_report_in_use(ofs, "workdir");
+		if (err)
+			goto out;
 	}
 
+	err = ovl_setup_trap(sb, ofs->workbasedir, &ofs->workbasedir_trap,
+			     "workdir");
+	if (err)
+		goto out;
+
 	err = ovl_make_workdir(sb, ofs, &workpath);
 
 out:
@@ -1313,16 +1334,16 @@ static int ovl_get_lower_layers(struct super_block *sb, struct ovl_fs *ofs,
 		if (err < 0)
 			goto out;
 
-		err = -EBUSY;
-		if (ovl_is_inuse(stack[i].dentry)) {
-			pr_err("overlayfs: lowerdir is in-use as upperdir/workdir\n");
-			goto out;
-		}
-
 		err = ovl_setup_trap(sb, stack[i].dentry, &trap, "lowerdir");
 		if (err)
 			goto out;
 
+		if (ovl_is_inuse(stack[i].dentry)) {
+			err = ovl_report_in_use(ofs, "lowerdir");
+			if (err)
+				goto out;
+		}
+
 		mnt = clone_private_mount(&stack[i]);
 		err = PTR_ERR(mnt);
 		if (IS_ERR(mnt)) {
@@ -1469,8 +1490,8 @@ static struct ovl_entry *ovl_get_lowerstack(struct super_block *sb,
  * - another layer of this overlayfs instance
  * - upper/work dir of any overlayfs instance
  */
-static int ovl_check_layer(struct super_block *sb, struct dentry *dentry,
-			   const char *name)
+static int ovl_check_layer(struct super_block *sb, struct ovl_fs *ofs,
+			   struct dentry *dentry, const char *name)
 {
 	struct dentry *next = dentry, *parent;
 	int err = 0;
@@ -1482,13 +1503,11 @@ static int ovl_check_layer(struct super_block *sb, struct dentry *dentry,
 
 	/* Walk back ancestors to root (inclusive) looking for traps */
 	while (!err && parent != next) {
-		if (ovl_is_inuse(parent)) {
-			err = -EBUSY;
-			pr_err("overlayfs: %s path overlapping in-use upperdir/workdir\n",
-			       name);
-		} else if (ovl_lookup_trap_inode(sb, parent)) {
+		if (ovl_lookup_trap_inode(sb, parent)) {
 			err = -ELOOP;
 			pr_err("overlayfs: overlapping %s path\n", name);
+		} else if (ovl_is_inuse(parent)) {
+			err = ovl_report_in_use(ofs, name);
 		}
 		next = parent;
 		parent = dget_parent(next);
@@ -1509,7 +1528,8 @@ static int ovl_check_overlapping_layers(struct super_block *sb,
 	int i, err;
 
 	if (ofs->upper_mnt) {
-		err = ovl_check_layer(sb, ofs->upper_mnt->mnt_root, "upperdir");
+		err = ovl_check_layer(sb, ofs, ofs->upper_mnt->mnt_root,
+				      "upperdir");
 		if (err)
 			return err;
 
@@ -1520,13 +1540,14 @@ static int ovl_check_overlapping_layers(struct super_block *sb,
 		 * workbasedir.  In that case, we already have their traps in
 		 * inode cache and we will catch that case on lookup.
 		 */
-		err = ovl_check_layer(sb, ofs->workbasedir, "workdir");
+		err = ovl_check_layer(sb, ofs, ofs->workbasedir, "workdir");
 		if (err)
 			return err;
 	}
 
 	for (i = 0; i < ofs->numlower; i++) {
-		err = ovl_check_layer(sb, ofs->lower_layers[i].mnt->mnt_root,
+		err = ovl_check_layer(sb, ofs,
+				      ofs->lower_layers[i].mnt->mnt_root,
 				      "lowerdir");
 		if (err)
 			return err;