Merge tag 'ovl-fixes-5.5-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs

Pull overlayfs fixes from Miklos Szeredi:
"Fix some bugs and documentation"

* tag 'ovl-fixes-5.5-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs:
docs: filesystems: overlayfs: Fix restview warnings
docs: filesystems: overlayfs: Rename overlayfs.txt to .rst
ovl: relax WARN_ON() on rename to self
ovl: fix corner case of non-unique st_dev;st_ino
ovl: don't use a temp buf for encoding real fh
ovl: make sure that real fid is 32bit aligned in memory
ovl: fix lookup failure on multi lower squashfs

Documentation/filesystems/overlayfs.txt Documentation/filesystems/overlayfs.rst

···
+.. SPDX-License-Identifier: GPL-2.0
+
 Written by: Neil Brown
 Please see MAINTAINERS file for where to send questions.
 
···
     worried about backward compatibility with kernels that have the redirect_dir
     feature and follow redirects even if turned off.
 
-Module options (can also be changed through /sys/module/overlay/parameters/*):
+Module options (can also be changed through /sys/module/overlay/parameters/):
 
 - "redirect_dir=BOOL":
     See OVERLAY_FS_REDIRECT_DIR kernel config option above.
···
 
 
 Metadata only copy up
---------------------
+---------------------
 
 When metadata only copy up feature is enabled, overlayfs will only copy
 up metadata (as opposed to whole file), when a metadata specific operation
···
 "trusted." xattrs will require CAP_SYS_ADMIN. But it should be possible
 for untrusted layers like from a pen drive.
 
-Note: redirect_dir={off|nofollow|follow(*)} conflicts with metacopy=on, and
+Note: redirect_dir={off|nofollow|follow[*]} conflicts with metacopy=on, and
 results in an error.
 
-(*) redirect_dir=follow only conflicts with metacopy=on if upperdir=... is
+[*] redirect_dir=follow only conflicts with metacopy=on if upperdir=... is
 given.
 
 Sharing and copying layers

MAINTAINERS

···
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs.git
 S:	Supported
 F:	fs/overlayfs/
-F:	Documentation/filesystems/overlayfs.txt
+F:	Documentation/filesystems/overlayfs.rst
 
 P54 WIRELESS DRIVER
 M:	Christian Lamparter <[e-mail]>

fs/overlayfs/copy_up.c

···
 struct ovl_fh *ovl_encode_real_fh(struct dentry *real, bool is_upper)
 {
 	struct ovl_fh *fh;
-	int fh_type, fh_len, dwords;
-	void *buf;
+	int fh_type, dwords;
 	int buflen = MAX_HANDLE_SZ;
 	uuid_t *uuid = &real->d_sb->s_uuid;
+	int err;
 
-	buf = kmalloc(buflen, GFP_KERNEL);
-	if (!buf)
+	/* Make sure the real fid stays 32bit aligned */
+	BUILD_BUG_ON(OVL_FH_FID_OFFSET % 4);
+	BUILD_BUG_ON(MAX_HANDLE_SZ + OVL_FH_FID_OFFSET > 255);
+
+	fh = kzalloc(buflen + OVL_FH_FID_OFFSET, GFP_KERNEL);
+	if (!fh)
 		return ERR_PTR(-ENOMEM);
 
 	/*
···
 	 * the price or reconnecting the dentry.
 	 */
 	dwords = buflen >> 2;
-	fh_type = exportfs_encode_fh(real, buf, &dwords, 0);
+	fh_type = exportfs_encode_fh(real, (void *)fh->fb.fid, &dwords, 0);
 	buflen = (dwords << 2);
 
-	fh = ERR_PTR(-EIO);
+	err = -EIO;
 	if (WARN_ON(fh_type < 0) ||
 	    WARN_ON(buflen > MAX_HANDLE_SZ) ||
 	    WARN_ON(fh_type == FILEID_INVALID))
-		goto out;
+		goto out_err;
 
-	BUILD_BUG_ON(MAX_HANDLE_SZ + offsetof(struct ovl_fh, fid) > 255);
-	fh_len = offsetof(struct ovl_fh, fid) + buflen;
-	fh = kmalloc(fh_len, GFP_KERNEL);
-	if (!fh) {
-		fh = ERR_PTR(-ENOMEM);
-		goto out;
-	}
-
-	fh->version = OVL_FH_VERSION;
-	fh->magic = OVL_FH_MAGIC;
-	fh->type = fh_type;
-	fh->flags = OVL_FH_FLAG_CPU_ENDIAN;
+	fh->fb.version = OVL_FH_VERSION;
+	fh->fb.magic = OVL_FH_MAGIC;
+	fh->fb.type = fh_type;
+	fh->fb.flags = OVL_FH_FLAG_CPU_ENDIAN;
 	/*
 	 * When we will want to decode an overlay dentry from this handle
 	 * and all layers are on the same fs, if we get a disconncted real
···
 	 * it to upperdentry or to lowerstack is by checking this flag.
 	 */
 	if (is_upper)
-		fh->flags |= OVL_FH_FLAG_PATH_UPPER;
-	fh->len = fh_len;
-	fh->uuid = *uuid;
-	memcpy(fh->fid, buf, buflen);
+		fh->fb.flags |= OVL_FH_FLAG_PATH_UPPER;
+	fh->fb.len = sizeof(fh->fb) + buflen;
+	fh->fb.uuid = *uuid;
 
-out:
-	kfree(buf);
 	return fh;
+
+out_err:
+	kfree(fh);
+	return ERR_PTR(err);
 }
 
 int ovl_set_origin(struct dentry *dentry, struct dentry *lower,
···
 	/*
 	 * Do not fail when upper doesn't support xattrs.
 	 */
-	err = ovl_check_setxattr(dentry, upper, OVL_XATTR_ORIGIN, fh,
-				 fh ? fh->len : 0, 0);
+	err = ovl_check_setxattr(dentry, upper, OVL_XATTR_ORIGIN, fh->buf,
+				 fh ? fh->fb.len : 0, 0);
 	kfree(fh);
 
 	return err;
···
 	if (IS_ERR(fh))
 		return PTR_ERR(fh);
 
-	err = ovl_do_setxattr(index, OVL_XATTR_UPPER, fh, fh->len, 0);
+	err = ovl_do_setxattr(index, OVL_XATTR_UPPER, fh->buf, fh->fb.len, 0);
 
 	kfree(fh);
 	return err;

fs/overlayfs/dir.c

···
 	if (newdentry == trap)
 		goto out_dput;
 
-	if (WARN_ON(olddentry->d_inode == newdentry->d_inode))
+	if (olddentry->d_inode == newdentry->d_inode)
 		goto out_dput;
 
 	err = 0;

fs/overlayfs/export.c

···
 	return 1;
 }
 
-static int ovl_d_to_fh(struct dentry *dentry, char *buf, int buflen)
+static int ovl_dentry_to_fid(struct dentry *dentry, u32 *fid, int buflen)
 {
 	struct ovl_fh *fh = NULL;
 	int err, enc_lower;
+	int len;
 
 	/*
 	 * Check if we should encode a lower or upper file handle and maybe
···
 		return PTR_ERR(fh);
 
 	err = -EOVERFLOW;
-	if (fh->len > buflen)
+	len = OVL_FH_LEN(fh);
+	if (len > buflen)
 		goto fail;
 
-	memcpy(buf, (char *)fh, fh->len);
-	err = fh->len;
+	memcpy(fid, fh, len);
+	err = len;
 
 out:
 	kfree(fh);
···
 
 fail:
 	pr_warn_ratelimited("overlayfs: failed to encode file handle (%pd2, err=%i, buflen=%d, len=%d, type=%d)\n",
-			    dentry, err, buflen, fh ? (int)fh->len : 0,
-			    fh ? fh->type : 0);
+			    dentry, err, buflen, fh ? (int)fh->fb.len : 0,
+			    fh ? fh->fb.type : 0);
 	goto out;
-}
-
-static int ovl_dentry_to_fh(struct dentry *dentry, u32 *fid, int *max_len)
-{
-	int res, len = *max_len << 2;
-
-	res = ovl_d_to_fh(dentry, (char *)fid, len);
-	if (res <= 0)
-		return FILEID_INVALID;
-
-	len = res;
-
-	/* Round up to dwords */
-	*max_len = (len + 3) >> 2;
-	return OVL_FILEID;
 }
 
 static int ovl_encode_fh(struct inode *inode, u32 *fid, int *max_len,
 			 struct inode *parent)
 {
 	struct dentry *dentry;
-	int type;
+	int bytes = *max_len << 2;
 
 	/* TODO: encode connectable file handles */
 	if (parent)
···
 	if (WARN_ON(!dentry))
 		return FILEID_INVALID;
 
-	type = ovl_dentry_to_fh(dentry, fid, max_len);
-
+	bytes = ovl_dentry_to_fid(dentry, fid, bytes);
 	dput(dentry);
-	return type;
+	if (bytes <= 0)
+		return FILEID_INVALID;
+
+	*max_len = bytes >> 2;
+
+	return OVL_FILEID_V1;
 }
 
 /*
···
 	goto out;
 }
 
+static struct ovl_fh *ovl_fid_to_fh(struct fid *fid, int buflen, int fh_type)
+{
+	struct ovl_fh *fh;
+
+	/* If on-wire inner fid is aligned - nothing to do */
+	if (fh_type == OVL_FILEID_V1)
+		return (struct ovl_fh *)fid;
+
+	if (fh_type != OVL_FILEID_V0)
+		return ERR_PTR(-EINVAL);
+
+	fh = kzalloc(buflen, GFP_KERNEL);
+	if (!fh)
+		return ERR_PTR(-ENOMEM);
+
+	/* Copy unaligned inner fh into aligned buffer */
+	memcpy(&fh->fb, fid, buflen - OVL_FH_WIRE_OFFSET);
+	return fh;
+}
+
 static struct dentry *ovl_fh_to_dentry(struct super_block *sb, struct fid *fid,
 				       int fh_len, int fh_type)
 {
 	struct dentry *dentry = NULL;
-	struct ovl_fh *fh = (struct ovl_fh *) fid;
+	struct ovl_fh *fh = NULL;
 	int len = fh_len << 2;
 	unsigned int flags = 0;
 	int err;
 
-	err = -EINVAL;
-	if (fh_type != OVL_FILEID)
+	fh = ovl_fid_to_fh(fid, len, fh_type);
+	err = PTR_ERR(fh);
+	if (IS_ERR(fh))
 		goto out_err;
 
 	err = ovl_check_fh_len(fh, len);
 	if (err)
 		goto out_err;
 
-	flags = fh->flags;
+	flags = fh->fb.flags;
 	dentry = (flags & OVL_FH_FLAG_PATH_UPPER) ?
 		 ovl_upper_fh_to_d(sb, fh) :
 		 ovl_lower_fh_to_d(sb, fh);
···
 	if (IS_ERR(dentry) && err != -ESTALE)
 		goto out_err;
 
+out:
+	/* We may have needed to re-align OVL_FILEID_V0 */
+	if (!IS_ERR_OR_NULL(fh) && fh != (void *)fid)
+		kfree(fh);
+
 	return dentry;
 
 out_err:
 	pr_warn_ratelimited("overlayfs: failed to decode file handle (len=%d, type=%d, flags=%x, err=%i)\n",
-			    len, fh_type, flags, err);
-	return ERR_PTR(err);
+			    fh_len, fh_type, flags, err);
+	dentry = ERR_PTR(err);
+	goto out;
 }
 
 static struct dentry *ovl_fh_to_parent(struct super_block *sb, struct fid *fid,

fs/overlayfs/inode.c

···
 			if (ovl_test_flag(OVL_INDEX, d_inode(dentry)) ||
 			    (!ovl_verify_lower(dentry->d_sb) &&
 			     (is_dir || lowerstat.nlink == 1))) {
-				stat->ino = lowerstat.ino;
 				lower_layer = ovl_layer_lower(dentry);
+				/*
+				 * Cannot use origin st_dev;st_ino because
+				 * origin inode content may differ from overlay
+				 * inode content.
+				 */
+				if (samefs || lower_layer->fsid)
+					stat->ino = lowerstat.ino;
 			}
 
 			/*

fs/overlayfs/namei.c

···
  * Return -ENODATA for "origin unknown".
  * Return <0 for an invalid file handle.
  */
-int ovl_check_fh_len(struct ovl_fh *fh, int fh_len)
+int ovl_check_fb_len(struct ovl_fb *fb, int fb_len)
 {
-	if (fh_len < sizeof(struct ovl_fh) || fh_len < fh->len)
+	if (fb_len < sizeof(struct ovl_fb) || fb_len < fb->len)
 		return -EINVAL;
 
-	if (fh->magic != OVL_FH_MAGIC)
+	if (fb->magic != OVL_FH_MAGIC)
 		return -EINVAL;
 
 	/* Treat larger version and unknown flags as "origin unknown" */
-	if (fh->version > OVL_FH_VERSION || fh->flags & ~OVL_FH_FLAG_ALL)
+	if (fb->version > OVL_FH_VERSION || fb->flags & ~OVL_FH_FLAG_ALL)
 		return -ENODATA;
 
 	/* Treat endianness mismatch as "origin unknown" */
-	if (!(fh->flags & OVL_FH_FLAG_ANY_ENDIAN) &&
-	    (fh->flags & OVL_FH_FLAG_BIG_ENDIAN) != OVL_FH_FLAG_CPU_ENDIAN)
+	if (!(fb->flags & OVL_FH_FLAG_ANY_ENDIAN) &&
+	    (fb->flags & OVL_FH_FLAG_BIG_ENDIAN) != OVL_FH_FLAG_CPU_ENDIAN)
 		return -ENODATA;
 
 	return 0;
···
 	if (res == 0)
 		return NULL;
 
-	fh = kzalloc(res, GFP_KERNEL);
+	fh = kzalloc(res + OVL_FH_WIRE_OFFSET, GFP_KERNEL);
 	if (!fh)
 		return ERR_PTR(-ENOMEM);
 
-	res = vfs_getxattr(dentry, name, fh, res);
+	res = vfs_getxattr(dentry, name, fh->buf, res);
 	if (res < 0)
 		goto fail;
 
-	err = ovl_check_fh_len(fh, res);
+	err = ovl_check_fb_len(&fh->fb, res);
 	if (err < 0) {
 		if (err == -ENODATA)
 			goto out;
···
 	 * Make sure that the stored uuid matches the uuid of the lower
 	 * layer where file handle will be decoded.
 	 */
-	if (!uuid_equal(&fh->uuid, &mnt->mnt_sb->s_uuid))
+	if (!uuid_equal(&fh->fb.uuid, &mnt->mnt_sb->s_uuid))
 		return NULL;
 
-	bytes = (fh->len - offsetof(struct ovl_fh, fid));
-	real = exportfs_decode_fh(mnt, (struct fid *)fh->fid,
-				  bytes >> 2, (int)fh->type,
+	bytes = (fh->fb.len - offsetof(struct ovl_fb, fid));
+	real = exportfs_decode_fh(mnt, (struct fid *)fh->fb.fid,
+				  bytes >> 2, (int)fh->fb.type,
 				  connected ? ovl_acceptable : NULL, mnt);
 	if (IS_ERR(real)) {
 		/*
···
 		 * index entries correctly.
 		 */
 		if (real == ERR_PTR(-ESTALE) &&
-		    !(fh->flags & OVL_FH_FLAG_PATH_UPPER))
+		    !(fh->fb.flags & OVL_FH_FLAG_PATH_UPPER))
 			real = NULL;
 		return real;
 	}
···
 	int i;
 
 	for (i = 0; i < ofs->numlower; i++) {
+		/*
+		 * If lower fs uuid is not unique among lower fs we cannot match
+		 * fh->uuid to layer.
+		 */
+		if (ofs->lower_layers[i].fsid &&
+		    ofs->lower_layers[i].fs->bad_uuid)
+			continue;
+
 		origin = ovl_decode_real_fh(fh, ofs->lower_layers[i].mnt,
 					    connected);
 		if (origin)
···
 	if (IS_ERR(ofh))
 		return PTR_ERR(ofh);
 
-	if (fh->len != ofh->len || memcmp(fh, ofh, fh->len))
+	if (fh->fb.len != ofh->fb.len || memcmp(&fh->fb, &ofh->fb, fh->fb.len))
 		err = -ESTALE;
 
 	kfree(ofh);
···
 
 	err = ovl_verify_fh(dentry, name, fh);
 	if (set && err == -ENODATA)
-		err = ovl_do_setxattr(dentry, name, fh, fh->len, 0);
+		err = ovl_do_setxattr(dentry, name, fh->buf, fh->fb.len, 0);
 	if (err)
 		goto fail;
 
···
 		goto fail;
 
 	err = -EINVAL;
-	if (index->d_name.len < sizeof(struct ovl_fh)*2)
+	if (index->d_name.len < sizeof(struct ovl_fb)*2)
 		goto fail;
 
 	err = -ENOMEM;
 	len = index->d_name.len / 2;
-	fh = kzalloc(len, GFP_KERNEL);
+	fh = kzalloc(len + OVL_FH_WIRE_OFFSET, GFP_KERNEL);
 	if (!fh)
 		goto fail;
 
 	err = -EINVAL;
-	if (hex2bin((u8 *)fh, index->d_name.name, len))
+	if (hex2bin(fh->buf, index->d_name.name, len))
 		goto fail;
 
-	err = ovl_check_fh_len(fh, len);
+	err = ovl_check_fb_len(&fh->fb, len);
 	if (err)
 		goto fail;
 
···
 {
 	char *n, *s;
 
-	n = kcalloc(fh->len, 2, GFP_KERNEL);
+	n = kcalloc(fh->fb.len, 2, GFP_KERNEL);
 	if (!n)
 		return -ENOMEM;
 
-	s  = bin2hex(n, fh, fh->len);
+	s  = bin2hex(n, fh->buf, fh->fb.len);
 	*name = (struct qstr) QSTR_INIT(n, s - n);
 
 	return 0;

fs/overlayfs/overlayfs.h

···
 #error Endianness not defined
 #endif
 
-/* The type returned by overlay exportfs ops when encoding an ovl_fh handle */
-#define OVL_FILEID	0xfb
+/* The type used to be returned by overlay exportfs for misaligned fid */
+#define OVL_FILEID_V0	0xfb
+/* The type returned by overlay exportfs for 32bit aligned fid */
+#define OVL_FILEID_V1	0xf8
 
-/* On-disk and in-memeory format for redirect by file handle */
-struct ovl_fh {
+/* On-disk format for "origin" file handle */
+struct ovl_fb {
 	u8 version;	/* 0 */
 	u8 magic;	/* 0xfb */
 	u8 len;		/* size of this header + size of fid */
 	u8 flags;	/* OVL_FH_FLAG_* */
 	u8 type;	/* fid_type of fid */
 	uuid_t uuid;	/* uuid of filesystem */
-	u8 fid[0];	/* file identifier */
+	u32 fid[0];	/* file identifier should be 32bit aligned in-memory */
 } __packed;
+
+/* In-memory and on-wire format for overlay file handle */
+struct ovl_fh {
+	u8 padding[3];	/* make sure fb.fid is 32bit aligned */
+	union {
+		struct ovl_fb fb;
+		u8 buf[0];
+	};
+} __packed;
+
+#define OVL_FH_WIRE_OFFSET	offsetof(struct ovl_fh, fb)
+#define OVL_FH_LEN(fh)		(OVL_FH_WIRE_OFFSET + (fh)->fb.len)
+#define OVL_FH_FID_OFFSET	(OVL_FH_WIRE_OFFSET + \
+				 offsetof(struct ovl_fb, fid))
 
 static inline int ovl_do_rmdir(struct inode *dir, struct dentry *dentry)
 {
···
 
 
 /* namei.c */
-int ovl_check_fh_len(struct ovl_fh *fh, int fh_len);
+int ovl_check_fb_len(struct ovl_fb *fb, int fb_len);
+
+static inline int ovl_check_fh_len(struct ovl_fh *fh, int fh_len)
+{
+	return ovl_check_fb_len(&fh->fb, fh_len - OVL_FH_WIRE_OFFSET);
+}
+
 struct dentry *ovl_decode_real_fh(struct ovl_fh *fh, struct vfsmount *mnt,
 				  bool connected);
 int ovl_check_origin_fh(struct ovl_fs *ofs, struct ovl_fh *fh, bool connected,

fs/overlayfs/ovl_entry.h

···
 struct ovl_sb {
 	struct super_block *sb;
 	dev_t pseudo_dev;
+	/* Unusable (conflicting) uuid */
+	bool bad_uuid;
 };
 
 struct ovl_layer {

fs/overlayfs/super.c

···
 {
 	unsigned int i;
 
-	if (!ofs->config.nfs_export && !(ofs->config.index && ofs->upper_mnt))
+	if (!ofs->config.nfs_export && !ofs->upper_mnt)
 		return true;
 
 	for (i = 0; i < ofs->numlowerfs; i++) {
···
 		 * We use uuid to associate an overlay lower file handle with a
 		 * lower layer, so we can accept lower fs with null uuid as long
 		 * as all lower layers with null uuid are on the same fs.
+		 * if we detect multiple lower fs with the same uuid, we
+		 * disable lower file handle decoding on all of them.
 		 */
-		if (uuid_equal(&ofs->lower_fs[i].sb->s_uuid, uuid))
+		if (uuid_equal(&ofs->lower_fs[i].sb->s_uuid, uuid)) {
+			ofs->lower_fs[i].bad_uuid = true;
 			return false;
+		}
 	}
 	return true;
 }
···
 	unsigned int i;
 	dev_t dev;
 	int err;
+	bool bad_uuid = false;
 
 	/* fsid 0 is reserved for upper fs even with non upper overlay */
 	if (ofs->upper_mnt && ofs->upper_mnt->mnt_sb == sb)
···
 	}
 
 	if (!ovl_lower_uuid_ok(ofs, &sb->s_uuid)) {
-		ofs->config.index = false;
-		ofs->config.nfs_export = false;
-		pr_warn("overlayfs: %s uuid detected in lower fs '%pd2', falling back to index=off,nfs_export=off.\n",
-			uuid_is_null(&sb->s_uuid) ? "null" : "conflicting",
-			path->dentry);
+		bad_uuid = true;
+		if (ofs->config.index || ofs->config.nfs_export) {
+			ofs->config.index = false;
+			ofs->config.nfs_export = false;
+			pr_warn("overlayfs: %s uuid detected in lower fs '%pd2', falling back to index=off,nfs_export=off.\n",
+				uuid_is_null(&sb->s_uuid) ? "null" :
+							    "conflicting",
+				path->dentry);
+		}
 	}
 
 	err = get_anon_bdev(&dev);
···
 
 	ofs->lower_fs[ofs->numlowerfs].sb = sb;
 	ofs->lower_fs[ofs->numlowerfs].pseudo_dev = dev;
+	ofs->lower_fs[ofs->numlowerfs].bad_uuid = bad_uuid;
 	ofs->numlowerfs++;
 
 	return ofs->numlowerfs;