overlayfs fixes for 6.6-rc5

-----BEGIN PGP SIGNATURE-----
 
 iQIzBAABCAAdFiEE9zuTYTs0RXF+Ke33EVvVyTe/1WoFAmUeSFQACgkQEVvVyTe/
 1WrKdxAAoMtgK5/GcNO3BijxtU1M7onu70Jm4froOFR0R2hdDFG76vfHrPHpavha
 gnSOAOpDZYNkkJTKNmN2b047dRvrtgYiZ1rrSxlLjXY+2ojVoyAFXddVPNpT4nSy
 2xka8KqLeU8Mtzt0GQq/PVe0JLgjbEEejn4wFdoilSBPZhf4sik24ReQVyxhYG3B
 aILG83GZAuqokzO3HkhYaMo+qgm0BVLcHwZ0zd5UQ404srB7nmE4JxS4+sh0mrsc
 LQmZdlXnzEj46nJ9pluWvUH2bGAA1+rtSg182mbcrwnXOlct40x0TI8MzzFcnmaj
 mue37V+iLHLRr6BeIzNghbTflF0SeUH8Qbn7mjYkv4+VTfomYU0OzV3Wy54yUqA9
 VrugqHAVpgv6r4gNPgkvNpp4YKTeFUd14FI4sO+Ny/f0CHCA+TeEP5g3IH5QguxF
 vf23Uww9QyNgRp/JaV0kQaJfRqbMtUylMOiTo7Kaou+XV3A6AfnHhyyeS2qVmYI8
 bsXjG4QzYrlxmvcFuVBXrLuiigVEI8DcTHSwXZ5NqpqIb/dCmXexZ48loL73f0Qj
 Us8U1w7JymQDscug74F4S47CE3JoUH4hsu25StQ1LTvfIc+leQy1GBU3tVJv/d7U
 cgD7OkktwjuQ5ClS+8gdP9fQ9AzI9/vGEhQL9+peKW6FK5EfWPc=
 =r3VL
 -----END PGP SIGNATURE-----

Merge tag 'ovl-fixes-6.6-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/overlayfs/vfs

Pull overlayfs fixes from Amir Goldstein:

 - Fix for file reference leak regression

 - Fix for NULL pointer deref regression

 - Fixes for RCU-walk race regressions:

   Two of the fixes were taken from Al's RCU pathwalk race fixes series
   with his consent [1].

   Note that unlike most of Al's series, these two patches are not about
   racing with ->kill_sb() and they are also very recent regressions
   from v6.5, so I think it's worth getting them into v6.5.y.

   There is also a fix for an RCU pathwalk race with ->kill_sb(), which
   may have been solved in vfs generic code as you suggested, but it
   also rids overlayfs from a nasty hack, so I think it's worth anyway.

Link: https://lore.kernel.org/linux-fsdevel/20231003204749.GA800259@ZenIV/ [1]

* tag 'ovl-fixes-6.6-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/overlayfs/vfs:
  ovl: fix NULL pointer defer when encoding non-decodable lower fid
  ovl: make use of ->layers safe in rcu pathwalk
  ovl: fetch inode once in ovl_dentry_revalidate_common()
  ovl: move freeing ovl_entry past rcu delay
  ovl: fix file reference leak when submitting aio

fs/overlayfs/export.c

@@ -188,7 +188,7 @@ static int ovl_check_encode_origin(struct dentry *dentry)
 
 	/* Lower file handle for non-upper non-decodable */
 	if (!ovl_dentry_upper(dentry) && !decodable)
-		return 0;
+		return 1;
 
 	/* Upper file handle for pure upper */
 	if (!ovl_dentry_lower(dentry))

fs/overlayfs/file.c

@@ -341,7 +341,6 @@ static ssize_t ovl_read_iter(struct kiocb *iocb, struct iov_iter *iter)
 		if (!aio_req)
 			goto out;
 
-		real.flags = 0;
 		aio_req->orig_iocb = iocb;
 		kiocb_clone(&aio_req->iocb, iocb, get_file(real.file));
 		aio_req->iocb.ki_complete = ovl_aio_rw_complete;
@@ -413,7 +412,6 @@ static ssize_t ovl_write_iter(struct kiocb *iocb, struct iov_iter *iter)
 		if (!aio_req)
 			goto out;
 
-		real.flags = 0;
 		aio_req->orig_iocb = iocb;
 		kiocb_clone(&aio_req->iocb, iocb, get_file(real.file));
 		aio_req->iocb.ki_flags = ifl;

fs/overlayfs/ovl_entry.h

@@ -8,6 +8,7 @@
 struct ovl_config {
 	char *upperdir;
 	char *workdir;
+	char **lowerdirs;
 	bool default_permissions;
 	int redirect_mode;
 	int verity_mode;
@@ -39,17 +40,8 @@ struct ovl_layer {
 	int idx;
 	/* One fsid per unique underlying sb (upper fsid == 0) */
 	int fsid;
-	char *name;
 };
 
-/*
- * ovl_free_fs() relies on @mnt being the first member when unmounting
- * the private mounts created for each layer. Let's check both the
- * offset and type.
- */
-static_assert(offsetof(struct ovl_layer, mnt) == 0);
-static_assert(__same_type(typeof_member(struct ovl_layer, mnt), struct vfsmount *));
-
 struct ovl_path {
 	const struct ovl_layer *layer;
 	struct dentry *dentry;

fs/overlayfs/params.c

@@ -752,12 +752,12 @@ void ovl_free_fs(struct ovl_fs *ofs)
 	if (ofs->upperdir_locked)
 		ovl_inuse_unlock(ovl_upper_mnt(ofs)->mnt_root);
 
-	/* Hack!  Reuse ofs->layers as a vfsmount array before freeing it */
-	mounts = (struct vfsmount **) ofs->layers;
+	/* Reuse ofs->config.lowerdirs as a vfsmount array before freeing it */
+	mounts = (struct vfsmount **) ofs->config.lowerdirs;
 	for (i = 0; i < ofs->numlayer; i++) {
 		iput(ofs->layers[i].trap);
+		kfree(ofs->config.lowerdirs[i]);
 		mounts[i] = ofs->layers[i].mnt;
-		kfree(ofs->layers[i].name);
 	}
 	kern_unmount_array(mounts, ofs->numlayer);
 	kfree(ofs->layers);
@@ -765,6 +765,7 @@ void ovl_free_fs(struct ovl_fs *ofs)
 		free_anon_bdev(ofs->fs[i].pseudo_dev);
 	kfree(ofs->fs);
 
+	kfree(ofs->config.lowerdirs);
 	kfree(ofs->config.upperdir);
 	kfree(ofs->config.workdir);
 	if (ofs->creator_cred)
@@ -949,16 +950,16 @@ int ovl_show_options(struct seq_file *m, struct dentry *dentry)
 	struct super_block *sb = dentry->d_sb;
 	struct ovl_fs *ofs = OVL_FS(sb);
 	size_t nr, nr_merged_lower = ofs->numlayer - ofs->numdatalayer;
-	const struct ovl_layer *data_layers = &ofs->layers[nr_merged_lower];
+	char **lowerdatadirs = &ofs->config.lowerdirs[nr_merged_lower];
 
-	/* ofs->layers[0] is the upper layer */
-	seq_printf(m, ",lowerdir=%s", ofs->layers[1].name);
+	/* lowerdirs[] starts from offset 1 */
+	seq_printf(m, ",lowerdir=%s", ofs->config.lowerdirs[1]);
 	/* dump regular lower layers */
 	for (nr = 2; nr < nr_merged_lower; nr++)
-		seq_printf(m, ":%s", ofs->layers[nr].name);
+		seq_printf(m, ":%s", ofs->config.lowerdirs[nr]);
 	/* dump data lower layers */
 	for (nr = 0; nr < ofs->numdatalayer; nr++)
-		seq_printf(m, "::%s", data_layers[nr].name);
+		seq_printf(m, "::%s", lowerdatadirs[nr]);
 	if (ofs->config.upperdir) {
 		seq_show_option(m, "upperdir", ofs->config.upperdir);
 		seq_show_option(m, "workdir", ofs->config.workdir);

fs/overlayfs/super.c

@@ -104,8 +104,8 @@ static int ovl_revalidate_real(struct dentry *d, unsigned int flags, bool weak)
 static int ovl_dentry_revalidate_common(struct dentry *dentry,
 					unsigned int flags, bool weak)
 {
-	struct ovl_entry *oe = OVL_E(dentry);
-	struct ovl_path *lowerstack = ovl_lowerstack(oe);
+	struct ovl_entry *oe;
+	struct ovl_path *lowerstack;
 	struct inode *inode = d_inode_rcu(dentry);
 	struct dentry *upper;
 	unsigned int i;
@@ -115,6 +115,8 @@ static int ovl_dentry_revalidate_common(struct dentry *dentry,
 	if (!inode)
 		return -ECHILD;
 
+	oe = OVL_I_E(inode);
+	lowerstack = ovl_lowerstack(oe);
 	upper = ovl_i_dentry_upper(inode);
 	if (upper)
 		ret = ovl_revalidate_real(upper, flags, weak);
@@ -167,6 +169,7 @@ static void ovl_free_inode(struct inode *inode)
 	struct ovl_inode *oi = OVL_I(inode);
 
 	kfree(oi->redirect);
+	kfree(oi->oe);
 	mutex_destroy(&oi->lock);
 	kmem_cache_free(ovl_inode_cachep, oi);
 }
@@ -176,7 +179,7 @@ static void ovl_destroy_inode(struct inode *inode)
 	struct ovl_inode *oi = OVL_I(inode);
 
 	dput(oi->__upperdentry);
-	ovl_free_entry(oi->oe);
+	ovl_stack_put(ovl_lowerstack(oi->oe), ovl_numlower(oi->oe));
 	if (S_ISDIR(inode->i_mode))
 		ovl_dir_cache_free(inode);
 	else
@@ -569,11 +572,6 @@ static int ovl_get_upper(struct super_block *sb, struct ovl_fs *ofs,
 	upper_layer->idx = 0;
 	upper_layer->fsid = 0;
 
-	err = -ENOMEM;
-	upper_layer->name = kstrdup(ofs->config.upperdir, GFP_KERNEL);
-	if (!upper_layer->name)
-		goto out;
-
 	/*
 	 * Inherit SB_NOSEC flag from upperdir.
 	 *
@@ -1122,7 +1120,8 @@ static int ovl_get_layers(struct super_block *sb, struct ovl_fs *ofs,
 		layers[ofs->numlayer].idx = ofs->numlayer;
 		layers[ofs->numlayer].fsid = fsid;
 		layers[ofs->numlayer].fs = &ofs->fs[fsid];
-		layers[ofs->numlayer].name = l->name;
+		/* Store for printing lowerdir=... in ovl_show_options() */
+		ofs->config.lowerdirs[ofs->numlayer] = l->name;
 		l->name = NULL;
 		ofs->numlayer++;
 		ofs->fs[fsid].is_lower = true;
@@ -1367,8 +1366,16 @@ int ovl_fill_super(struct super_block *sb, struct fs_context *fc)
 	if (!layers)
 		goto out_err;
 
+	ofs->config.lowerdirs = kcalloc(ctx->nr + 1, sizeof(char *), GFP_KERNEL);
+	if (!ofs->config.lowerdirs) {
+		kfree(layers);
+		goto out_err;
+	}
 	ofs->layers = layers;
-	/* Layer 0 is reserved for upper even if there's no upper */
+	/*
+	 * Layer 0 is reserved for upper even if there's no upper.
+	 * For consistency, config.lowerdirs[0] is NULL.
+	 */
 	ofs->numlayer = 1;
 
 	sb->s_stack_depth = 0;