[PATCH 2/2] hfsplus: always return error if fill_super fails

["Ernesto A. Fernández" <ernesto.mnd.fernandez@gmail.com>]

If the filesystem is corrupted in such a way that the HFS+ Private Data
directory is not of type HFSPLUS_FOLDER, the mount will fail without
throwing an error code. The vfs layer is then forced to dereference a
NULL root dentry.

This bug was found by KASAN while running the reproducer provided by
syzbot+4f2e5f086147d543ab03@syzkaller.appspotmail.com for a separate
issue.

Fixes: 5bd9d99d107c ("hfsplus: add error checking for hfs_find_init()")
Signed-off-by: Ernesto A. Fernández <ernesto.mnd.fernandez@gmail.com>
---
 fs/hfsplus/super.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c
index 4bc49e3f171d..4f62634c5666 100644
--- a/fs/hfsplus/super.c
+++ b/fs/hfsplus/super.c
@@ -524,8 +524,10 @@ static int hfsplus_fill_super(struct super_block *sb, void *data, int silent)
 		goto out_put_root;
 	if (!hfs_brec_read(&fd, &entry, sizeof(entry))) {
 		hfs_find_exit(&fd);
-		if (entry.type != cpu_to_be16(HFSPLUS_FOLDER))
+		if (entry.type != cpu_to_be16(HFSPLUS_FOLDER)) {
+			err = -EINVAL;
 			goto out_put_root;
+		}
 		inode = hfsplus_iget(sb, be32_to_cpu(entry.folder.id));
 		if (IS_ERR(inode)) {
 			err = PTR_ERR(inode);
-- 
2.11.0