ext4: miss an unlock in ext4_check_descriptors()

 # mkfs.ext3 -m 0 /dev/sda6
 # tune2fs -E test_fs -O extents,uninit_bg
 (no e2fsck after tune2fs)
 # mount -t ext4dev /dev/sda6

oops immediately:

EXT4-fs: ext4_check_descriptors: Checksum for group 0 failed (20091!=0)
EXT4-fs: group descriptors corrupted!

=========================
[ BUG: held lock freed! ]
-------------------------
mount/9422 is freeing memory c6ca0000-c6ca7fff, with a lock still held there!
(&bgl->locks[i].lock#2){--..}, at: [<f8996589>] ext4_fill_super+0xde8/0x1dcf [ext4dev]
2 locks held by mount/9422:
#0: (&type->s_umount_key#19){----}, at: [<c0488258>] sget+0x1dd/0x31c
#1: (&bgl->locks[i].lock#2){--..}, at: [<f8996589>] ext4_fill_super+0xde8/0x1dcf [ext4dev]

---
 fs/ext4/super.c |    4 +++-
 1 files changed, 3 insertions(+), 1 deletions(-)

index 566344b..24e7df6 100644
@@ -1623,8 +1623,10 @@ static int ext4_check_descriptors(struct super_block *sb)
 			       "Checksum for group %lu failed (%u!=%u)\n",
 			       i, le16_to_cpu(ext4_group_desc_csum(sbi, i,
 			       gdp)), le16_to_cpu(gdp->bg_checksum));
-			if (!(sb->s_flags & MS_RDONLY))
+			if (!(sb->s_flags & MS_RDONLY)) {
+				spin_unlock(sb_bgl_lock(sbi, i));
 				return 0;
+			}
 		}
 		spin_unlock(sb_bgl_lock(sbi, i));
 		if (!flexbg_flag)

-- 1.5.4.rc3

Thanks, I've applied it to the ext4 patch queue.

- Ted