ext4: fix xattr shifting when expanding inodes part 2

When multiple xattrs need to be moved out of inode, we did not properly
recompute total size of xattr headers in the inode and the new header
position. Thus when moving the second and further xattr we asked
ext4_xattr_shift_entries() to move too much and from the wrong place,
resulting in possible xattr value corruption or general memory
corruption.

CC: stable@vger.kernel.org  # 4.4.x
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
(cherry picked from commit 418c12d08dc64a45107c467ec1ba29b5e69b0715)

Orabug: 25718971

Signed-off-by: Shuning Zhang <sunny.s.zhang@oracle.com>
Reviewed-by: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Brian Maly <brian.maly@oracle.com>

diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c
index 6641da6942f2d66ec3ceb49585c0214d96b05c38..e899c453683cb6e52468f501c178d3fbae3f19eb 100644 (file)
--- a/fs/ext4/xattr.c
+++ b/fs/ext4/xattr.c
@@ -1450,6 +1450,7 @@ retry:
                error = ext4_xattr_ibody_set(handle, inode, &i, is);
                if (error)
                        goto cleanup;
+               total_ino -= entry_size;
 
                entry = IFIRST(header);
                if (entry_size + EXT4_XATTR_SIZE(size) >= isize_diff)
@@ -1460,11 +1461,11 @@ retry:
                ext4_xattr_shift_entries(entry, -shift_bytes,
                        (void *)raw_inode + EXT4_GOOD_OLD_INODE_SIZE +
                        EXT4_I(inode)->i_extra_isize + shift_bytes,
-                       (void *)header, total_ino - entry_size,
-                       inode->i_sb->s_blocksize);
+                       (void *)header, total_ino, inode->i_sb->s_blocksize);
 
                isize_diff -= shift_bytes;
                EXT4_I(inode)->i_extra_isize += shift_bytes;
+               header = IHDR(inode, raw_inode);
 
                i.name = b_entry_name;
                i.value = buffer;