nilfs2: fix sanity check of btree level in nilfs_btree_root_broken()
authorRyusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
Tue, 5 May 2015 23:24:00 +0000 (16:24 -0700)
committerBen Hutchings <ben@decadent.org.uk>
Thu, 6 Aug 2015 23:32:09 +0000 (00:32 +0100)
commit74e08b1cc7be2c4eecdd277484ed89fcacb0d3dd
tree37038e39ef8c07e61450fbb6d34f3fc91c74c5a0
parent38787a8ce9e6e053070b12ea19945b962d79d579
nilfs2: fix sanity check of btree level in nilfs_btree_root_broken()

commit d8fd150fe3935e1692bf57c66691e17409ebb9c1 upstream.

The range check for b-tree level parameter in nilfs_btree_root_broken()
is wrong; it accepts the case of "level == NILFS_BTREE_LEVEL_MAX" even
though the level is limited to values in the range of 0 to
(NILFS_BTREE_LEVEL_MAX - 1).

Since the level parameter is read from storage device and used to index
nilfs_btree_path array whose element count is NILFS_BTREE_LEVEL_MAX, it
can cause memory overrun during btree operations if the boundary value
is set to the level parameter on device.

This fixes the broken sanity check and adds a comment to clarify that
the upper bound NILFS_BTREE_LEVEL_MAX is exclusive.

Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
fs/nilfs2/btree.c
include/linux/nilfs2_fs.h