f2fs: check entire encrypted bigname when finding a dentry
authorJaegeuk Kim <jaegeuk@kernel.org>
Mon, 24 Apr 2017 17:00:08 +0000 (10:00 -0700)
committerTheodore Ts'o <tytso@mit.edu>
Thu, 4 May 2017 15:44:35 +0000 (11:44 -0400)
commit6332cd32c8290a80e929fc044dc5bdba77396e33
tree24f76cfe6d3dddfd9aa15189ee6445584f71fee6
parent413d5a9edbb45eec4dbb1904fe689b561ea5d143
f2fs: check entire encrypted bigname when finding a dentry

If user has no key under an encrypted dir, fscrypt gives digested dentries.
Previously, when looking up a dentry, f2fs only checks its hash value with
first 4 bytes of the digested dentry, which didn't handle hash collisions fully.
This patch enhances to check entire dentry bytes likewise ext4.

Eric reported how to reproduce this issue by:

 # seq -f "edir/abcdefghijklmnopqrstuvwxyz012345%.0f" 100000 | xargs touch
 # find edir -type f | xargs stat -c %i | sort | uniq | wc -l
100000
 # sync
 # echo 3 > /proc/sys/vm/drop_caches
 # keyctl new_session
 # find edir -type f | xargs stat -c %i | sort | uniq | wc -l
99999

Cc: <stable@vger.kernel.org>
Reported-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
(fixed f2fs_dentry_hash() to work even when the hash is 0)
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
fs/f2fs/dir.c
fs/f2fs/f2fs.h
fs/f2fs/hash.c
fs/f2fs/inline.c