From: Eric Biggers Subject: [PATCH] fscrypto: lock inode while setting encryption policy Date: Wed, 28 Sep 2016 11:34:48 -0700 Message-ID: <1475087688-4048-1-git-send-email-ebiggers@google.com> Cc: linux-ext4@vger.kernel.org, linux-f2fs-devel@lists.sourceforge.net, tytso@mit.edu, jaegeuk@kernel.org, richard@nod.at, Eric Biggers To: linux-fsdevel@vger.kernel.org Return-path: Sender: linux-fsdevel-owner@vger.kernel.org List-Id: linux-ext4.vger.kernel.org i_rwsem needs to be acquired while setting an encryption policy so that concurrent calls to FS_IOC_SET_ENCRYPTION_POLICY are correctly serialized (especially the ->get_context() + ->set_context() pair), and so that new files cannot be created in the directory during or after the ->empty_dir() check. Signed-off-by: Eric Biggers --- Note: to be applied without conflict, this patch needs the changes from v4.8-rc6. --- fs/crypto/policy.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/fs/crypto/policy.c b/fs/crypto/policy.c index ed115ac..6865663 100644 --- a/fs/crypto/policy.c +++ b/fs/crypto/policy.c @@ -109,6 +109,8 @@ int fscrypt_process_policy(struct file *filp, if (ret) return ret; + inode_lock(inode); + if (!inode_has_encryption_context(inode)) { if (!S_ISDIR(inode->i_mode)) ret = -EINVAL; @@ -127,6 +129,8 @@ int fscrypt_process_policy(struct file *filp, ret = -EINVAL; } + inode_unlock(inode); + mnt_drop_write_file(filp); return ret; } -- 2.8.0.rc3.226.g39d4020