kernel/linux.git/net/ceph/auth_x.c, branch linux-4.13.y

libceph: fix error handling in process_one_ticket()

2017-05-23T18:32:28+00:00

Don't leak key internals after new_session_key is populated. Signed-off-by: Ilya Dryomov Reviewed-by: Alex Elder

libceph: validate blob_struct_v in process_one_ticket()

2017-05-23T18:32:25+00:00

None of these are validated in userspace, but since we do validate reply_struct_v in ceph_x_proc_ticket_reply(), tkt_struct_v (first) and CephXServiceTicket struct_v (second) in process_one_ticket(), validate CephXTicketBlob struct_v as well. Signed-off-by: Ilya Dryomov Reviewed-by: Alex Elder

libceph: drop len argument of *verify_authorizer_reply()

2016-12-12T22:09:21+00:00

The length of the reply is protocol-dependent - for cephx it's ceph_x_authorize_reply. Nothing sensible can be passed from the messenger layer anyway. Signed-off-by: Ilya Dryomov Reviewed-by: Sage Weil

libceph: switch ceph_x_decrypt() to ceph_crypt()

2016-12-12T22:09:19+00:00

Signed-off-by: Ilya Dryomov Reviewed-by: Sage Weil

libceph: switch ceph_x_encrypt() to ceph_crypt()

2016-12-12T22:09:19+00:00

Signed-off-by: Ilya Dryomov Reviewed-by: Sage Weil

libceph: tweak calcu_signature() a little

2016-12-12T22:09:19+00:00

- replace an ad-hoc array with a struct - rename to calc_signature() for consistency Signed-off-by: Ilya Dryomov Reviewed-by: Sage Weil

libceph: rename and align ceph_x_authorizer::reply_buf

2016-12-12T22:09:19+00:00

It's going to be used as a temporary buffer for in-place en/decryption with ceph_crypt() instead of on-stack buffers, so rename to enc_buf. Ensure alignment to avoid GFP_ATOMIC allocations in the crypto stack. Signed-off-by: Ilya Dryomov Reviewed-by: Sage Weil

libceph: introduce ceph_x_encrypt_offset()

2016-12-12T22:09:19+00:00

Signed-off-by: Ilya Dryomov Reviewed-by: Sage Weil

libceph: old_key in process_one_ticket() is redundant

2016-12-12T22:09:19+00:00

Since commit 0a990e709356 ("ceph: clean up service ticket decoding"), th->session_key isn't assigned until everything is decoded. Signed-off-by: Ilya Dryomov Reviewed-by: Sage Weil

libceph: ceph_x_encrypt_buflen() takes in_len

2016-12-12T22:09:19+00:00

Pass what's going to be encrypted - that's msg_b, not ticket_blob. ceph_x_encrypt_buflen() returns the upper bound, so this doesn't change the maxlen calculation, but makes it a bit clearer. Signed-off-by: Ilya Dryomov Reviewed-by: Sage Weil