Commit graph

69 commits

Author SHA1 Message Date
Mark Haines
d7bc00c81d Merge pull request #23 from matrix-org/markjh/remove_message_index
Remove the messsage index from olm_init_inbound_group_session
2016-09-13 17:54:14 +01:00
Mark Haines
a628ef41bd Remove the messsage index from olm_init_inbound_group_session since it is read from the session_key 2016-09-13 17:51:02 +01:00
Mark Haines
71bcaa5d45 Add a test to check the equivalence of session ids for inbound and outbound sessions 2016-09-13 17:15:28 +01:00
Mark Haines
d62e344db7 Use the ed22519 public key as the group session id.
Some clients expect the session id to be globally unique,
so allowing the end devices to pick the session id will cause
problems.

Include the current ratchet index with the initial keys, this decreases
the risk that the client will supply the wrong index causing problems.

Sign the initial keys with the ratchet ed25519 key, this reduces the
risk of a client claiming a session that they didn't create.
2016-09-13 15:42:47 +01:00
Richard van der Hoff
2fc83aa9ac Sign megolm messages
Add ed25519 keys to the inbound and outbound sessions, and use them to sign and
verify megolm messages.

We just stuff the ed25519 public key in alongside the megolm session key (and
add a version byte), to save adding more boilerplate to the JS/python/etc
layers.
2016-09-06 15:26:26 +01:00
Richard van der Hoff
69f269ffaf Convert AES functions to plain C 2016-09-05 10:40:39 +01:00
Richard van der Hoff
f0acf6582f Convert Ed25519 and Curve25519 functions to plain C 2016-09-05 10:40:39 +01:00
Richard van der Hoff
39212987bd Create new constants for key lengths, etc
We were using olm::KEY_LENGTH for everything under the sun which happened to be
32 bytes long, and making a bunch of assumptions in the process. Create a bunch
of new constants (as C #defines rather than C++ consts so that I can use them
in another forthcoming refactor).
2016-09-02 15:11:14 +01:00
Richard van der Hoff
0c462cff11 Fix Ed25519 keypair generation
Ed25519 private keys, it turns out, have 64 bytes, not 32.

We were previously generating only 32 bytes (which is all that is required to
generate the public key), and then using the public key as the upper 32 bytes
when generating the per-message session key. This meant that everything
appeared to work, but the security of the private key was severely compromised.

By way of fixes:

 * Use the correct algorithm for generating the Ed25519 private key, and store
   all 512 bits of it.

 * Update the account pickle format and refuse to load the old format (since we
   should consider it compromised).

 * Bump the library version, and add a function to retrieve the library
   version, so that applications can verify that they are linked against a
   fixed version of the library.

 * Remove the curve25519_{sign, verify} functions which were unused and of
   dubious quality.
2016-09-01 13:35:23 +01:00
Richard van der Hoff
d8b24acb77 PR feedback
* write V1 pickles on the master branch
* the logging branch is going to write v0x80000001
2016-06-30 14:01:02 +01:00
Richard van der Hoff
8dd3c182ee Make space in the session pickle for chain index
Keeping track of the chain index is a useful thing to do, but is only required
if we've enabled diagnostics. Extend the session pickle format to make a space
for it, so that pickles can be transferred between the logging_enabled branch
and the master branch without loss of information.

Also add some tests for session pickling which explicitly check that we can
unpickle both formats of pickle.
2016-06-30 11:38:01 +01:00
Richard van der Hoff
757c422578 Remove unused 'chain_index' from Ratchet
This was introduced when I was experimenting with support for logging progress
in Olm. That is now relegated to the logging_enabled branch, so this should
probably be removed.

This also fixes the incompatibility of session pickles from the current master
branch with those from olm 0.1.0.
2016-06-30 11:31:36 +01:00
Richard van der Hoff
708fddd747 Remove session_id from group messages
Putting the session_id inside the packed message body makes it hard to extract
so that we can decide which session to use. We don't think there is any
advantage to having thes sesion_id protected by the HMACs, so we're going to
move it to the JSON framing.
2016-05-25 17:42:32 +01:00
Mark Haines
19a7fb5df5 Fix an integer wrap around bug and add a couple more tests 2016-05-25 15:00:05 +01:00
Richard van der Hoff
01ea3d4b9a Fix handling of integer wraparound in megolm.c 2016-05-24 17:52:35 +01:00
Richard van der Hoff
fc4756ddf1 Fix up some names, and protobuf tags
Make names (of session_key and message_index) more consistent.

Use our own protobuf tags rather than trying to piggyback on the one-to-one
structure.
2016-05-24 13:40:21 +01:00
Richard van der Hoff
a073d12d83 Support for pickling inbound group sessions 2016-05-24 13:40:21 +01:00
Richard van der Hoff
39ad75314b Implement decrypting inbound group messages
Includes creation of inbound sessions, etc
2016-05-24 13:39:34 +01:00
Richard van der Hoff
c058554132 Implement pickling/unpickling for outbound group sessions 2016-05-24 13:39:34 +01:00
Richard van der Hoff
caaed796ad Implementation of an outbound group session 2016-05-24 13:39:34 +01:00
Richard van der Hoff
68d3c7bfa9 Implementation of the megolm ratchet 2016-05-24 13:39:34 +01:00
Richard van der Hoff
2fd28a6682 Rewrite _olm_cipher_aes_sha_256 initialisation
Replace the init-static-var dance with some preprocessor macros
2016-05-24 12:06:47 +01:00
Richard van der Hoff
444ef1f706 Prefix for internal symbols
Give a load of internal symbols "_olm_" prefixes. This better delineates the
public and private interfaces in the module, and helps avoid internal symbols
leaking out and possibly being abused.
2016-05-23 18:55:06 +01:00
Richard van der Hoff
c57b2b71c5 C bindings for base64 functions 2016-05-23 18:55:06 +01:00
Richard van der Hoff
294cf482ea Convert cipher.hh to plain C 2016-05-23 18:55:05 +01:00
Richard van der Hoff
e533b0dc8e Give SHA256 functions C bindings 2016-05-23 18:55:05 +01:00
Richard van der Hoff
4f1bb49d20 Rename olm.hh to olm.h 2016-05-23 18:55:05 +01:00
Richard van der Hoff
e21d5cb222 Fix warnings and set -Werror 2016-05-23 18:08:22 +01:00
Mark Haines
b318055185 Replace hard coded references to the 32-byte key length with a constant, add utilities for copying data to and from fixed sized arrays 2015-08-19 17:32:06 +01:00
Mark Haines
158f7ee891 Fix crash where the message length was shorter than the length of the mac 2015-08-07 19:33:48 +01:00
Mark Haines
a4b2927884 Initialise the length fields of the reader struct in decode_message, even if the message is invalid, fixes a crash where the message was too short 2015-08-07 18:58:42 +01:00
Mark Haines
76ecd85c2c Fix a crash when decoding messages that are too short 2015-08-07 18:25:21 +01:00
Mark Haines
39c1f3b355 Add methods for computing sha256 hashes and validating ed25519 signatures 2015-07-24 14:29:52 +01:00
Mark Haines
3468886e27 Add method getting a session id. Update the python and javascript bindings 2015-07-16 11:45:20 +01:00
Mark Haines
7523b700cf Add a test of olm which uses malloc to create it's buffers so that it is possible to check of out of bounds reads and writes using valgrind 2015-07-15 17:34:48 +01:00
Mark Haines
3bfa8e5863 Fix bug in list where the wrong value was copied if an item was inserted at the beinging of the list 2015-07-15 16:31:45 +01:00
Mark Haines
a59fbdfe7f Add a test for pickling and unpickling sessions, fix off by one error when unpickling sessions 2015-07-14 11:32:11 +01:00
Mark Haines
b6e248c9a5 Output simpler JSON for the account keys, don't sign the JSON but instead provide a olm_account_sign method so that the user of the library can sign the JSON themselves 2015-07-10 11:57:53 +01:00
Mark Haines
373acefde7 Add c bindings for the methods for managing one time keys 2015-07-09 18:35:54 +01:00
Mark Haines
532dc0d4e7 Change the JSON format for one time keys to include what algorithm they are for 2015-07-08 15:30:34 +01:00
Mark Haines
5291ec78b5 Send the public part of the one time key rather than passing an identifier 2015-07-08 14:53:25 +01:00
Mark Haines
3a382aec59 Encode the account keys as a signed JSON object 2015-07-07 16:42:03 +01:00
Mark Haines
2a873fd4e1 Add functions for creating and verifying ed25519 signatures 2015-07-07 09:50:32 +01:00
Mark Haines
a30a64d179 Rename generate_key to curve25519_generate_key 2015-07-07 09:03:12 +01:00
Matthew Hodgson
09d4125ff1 Rename axolotlpp as olm to avoid confusion with Axolotl-the-spec and Axolotl-the-OWS-libraries at moxie's request 2015-06-27 01:15:23 +02:00
Mark Haines
84d6fbb0dc Add some tests which advance the ratchet on both sides a few times 2015-06-26 14:42:24 +01:00
Mark Haines
0ca3797d2d Make the 'random' numbers different for each key in the tests 2015-06-21 18:33:46 +01:00
Mark Haines
9cd53394e7 Fix a few valgrind warnings in the tests 2015-06-20 20:13:31 +01:00
Mark Haines
76f49cf177 Add a test for the axolotl API 2015-06-16 15:15:40 +01:00
Mark Haines
7cdde73560 Add base64 encoders and decoders 2015-06-15 17:48:09 +01:00