59 add hmac support #93
Conversation
src/src_main/crypto_tc.c
Outdated
| status = cryptography_if->cryptography_aead_encrypt(&p_new_enc_frame[index], // ciphertext output | ||
| (size_t)tf_payload_len, // length of data | ||
| (uint8_t*)(p_in_frame + TC_FRAME_HEADER_SIZE + segment_hdr_len), // plaintext input | ||
| (size_t)tf_payload_len, // in data length | ||
| NULL, // Using SA key reference, key is null | ||
| KEY_SIZE, // Length of key. TODO - why is this hard-coded? | ||
| ek_ring[sa_ptr->ekid].key_len, // Length of key derived from sa_ptr key_ref |
There was a problem hiding this comment.
I thought we agreed in our weekly meetings that key length would be a function of the ECS/ACS algorithm enums being chosen -- EG, if ecs is CRYPTO_CIPHER_AES256_GCM, then that means we have a 32 byte key. (256/8 = 32). AES256 must have a 32 byte key, since we know that's the algorithm, we don't need to reference the ek_ring.
There was a problem hiding this comment.
<not duplicating my comments for all new key_len functions in this PR, this applies anywhere key_len is referenced>
Codecov Report
@@ Coverage Diff @@
## dev #93 +/- ##
==========================================
+ Coverage 83.51% 84.59% +1.07%
==========================================
Files 18 18
Lines 4229 4954 +725
==========================================
+ Hits 3532 4191 +659
- Misses 697 763 +66
Continue to review full report at Codecov.
|
No description provided.