- Generate "semi random" values for TVLA using biased values in middle rounds.
- ~2000 inputs per second on my laptop.
- Use
scipy'struncnorm(truncated normal distribution) to respect normal hamming weight distribution of values.
Example:
from tvla_semirandom import aes_semi_random_generator
from os import urandom
key = urandom(16)
tvla_generator = aes_semi_random_generator(key, 0, 8, 5, 'mix_columns')
for _ in range(1_000):
(plain, cipher) = next(tvla_generator)
Investigate the intermediate states:
from tvla_semirandom import aes_semi_random_generator, AESSemi from os import urandom tvla_generator = aes_semi_random_generator(key, 0, 8, 12, 'sub_bytes') plain, cipher = next(tvla_generator) AESSemi(key, print_states=True).encrypt_block(plain) # Outputs: # 0 add_round_key 65f44af03eba41f564c28fff3a96abeb # 1 sub_bytes 681960c160d74526a603a613d8ffd33c # 1 shift_rows 45d4d078d00e6ef7247b247d611666eb # 1 mix_columns 450e24ebd07b66782416d0f761d46e7d # 1 add_round_key 57de252828f4ef8655948b5fb61deee3 # 2 sub_bytes f3d646623215d199ecdf6abc84ebcd7a # 2 shift_rows 0df65aaa23593eeece9e02655fe9bdda # 2 mix_columns 0d5902da239ebdaacee95aee5ff63e65 # 2 add_round_key 296325e3e87239091307ba3de48f7ae3 # 3 sub_bytes 67a8e1f1f8d4f9c8c1605310d481eb19 # 3 shift_rows 85c2f8a1414899e878d0edca480ce9d4 # 3 mix_columns 8548edd441d0e9a1780cf8e848c299ca # 3 add_round_key f0ed6b82a17ba0a3f49bbcb79eade60c # 4 sub_bytes 504e89e51b397cdbf792812caf52f80e # 4 shift_rows 532fa7d9af1210b9684f0c71790041ab # 4 mix_columns 53120cabaf4f41d96800a7b9792f1071 # 4 add_round_key 37c8bfa60c2b124dce23ed76e266e556 # 5 sub_bytes 6d710c7346346159565b774f4a10ee95 # 5 shift_rows 3ca3fe8f5a18efcbb139f584d6ca282a # 5 mix_columns 3c18f52a5a39288fb1cafecbd6a3ef84 # 5 add_round_key 8f22abfd58dfb9fa09ecda712225273e # 6 sub_bytes edb962b48006accbd23cf2dbc80a1196 # 6 shift_rows 5556aa8dcd6f911fb5eb89b9e8678290 # 6 mix_columns 556f8990cdeb828db567aa1fe85691b9 # 6 add_round_key 029b9822a810b5246d81bc371955578d # 7 sub_bytes 4927e970a9b3b762f45a244828f8c431 # 7 shift_rows 3bcc1e51d36da9aabfbe365234411cc7 # 7 mix_columns 3b6d36c7d3be1c51bf411eaa34cca952 # 7 add_round_key 307c688329c1a66e12b527cadc05479d # 8 sub_bytes 95727daf5416a673b4b20f7d902d5982 # 8 shift_rows 2a40ff792047248f8d3776ff60d8cb13 # 8 mix_columns 2a4776132037cb798dd8ff8f604024ff # 8 add_round_key f82db469ab71116e02b33aaedb737221 # 9 sub_bytes 8fe30512dd1ca253ed0511ec0568cadf # 9 shift_rows 73116bc9c19c3aed556b82ce6b45749e # 9 mix_columns 739c829ec16b74c955456bed6b113ace # 9 add_round_key 455349ac994202cee38fea1011c94711 # 10 sub_bytes fdf2303b5c347b4480febb2d3e900833 # 10 shift_rows 548904e24a18211bcdbbead8b26030c3 # 10 mix_columns 5418eac34abb30e2cd60041bb28921d8 # 10 add_round_key a982dd939095ac8a3e1a881e06000ace # 11 sub_bytes 05c8fffd4ab23dd90b8b320fed8a0821 # 11 shift_rows 6be81654d63727352b3d2376557e30fd # 11 mix_columns 6b3723fdd63d30542b7e163555e82776 # 11 add_round_key 519d064894a877c4f7d8267fd88169dc # 12 sub_bytes 00420800000100060000008000000001 # 12 shift_rows 632c3063637c636f636363cd6363637c # 12 mix_columns 637c637c636363636363306f632c63cd # 12 add_round_key 5d425d42636363633c9ad1281c53c56b # 13 sub_bytes ddf1be4d39f7113f539f196598dc0fc9 # 13 shift_rows c1a1aee312688275eddbd44d468676dd # 13 mix_columns c168d4dd12db76e3ed86ae7546a1824d # 13 add_round_key 28ab6645c7c61b468b66b3eebbcf2f73 # 14 sub_bytes 26071c375dc316f6e6bb98a10e936de1 # 14 shift_rows f7c59c9a4c2e47428eea4632abdc3cf8 # 14 add_round_key f72e46f84cea3c9a8edc9c42abc54732 # b'}\xb1\xea"\x9c\xe1\xe2\x1c1\xd2\x8a\x89\x90D\x9b['
python -m unittest
python -m build
pip install tvla_semirandom-x.x.x.tar.gz or pip install tvla_semirandom-x.x.x.whl