rvsim-core 2.0.0

//! AES rounds and key schedule (Zvkned).

/// AES S-box (forward).
pub(super) const AES_SBOX: [u8; 256] = [
    0x63, 0x7c, 0x77, 0x7b, 0xf2, 0x6b, 0x6f, 0xc5, 0x30, 0x01, 0x67, 0x2b, 0xfe, 0xd7, 0xab, 0x76,
    0xca, 0x82, 0xc9, 0x7d, 0xfa, 0x59, 0x47, 0xf0, 0xad, 0xd4, 0xa2, 0xaf, 0x9c, 0xa4, 0x72, 0xc0,
    0xb7, 0xfd, 0x93, 0x26, 0x36, 0x3f, 0xf7, 0xcc, 0x34, 0xa5, 0xe5, 0xf1, 0x71, 0xd8, 0x31, 0x15,
    0x04, 0xc7, 0x23, 0xc3, 0x18, 0x96, 0x05, 0x9a, 0x07, 0x12, 0x80, 0xe2, 0xeb, 0x27, 0xb2, 0x75,
    0x09, 0x83, 0x2c, 0x1a, 0x1b, 0x6e, 0x5a, 0xa0, 0x52, 0x3b, 0xd6, 0xb3, 0x29, 0xe3, 0x2f, 0x84,
    0x53, 0xd1, 0x00, 0xed, 0x20, 0xfc, 0xb1, 0x5b, 0x6a, 0xcb, 0xbe, 0x39, 0x4a, 0x4c, 0x58, 0xcf,
    0xd0, 0xef, 0xaa, 0xfb, 0x43, 0x4d, 0x33, 0x85, 0x45, 0xf9, 0x02, 0x7f, 0x50, 0x3c, 0x9f, 0xa8,
    0x51, 0xa3, 0x40, 0x8f, 0x92, 0x9d, 0x38, 0xf5, 0xbc, 0xb6, 0xda, 0x21, 0x10, 0xff, 0xf3, 0xd2,
    0xcd, 0x0c, 0x13, 0xec, 0x5f, 0x97, 0x44, 0x17, 0xc4, 0xa7, 0x7e, 0x3d, 0x64, 0x5d, 0x19, 0x73,
    0x60, 0x81, 0x4f, 0xdc, 0x22, 0x2a, 0x90, 0x88, 0x46, 0xee, 0xb8, 0x14, 0xde, 0x5e, 0x0b, 0xdb,
    0xe0, 0x32, 0x3a, 0x0a, 0x49, 0x06, 0x24, 0x5c, 0xc2, 0xd3, 0xac, 0x62, 0x91, 0x95, 0xe4, 0x79,
    0xe7, 0xc8, 0x37, 0x6d, 0x8d, 0xd5, 0x4e, 0xa9, 0x6c, 0x56, 0xf4, 0xea, 0x65, 0x7a, 0xae, 0x08,
    0xba, 0x78, 0x25, 0x2e, 0x1c, 0xa6, 0xb4, 0xc6, 0xe8, 0xdd, 0x74, 0x1f, 0x4b, 0xbd, 0x8b, 0x8a,
    0x70, 0x3e, 0xb5, 0x66, 0x48, 0x03, 0xf6, 0x0e, 0x61, 0x35, 0x57, 0xb9, 0x86, 0xc1, 0x1d, 0x9e,
    0xe1, 0xf8, 0x98, 0x11, 0x69, 0xd9, 0x8e, 0x94, 0x9b, 0x1e, 0x87, 0xe9, 0xce, 0x55, 0x28, 0xdf,
    0x8c, 0xa1, 0x89, 0x0d, 0xbf, 0xe6, 0x42, 0x68, 0x41, 0x99, 0x2d, 0x0f, 0xb0, 0x54, 0xbb, 0x16,
];

/// AES inverse S-box.
pub(super) const AES_INV_SBOX: [u8; 256] = [
    0x52, 0x09, 0x6a, 0xd5, 0x30, 0x36, 0xa5, 0x38, 0xbf, 0x40, 0xa3, 0x9e, 0x81, 0xf3, 0xd7, 0xfb,
    0x7c, 0xe3, 0x39, 0x82, 0x9b, 0x2f, 0xff, 0x87, 0x34, 0x8e, 0x43, 0x44, 0xc4, 0xde, 0xe9, 0xcb,
    0x54, 0x7b, 0x94, 0x32, 0xa6, 0xc2, 0x23, 0x3d, 0xee, 0x4c, 0x95, 0x0b, 0x42, 0xfa, 0xc3, 0x4e,
    0x08, 0x2e, 0xa1, 0x66, 0x28, 0xd9, 0x24, 0xb2, 0x76, 0x5b, 0xa2, 0x49, 0x6d, 0x8b, 0xd1, 0x25,
    0x72, 0xf8, 0xf6, 0x64, 0x86, 0x68, 0x98, 0x16, 0xd4, 0xa4, 0x5c, 0xcc, 0x5d, 0x65, 0xb6, 0x92,
    0x6c, 0x70, 0x48, 0x50, 0xfd, 0xed, 0xb9, 0xda, 0x5e, 0x15, 0x46, 0x57, 0xa7, 0x8d, 0x9d, 0x84,
    0x90, 0xd8, 0xab, 0x00, 0x8c, 0xbc, 0xd3, 0x0a, 0xf7, 0xe4, 0x58, 0x05, 0xb8, 0xb3, 0x45, 0x06,
    0xd0, 0x2c, 0x1e, 0x8f, 0xca, 0x3f, 0x0f, 0x02, 0xc1, 0xaf, 0xbd, 0x03, 0x01, 0x13, 0x8a, 0x6b,
    0x3a, 0x91, 0x11, 0x41, 0x4f, 0x67, 0xdc, 0xea, 0x97, 0xf2, 0xcf, 0xce, 0xf0, 0xb4, 0xe6, 0x73,
    0x96, 0xac, 0x74, 0x22, 0xe7, 0xad, 0x35, 0x85, 0xe2, 0xf9, 0x37, 0xe8, 0x1c, 0x75, 0xdf, 0x6e,
    0x47, 0xf1, 0x1a, 0x71, 0x1d, 0x29, 0xc5, 0x89, 0x6f, 0xb7, 0x62, 0x0e, 0xaa, 0x18, 0xbe, 0x1b,
    0xfc, 0x56, 0x3e, 0x4b, 0xc6, 0xd2, 0x79, 0x20, 0x9a, 0xdb, 0xc0, 0xfe, 0x78, 0xcd, 0x5a, 0xf4,
    0x1f, 0xdd, 0xa8, 0x33, 0x88, 0x07, 0xc7, 0x31, 0xb1, 0x12, 0x10, 0x59, 0x27, 0x80, 0xec, 0x5f,
    0x60, 0x51, 0x7f, 0xa9, 0x19, 0xb5, 0x4a, 0x0d, 0x2d, 0xe5, 0x7a, 0x9f, 0x93, 0xc9, 0x9c, 0xef,
    0xa0, 0xe0, 0x3b, 0x4d, 0xae, 0x2a, 0xf5, 0xb0, 0xc8, 0xeb, 0xbb, 0x3c, 0x83, 0x53, 0x99, 0x61,
    0x17, 0x2b, 0x04, 0x7e, 0xba, 0x77, 0xd6, 0x26, 0xe1, 0x69, 0x14, 0x63, 0x55, 0x21, 0x0c, 0x7d,
];

/// AES round constants for key schedule (Rcon[i] for i=1..10).
pub(super) const AES_RCON: [u8; 11] =
    [0x00, 0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80, 0x1b, 0x36];

/// xtime: multiply by x in GF(2^8) with the AES reduction polynomial 0x11b.
#[inline]
pub(super) const fn xtime(b: u8) -> u8 {
    let high = b >> 7;
    (b << 1) ^ (high.wrapping_neg() & 0x1b)
}

/// AES `SubBytes` on a 16-byte state.
pub(super) fn sub_bytes(state: &mut [u8; 16]) {
    for b in state.iter_mut() {
        *b = AES_SBOX[*b as usize];
    }
}

/// AES `InvSubBytes` on a 16-byte state.
pub(super) fn inv_sub_bytes(state: &mut [u8; 16]) {
    for b in state.iter_mut() {
        *b = AES_INV_SBOX[*b as usize];
    }
}

/// AES `ShiftRows`: rotate row r left by r bytes (column-major state).
pub(super) fn shift_rows(state: &mut [u8; 16]) {
    let s = *state;
    for c in 0..4 {
        for r in 0..4 {
            state[c * 4 + r] = s[((c + r) % 4) * 4 + r];
        }
    }
}

/// AES `InvShiftRows`: rotate row r right by r bytes.
pub(super) fn inv_shift_rows(state: &mut [u8; 16]) {
    let s = *state;
    for c in 0..4 {
        for r in 0..4 {
            state[c * 4 + r] = s[((c + 4 - r) % 4) * 4 + r];
        }
    }
}

/// AES `MixColumns` on a single column [s0, s1, s2, s3].
#[inline]
pub(super) const fn mix_column(c: [u8; 4]) -> [u8; 4] {
    let s0 = c[0];
    let s1 = c[1];
    let s2 = c[2];
    let s3 = c[3];
    [
        xtime(s0) ^ xtime(s1) ^ s1 ^ s2 ^ s3,
        s0 ^ xtime(s1) ^ xtime(s2) ^ s2 ^ s3,
        s0 ^ s1 ^ xtime(s2) ^ xtime(s3) ^ s3,
        xtime(s0) ^ s0 ^ s1 ^ s2 ^ xtime(s3),
    ]
}

/// AES `MixColumns` on full state.
pub(super) fn mix_columns(state: &mut [u8; 16]) {
    for c in 0..4 {
        let col = [state[c * 4], state[c * 4 + 1], state[c * 4 + 2], state[c * 4 + 3]];
        let r = mix_column(col);
        state[c * 4..c * 4 + 4].copy_from_slice(&r);
    }
}

/// AES `InvMixColumns` on a single column.
#[inline]
pub(super) const fn inv_mix_column(c: [u8; 4]) -> [u8; 4] {
    // M^-1 mult by 0x0e, 0x0b, 0x0d, 0x09 in GF(2^8).
    let s0 = c[0];
    let s1 = c[1];
    let s2 = c[2];
    let s3 = c[3];
    [
        gf_mul(0x0e, s0) ^ gf_mul(0x0b, s1) ^ gf_mul(0x0d, s2) ^ gf_mul(0x09, s3),
        gf_mul(0x09, s0) ^ gf_mul(0x0e, s1) ^ gf_mul(0x0b, s2) ^ gf_mul(0x0d, s3),
        gf_mul(0x0d, s0) ^ gf_mul(0x09, s1) ^ gf_mul(0x0e, s2) ^ gf_mul(0x0b, s3),
        gf_mul(0x0b, s0) ^ gf_mul(0x0d, s1) ^ gf_mul(0x09, s2) ^ gf_mul(0x0e, s3),
    ]
}

/// AES `InvMixColumns` on full state.
pub(super) fn inv_mix_columns(state: &mut [u8; 16]) {
    for c in 0..4 {
        let col = [state[c * 4], state[c * 4 + 1], state[c * 4 + 2], state[c * 4 + 3]];
        let r = inv_mix_column(col);
        state[c * 4..c * 4 + 4].copy_from_slice(&r);
    }
}

/// GF(2^8) multiply with AES reduction polynomial.
#[inline]
pub(super) const fn gf_mul(mut a: u8, mut b: u8) -> u8 {
    let mut acc: u8 = 0;
    let mut i = 0;
    while i < 8 {
        if b & 1 != 0 {
            acc ^= a;
        }
        let high = a >> 7;
        a = (a << 1) ^ (high.wrapping_neg() & 0x1b);
        b >>= 1;
        i += 1;
    }
    acc
}

/// XOR `a` with `b` in-place.
#[inline]
pub(super) fn xor_state(a: &mut [u8; 16], b: &[u8; 16]) {
    for i in 0..16 {
        a[i] ^= b[i];
    }
}

/// Convert 4 little-endian u32 words to a 16-byte state (state[col*4 + row]).
pub(super) fn words_to_state(w: [u32; 4]) -> [u8; 16] {
    let mut s = [0u8; 16];
    for (i, word) in w.iter().enumerate() {
        s[i * 4..i * 4 + 4].copy_from_slice(&word.to_le_bytes());
    }
    s
}

/// Convert 16-byte state back to 4 little-endian u32 words.
pub(super) fn state_to_words(s: [u8; 16]) -> [u32; 4] {
    let mut w = [0u32; 4];
    for i in 0..4 {
        w[i] = u32::from_le_bytes([s[i * 4], s[i * 4 + 1], s[i * 4 + 2], s[i * 4 + 3]]);
    }
    w
}

/// AES middle-round encryption: state = AddRoundKey(MixColumns(ShiftRows(SubBytes(state))), key)
pub(super) fn aes_round_enc(state: [u32; 4], key: [u32; 4]) -> [u32; 4] {
    let mut s = words_to_state(state);
    sub_bytes(&mut s);
    shift_rows(&mut s);
    mix_columns(&mut s);
    let k = words_to_state(key);
    xor_state(&mut s, &k);
    state_to_words(s)
}

/// AES final-round encryption (no `MixColumns`).
pub(super) fn aes_round_enc_final(state: [u32; 4], key: [u32; 4]) -> [u32; 4] {
    let mut s = words_to_state(state);
    sub_bytes(&mut s);
    shift_rows(&mut s);
    let k = words_to_state(key);
    xor_state(&mut s, &k);
    state_to_words(s)
}

/// AES middle-round decryption (per Zvkned spec — equivalent inverse round):
/// state = InvMixColumns(AddRoundKey(InvSubBytes(InvShiftRows(state)), key))
pub(super) fn aes_round_dec(state: [u32; 4], key: [u32; 4]) -> [u32; 4] {
    let mut s = words_to_state(state);
    inv_shift_rows(&mut s);
    inv_sub_bytes(&mut s);
    let k = words_to_state(key);
    xor_state(&mut s, &k);
    inv_mix_columns(&mut s);
    state_to_words(s)
}

/// AES final-round decryption (no `InvMixColumns`).
pub(super) fn aes_round_dec_final(state: [u32; 4], key: [u32; 4]) -> [u32; 4] {
    let mut s = words_to_state(state);
    inv_shift_rows(&mut s);
    inv_sub_bytes(&mut s);
    let k = words_to_state(key);
    xor_state(&mut s, &k);
    state_to_words(s)
}

/// AES round-zero: just XOR with key.
pub(super) const fn aes_round_zero(state: [u32; 4], key: [u32; 4]) -> [u32; 4] {
    [state[0] ^ key[0], state[1] ^ key[1], state[2] ^ key[2], state[3] ^ key[3]]
}

/// `SubWord`: AES S-box applied to each byte of a u32.
#[inline]
pub(super) const fn sub_word(w: u32) -> u32 {
    let bytes = w.to_le_bytes();
    u32::from_le_bytes([
        AES_SBOX[bytes[0] as usize],
        AES_SBOX[bytes[1] as usize],
        AES_SBOX[bytes[2] as usize],
        AES_SBOX[bytes[3] as usize],
    ])
}

/// `RotWord`: rotate a u32 right by 8 bits (= rotate the 4 bytes left by 1).
#[inline]
pub(super) const fn rot_word(w: u32) -> u32 {
    w.rotate_right(8)
}

/// AES-128 forward key schedule round (Zvkned §vaeskf1).
///
/// Per the RVV crypto spec, the round number comes from `uimm[3:0]`
/// (uimm[4] is reserved/ignored). The valid range is 1..=10; if the
/// 4-bit value falls outside that range, the spec says the round
/// number is `uimm[3:0] XOR 0x8`.
pub(super) fn aes_kf1(prev_key: [u32; 4], rnd: u32) -> [u32; 4] {
    let mut r = rnd & 0xf;
    if !(1..=10).contains(&r) {
        r ^= 0x8;
    }
    let r = r as usize;
    let temp = sub_word(rot_word(prev_key[3])) ^ u32::from(AES_RCON[r]);
    let w0 = prev_key[0] ^ temp;
    let w1 = prev_key[1] ^ w0;
    let w2 = prev_key[2] ^ w1;
    let w3 = prev_key[3] ^ w2;
    [w0, w1, w2, w3]
}

/// AES-256 forward key schedule round (Zvkned §vaeskf2).
///
/// Takes the current round key (vd, 4 words) and the previous round key
/// (vs2, 4 words) and a round number `rnd` (2–14). The round number comes
/// from `uimm[3:0]` (uimm[4] is reserved). Out-of-range values are
/// normalised by XOR with 0x8.
///
/// Even rounds apply SubWord(RotWord) ⊕ Rcon, odd rounds just `SubWord`
/// (no rotate, no Rcon).
pub(super) fn aes_kf2(curr_key: [u32; 4], prev_key: [u32; 4], rnd: u32) -> [u32; 4] {
    let mut r = rnd & 0xf;
    if !(2..=14).contains(&r) {
        r ^= 0x8;
    }
    let temp = if r.is_multiple_of(2) {
        let rcon_idx = ((r / 2) as usize).min(AES_RCON.len() - 1);
        sub_word(rot_word(prev_key[3])) ^ u32::from(AES_RCON[rcon_idx])
    } else {
        sub_word(prev_key[3])
    };
    let w0 = curr_key[0] ^ temp;
    let w1 = curr_key[1] ^ w0;
    let w2 = curr_key[2] ^ w1;
    let w3 = curr_key[3] ^ w2;
    [w0, w1, w2, w3]
}