//encryption
void aes_block_enc(uint8 *data, uint8 *key)
{
uint32 i;
uint32 k0, k1, k2, k3;
uint32 s0, s1, s2, s3;
uint32 t0, t1, t2, t3;
//load 128-bit message into 32-bit words
LOAD32(data, &s0);
LOAD32(data + 4, &s1);
LOAD32(data + 8, &s2);
LOAD32(data + 12, &s3);
//load 128-bit key into 32-bit words
LOAD32(key, &k0);
LOAD32(key + 4, &k1);
LOAD32(key + 8, &k2);
LOAD32(key + 12, &k3);
//add round key
s0 ^= k0;
s1 ^= k1;
s2 ^= k2;
s3 ^= k3;
//9 regular rounds
for (i = 9; i != 0; i--)
{
k0 ^= rcon[9 - i] ^ Te4_3(byte(k3, 2)) ^ Te4_2(byte(k3, 1)) ^ Te4_1(byte(k3, 0)) ^ Te4_0(byte(k3, 3));
k1 ^= k0;
k2 ^= k1;
k3 ^= k2;
t0 = Te0(byte(s0, 0)) ^ Te1(byte(s1, 1)) ^ Te2(byte(s2, 2)) ^ Te3(byte(s3, 3)) ^ k0;
t1 = Te0(byte(s1, 3)) ^ Te1(byte(s2, 2)) ^ Te2(byte(s3, 1)) ^ Te3(byte(s0, 0)) ^ k1;
t2 = Te0(byte(s2, 3)) ^ Te1(byte(s3, 2)) ^ Te2(byte(s0, 1)) ^ Te3(byte(s1, 0)) ^ k2;
t3 = Te0(byte(s3, 3)) ^ Te1(byte(s0, 2)) ^ Te2(byte(s1, 1)) ^ Te3(byte(s2, 0)) ^ k3;
s0 = t0;
s1 = t1;
s2 = t2;
s3 = t3;
}
// the last round for the full AES
k0 ^= rcon[9] ^ Te4_3(byte(k3, 2)) ^ Te4_2(byte(k3, 1)) ^ Te4_1(byte(k3, 0)) ^ Te4_0(byte(k3, 3));
k1 ^= k0;
k2 ^= k1;
k3 ^= k2;
s0 = Te4_3(byte(t0, 3)) ^ Te4_2(byte(t1, 2)) ^ Te4_1(byte(t2, 1)) ^ Te4_0(byte(t3, 0)) ^ k0;
s1 = Te4_3(byte(t1, 3)) ^ Te4_2(byte(t2, 2)) ^ Te4_1(byte(t3, 1)) ^ Te4_0(byte(t0, 0)) ^ k1;
s2 = Te4_3(byte(t2, 3)) ^ Te4_2(byte(t3, 2)) ^ Te4_1(byte(t0, 1)) ^ Te4_0(byte(t1, 0)) ^ k2;
s3 = Te4_3(byte(t3, 3)) ^ Te4_2(byte(t0, 2)) ^ Te4_1(byte(t1, 1)) ^ Te4_0(byte(t2, 0)) ^ k3;
STORE32(s0, data);
STORE32(s1, data + 4);
STORE32(s2, data + 8);
STORE32(s3, data + 12);
}
static void four_rounds(pelican_state *pelmac)
{
ulong32 s0, s1, s2, s3, t0, t1, t2, t3;
int r;
LOAD32H(s0, pelmac->state );
LOAD32H(s1, pelmac->state + 4);
LOAD32H(s2, pelmac->state + 8);
LOAD32H(s3, pelmac->state + 12);
for (r = 0; r < 4; r++) {
t0 =
Te0(byte(s0, 3)) ^
Te1(byte(s1, 2)) ^
Te2(byte(s2, 1)) ^
Te3(byte(s3, 0));
t1 =
Te0(byte(s1, 3)) ^
Te1(byte(s2, 2)) ^
Te2(byte(s3, 1)) ^
Te3(byte(s0, 0));
t2 =
Te0(byte(s2, 3)) ^
Te1(byte(s3, 2)) ^
Te2(byte(s0, 1)) ^
Te3(byte(s1, 0));
t3 =
Te0(byte(s3, 3)) ^
Te1(byte(s0, 2)) ^
Te2(byte(s1, 1)) ^
Te3(byte(s2, 0));
s0 = t0; s1 = t1; s2 = t2; s3 = t3;
}
STORE32H(s0, pelmac->state );
STORE32H(s1, pelmac->state + 4);
STORE32H(s2, pelmac->state + 8);
STORE32H(s3, pelmac->state + 12);
}
int rijndael_ecb_encrypt(const unsigned char *pt, unsigned char *ct, rijndael_key *rijndael)
{
unsigned long s0, s1, s2, s3, t0, t1, t2, t3, *rk;
int Nr, r;
LTC_ARGCHK(pt != NULL);
LTC_ARGCHK(ct != NULL);
LTC_ARGCHK(rijndael != NULL);
Nr = rijndael->Nr;
rk = rijndael->eK;
/*
* map byte array block to cipher state
* and add initial round key:
*/
LOAD32H(s0, pt); s0 ^= rk[0];
LOAD32H(s1, pt + 4); s1 ^= rk[1];
LOAD32H(s2, pt + 8); s2 ^= rk[2];
LOAD32H(s3, pt + 12); s3 ^= rk[3];
/*
* Nr - 1 full rounds:
*/
r = Nr >> 1;
for (;;) {
t0 =
Te0(byte(s0, 3)) ^
Te1(byte(s1, 2)) ^
Te2(byte(s2, 1)) ^
Te3(byte(s3, 0)) ^
rk[4];
t1 =
Te0(byte(s1, 3)) ^
Te1(byte(s2, 2)) ^
Te2(byte(s3, 1)) ^
Te3(byte(s0, 0)) ^
rk[5];
t2 =
Te0(byte(s2, 3)) ^
Te1(byte(s3, 2)) ^
Te2(byte(s0, 1)) ^
Te3(byte(s1, 0)) ^
rk[6];
t3 =
Te0(byte(s3, 3)) ^
Te1(byte(s0, 2)) ^
Te2(byte(s1, 1)) ^
Te3(byte(s2, 0)) ^
rk[7];
rk += 8;
if (--r == 0)
break;
s0 =
Te0(byte(t0, 3)) ^
Te1(byte(t1, 2)) ^
Te2(byte(t2, 1)) ^
Te3(byte(t3, 0)) ^
rk[0];
s1 =
Te0(byte(t1, 3)) ^
Te1(byte(t2, 2)) ^
Te2(byte(t3, 1)) ^
Te3(byte(t0, 0)) ^
rk[1];
s2 =
Te0(byte(t2, 3)) ^
Te1(byte(t3, 2)) ^
Te2(byte(t0, 1)) ^
Te3(byte(t1, 0)) ^
rk[2];
s3 =
Te0(byte(t3, 3)) ^
Te1(byte(t0, 2)) ^
Te2(byte(t1, 1)) ^
Te3(byte(t2, 0)) ^
rk[3];
}
/*
* apply last round and
* map cipher state to byte array block:
*/
s0 =
(Te4_3[byte(t0, 3)]) ^
(Te4_2[byte(t1, 2)]) ^
(Te4_1[byte(t2, 1)]) ^
(Te4_0[byte(t3, 0)]) ^
rk[0];
STORE32H(s0, ct);
s1 =
(Te4_3[byte(t1, 3)]) ^
(Te4_2[byte(t2, 2)]) ^
(Te4_1[byte(t3, 1)]) ^
(Te4_0[byte(t0, 0)]) ^
rk[1];
STORE32H(s1, ct + 4);
s2 =
(Te4_3[byte(t2, 3)]) ^
(Te4_2[byte(t3, 2)]) ^
(Te4_1[byte(t0, 1)]) ^
(Te4_0[byte(t1, 0)]) ^
rk[2];
STORE32H(s2, ct + 8);
s3 =
(Te4_3[byte(t3, 3)]) ^
(Te4_2[byte(t0, 2)]) ^
(Te4_1[byte(t1, 1)]) ^
(Te4_0[byte(t2, 0)]) ^
rk[3];
STORE32H(s3, ct + 12);
return CRYPT_OK;
}
int ECB_ENC(const unsigned char *pt, unsigned char *ct, symmetric_key *skey)
#endif
{
ulong32 s0, s1, s2, s3, t0, t1, t2, t3, *rk;
int Nr, r;
LTC_ARGCHK(pt != NULL);
LTC_ARGCHK(ct != NULL);
LTC_ARGCHK(skey != NULL);
Nr = skey->rijndael.Nr;
rk = skey->rijndael.eK;
/*
* map byte array block to cipher state
* and add initial round key:
*/
LOAD32H(s0, pt ); s0 ^= rk[0];
LOAD32H(s1, pt + 4); s1 ^= rk[1];
LOAD32H(s2, pt + 8); s2 ^= rk[2];
LOAD32H(s3, pt + 12); s3 ^= rk[3];
#ifdef LTC_SMALL_CODE
for (r = 0; ; r++) {
rk += 4;
t0 =
Te0(byte(s0, 3)) ^
Te1(byte(s1, 2)) ^
Te2(byte(s2, 1)) ^
Te3(byte(s3, 0)) ^
rk[0];
t1 =
Te0(byte(s1, 3)) ^
Te1(byte(s2, 2)) ^
Te2(byte(s3, 1)) ^
Te3(byte(s0, 0)) ^
rk[1];
t2 =
Te0(byte(s2, 3)) ^
Te1(byte(s3, 2)) ^
Te2(byte(s0, 1)) ^
Te3(byte(s1, 0)) ^
rk[2];
t3 =
Te0(byte(s3, 3)) ^
Te1(byte(s0, 2)) ^
Te2(byte(s1, 1)) ^
Te3(byte(s2, 0)) ^
rk[3];
if (r == Nr-2) {
break;
}
s0 = t0; s1 = t1; s2 = t2; s3 = t3;
}
rk += 4;
#else
/*
* Nr - 1 full rounds:
*/
r = Nr >> 1;
for (;;) {
t0 =
Te0(byte(s0, 3)) ^
Te1(byte(s1, 2)) ^
Te2(byte(s2, 1)) ^
Te3(byte(s3, 0)) ^
rk[4];
t1 =
Te0(byte(s1, 3)) ^
Te1(byte(s2, 2)) ^
Te2(byte(s3, 1)) ^
Te3(byte(s0, 0)) ^
rk[5];
t2 =
Te0(byte(s2, 3)) ^
Te1(byte(s3, 2)) ^
Te2(byte(s0, 1)) ^
Te3(byte(s1, 0)) ^
rk[6];
t3 =
Te0(byte(s3, 3)) ^
Te1(byte(s0, 2)) ^
Te2(byte(s1, 1)) ^
Te3(byte(s2, 0)) ^
rk[7];
rk += 8;
if (--r == 0) {
break;
}
s0 =
Te0(byte(t0, 3)) ^
Te1(byte(t1, 2)) ^
Te2(byte(t2, 1)) ^
Te3(byte(t3, 0)) ^
rk[0];
s1 =
Te0(byte(t1, 3)) ^
Te1(byte(t2, 2)) ^
Te2(byte(t3, 1)) ^
Te3(byte(t0, 0)) ^
rk[1];
s2 =
Te0(byte(t2, 3)) ^
Te1(byte(t3, 2)) ^
Te2(byte(t0, 1)) ^
Te3(byte(t1, 0)) ^
rk[2];
s3 =
Te0(byte(t3, 3)) ^
Te1(byte(t0, 2)) ^
Te2(byte(t1, 1)) ^
Te3(byte(t2, 0)) ^
rk[3];
}
#endif
/*
* apply last round and
* map cipher state to byte array block:
*/
s0 =
(Te4_3[byte(t0, 3)]) ^
(Te4_2[byte(t1, 2)]) ^
(Te4_1[byte(t2, 1)]) ^
(Te4_0[byte(t3, 0)]) ^
rk[0];
STORE32H(s0, ct);
s1 =
(Te4_3[byte(t1, 3)]) ^
(Te4_2[byte(t2, 2)]) ^
(Te4_1[byte(t3, 1)]) ^
(Te4_0[byte(t0, 0)]) ^
rk[1];
STORE32H(s1, ct+4);
s2 =
(Te4_3[byte(t2, 3)]) ^
(Te4_2[byte(t3, 2)]) ^
(Te4_1[byte(t0, 1)]) ^
(Te4_0[byte(t1, 0)]) ^
rk[2];
STORE32H(s2, ct+8);
s3 =
(Te4_3[byte(t3, 3)]) ^
(Te4_2[byte(t0, 2)]) ^
(Te4_1[byte(t1, 1)]) ^
(Te4_0[byte(t2, 0)]) ^
rk[3];
STORE32H(s3, ct+12);
return CRYPT_OK;
}
/**
Encrypts a block of text with AES
@param pt The input plaintext (16 bytes)
@param ct The output ciphertext (16 bytes)
@param skey The key as scheduled
@return CRYPT_OK if successful
*/
static int aes_encrypt(aes_key_t* ctx, uint8_t* pt, uint8_t* ct)
{
ulong32 s0, s1, s2, s3, t0, t1, t2, t3, *rk;
int Nr, r;
Nr = ctx->Nr;
rk = ctx->eK;
/*
* map byte array block to cipher state
* and add initial round key:
*/
LOAD32H(s0, pt ); s0 ^= rk[0];
LOAD32H(s1, pt + 4); s1 ^= rk[1];
LOAD32H(s2, pt + 8); s2 ^= rk[2];
LOAD32H(s3, pt + 12); s3 ^= rk[3];
/*
* Nr - 1 full rounds:
*/
r = Nr >> 1;
for (;;) {
t0 =
Te0(byte(s0, 3)) ^
Te1(byte(s1, 2)) ^
Te2(byte(s2, 1)) ^
Te3(byte(s3, 0)) ^
rk[4];
t1 =
Te0(byte(s1, 3)) ^
Te1(byte(s2, 2)) ^
Te2(byte(s3, 1)) ^
Te3(byte(s0, 0)) ^
rk[5];
t2 =
Te0(byte(s2, 3)) ^
Te1(byte(s3, 2)) ^
Te2(byte(s0, 1)) ^
Te3(byte(s1, 0)) ^
rk[6];
t3 =
Te0(byte(s3, 3)) ^
Te1(byte(s0, 2)) ^
Te2(byte(s1, 1)) ^
Te3(byte(s2, 0)) ^
rk[7];
rk += 8;
if (--r == 0) {
break;
}
s0 =
Te0(byte(t0, 3)) ^
Te1(byte(t1, 2)) ^
Te2(byte(t2, 1)) ^
Te3(byte(t3, 0)) ^
rk[0];
s1 =
Te0(byte(t1, 3)) ^
Te1(byte(t2, 2)) ^
Te2(byte(t3, 1)) ^
Te3(byte(t0, 0)) ^
rk[1];
s2 =
Te0(byte(t2, 3)) ^
Te1(byte(t3, 2)) ^
Te2(byte(t0, 1)) ^
Te3(byte(t1, 0)) ^
rk[2];
s3 =
Te0(byte(t3, 3)) ^
Te1(byte(t0, 2)) ^
Te2(byte(t1, 1)) ^
Te3(byte(t2, 0)) ^
rk[3];
}
/*
* apply last round and
* map cipher state to byte array block:
*/
s0 =
(Te4_3[byte(t0, 3)]) ^
(Te4_2[byte(t1, 2)]) ^
(Te4_1[byte(t2, 1)]) ^
(Te4_0[byte(t3, 0)]) ^
rk[0];
STORE32H(s0, ct);
s1 =
(Te4_3[byte(t1, 3)]) ^
(Te4_2[byte(t2, 2)]) ^
(Te4_1[byte(t3, 1)]) ^
(Te4_0[byte(t0, 0)]) ^
rk[1];
STORE32H(s1, ct+4);
s2 =
(Te4_3[byte(t2, 3)]) ^
(Te4_2[byte(t3, 2)]) ^
(Te4_1[byte(t0, 1)]) ^
(Te4_0[byte(t1, 0)]) ^
rk[2];
STORE32H(s2, ct+8);
s3 =
(Te4_3[byte(t3, 3)]) ^
(Te4_2[byte(t0, 2)]) ^
(Te4_1[byte(t1, 1)]) ^
(Te4_0[byte(t2, 0)]) ^
rk[3];
STORE32H(s3, ct+12);
}
