static AYCW_INLINE void aycw_bs_stream_sbox2(dvbcsa_bs_word_t *fa, dvbcsa_bs_word_t *fb,
dvbcsa_bs_word_t *fc, dvbcsa_bs_word_t *fd,
dvbcsa_bs_word_t *fe,
dvbcsa_bs_word_t *sa, dvbcsa_bs_word_t *sb)
{
dvbcsa_bs_word_t tmp0, tmp1, tmp2, tmp3;
tmp0 = BS_XOR (*fa, BS_XOR (BS_AND (*fb, BS_OR (*fc, *fd)), BS_XOR (*fc, BS_NOT (*fd))));
tmp1 = BS_OR (BS_AND (*fa, BS_XOR (*fb, *fd)), BS_AND (BS_OR (*fa, *fb), *fc));
tmp2 = BS_XOR (BS_AND (*fb, *fd), BS_OR (BS_AND (*fa, *fd), BS_XOR (*fb, BS_NOT (*fc))));
tmp3 = BS_OR (BS_AND (*fa, *fd), BS_XOR (*fa, BS_XOR (*fb, BS_AND (*fc, *fd))));
*sa = BS_XOR (tmp0, BS_AND (*fe, tmp1));
*sb = BS_XOR (tmp2, BS_AND (*fe, tmp3));
}
static void DVBCSA_INLINE
dvbcsa_bs_stream_sbox2(dvbcsa_bs_word_t fa, dvbcsa_bs_word_t fb,
dvbcsa_bs_word_t fc, dvbcsa_bs_word_t fd,
dvbcsa_bs_word_t fe,
dvbcsa_bs_word_t *sa, dvbcsa_bs_word_t *sb)
{
dvbcsa_bs_word_t tmp0, tmp1, tmp2, tmp3;
tmp0 = BS_XOR (fa, BS_XOR (BS_AND (fb, BS_OR (fc, fd)), BS_XOR (fc, BS_NOT (fd))));
tmp1 = BS_OR (BS_AND (fa, BS_XOR (fb, fd)), BS_AND (BS_OR (fa, fb), fc));
tmp2 = BS_XOR (BS_AND (fb, fd), BS_OR (BS_AND (fa, fd), BS_XOR (fb, BS_NOT (fc))));
tmp3 = BS_OR (BS_AND (fa, fd), BS_XOR (fa, BS_XOR (fb, BS_AND (fc, fd))));
*sa = BS_XOR (tmp0, BS_AND (fe, tmp1));
*sb = BS_XOR (tmp2, BS_AND (fe, tmp3));
}
/**
set up data used for stream. Depends on scrambled data only, so can be global
@param data_return[out] bit sliced output data IB1
@param outbits[in] number of bits to calculate
@param BS_key[in] bit sliced key array 64 elements
@param bs_data_sb0[in] bit sliced 1st data block used for initialization
*/
void aycw_stream_decrypt(dvbcsa_bs_word_t * data_return, unsigned int outbits, dvbcsa_bs_word_t * BS_key, dvbcsa_bs_word_t *bs_data_sb0)
{
unsigned int i;
dvbcsa_bs_word_t BS_Streambit0, BS_Streambit1;
aycw_tstRegister stRegister/*=
{
{0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0, 0,0,0,0/*,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0* /},
{0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0, 0,0,0,0/*,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0* /},
}*/;
aycw_tstPQXYZ stPQXYZ =
{
0,
0,
0, 0, 0, 0,
0, 0, 0, 0,
0, 0, 0, 0
};
aycw_tstCDEF stCDEF =
{
0,
0, 0, 0, 0,
0, 0, 0, 0,
0, 0, 0, 0
};
// aycw__vInitVariables( &stPQXYZ, &stCDEF , & BS_Streambit0, & BS_Streambit1);
/* aycw_vInitVariables();*/
/* BS_vTransformMultiKey((char *) cw);*/
/* init A and B - bs */
aycw__vInitShiftRegister(BS_key, &stRegister);
/* init A and B - bs */
for (i = 0; i < 8; i++)
{
aycw__vInitRound(0, i, &stPQXYZ, &stCDEF, &stRegister, bs_data_sb0);
aycw__vInitRound(1, i, &stPQXYZ, &stCDEF, &stRegister, bs_data_sb0);
aycw__vInitRound(2, i, &stPQXYZ, &stCDEF, &stRegister, bs_data_sb0);
aycw__vInitRound(3, i, &stPQXYZ, &stCDEF, &stRegister, bs_data_sb0);
}
for (i = 0; i < outbits; i += 8)
{
aycw__vRound(&stPQXYZ, &stCDEF, &stRegister, &BS_Streambit0, &BS_Streambit1);
data_return[i + 6] = BS_XOR(bs_data_sb0[i + 64 + 6], BS_Streambit0);
data_return[i + 7] = BS_XOR(bs_data_sb0[i + 64 + 7], BS_Streambit1);
aycw__vRound(&stPQXYZ, &stCDEF, &stRegister, &BS_Streambit0, &BS_Streambit1);
data_return[i + 4] = BS_XOR(bs_data_sb0[i + 64 + 4], BS_Streambit0);
data_return[i + 5] = BS_XOR(bs_data_sb0[i + 64 + 5], BS_Streambit1);
aycw__vRound( &stPQXYZ, &stCDEF, &stRegister, &BS_Streambit0, &BS_Streambit1);
data_return[i + 2] = BS_XOR(bs_data_sb0[i + 64 + 2], BS_Streambit0);
data_return[i + 3] = BS_XOR(bs_data_sb0[i + 64 + 3], BS_Streambit1);
aycw__vRound( &stPQXYZ, &stCDEF, &stRegister, &BS_Streambit0, &BS_Streambit1);
data_return[i + 0] = BS_XOR(bs_data_sb0[i + 64 + 0], BS_Streambit0);
data_return[i + 1] = BS_XOR(bs_data_sb0[i + 64 + 1], BS_Streambit1);
}
}
void
dvbcsa_bs_stream_cipher_batch(const struct dvbcsa_bs_key_s *key,
const struct dvbcsa_bs_batch_s *pcks,
unsigned int maxlen)
{
dvbcsa_bs_word_t A[32 + 10][4]; // 32 because we will move back (virtual shift register)
dvbcsa_bs_word_t B[32 + 10][4]; // 32 because we will move back (virtual shift register)
dvbcsa_bs_word_t X[4];
dvbcsa_bs_word_t Y[4];
dvbcsa_bs_word_t Z[4];
dvbcsa_bs_word_t D[4];
dvbcsa_bs_word_t E[4];
dvbcsa_bs_word_t F[4];
dvbcsa_bs_word_t p;
dvbcsa_bs_word_t q;
dvbcsa_bs_word_t r;
dvbcsa_bs_word_t in1[4];
dvbcsa_bs_word_t in2[4];
dvbcsa_bs_word_t extra_B[4];
dvbcsa_bs_word_t s1a, s1b, s2a, s2b, s3a, s3b, s4a, s4b, s5a, s5b, s6a, s6b, s7a, s7b;
dvbcsa_bs_word_t next_E[4];
dvbcsa_bs_word_t tmp0, tmp1, tmp3, tmp4;
dvbcsa_bs_word_t sb[64];
int h, i, j, k, b;
dvbcsa_bs_stream_transpose_in(pcks, sb);
for (b = 0; b < 4; b++)
{
for (i = 0; i < 8; i++)
{
A[i][b] = key->stream[b + i * 4];
B[i][b] = key->stream[b + i * 4 + 32];
}
// all other regs = 0
A[8][b] = BS_VAL8(00);
A[9][b] = BS_VAL8(00);
B[8][b] = BS_VAL8(00);
B[9][b] = BS_VAL8(00);
X[b] = BS_VAL8(00);
Y[b] = BS_VAL8(00);
Z[b] = BS_VAL8(00);
D[b] = BS_VAL8(00);
E[b] = BS_VAL8(00);
F[b] = BS_VAL8(00);
}
p = BS_VAL8(00);
q = BS_VAL8(00);
r = BS_VAL8(00);
/* Stream INIT */
for (i = 0; i < 8; i++)
{
for (b = 0; b < 4; b++)
{
in1[b] = sb[8 * i + 4 + b];
in2[b] = sb[8 * i + b];
}
for (j = 0; j < 4; j++)
{
dvbcsa_bs_stream_sbox1(A[0][2], A[5][1], A[6][3], A[8][0], A[3][0], &s1a, &s1b);
dvbcsa_bs_stream_sbox2(A[2][2], A[5][3], A[6][0], A[8][1], A[1][1], &s2a, &s2b);
dvbcsa_bs_stream_sbox3(A[1][0], A[4][1], A[4][3], A[5][2], A[0][3], &s3a, &s3b);
dvbcsa_bs_stream_sbox4(A[0][1], A[1][3], A[3][2], A[7][0], A[2][3], &s4a, &s4b);
dvbcsa_bs_stream_sbox5(A[3][3], A[5][0], A[7][1], A[8][2], A[4][2], &s5a, &s5b);
dvbcsa_bs_stream_sbox6(A[3][1], A[4][0], A[6][2], A[8][3], A[2][1], &s6a, &s6b);
dvbcsa_bs_stream_sbox7(A[2][0], A[6][1], A[7][2], A[7][3], A[1][2], &s7a, &s7b);
extra_B[3] = BS_XOR (BS_XOR (BS_XOR (B[2][0], B[5][1]), B[6][2]), B[8][3]);
extra_B[2] = BS_XOR (BS_XOR (BS_XOR (B[5][0], B[7][1]), B[2][3]), B[3][2]);
extra_B[1] = BS_XOR (BS_XOR (BS_XOR (B[4][3], B[7][2]), B[3][0]), B[4][1]);
extra_B[0] = BS_XOR (BS_XOR (BS_XOR (B[8][2], B[5][3]), B[2][1]), B[7][0]);
for (b = 0; b < 4; b++)
{
dvbcsa_bs_word_t A_next;
A_next = BS_XOR (A[9][b], X[b]);
A_next = BS_XOR (BS_XOR (A_next, D[b]), ((j % 2) ? in2[b] : in1[b]));
for (k = 9; k > 0; k--)
A[k][b] = A[k - 1][b];
A[0][b] = A_next;
}
dvbcsa_bs_word_t B_next[4];
for (b = 0; b < 4; b++)
{
B_next[b] = BS_XOR (BS_XOR (B[6][b], B[9][b]), Y[b]);
B_next[b] = BS_XOR (B_next[b], ((j % 2) ? in1[b] : in2[b]));
}
tmp3 = B_next[3];
B_next[3] = BS_XOR (B_next[3], BS_AND (BS_XOR (B_next[3], B_next[2]), p));
B_next[2] = BS_XOR (B_next[2], BS_AND (BS_XOR (B_next[2], B_next[1]), p));
B_next[1] = BS_XOR (B_next[1], BS_AND (BS_XOR (B_next[1], B_next[0]), p));
B_next[0] = BS_XOR (B_next[0], BS_AND (BS_XOR (B_next[0], tmp3), p));
for (b = 0; b < 4; b++)
{
for (k = 9; k > 0; k--)
B[k][b] = B[k - 1][b];
B[0][b] = B_next[b];
}
for (b = 0; b < 4; b++)
D[b] = BS_XOR (BS_XOR (E[b], Z[b]), extra_B[b]);
for (b = 0; b < 4; b++)
next_E[b] = F[b];
tmp0 = BS_XOR (Z[0], E[0]);
tmp1 = BS_AND (Z[0], E[0]);
F[0] = BS_XOR (E[0], BS_AND (q, BS_XOR (Z[0], r)));
tmp3 = BS_AND (tmp0, r);
tmp4 = BS_OR (tmp1, tmp3);
tmp0 = BS_XOR (Z[1], E[1]);
tmp1 = BS_AND (Z[1], E[1]);
F[1] = BS_XOR (E[1], BS_AND (q, BS_XOR (Z[1], tmp4)));
tmp3 = BS_AND (tmp0, tmp4);
tmp4 = BS_OR (tmp1, tmp3);
tmp0 = BS_XOR (Z[2], E[2]);
tmp1 = BS_AND (Z[2], E[2]);
F[2] = BS_XOR (E[2], BS_AND (q, BS_XOR (Z[2], tmp4)));
tmp3 = BS_AND (tmp0, tmp4);
tmp4 = BS_OR (tmp1, tmp3);
tmp0 = BS_XOR (Z[3], E[3]);
tmp1 = BS_AND (Z[3], E[3]);
F[3] = BS_XOR (E[3], BS_AND (q, BS_XOR (Z[3], tmp4)));
tmp3 = BS_AND (tmp0, tmp4);
r = BS_XOR (r, BS_AND (q, BS_XOR (BS_OR (tmp1, tmp3), r))); // ultimate carry
for (b = 0; b < 4; b++)
E[b] = next_E[b];
X[0] = s1a;
X[1] = s2a;
X[2] = s3b;
X[3] = s4b;
Y[0] = s3a;
Y[1] = s4a;
Y[2] = s5b;
Y[3] = s6b;
Z[0] = s5a;
Z[1] = s6a;
Z[2] = s1b;
Z[3] = s2b;
p = s7a;
q = s7b;
}
}
/* Stream GEN */
for (h = 8; h < maxlen; h++)
{
dvbcsa_bs_word_t cb[8];
for (j = 0; j < 4; j++)
{
dvbcsa_bs_stream_sbox1(A[0][2], A[5][1], A[6][3], A[8][0], A[3][0], &s1a, &s1b);
dvbcsa_bs_stream_sbox2(A[2][2], A[5][3], A[6][0], A[8][1], A[1][1], &s2a, &s2b);
dvbcsa_bs_stream_sbox3(A[1][0], A[4][1], A[4][3], A[5][2], A[0][3], &s3a, &s3b);
dvbcsa_bs_stream_sbox4(A[0][1], A[1][3], A[3][2], A[7][0], A[2][3], &s4a, &s4b);
dvbcsa_bs_stream_sbox5(A[3][3], A[5][0], A[7][1], A[8][2], A[4][2], &s5a, &s5b);
dvbcsa_bs_stream_sbox6(A[3][1], A[4][0], A[6][2], A[8][3], A[2][1], &s6a, &s6b);
dvbcsa_bs_stream_sbox7(A[2][0], A[6][1], A[7][2], A[7][3], A[1][2], &s7a, &s7b);
// use 4x4 xor to produce extra nibble for T3
extra_B[3] = BS_XOR (BS_XOR (BS_XOR (B[2][0], B[5][1]), B[6][2]), B[8][3]);
extra_B[2] = BS_XOR (BS_XOR (BS_XOR (B[5][0], B[7][1]), B[2][3]), B[3][2]);
extra_B[1] = BS_XOR (BS_XOR (BS_XOR (B[4][3], B[7][2]), B[3][0]), B[4][1]);
extra_B[0] = BS_XOR (BS_XOR (BS_XOR (B[8][2], B[5][3]), B[2][1]), B[7][0]);
// T1 = xor all inputs
// in1, in2, D are only used in T1 during initialisation, not generation
for (b = 0; b < 4; b++)
{
dvbcsa_bs_word_t A_next;
A_next = BS_XOR (A[9][b], X[b]);
for (k = 9; k > 0; k--)
A[k][b] = A[k - 1][b];
A[0][b] = A_next;
}
dvbcsa_bs_word_t B_next[4];
// T2 = xor all inputs
// in1, in2 are only used in T1 during initialisation, not generation
// if p=0, use this, if p=1, rotate the result left
for (b = 0; b < 4; b++)
B_next[b] = BS_XOR (BS_XOR (B[6][b], B[9][b]), Y[b]);
// if p=1, rotate left (yes, this is what we're doing)
tmp3 = B_next[3];
B_next[3] = BS_XOR (B_next[3], BS_AND (BS_XOR (B_next[3], B_next[2]), p));
B_next[2] = BS_XOR (B_next[2], BS_AND (BS_XOR (B_next[2], B_next[1]), p));
B_next[1] = BS_XOR (B_next[1], BS_AND (BS_XOR (B_next[1], B_next[0]), p));
B_next[0] = BS_XOR (B_next[0], BS_AND (BS_XOR (B_next[0], tmp3), p));
for (b = 0; b < 4; b++)
{
for (k = 9; k > 0; k--)
B[k][b] = B[k - 1][b];
B[0][b] = B_next[b];
}
// T3 = xor all inputs
for (b = 0; b < 4; b++)
D[b] = BS_XOR (BS_XOR (E[b], Z[b]), extra_B[b]);
// T4 = sum, carry of Z + E + r
for (b = 0; b < 4; b++)
next_E[b] = F[b];
tmp0 = BS_XOR (Z[0], E[0]);
tmp1 = BS_AND (Z[0], E[0]);
F[0] = BS_XOR (E[0], BS_AND (q, BS_XOR (Z[0], r)));
tmp3 = BS_AND (tmp0, r);
tmp4 = BS_OR (tmp1, tmp3);
tmp0 = BS_XOR (Z[1], E[1]);
tmp1 = BS_AND (Z[1], E[1]);
F[1] = BS_XOR (E[1], BS_AND (q, BS_XOR (Z[1], tmp4)));
tmp3 = BS_AND (tmp0, tmp4);
tmp4 = BS_OR (tmp1, tmp3);
tmp0 = BS_XOR (Z[2], E[2]);
tmp1 = BS_AND (Z[2], E[2]);
F[2] = BS_XOR (E[2], BS_AND (q, BS_XOR (Z[2], tmp4)));
tmp3 = BS_AND (tmp0, tmp4);
tmp4 = BS_OR (tmp1, tmp3);
tmp0 = BS_XOR (Z[3], E[3]);
tmp1 = BS_AND (Z[3], E[3]);
F[3] = BS_XOR (E[3], BS_AND (q, BS_XOR (Z[3], tmp4)));
tmp3 = BS_AND (tmp0, tmp4);
r = BS_XOR (r, BS_AND (q, BS_XOR (BS_OR (tmp1, tmp3), r))); // ultimate carry
for (b = 0; b < 4; b++)
E[b] = next_E[b];
X[0] = s1a;
X[1] = s2a;
X[2] = s3b;
X[3] = s4b;
Y[0] = s3a;
Y[1] = s4a;
Y[2] = s5b;
Y[3] = s6b;
Z[0] = s5a;
Z[1] = s6a;
Z[2] = s1b;
Z[3] = s2b;
p = s7a;
q = s7b;
// require 4 loops per output byte
// 2 output bits are a function of the 4 bits of D
// xor 2 by 2
cb[7 - 2 * j] = BS_XOR (D[2], D[3]);
cb[6 - 2 * j] = BS_XOR (D[0], D[1]);
} // EXTERNAL LOOP
////////////////////////////////////////////////////////////////////////////////
dvbcsa_bs_stream_transpose_out(pcks, h, cb);
}
}
/****** execute round **************/
AYCW_INLINE void aycw__vRound(aycw_tstPQXYZ *stPQXYZ,
aycw_tstCDEF *stCDEF ,
aycw_tstRegister *stRegister,
dvbcsa_bs_word_t *BS_Streambit0,
dvbcsa_bs_word_t *BS_Streambit1)
{
uint8 k;
dvbcsa_bs_word_t tmp0,tmp4,tmp1,tmp3;
uint8 b;
dvbcsa_bs_word_t BS_TMP_B[5];
dvbcsa_bs_word_t BS_TMP_B_Result[4];
dvbcsa_bs_word_t BS_Bout[4];
dvbcsa_bs_word_t BS_Enew[4];
aycw__ShiftRegisterLeft( stRegister->A_BS, 4, 80);
for (k = 0; k < 4; k++)
{
stRegister->A_BS[k] = stRegister->A_BS[k+40];
BS_XOREQ(stRegister->A_BS[k], stPQXYZ->BS_X[k]);
}
for (k = 0; k < 4; k++)
{
//A: 0x0000000bb6510468
//B: 0x00000006baf6a610
// BS_ConverToDoubleArray(&tmpDoubleArray,B_BS,40);
BS_TMP_B[k] = BS_XOR(stRegister->B_BS[k+6*4], stRegister->B_BS[k+9*4]);
// BS_ConverToByteArray(&tmpByteArray,BS_TMP_B,4);
BS_TMP_B[k] = BS_XOR(BS_TMP_B[k], stPQXYZ->BS_Y[k]);
// BS_ConverToIntArray(tmpIntArray,B_BS,40);
#if Init
BS_TMP_B[k] = BS_TMP_B[k] ^ A_tBITVALUE(&u32IV, 0, k);
BS_ConverToIntArray(tmpIntArray,B_BS,40);
#endif
BS_TMP_B_Result[k] = BS_AND(BS_TMP_B[k], BS_NOT(stPQXYZ->BS_P)); //Die nicht zu rotierenden Daten zunächst zwischenspeichern.
// BS_ConverToIntArray(tmpIntArray,B_BS,40);
//B: 0x00000006baf6a617
}
for (k = 4; k > 0; k--)
{
BS_TMP_B[k] = BS_TMP_B[k-1];
}
BS_TMP_B[0] = BS_TMP_B[4];
aycw__ShiftRegisterLeft( stRegister->B_BS, 4, 80);
for ( k = 0; k < 4; k++)
{
stRegister->B_BS[k] = BS_OR(BS_AND(BS_TMP_B[k], stPQXYZ->BS_P), BS_TMP_B_Result[k]);
}
// BS_ConverToIntArray(tmpIntArray,B_BS,40);
// BS_ConverToDoubleArray(&tmpDoubleArray,B_BS,40);
/********** execute combiner calc **********/
/* calc of bout*/
/*
Bout 3 := b2,0 b5,1 b6,2 b8,3
Bout 2 := b5,0 b7,1 b2,3 b3,2
Bout 1 := b4,3 b7,2 b3,0 b4,1
Bout 0 := b8,2 b5,3 b2,1 b7,0
*/
/*12 25 30 37*/
/* 10000000 01000010000000000001000000000000 */
/*15 18 24 33*/
/* 00000010 00000001000001001000000000000000 */
/*16 21 23 34*/
/* 00000100 00000000101000010000000000000000 */
/*13 27 32 38*/
/* 01000001 00001000000000000010000000000000 */
/*8 21 26 35*/
/*extra_B[3] = BS_XOR (BS_XOR (BS_XOR (B[2][0], B[5][1]), B[6][2]), B[8][3]);
extra_B[2] = BS_XOR (BS_XOR (BS_XOR (B[5][0], B[7][1]), B[2][3]), B[3][2]);
extra_B[1] = BS_XOR (BS_XOR (BS_XOR (B[4][3], B[7][2]), B[3][0]), B[4][1]);
extra_B[0] = BS_XOR (BS_XOR (BS_XOR (B[8][2], B[5][3]), B[2][1]), B[7][0]);*/
/*
#define shiftB 4
BS_Bout[3] = stRegister->B_BS[2*4 + 0 + shiftB ] ^ stRegister->B_BS[5*4 + 1 + shiftB ] ^ stRegister->B_BS[6*4 + 2 + shiftB ] ^ stRegister->B_BS[8*4 + 3 + shiftB ] ;
BS_Bout[2] = stRegister->B_BS[5*4 + 0 + shiftB ] ^ stRegister->B_BS[7*4 + 1 + shiftB ] ^ stRegister->B_BS[2*4 + 3 + shiftB ] ^ stRegister->B_BS[3*4 + 2 + shiftB ] ;
BS_Bout[1] = stRegister->B_BS[4*4 + 3 + shiftB ] ^ stRegister->B_BS[7*4 + 2 + shiftB ] ^ stRegister->B_BS[3*4 + 0 + shiftB ] ^ stRegister->B_BS[4*4 + 1 + shiftB ] ;
BS_Bout[0] = stRegister->B_BS[8*4 + 2 + shiftB ] ^ stRegister->B_BS[5*4 + 3 + shiftB ] ^ stRegister->B_BS[2*4 + 1 + shiftB ] ^ stRegister->B_BS[7*4 + 0 + shiftB ] ;
*/
BS_Bout[3] = BS_XOR(BS_XOR(stRegister->B_BS[12], stRegister->B_BS[25]), BS_XOR(stRegister->B_BS[30], stRegister->B_BS[39]));
BS_Bout[2] = BS_XOR(BS_XOR(stRegister->B_BS[24], stRegister->B_BS[33]), BS_XOR(stRegister->B_BS[15], stRegister->B_BS[18]));
BS_Bout[1] = BS_XOR(BS_XOR(stRegister->B_BS[23], stRegister->B_BS[34]), BS_XOR(stRegister->B_BS[16], stRegister->B_BS[21]));
BS_Bout[0] = BS_XOR(BS_XOR(stRegister->B_BS[38], stRegister->B_BS[27]), BS_XOR(stRegister->B_BS[13], stRegister->B_BS[32]));
/*
Bout[3] = B_BS[1*4 + 0 ] ^ B_BS[4*4 + 1 ] ^ B_BS[5*4 + 2 ] ^ B_BS[7*4 + 3 ] ;
Bout[2] = B_BS[4*4 + 0 ] ^ B_BS[6*4 + 1 ] ^ B_BS[1*4 + 3 ] ^ B_BS[2*4 + 2 ] ;
Bout[1] = B_BS[3*4 + 3 ] ^ B_BS[6*4 + 2 ] ^ B_BS[2*4 + 0 ] ^ B_BS[3*4 + 1 ] ;
Bout[0] = B_BS[7*4 + 2 ] ^ B_BS[4*4 + 3 ] ^ B_BS[1*4 + 1 ] ^ B_BS[6*4 + 0 ] ;
*/
/* calc of D */
for (k = 0; k < 4; k++)
{
/* use old E????*/
stCDEF->BS_D[k] = BS_XOR(BS_XOR(BS_Bout[k], stCDEF->BS_E [k]), stPQXYZ->BS_Z[k]);
// BS_ConverToByteArray(&tmpByteArray,BS_E,4);
// BS_ConverToByteArray(&tmpByteArray,BS_Z,4);
}
// BS_ConverToByteArray(&tmpByteArray,BS_D,4);
for (b = 0; b < 4; b++)
BS_Enew[b] = stCDEF->BS_F[b];
tmp0 = BS_XOR (stPQXYZ->BS_Z[0], stCDEF->BS_E[0]);
tmp1 = BS_AND (stPQXYZ->BS_Z[0], stCDEF->BS_E[0]);
stCDEF->BS_F[0] = BS_XOR (stCDEF->BS_E[0], BS_AND (stPQXYZ->BS_Q, BS_XOR (stPQXYZ->BS_Z[0], stCDEF->BS_C)));
tmp3 = BS_AND (tmp0, stCDEF->BS_C);
tmp4 = BS_OR (tmp1, tmp3);
tmp0 = BS_XOR (stPQXYZ->BS_Z[1], stCDEF->BS_E[1]);
tmp1 = BS_AND (stPQXYZ->BS_Z[1], stCDEF->BS_E[1]);
stCDEF->BS_F[1] = BS_XOR (stCDEF->BS_E[1], BS_AND (stPQXYZ->BS_Q, BS_XOR (stPQXYZ->BS_Z[1], tmp4)));
tmp3 = BS_AND (tmp0, tmp4);
tmp4 = BS_OR (tmp1, tmp3);
tmp0 = BS_XOR (stPQXYZ->BS_Z[2], stCDEF->BS_E[2]);
tmp1 = BS_AND (stPQXYZ->BS_Z[2], stCDEF->BS_E[2]);
stCDEF->BS_F[2] = BS_XOR (stCDEF->BS_E[2], BS_AND (stPQXYZ->BS_Q, BS_XOR (stPQXYZ->BS_Z[2], tmp4)));
tmp3 = BS_AND (tmp0, tmp4);
tmp4 = BS_OR (tmp1, tmp3);
tmp0 = BS_XOR (stPQXYZ->BS_Z[3], stCDEF->BS_E[3]);
tmp1 = BS_AND (stPQXYZ->BS_Z[3], stCDEF->BS_E[3]);
stCDEF->BS_F[3] = BS_XOR (stCDEF->BS_E[3], BS_AND (stPQXYZ->BS_Q, BS_XOR (stPQXYZ->BS_Z[3], tmp4)));
tmp3 = BS_AND (tmp0, tmp4);
stCDEF->BS_C = BS_XOR (stCDEF->BS_C, BS_AND (stPQXYZ->BS_Q, BS_XOR (BS_OR (tmp1, tmp3), stCDEF->BS_C))); // ultimate carry
for (b = 0; b < 4; b++)
stCDEF->BS_E[b] = BS_Enew[b];
aycw__vCaculatePQXYZ(stRegister->A_BS, stPQXYZ);
/* calc stream bit */
*BS_Streambit0 = BS_XOR(stCDEF->BS_D[0], stCDEF->BS_D[1]);
*BS_Streambit1 = BS_XOR(stCDEF->BS_D[2], stCDEF->BS_D[3]);
}
/****** execute init round **************/
AYCW_INLINE void aycw__vInitRound(uint8 j, uint8 u8Byte,
aycw_tstPQXYZ * stPQXYZ ,
aycw_tstCDEF *stCDEF,
aycw_tstRegister *stRegister,
dvbcsa_bs_word_t *bs_data_sb0)
{
uint8 k;
dvbcsa_bs_word_t /*tmp,*/tmp0,tmp4,tmp1,tmp3;
uint8 b;
/*uint8 u32IV;*/
dvbcsa_bs_word_t BS_TMP_B[5];
dvbcsa_bs_word_t BS_TMP_B_Result[4];
dvbcsa_bs_word_t BS_Bout[4];
dvbcsa_bs_word_t BS_Enew[4];
/*
if (j & 0x1) // all odd
{
u32IV = (iv >> 4 | (iv << 4));
}
else
{
u32IV = iv;
}
*/
aycw__ShiftRegisterLeft( stRegister->A_BS, 4, 44);
for (k = 0; k < 4; k++)
{
stRegister->A_BS[k] = stRegister->A_BS[k+40];
BS_XOREQ(stRegister->A_BS[k],stPQXYZ->BS_X[k]);
BS_XOREQ(stRegister->A_BS[k],stCDEF->BS_D[k]); /* data, byte, bit */
BS_XOREQ(stRegister->A_BS[k],( (j & 0x1)?bs_data_sb0[k+u8Byte*8]:bs_data_sb0[k+4+u8Byte*8]));
}
for (k = 0; k < 4; k++)
{
BS_TMP_B[k] = BS_XOR(stRegister->B_BS[k+6*4], stRegister->B_BS[k+9*4]);
BS_TMP_B[k] = BS_XOR(BS_TMP_B[k], stPQXYZ->BS_Y[k]);
// OPTIMIZEME: optimizeable by using the upper A_tBITVALUE calculation
/*tmp = (j & 0x1)?bs_data_sb0[k+4+u8Byte*8]:bs_data_sb0[k+u8Byte*8];
tmp0 = aycw__BitExpandOfByteToBsWord(&u32IV, 0, k);
if (tmp0 != tmp)
return 0;*/
BS_TMP_B[k] = BS_XOR(BS_TMP_B[k], ( (j & 0x1)?bs_data_sb0[k+4+u8Byte*8]:bs_data_sb0[k+u8Byte*8]));
BS_TMP_B_Result[k] = BS_AND(BS_TMP_B[k], BS_NOT(stPQXYZ->BS_P)); //Die nicht zu rotierenden Daten zunächst zwischenspeichern.
}
/* yet rotate B */
/* OPTIMIZEME: write into B directly?? */
for (k = 4; k > 0; k--)
{
BS_TMP_B[k] = BS_TMP_B[k-1];
}
BS_TMP_B[0] = BS_TMP_B[4];
/* B must not be moved by 4 before handling is finished */
aycw__ShiftRegisterLeft(stRegister->B_BS, 4, 44);
/* now write the result, both rotated and unrotated */
for ( k = 0; k < 4; k++)
{ // rotated //not rotated
stRegister->B_BS[k] = BS_OR(BS_AND(BS_TMP_B[k], stPQXYZ->BS_P), BS_TMP_B_Result[k]);
}
/********** Combiner calculation **********/
/* calc bout
Bout 3 := b2,0 b5,1 b6,2 b8,3
Bout 2 := b5,0 b7,1 b2,3 b3,2
Bout 1 := b4,3 b7,2 b3,0 b4,1
Bout 0 := b8,2 b5,3 b2,1 b7,0
*/
/*
#define shiftB 4
BS_Bout[3] = stRegister->B_BS[2*4 + 0 + shiftB ] ^ stRegister->B_BS[5*4 + 1 + shiftB ] ^ stRegister->B_BS[6*4 + 2 + shiftB ] ^ stRegister->B_BS[8*4 + 3 + shiftB ] ;
BS_Bout[2] = stRegister->B_BS[5*4 + 0 + shiftB ] ^ stRegister->B_BS[7*4 + 1 + shiftB ] ^ stRegister->B_BS[2*4 + 3 + shiftB ] ^ stRegister->B_BS[3*4 + 2 + shiftB ] ;
BS_Bout[1] = stRegister->B_BS[4*4 + 3 + shiftB ] ^ stRegister->B_BS[7*4 + 2 + shiftB ] ^ stRegister->B_BS[3*4 + 0 + shiftB ] ^ stRegister->B_BS[4*4 + 1 + shiftB ] ;
BS_Bout[0] = stRegister->B_BS[8*4 + 2 + shiftB ] ^ stRegister->B_BS[5*4 + 3 + shiftB ] ^ stRegister->B_BS[2*4 + 1 + shiftB ] ^ stRegister->B_BS[7*4 + 0 + shiftB ] ;
*/
BS_Bout[3] = BS_XOR(BS_XOR(stRegister->B_BS[12], stRegister->B_BS[25]), BS_XOR(stRegister->B_BS[30], stRegister->B_BS[39]));
BS_Bout[2] = BS_XOR(BS_XOR(stRegister->B_BS[24], stRegister->B_BS[33]), BS_XOR(stRegister->B_BS[15], stRegister->B_BS[18]));
BS_Bout[1] = BS_XOR(BS_XOR(stRegister->B_BS[23], stRegister->B_BS[34]), BS_XOR(stRegister->B_BS[16], stRegister->B_BS[21]));
BS_Bout[0] = BS_XOR(BS_XOR(stRegister->B_BS[38], stRegister->B_BS[27]), BS_XOR(stRegister->B_BS[13], stRegister->B_BS[32]));
/* calc D */
for (k = 0; k < 4; k++)
{
stCDEF->BS_D[k] = BS_XOR(BS_XOR(BS_Bout[k], stCDEF->BS_E[k]), stPQXYZ->BS_Z[k]);
}
for (b = 0; b < 4; b++)
BS_Enew[b] = stCDEF->BS_F[b];
tmp0 = BS_XOR (stPQXYZ->BS_Z[0], stCDEF->BS_E[0]);
tmp1 = BS_AND (stPQXYZ->BS_Z[0], stCDEF->BS_E[0]);
stCDEF->BS_F[0] = BS_XOR (stCDEF->BS_E[0], BS_AND (stPQXYZ->BS_Q, BS_XOR (stPQXYZ->BS_Z[0], stCDEF->BS_C)));
tmp3 = BS_AND (tmp0, stCDEF->BS_C);
tmp4 = BS_OR (tmp1, tmp3);
tmp0 = BS_XOR (stPQXYZ->BS_Z[1], stCDEF->BS_E[1]);
tmp1 = BS_AND (stPQXYZ->BS_Z[1], stCDEF->BS_E[1]);
stCDEF->BS_F[1] = BS_XOR (stCDEF->BS_E[1], BS_AND (stPQXYZ->BS_Q, BS_XOR (stPQXYZ->BS_Z[1], tmp4)));
tmp3 = BS_AND (tmp0, tmp4);
tmp4 = BS_OR (tmp1, tmp3);
tmp0 = BS_XOR (stPQXYZ->BS_Z[2], stCDEF->BS_E[2]);
tmp1 = BS_AND (stPQXYZ->BS_Z[2], stCDEF->BS_E[2]);
stCDEF->BS_F[2] = BS_XOR (stCDEF->BS_E[2], BS_AND (stPQXYZ->BS_Q, BS_XOR (stPQXYZ->BS_Z[2], tmp4)));
tmp3 = BS_AND (tmp0, tmp4);
tmp4 = BS_OR (tmp1, tmp3);
tmp0 = BS_XOR (stPQXYZ->BS_Z[3], stCDEF->BS_E[3]);
tmp1 = BS_AND (stPQXYZ->BS_Z[3], stCDEF->BS_E[3]);
stCDEF->BS_F[3] = BS_XOR (stCDEF->BS_E[3], BS_AND (stPQXYZ->BS_Q, BS_XOR (stPQXYZ->BS_Z[3], tmp4)));
tmp3 = BS_AND (tmp0, tmp4);
stCDEF->BS_C = BS_XOR (stCDEF->BS_C, BS_AND (stPQXYZ->BS_Q, BS_XOR (BS_OR (tmp1, tmp3), stCDEF->BS_C))); // ultimate carry
for (b = 0; b < 4; b++)
stCDEF->BS_E[b] = BS_Enew[b];
aycw__vCaculatePQXYZ(stRegister->A_BS, stPQXYZ);
}