void sha512_compress(psDigestContext_t * md, unsigned char *buf)
#endif
{
uint64 S[8], W[80], t0, t1;
int i;
/* copy state into S */
for (i = 0; i < 8; i++) {
S[i] = md->sha512.state[i];
}
/* copy the state into 1024-bits into W[0..15] */
for (i = 0; i < 16; i++) {
LOAD64H(W[i], buf + (8*i));
}
/* fill W[16..79] */
for (i = 16; i < 80; i++) {
W[i] = Gamma1(W[i - 2]) + W[i - 7] + Gamma0(W[i - 15]) + W[i - 16];
}
/* Compress */
#ifndef PS_SHA512_IMPROVE_PERF_INCREASE_CODESIZE
for (i = 0; i < 80; i++) {
t0 = S[7] + Sigma1(S[4]) + Ch(S[4], S[5], S[6]) + K[i] + W[i];
t1 = Sigma0(S[0]) + Maj(S[0], S[1], S[2]);
S[7] = S[6];
S[6] = S[5];
S[5] = S[4];
S[4] = S[3] + t0;
S[3] = S[2];
S[2] = S[1];
S[1] = S[0];
S[0] = t0 + t1;
}
#else
#define RND(a,b,c,d,e,f,g,h,i) \
t0 = h + Sigma1(e) + Ch(e, f, g) + K[i] + W[i]; \
t1 = Sigma0(a) + Maj(a, b, c); \
d += t0; \
h = t0 + t1;
for (i = 0; i < 80; i += 8) {
RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],i+0);
RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],i+1);
RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],i+2);
RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],i+3);
RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],i+4);
RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],i+5);
RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],i+6);
RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],i+7);
}
#endif /* PS_SHA512_IMPROVE_PERF_INCREASE_CODESIZE */
/* feedback */
for (i = 0; i < 8; i++) {
md->sha512.state[i] = md->sha512.state[i] + S[i];
}
}
static int whirlpool_compress(hash_state *md, unsigned char *buf)
#endif
{
ulong64 K[2][8], T[3][8];
int x, y;
/* load the block/state */
for (x = 0; x < 8; x++) {
K[0][x] = md->whirlpool.state[x];
LOAD64H(T[0][x], buf + (8 * x));
T[2][x] = T[0][x];
T[0][x] ^= K[0][x];
}
/* do rounds 1..10 */
for (x = 0; x < 10; x += 2) {
/* odd round */
/* apply main transform to K[0] into K[1] */
for (y = 0; y < 8; y++) {
K[1][y] = theta_pi_gamma(K[0], y);
}
/* xor the constant */
K[1][0] ^= cont[x];
/* apply main transform to T[0] into T[1] */
for (y = 0; y < 8; y++) {
T[1][y] = theta_pi_gamma(T[0], y) ^ K[1][y];
}
/* even round */
/* apply main transform to K[1] into K[0] */
for (y = 0; y < 8; y++) {
K[0][y] = theta_pi_gamma(K[1], y);
}
/* xor the constant */
K[0][0] ^= cont[x+1];
/* apply main transform to T[1] into T[0] */
for (y = 0; y < 8; y++) {
T[0][y] = theta_pi_gamma(T[1], y) ^ K[0][y];
}
}
/* store state */
for (x = 0; x < 8; x++) {
md->whirlpool.state[x] ^= T[0][x] ^ T[2][x];
}
return CRYPT_OK;
}
/* compress 1024-bits */
static int sha512_compress(sha512_context *md, unsigned char *buf)
{
u64 S[8], W[80], t0, t1;
int i;
/* copy state into S */
for (i = 0; i < 8; i++) {
S[i] = md->state[i];
}
/* copy the state into 1024-bits into W[0..15] */
for (i = 0; i < 16; i++) {
LOAD64H(W[i], buf + (8*i));
}
/* fill W[16..79] */
for (i = 16; i < 80; i++) {
W[i] = Gamma1(W[i - 2]) + W[i - 7] + Gamma0(W[i - 15]) + W[i - 16];
}
/* Compress */
#define RND(a,b,c,d,e,f,g,h,i) \
t0 = h + Sigma1(e) + Ch(e, f, g) + K[i] + W[i]; \
t1 = Sigma0(a) + Maj(a, b, c);\
d += t0; \
h = t0 + t1;
for (i = 0; i < 80; i += 8) {
RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],i+0);
RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],i+1);
RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],i+2);
RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],i+3);
RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],i+4);
RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],i+5);
RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],i+6);
RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],i+7);
}
#undef RND
/* feedback */
for (i = 0; i < 8; i++) {
md->state[i] = md->state[i] + S[i];
}
return 0;
}
static int sha512_compress(hash_state * md, unsigned char *buf)
#endif
{
ulong64 S[8], W[80], t0, t1;
int i;
/* copy state into S */
for (i = 0; i < 8; i++) {
S[i] = md->sha512.state[i];
}
/* copy the state into 1024-bits into W[0..15] */
for (i = 0; i < 16; i++) {
LOAD64H(W[i], buf + (8*i));
}
/* fill W[16..79] */
for (i = 16; i < 80; i++) {
W[i] = Gamma1(W[i - 2]) + W[i - 7] + Gamma0(W[i - 15]) + W[i - 16];
}
/* Compress */
#ifdef LTC_SMALL_CODE
for (i = 0; i < 80; i++) {
t0 = S[7] + Sigma1(S[4]) + Ch(S[4], S[5], S[6]) + K[i] + W[i];
t1 = Sigma0(S[0]) + Maj(S[0], S[1], S[2]);
S[7] = S[6];
S[6] = S[5];
S[5] = S[4];
S[4] = S[3] + t0;
S[3] = S[2];
S[2] = S[1];
S[1] = S[0];
S[0] = t0 + t1;
}
#else
#define RND(a,b,c,d,e,f,g,h,i) \
t0 = h + Sigma1(e) + Ch(e, f, g) + K[i] + W[i]; \
t1 = Sigma0(a) + Maj(a, b, c); \
d += t0; \
h = t0 + t1;
for (i = 0; i < 80; i += 8) {
RND(S[0],S[1],S[2],S[3],S[4],S[5],S[6],S[7],i+0);
RND(S[7],S[0],S[1],S[2],S[3],S[4],S[5],S[6],i+1);
RND(S[6],S[7],S[0],S[1],S[2],S[3],S[4],S[5],i+2);
RND(S[5],S[6],S[7],S[0],S[1],S[2],S[3],S[4],i+3);
RND(S[4],S[5],S[6],S[7],S[0],S[1],S[2],S[3],i+4);
RND(S[3],S[4],S[5],S[6],S[7],S[0],S[1],S[2],i+5);
RND(S[2],S[3],S[4],S[5],S[6],S[7],S[0],S[1],i+6);
RND(S[1],S[2],S[3],S[4],S[5],S[6],S[7],S[0],i+7);
}
#endif
/* feedback */
for (i = 0; i < 8; i++) {
md->sha512.state[i] = md->sha512.state[i] + S[i];
}
return CRYPT_OK;
}
/* Test store/load macros with offsets */
int store_test(void)
{
unsigned char buf[256];
int y;
ulong32 L, L1;
ulong64 LL, LL1;
#ifdef LTC_FAST
int x, z;
#endif
for (y = 0; y < 4; y++) {
L = 0x12345678UL;
L1 = 0;
STORE32L(L, buf + y);
LOAD32L(L1, buf + y);
if (L1 != L) {
fprintf(stderr, "\n32L failed at offset %d\n", y);
return 1;
}
STORE32H(L, buf + y);
LOAD32H(L1, buf + y);
if (L1 != L) {
fprintf(stderr, "\n32H failed at offset %d\n", y);
return 1;
}
}
for (y = 0; y < 8; y++) {
LL = CONST64 (0x01020304050607);
LL1 = 0;
STORE64L(LL, buf + y);
LOAD64L(LL1, buf + y);
if (LL1 != LL) {
fprintf(stderr, "\n64L failed at offset %d\n", y);
return 1;
}
STORE64H(LL, buf + y);
LOAD64H(LL1, buf + y);
if (LL1 != LL) {
fprintf(stderr, "\n64H failed at offset %d\n", y);
return 1;
}
}
/* test LTC_FAST */
#ifdef LTC_FAST
y = 16;
for (z = 0; z < y; z++) {
/* fill y bytes with random */
yarrow_read(buf+z, y, &yarrow_prng);
yarrow_read(buf+z+y, y, &yarrow_prng);
/* now XOR it byte for byte */
for (x = 0; x < y; x++) {
buf[2*y+z+x] = buf[z+x] ^ buf[z+y+x];
}
/* now XOR it word for word */
for (x = 0; x < y; x += sizeof(LTC_FAST_TYPE)) {
*((LTC_FAST_TYPE*)(&buf[3*y+z+x])) = *((LTC_FAST_TYPE*)(&buf[z+x])) ^ *((LTC_FAST_TYPE*)(&buf[z+y+x]));
}
if (memcmp(&buf[2*y+z], &buf[3*y+z], y)) {
fprintf(stderr, "\nLTC_FAST failed at offset %d\n", z);
return 1;
}
}
#endif
return 0;
}
