static void
hamsi_big(sph_hamsi_big_context *sc, const unsigned char *buf, size_t num)
{
DECL_STATE_BIG
#if !SPH_64
sph_u32 tmp;
#endif
#if SPH_64
sc->count += (sph_u64)num << 6;
#else
tmp = SPH_T32((sph_u32)num << 6);
sc->count_low = SPH_T32(sc->count_low + tmp);
sc->count_high += (sph_u32)((num >> 13) >> 13);
if (sc->count_low < tmp)
sc->count_high ++;
#endif
READ_STATE_BIG(sc);
while (num -- > 0) {
sph_u32 m0, m1, m2, m3, m4, m5, m6, m7;
sph_u32 m8, m9, mA, mB, mC, mD, mE, mF;
INPUT_BIG;
P_BIG;
T_BIG;
buf += 8;
}
WRITE_STATE_BIG(sc);
}
static void shavite_big_core(sph_shavite_big_context *sc, const void *data, size_t len)
{
unsigned char *buf;
size_t ptr;
buf = sc->buf;
ptr = sc->ptr;
while (len > 0) {
size_t clen;
clen = (sizeof sc->buf) - ptr;
if (clen > len)
clen = len;
memcpy(buf + ptr, data, clen);
data = (const unsigned char *)data + clen;
ptr += clen;
len -= clen;
if (ptr == sizeof sc->buf) {
if ((sc->count0 = SPH_T32(sc->count0 + 1024)) == 0) {
sc->count1 = SPH_T32(sc->count1 + 1);
if (sc->count1 == 0) {
sc->count2 = SPH_T32(sc->count2 + 1);
if (sc->count2 == 0) {
sc->count3 = SPH_T32(
sc->count3 + 1);
}
}
}
c512(sc, buf);
ptr = 0;
}
}
sc->ptr = ptr;
}
static void
hamsi_small(sph_hamsi_small_context *sc, const unsigned char *buf, size_t num)
{
DECL_STATE_SMALL
#if !SPH_64
sph_u32 tmp;
#endif
#if SPH_64
sc->count += (sph_u64)num << 5;
#else
tmp = SPH_T32((sph_u32)num << 5);
sc->count_low = SPH_T32(sc->count_low + tmp);
sc->count_high += (sph_u32)((num >> 13) >> 14);
if (sc->count_low < tmp)
sc->count_high ++;
#endif
READ_STATE_SMALL(sc);
while (num -- > 0) {
sph_u32 m0, m1, m2, m3, m4, m5, m6, m7;
INPUT_SMALL;
P_SMALL;
T_SMALL;
buf += 4;
}
WRITE_STATE_SMALL(sc);
}
static void
jh_close(sph_jh_context *sc, unsigned ub, unsigned n,
void *dst, size_t out_size_w32, const void *iv)
{
unsigned z;
unsigned char buf[128];
size_t numz, u;
#if SPH_64
sph_u64 l0, l1;
#else
sph_u32 l0, l1, l2, l3;
#endif
z = 0x80 >> n;
buf[0] = ((ub & -z) | z) & 0xFF;
if (sc->ptr == 0 && n == 0) {
numz = 47;
} else {
numz = 111 - sc->ptr;
}
memset(buf + 1, 0, numz);
#if SPH_64
l0 = SPH_T64(sc->block_count << 9) + (sc->ptr << 3) + n;
l1 = SPH_T64(sc->block_count >> 55);
sph_enc64be(buf + numz + 1, l1);
sph_enc64be(buf + numz + 9, l0);
#else
l0 = SPH_T32(sc->block_count_low << 9) + (sc->ptr << 3) + n;
l1 = SPH_T32(sc->block_count_low >> 23)
+ SPH_T32(sc->block_count_high << 9);
l2 = SPH_T32(sc->block_count_high >> 23);
l3 = 0;
sph_enc32be(buf + numz + 1, l3);
sph_enc32be(buf + numz + 5, l2);
sph_enc32be(buf + numz + 9, l1);
sph_enc32be(buf + numz + 13, l0);
#endif
jh_core(sc, buf, numz + 17);
#if SPH_JH_64
for (u = 0; u < 8; u ++)
enc64e(buf + (u << 3), sc->H.wide[u + 8]);
#else
for (u = 0; u < 16; u ++)
enc32e(buf + (u << 2), sc->H.narrow[u + 16]);
#endif
memcpy(dst, buf + ((16 - out_size_w32) << 2), out_size_w32 << 2);
jh_init(sc, iv);
}
static void
blake32_close(sph_blake_small_context *sc,
unsigned ub, unsigned n, void *dst, size_t out_size_w32)
{
union {
unsigned char buf[64];
sph_u32 dummy;
} u;
size_t ptr, k;
unsigned bit_len;
unsigned z;
sph_u32 th, tl;
unsigned char *out;
ptr = sc->ptr;
bit_len = ((unsigned)ptr << 3) + n;
z = 0x80 >> n;
u.buf[ptr] = ((ub & -z) | z) & 0xFF;
tl = sc->T0 + bit_len;
th = sc->T1;
if (ptr == 0 && n == 0) {
sc->T0 = SPH_C32(0xFFFFFE00);
sc->T1 = SPH_C32(0xFFFFFFFF);
} else if (sc->T0 == 0) {
sc->T0 = SPH_C32(0xFFFFFE00) + bit_len;
sc->T1 = SPH_T32(sc->T1 - 1);
} else {
sc->T0 -= 512 - bit_len;
}
if (bit_len <= 446) {
memset(u.buf + ptr + 1, 0, 55 - ptr);
if (out_size_w32 == 8)
u.buf[55] |= 1;
sph_enc32be_aligned(u.buf + 56, th);
sph_enc32be_aligned(u.buf + 60, tl);
blake32(sc, u.buf + ptr, 64 - ptr);
} else {
memset(u.buf + ptr + 1, 0, 63 - ptr);
blake32(sc, u.buf + ptr, 64 - ptr);
sc->T0 = SPH_C32(0xFFFFFE00);
sc->T1 = SPH_C32(0xFFFFFFFF);
memset(u.buf, 0, 56);
if (out_size_w32 == 8)
u.buf[55] = 1;
sph_enc32be_aligned(u.buf + 56, th);
sph_enc32be_aligned(u.buf + 60, tl);
blake32(sc, u.buf, 64);
}
out = dst;
for (k = 0; k < out_size_w32; k ++)
sph_enc32be(out + (k << 2), sc->H[k]);
}
static void
blake32(sph_blake_small_context *sc, const void *data, size_t len)
{
unsigned char *buf;
size_t ptr;
DECL_STATE32
buf = sc->buf;
ptr = sc->ptr;
if (len < (sizeof sc->buf) - ptr) {
memcpy(buf + ptr, data, len);
ptr += len;
sc->ptr = ptr;
return;
}
READ_STATE32(sc);
while (len > 0) {
size_t clen;
clen = (sizeof sc->buf) - ptr;
if (clen > len)
clen = len;
memcpy(buf + ptr, data, clen);
ptr += clen;
data = (const unsigned char *)data + clen;
len -= clen;
if (ptr == sizeof sc->buf) {
if ((T0 = SPH_T32(T0 + 512)) < 512)
T1 = SPH_T32(T1 + 1);
COMPRESS32;
ptr = 0;
}
}
WRITE_STATE32(sc);
sc->ptr = ptr;
}
static void
bmw32(sph_bmw_small_context *sc, const void *data, size_t len)
{
unsigned char *buf;
size_t ptr;
sph_u32 htmp[16];
sph_u32 *h1, *h2;
#if !SPH_64
sph_u32 tmp;
#endif
#if SPH_64
sc->bit_count += (sph_u64)len << 3;
#else
tmp = sc->bit_count_low;
sc->bit_count_low = SPH_T32(tmp + ((sph_u32)len << 3));
if (sc->bit_count_low < tmp)
sc->bit_count_high ++;
sc->bit_count_high += len >> 29;
#endif
buf = sc->buf;
ptr = sc->ptr;
h1 = sc->H;
h2 = htmp;
while (len > 0) {
size_t clen;
clen = (sizeof sc->buf) - ptr;
if (clen > len)
clen = len;
memcpy(buf + ptr, data, clen);
data = (const unsigned char *)data + clen;
len -= clen;
ptr += clen;
if (ptr == sizeof sc->buf) {
sph_u32 *ht;
compress_small(buf, h1, h2);
ht = h1;
h1 = h2;
h2 = ht;
ptr = 0;
}
}
sc->ptr = ptr;
if (h1 != sc->H)
memcpy(sc->H, h1, sizeof sc->H);
}
static void
jh_core(sph_jh_context *sc, const void *data, size_t len)
{
unsigned char *buf;
size_t ptr;
DECL_STATE
buf = sc->buf;
ptr = sc->ptr;
if (len < (sizeof sc->buf) - ptr) {
memcpy(buf + ptr, data, len);
ptr += len;
sc->ptr = ptr;
return;
}
READ_STATE(sc);
while (len > 0) {
size_t clen;
clen = (sizeof sc->buf) - ptr;
if (clen > len)
clen = len;
memcpy(buf + ptr, data, clen);
ptr += clen;
data = (const unsigned char *)data + clen;
len -= clen;
if (ptr == sizeof sc->buf) {
INPUT_BUF1;
E8;
INPUT_BUF2;
#if SPH_64
sc->block_count ++;
#else
if ((sc->block_count_low = SPH_T32(
sc->block_count_low + 1)) == 0)
sc->block_count_high ++;
#endif
ptr = 0;
}
}
WRITE_STATE(sc);
sc->ptr = ptr;
}
static void
bmw32_close(sph_bmw_small_context *sc, unsigned ub, unsigned n,
void *dst, size_t out_size_w32)
{
unsigned char *buf, *out;
size_t ptr, u, v;
unsigned z;
sph_u32 h1[16], h2[16], *h;
buf = sc->buf;
ptr = sc->ptr;
z = 0x80 >> n;
buf[ptr ++] = ((ub & -z) | z) & 0xFF;
h = sc->H;
if (ptr > (sizeof sc->buf) - 8) {
memset(buf + ptr, 0, (sizeof sc->buf) - ptr);
compress_small(buf, h, h1);
ptr = 0;
h = h1;
}
memset(buf + ptr, 0, (sizeof sc->buf) - 8 - ptr);
#if SPH_64
sph_enc64le_aligned(buf + (sizeof sc->buf) - 8,
SPH_T64(sc->bit_count + n));
#else
sph_enc32le_aligned(buf + (sizeof sc->buf) - 8,
sc->bit_count_low + n);
sph_enc32le_aligned(buf + (sizeof sc->buf) - 4,
SPH_T32(sc->bit_count_high));
#endif
compress_small(buf, h, h2);
for (u = 0; u < 16; u ++)
sph_enc32le_aligned(buf + 4 * u, h2[u]);
compress_small(buf, final_s, h1);
out = dst;
for (u = 0, v = 16 - out_size_w32; u < out_size_w32; u ++, v ++)
sph_enc32le(out + 4 * u, h1[v]);
}
/*
* This function assumes that "msg" is aligned for 32-bit access.
*/
static void
c512(sph_shavite_big_context *sc, const void *msg)
{
sph_u32 p0, p1, p2, p3, p4, p5, p6, p7;
sph_u32 p8, p9, pA, pB, pC, pD, pE, pF;
sph_u32 x0, x1, x2, x3;
sph_u32 rk00, rk01, rk02, rk03, rk04, rk05, rk06, rk07;
sph_u32 rk08, rk09, rk0A, rk0B, rk0C, rk0D, rk0E, rk0F;
sph_u32 rk10, rk11, rk12, rk13, rk14, rk15, rk16, rk17;
sph_u32 rk18, rk19, rk1A, rk1B, rk1C, rk1D, rk1E, rk1F;
int r;
p0 = sc->h[0x0];
p1 = sc->h[0x1];
p2 = sc->h[0x2];
p3 = sc->h[0x3];
p4 = sc->h[0x4];
p5 = sc->h[0x5];
p6 = sc->h[0x6];
p7 = sc->h[0x7];
p8 = sc->h[0x8];
p9 = sc->h[0x9];
pA = sc->h[0xA];
pB = sc->h[0xB];
pC = sc->h[0xC];
pD = sc->h[0xD];
pE = sc->h[0xE];
pF = sc->h[0xF];
/* round 0 */
rk00 = sph_dec32le_aligned((const unsigned char *)msg + 0);
x0 = p4 ^ rk00;
rk01 = sph_dec32le_aligned((const unsigned char *)msg + 4);
x1 = p5 ^ rk01;
rk02 = sph_dec32le_aligned((const unsigned char *)msg + 8);
x2 = p6 ^ rk02;
rk03 = sph_dec32le_aligned((const unsigned char *)msg + 12);
x3 = p7 ^ rk03;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk04 = sph_dec32le_aligned((const unsigned char *)msg + 16);
x0 ^= rk04;
rk05 = sph_dec32le_aligned((const unsigned char *)msg + 20);
x1 ^= rk05;
rk06 = sph_dec32le_aligned((const unsigned char *)msg + 24);
x2 ^= rk06;
rk07 = sph_dec32le_aligned((const unsigned char *)msg + 28);
x3 ^= rk07;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk08 = sph_dec32le_aligned((const unsigned char *)msg + 32);
x0 ^= rk08;
rk09 = sph_dec32le_aligned((const unsigned char *)msg + 36);
x1 ^= rk09;
rk0A = sph_dec32le_aligned((const unsigned char *)msg + 40);
x2 ^= rk0A;
rk0B = sph_dec32le_aligned((const unsigned char *)msg + 44);
x3 ^= rk0B;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk0C = sph_dec32le_aligned((const unsigned char *)msg + 48);
x0 ^= rk0C;
rk0D = sph_dec32le_aligned((const unsigned char *)msg + 52);
x1 ^= rk0D;
rk0E = sph_dec32le_aligned((const unsigned char *)msg + 56);
x2 ^= rk0E;
rk0F = sph_dec32le_aligned((const unsigned char *)msg + 60);
x3 ^= rk0F;
AES_ROUND_NOKEY(x0, x1, x2, x3);
p0 ^= x0;
p1 ^= x1;
p2 ^= x2;
p3 ^= x3;
rk10 = sph_dec32le_aligned((const unsigned char *)msg + 64);
x0 = pC ^ rk10;
rk11 = sph_dec32le_aligned((const unsigned char *)msg + 68);
x1 = pD ^ rk11;
rk12 = sph_dec32le_aligned((const unsigned char *)msg + 72);
x2 = pE ^ rk12;
rk13 = sph_dec32le_aligned((const unsigned char *)msg + 76);
x3 = pF ^ rk13;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk14 = sph_dec32le_aligned((const unsigned char *)msg + 80);
x0 ^= rk14;
rk15 = sph_dec32le_aligned((const unsigned char *)msg + 84);
x1 ^= rk15;
rk16 = sph_dec32le_aligned((const unsigned char *)msg + 88);
x2 ^= rk16;
rk17 = sph_dec32le_aligned((const unsigned char *)msg + 92);
x3 ^= rk17;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk18 = sph_dec32le_aligned((const unsigned char *)msg + 96);
x0 ^= rk18;
rk19 = sph_dec32le_aligned((const unsigned char *)msg + 100);
x1 ^= rk19;
rk1A = sph_dec32le_aligned((const unsigned char *)msg + 104);
x2 ^= rk1A;
rk1B = sph_dec32le_aligned((const unsigned char *)msg + 108);
x3 ^= rk1B;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk1C = sph_dec32le_aligned((const unsigned char *)msg + 112);
x0 ^= rk1C;
rk1D = sph_dec32le_aligned((const unsigned char *)msg + 116);
x1 ^= rk1D;
rk1E = sph_dec32le_aligned((const unsigned char *)msg + 120);
x2 ^= rk1E;
rk1F = sph_dec32le_aligned((const unsigned char *)msg + 124);
x3 ^= rk1F;
AES_ROUND_NOKEY(x0, x1, x2, x3);
p8 ^= x0;
p9 ^= x1;
pA ^= x2;
pB ^= x3;
for (r = 0; r < 3; r ++) {
/* round 1, 5, 9 */
KEY_EXPAND_ELT(rk00, rk01, rk02, rk03);
rk00 ^= rk1C;
rk01 ^= rk1D;
rk02 ^= rk1E;
rk03 ^= rk1F;
if (r == 0) {
rk00 ^= sc->count0;
rk01 ^= sc->count1;
rk02 ^= sc->count2;
rk03 ^= SPH_T32(~sc->count3);
}
x0 = p0 ^ rk00;
x1 = p1 ^ rk01;
x2 = p2 ^ rk02;
x3 = p3 ^ rk03;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk04, rk05, rk06, rk07);
rk04 ^= rk00;
rk05 ^= rk01;
rk06 ^= rk02;
rk07 ^= rk03;
if (r == 1) {
rk04 ^= sc->count3;
rk05 ^= sc->count2;
rk06 ^= sc->count1;
rk07 ^= SPH_T32(~sc->count0);
}
x0 ^= rk04;
x1 ^= rk05;
x2 ^= rk06;
x3 ^= rk07;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk08, rk09, rk0A, rk0B);
rk08 ^= rk04;
rk09 ^= rk05;
rk0A ^= rk06;
rk0B ^= rk07;
x0 ^= rk08;
x1 ^= rk09;
x2 ^= rk0A;
x3 ^= rk0B;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk0C, rk0D, rk0E, rk0F);
rk0C ^= rk08;
rk0D ^= rk09;
rk0E ^= rk0A;
rk0F ^= rk0B;
x0 ^= rk0C;
x1 ^= rk0D;
x2 ^= rk0E;
x3 ^= rk0F;
AES_ROUND_NOKEY(x0, x1, x2, x3);
pC ^= x0;
pD ^= x1;
pE ^= x2;
pF ^= x3;
KEY_EXPAND_ELT(rk10, rk11, rk12, rk13);
rk10 ^= rk0C;
rk11 ^= rk0D;
rk12 ^= rk0E;
rk13 ^= rk0F;
x0 = p8 ^ rk10;
x1 = p9 ^ rk11;
x2 = pA ^ rk12;
x3 = pB ^ rk13;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk14, rk15, rk16, rk17);
rk14 ^= rk10;
rk15 ^= rk11;
rk16 ^= rk12;
rk17 ^= rk13;
x0 ^= rk14;
x1 ^= rk15;
x2 ^= rk16;
x3 ^= rk17;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk18, rk19, rk1A, rk1B);
rk18 ^= rk14;
rk19 ^= rk15;
rk1A ^= rk16;
rk1B ^= rk17;
x0 ^= rk18;
x1 ^= rk19;
x2 ^= rk1A;
x3 ^= rk1B;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk1C, rk1D, rk1E, rk1F);
rk1C ^= rk18;
rk1D ^= rk19;
rk1E ^= rk1A;
rk1F ^= rk1B;
if (r == 2) {
rk1C ^= sc->count2;
rk1D ^= sc->count3;
rk1E ^= sc->count0;
rk1F ^= SPH_T32(~sc->count1);
}
x0 ^= rk1C;
x1 ^= rk1D;
x2 ^= rk1E;
x3 ^= rk1F;
AES_ROUND_NOKEY(x0, x1, x2, x3);
p4 ^= x0;
p5 ^= x1;
p6 ^= x2;
p7 ^= x3;
/* round 2, 6, 10 */
rk00 ^= rk19;
x0 = pC ^ rk00;
rk01 ^= rk1A;
x1 = pD ^ rk01;
rk02 ^= rk1B;
x2 = pE ^ rk02;
rk03 ^= rk1C;
x3 = pF ^ rk03;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk04 ^= rk1D;
x0 ^= rk04;
rk05 ^= rk1E;
x1 ^= rk05;
rk06 ^= rk1F;
x2 ^= rk06;
rk07 ^= rk00;
x3 ^= rk07;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk08 ^= rk01;
x0 ^= rk08;
rk09 ^= rk02;
x1 ^= rk09;
rk0A ^= rk03;
x2 ^= rk0A;
rk0B ^= rk04;
x3 ^= rk0B;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk0C ^= rk05;
x0 ^= rk0C;
rk0D ^= rk06;
x1 ^= rk0D;
rk0E ^= rk07;
x2 ^= rk0E;
rk0F ^= rk08;
x3 ^= rk0F;
AES_ROUND_NOKEY(x0, x1, x2, x3);
p8 ^= x0;
p9 ^= x1;
pA ^= x2;
pB ^= x3;
rk10 ^= rk09;
x0 = p4 ^ rk10;
rk11 ^= rk0A;
x1 = p5 ^ rk11;
rk12 ^= rk0B;
x2 = p6 ^ rk12;
rk13 ^= rk0C;
x3 = p7 ^ rk13;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk14 ^= rk0D;
x0 ^= rk14;
rk15 ^= rk0E;
x1 ^= rk15;
rk16 ^= rk0F;
x2 ^= rk16;
rk17 ^= rk10;
x3 ^= rk17;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk18 ^= rk11;
x0 ^= rk18;
rk19 ^= rk12;
x1 ^= rk19;
rk1A ^= rk13;
x2 ^= rk1A;
rk1B ^= rk14;
x3 ^= rk1B;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk1C ^= rk15;
x0 ^= rk1C;
rk1D ^= rk16;
x1 ^= rk1D;
rk1E ^= rk17;
x2 ^= rk1E;
rk1F ^= rk18;
x3 ^= rk1F;
AES_ROUND_NOKEY(x0, x1, x2, x3);
p0 ^= x0;
p1 ^= x1;
p2 ^= x2;
p3 ^= x3;
/* round 3, 7, 11 */
KEY_EXPAND_ELT(rk00, rk01, rk02, rk03);
rk00 ^= rk1C;
rk01 ^= rk1D;
rk02 ^= rk1E;
rk03 ^= rk1F;
x0 = p8 ^ rk00;
x1 = p9 ^ rk01;
x2 = pA ^ rk02;
x3 = pB ^ rk03;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk04, rk05, rk06, rk07);
rk04 ^= rk00;
rk05 ^= rk01;
rk06 ^= rk02;
rk07 ^= rk03;
x0 ^= rk04;
x1 ^= rk05;
x2 ^= rk06;
x3 ^= rk07;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk08, rk09, rk0A, rk0B);
rk08 ^= rk04;
rk09 ^= rk05;
rk0A ^= rk06;
rk0B ^= rk07;
x0 ^= rk08;
x1 ^= rk09;
x2 ^= rk0A;
x3 ^= rk0B;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk0C, rk0D, rk0E, rk0F);
rk0C ^= rk08;
rk0D ^= rk09;
rk0E ^= rk0A;
rk0F ^= rk0B;
x0 ^= rk0C;
x1 ^= rk0D;
x2 ^= rk0E;
x3 ^= rk0F;
AES_ROUND_NOKEY(x0, x1, x2, x3);
p4 ^= x0;
p5 ^= x1;
p6 ^= x2;
p7 ^= x3;
KEY_EXPAND_ELT(rk10, rk11, rk12, rk13);
rk10 ^= rk0C;
rk11 ^= rk0D;
rk12 ^= rk0E;
rk13 ^= rk0F;
x0 = p0 ^ rk10;
x1 = p1 ^ rk11;
x2 = p2 ^ rk12;
x3 = p3 ^ rk13;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk14, rk15, rk16, rk17);
rk14 ^= rk10;
rk15 ^= rk11;
rk16 ^= rk12;
rk17 ^= rk13;
x0 ^= rk14;
x1 ^= rk15;
x2 ^= rk16;
x3 ^= rk17;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk18, rk19, rk1A, rk1B);
rk18 ^= rk14;
rk19 ^= rk15;
rk1A ^= rk16;
rk1B ^= rk17;
x0 ^= rk18;
x1 ^= rk19;
x2 ^= rk1A;
x3 ^= rk1B;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk1C, rk1D, rk1E, rk1F);
rk1C ^= rk18;
rk1D ^= rk19;
rk1E ^= rk1A;
rk1F ^= rk1B;
x0 ^= rk1C;
x1 ^= rk1D;
x2 ^= rk1E;
x3 ^= rk1F;
AES_ROUND_NOKEY(x0, x1, x2, x3);
pC ^= x0;
pD ^= x1;
pE ^= x2;
pF ^= x3;
/* round 4, 8, 12 */
rk00 ^= rk19;
x0 = p4 ^ rk00;
rk01 ^= rk1A;
x1 = p5 ^ rk01;
rk02 ^= rk1B;
x2 = p6 ^ rk02;
rk03 ^= rk1C;
x3 = p7 ^ rk03;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk04 ^= rk1D;
x0 ^= rk04;
rk05 ^= rk1E;
x1 ^= rk05;
rk06 ^= rk1F;
x2 ^= rk06;
rk07 ^= rk00;
x3 ^= rk07;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk08 ^= rk01;
x0 ^= rk08;
rk09 ^= rk02;
x1 ^= rk09;
rk0A ^= rk03;
x2 ^= rk0A;
rk0B ^= rk04;
x3 ^= rk0B;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk0C ^= rk05;
x0 ^= rk0C;
rk0D ^= rk06;
x1 ^= rk0D;
rk0E ^= rk07;
x2 ^= rk0E;
rk0F ^= rk08;
x3 ^= rk0F;
AES_ROUND_NOKEY(x0, x1, x2, x3);
p0 ^= x0;
p1 ^= x1;
p2 ^= x2;
p3 ^= x3;
rk10 ^= rk09;
x0 = pC ^ rk10;
rk11 ^= rk0A;
x1 = pD ^ rk11;
rk12 ^= rk0B;
x2 = pE ^ rk12;
rk13 ^= rk0C;
x3 = pF ^ rk13;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk14 ^= rk0D;
x0 ^= rk14;
rk15 ^= rk0E;
x1 ^= rk15;
rk16 ^= rk0F;
x2 ^= rk16;
rk17 ^= rk10;
x3 ^= rk17;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk18 ^= rk11;
x0 ^= rk18;
rk19 ^= rk12;
x1 ^= rk19;
rk1A ^= rk13;
x2 ^= rk1A;
rk1B ^= rk14;
x3 ^= rk1B;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk1C ^= rk15;
x0 ^= rk1C;
rk1D ^= rk16;
x1 ^= rk1D;
rk1E ^= rk17;
x2 ^= rk1E;
rk1F ^= rk18;
x3 ^= rk1F;
AES_ROUND_NOKEY(x0, x1, x2, x3);
p8 ^= x0;
p9 ^= x1;
pA ^= x2;
pB ^= x3;
}
/* round 13 */
KEY_EXPAND_ELT(rk00, rk01, rk02, rk03);
rk00 ^= rk1C;
rk01 ^= rk1D;
rk02 ^= rk1E;
rk03 ^= rk1F;
x0 = p0 ^ rk00;
x1 = p1 ^ rk01;
x2 = p2 ^ rk02;
x3 = p3 ^ rk03;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk04, rk05, rk06, rk07);
rk04 ^= rk00;
rk05 ^= rk01;
rk06 ^= rk02;
rk07 ^= rk03;
x0 ^= rk04;
x1 ^= rk05;
x2 ^= rk06;
x3 ^= rk07;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk08, rk09, rk0A, rk0B);
rk08 ^= rk04;
rk09 ^= rk05;
rk0A ^= rk06;
rk0B ^= rk07;
x0 ^= rk08;
x1 ^= rk09;
x2 ^= rk0A;
x3 ^= rk0B;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk0C, rk0D, rk0E, rk0F);
rk0C ^= rk08;
rk0D ^= rk09;
rk0E ^= rk0A;
rk0F ^= rk0B;
x0 ^= rk0C;
x1 ^= rk0D;
x2 ^= rk0E;
x3 ^= rk0F;
AES_ROUND_NOKEY(x0, x1, x2, x3);
pC ^= x0;
pD ^= x1;
pE ^= x2;
pF ^= x3;
KEY_EXPAND_ELT(rk10, rk11, rk12, rk13);
rk10 ^= rk0C;
rk11 ^= rk0D;
rk12 ^= rk0E;
rk13 ^= rk0F;
x0 = p8 ^ rk10;
x1 = p9 ^ rk11;
x2 = pA ^ rk12;
x3 = pB ^ rk13;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk14, rk15, rk16, rk17);
rk14 ^= rk10;
rk15 ^= rk11;
rk16 ^= rk12;
rk17 ^= rk13;
x0 ^= rk14;
x1 ^= rk15;
x2 ^= rk16;
x3 ^= rk17;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk18, rk19, rk1A, rk1B);
rk18 ^= rk14 ^ sc->count1;
rk19 ^= rk15 ^ sc->count0;
rk1A ^= rk16 ^ sc->count3;
rk1B ^= rk17 ^ SPH_T32(~sc->count2);
x0 ^= rk18;
x1 ^= rk19;
x2 ^= rk1A;
x3 ^= rk1B;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk1C, rk1D, rk1E, rk1F);
rk1C ^= rk18;
rk1D ^= rk19;
rk1E ^= rk1A;
rk1F ^= rk1B;
x0 ^= rk1C;
x1 ^= rk1D;
x2 ^= rk1E;
x3 ^= rk1F;
AES_ROUND_NOKEY(x0, x1, x2, x3);
p4 ^= x0;
p5 ^= x1;
p6 ^= x2;
p7 ^= x3;
sc->h[0x0] ^= p8;
sc->h[0x1] ^= p9;
sc->h[0x2] ^= pA;
sc->h[0x3] ^= pB;
sc->h[0x4] ^= pC;
sc->h[0x5] ^= pD;
sc->h[0x6] ^= pE;
sc->h[0x7] ^= pF;
sc->h[0x8] ^= p0;
sc->h[0x9] ^= p1;
sc->h[0xA] ^= p2;
sc->h[0xB] ^= p3;
sc->h[0xC] ^= p4;
sc->h[0xD] ^= p5;
sc->h[0xE] ^= p6;
sc->h[0xF] ^= p7;
}
/*
* This function assumes that "msg" is aligned for 32-bit access.
*/
static void
c512(sph_shavite_big_context *sc, const void *msg)
{
sph_u32 p0, p1, p2, p3, p4, p5, p6, p7;
sph_u32 p8, p9, pA, pB, pC, pD, pE, pF;
sph_u32 rk[448];
size_t u;
int r, s;
#if SPH_LITTLE_ENDIAN
memcpy(rk, msg, 128);
#else
for (u = 0; u < 32; u += 4) {
rk[u + 0] = sph_dec32le_aligned(
(const unsigned char *)msg + (u << 2) + 0);
rk[u + 1] = sph_dec32le_aligned(
(const unsigned char *)msg + (u << 2) + 4);
rk[u + 2] = sph_dec32le_aligned(
(const unsigned char *)msg + (u << 2) + 8);
rk[u + 3] = sph_dec32le_aligned(
(const unsigned char *)msg + (u << 2) + 12);
}
#endif
u = 32;
for (;;) {
for (s = 0; s < 4; s ++) {
sph_u32 x0, x1, x2, x3;
x0 = rk[u - 31];
x1 = rk[u - 30];
x2 = rk[u - 29];
x3 = rk[u - 32];
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk[u + 0] = x0 ^ rk[u - 4];
rk[u + 1] = x1 ^ rk[u - 3];
rk[u + 2] = x2 ^ rk[u - 2];
rk[u + 3] = x3 ^ rk[u - 1];
if (u == 32) {
rk[ 32] ^= sc->count0;
rk[ 33] ^= sc->count1;
rk[ 34] ^= sc->count2;
rk[ 35] ^= SPH_T32(~sc->count3);
} else if (u == 440) {
rk[440] ^= sc->count1;
rk[441] ^= sc->count0;
rk[442] ^= sc->count3;
rk[443] ^= SPH_T32(~sc->count2);
}
u += 4;
x0 = rk[u - 31];
x1 = rk[u - 30];
x2 = rk[u - 29];
x3 = rk[u - 32];
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk[u + 0] = x0 ^ rk[u - 4];
rk[u + 1] = x1 ^ rk[u - 3];
rk[u + 2] = x2 ^ rk[u - 2];
rk[u + 3] = x3 ^ rk[u - 1];
if (u == 164) {
rk[164] ^= sc->count3;
rk[165] ^= sc->count2;
rk[166] ^= sc->count1;
rk[167] ^= SPH_T32(~sc->count0);
} else if (u == 316) {
rk[316] ^= sc->count2;
rk[317] ^= sc->count3;
rk[318] ^= sc->count0;
rk[319] ^= SPH_T32(~sc->count1);
}
u += 4;
}
if (u == 448)
break;
for (s = 0; s < 8; s ++) {
rk[u + 0] = rk[u - 32] ^ rk[u - 7];
rk[u + 1] = rk[u - 31] ^ rk[u - 6];
rk[u + 2] = rk[u - 30] ^ rk[u - 5];
rk[u + 3] = rk[u - 29] ^ rk[u - 4];
u += 4;
}
}
p0 = sc->h[0x0];
p1 = sc->h[0x1];
p2 = sc->h[0x2];
p3 = sc->h[0x3];
p4 = sc->h[0x4];
p5 = sc->h[0x5];
p6 = sc->h[0x6];
p7 = sc->h[0x7];
p8 = sc->h[0x8];
p9 = sc->h[0x9];
pA = sc->h[0xA];
pB = sc->h[0xB];
pC = sc->h[0xC];
pD = sc->h[0xD];
pE = sc->h[0xE];
pF = sc->h[0xF];
u = 0;
for (r = 0; r < 14; r ++) {
#define C512_ELT(l0, l1, l2, l3, r0, r1, r2, r3) do { \
sph_u32 x0, x1, x2, x3; \
x0 = r0 ^ rk[u ++]; \
x1 = r1 ^ rk[u ++]; \
x2 = r2 ^ rk[u ++]; \
x3 = r3 ^ rk[u ++]; \
AES_ROUND_NOKEY(x0, x1, x2, x3); \
x0 ^= rk[u ++]; \
x1 ^= rk[u ++]; \
x2 ^= rk[u ++]; \
x3 ^= rk[u ++]; \
AES_ROUND_NOKEY(x0, x1, x2, x3); \
x0 ^= rk[u ++]; \
x1 ^= rk[u ++]; \
x2 ^= rk[u ++]; \
x3 ^= rk[u ++]; \
AES_ROUND_NOKEY(x0, x1, x2, x3); \
x0 ^= rk[u ++]; \
x1 ^= rk[u ++]; \
x2 ^= rk[u ++]; \
x3 ^= rk[u ++]; \
AES_ROUND_NOKEY(x0, x1, x2, x3); \
l0 ^= x0; \
l1 ^= x1; \
l2 ^= x2; \
l3 ^= x3; \
} while (0)
#define WROT(a, b, c, d) do { \
sph_u32 t = d; \
d = c; \
c = b; \
b = a; \
a = t; \
} while (0)
C512_ELT(p0, p1, p2, p3, p4, p5, p6, p7);
C512_ELT(p8, p9, pA, pB, pC, pD, pE, pF);
WROT(p0, p4, p8, pC);
WROT(p1, p5, p9, pD);
WROT(p2, p6, pA, pE);
WROT(p3, p7, pB, pF);
#undef C512_ELT
#undef WROT
}
sc->h[0x0] ^= p0;
sc->h[0x1] ^= p1;
sc->h[0x2] ^= p2;
sc->h[0x3] ^= p3;
sc->h[0x4] ^= p4;
sc->h[0x5] ^= p5;
sc->h[0x6] ^= p6;
sc->h[0x7] ^= p7;
sc->h[0x8] ^= p8;
sc->h[0x9] ^= p9;
sc->h[0xA] ^= pA;
sc->h[0xB] ^= pB;
sc->h[0xC] ^= pC;
sc->h[0xD] ^= pD;
sc->h[0xE] ^= pE;
sc->h[0xF] ^= pF;
}
/*
* This function assumes that "msg" is aligned for 32-bit access.
*/
static void
c256(sph_shavite_small_context *sc, const void *msg)
{
sph_u32 p0, p1, p2, p3, p4, p5, p6, p7;
sph_u32 x0, x1, x2, x3;
sph_u32 rk0, rk1, rk2, rk3, rk4, rk5, rk6, rk7;
sph_u32 rk8, rk9, rkA, rkB, rkC, rkD, rkE, rkF;
p0 = sc->h[0x0];
p1 = sc->h[0x1];
p2 = sc->h[0x2];
p3 = sc->h[0x3];
p4 = sc->h[0x4];
p5 = sc->h[0x5];
p6 = sc->h[0x6];
p7 = sc->h[0x7];
/* round 0 */
rk0 = sph_dec32le_aligned((const unsigned char *)msg + 0);
x0 = p4 ^ rk0;
rk1 = sph_dec32le_aligned((const unsigned char *)msg + 4);
x1 = p5 ^ rk1;
rk2 = sph_dec32le_aligned((const unsigned char *)msg + 8);
x2 = p6 ^ rk2;
rk3 = sph_dec32le_aligned((const unsigned char *)msg + 12);
x3 = p7 ^ rk3;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk4 = sph_dec32le_aligned((const unsigned char *)msg + 16);
x0 ^= rk4;
rk5 = sph_dec32le_aligned((const unsigned char *)msg + 20);
x1 ^= rk5;
rk6 = sph_dec32le_aligned((const unsigned char *)msg + 24);
x2 ^= rk6;
rk7 = sph_dec32le_aligned((const unsigned char *)msg + 28);
x3 ^= rk7;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk8 = sph_dec32le_aligned((const unsigned char *)msg + 32);
x0 ^= rk8;
rk9 = sph_dec32le_aligned((const unsigned char *)msg + 36);
x1 ^= rk9;
rkA = sph_dec32le_aligned((const unsigned char *)msg + 40);
x2 ^= rkA;
rkB = sph_dec32le_aligned((const unsigned char *)msg + 44);
x3 ^= rkB;
AES_ROUND_NOKEY(x0, x1, x2, x3);
p0 ^= x0;
p1 ^= x1;
p2 ^= x2;
p3 ^= x3;
/* round 1 */
rkC = sph_dec32le_aligned((const unsigned char *)msg + 48);
x0 = p0 ^ rkC;
rkD = sph_dec32le_aligned((const unsigned char *)msg + 52);
x1 = p1 ^ rkD;
rkE = sph_dec32le_aligned((const unsigned char *)msg + 56);
x2 = p2 ^ rkE;
rkF = sph_dec32le_aligned((const unsigned char *)msg + 60);
x3 = p3 ^ rkF;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk0, rk1, rk2, rk3);
rk0 ^= rkC ^ sc->count0;
rk1 ^= rkD ^ SPH_T32(~sc->count1);
rk2 ^= rkE;
rk3 ^= rkF;
x0 ^= rk0;
x1 ^= rk1;
x2 ^= rk2;
x3 ^= rk3;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk4, rk5, rk6, rk7);
rk4 ^= rk0;
rk5 ^= rk1;
rk6 ^= rk2;
rk7 ^= rk3;
x0 ^= rk4;
x1 ^= rk5;
x2 ^= rk6;
x3 ^= rk7;
AES_ROUND_NOKEY(x0, x1, x2, x3);
p4 ^= x0;
p5 ^= x1;
p6 ^= x2;
p7 ^= x3;
/* round 2 */
KEY_EXPAND_ELT(rk8, rk9, rkA, rkB);
rk8 ^= rk4;
rk9 ^= rk5;
rkA ^= rk6;
rkB ^= rk7;
x0 = p4 ^ rk8;
x1 = p5 ^ rk9;
x2 = p6 ^ rkA;
x3 = p7 ^ rkB;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rkC, rkD, rkE, rkF);
rkC ^= rk8;
rkD ^= rk9;
rkE ^= rkA;
rkF ^= rkB;
x0 ^= rkC;
x1 ^= rkD;
x2 ^= rkE;
x3 ^= rkF;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk0 ^= rkD;
x0 ^= rk0;
rk1 ^= rkE;
x1 ^= rk1;
rk2 ^= rkF;
x2 ^= rk2;
rk3 ^= rk0;
x3 ^= rk3;
AES_ROUND_NOKEY(x0, x1, x2, x3);
p0 ^= x0;
p1 ^= x1;
p2 ^= x2;
p3 ^= x3;
/* round 3 */
rk4 ^= rk1;
x0 = p0 ^ rk4;
rk5 ^= rk2;
x1 = p1 ^ rk5;
rk6 ^= rk3;
x2 = p2 ^ rk6;
rk7 ^= rk4;
x3 = p3 ^ rk7;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk8 ^= rk5;
x0 ^= rk8;
rk9 ^= rk6;
x1 ^= rk9;
rkA ^= rk7;
x2 ^= rkA;
rkB ^= rk8;
x3 ^= rkB;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rkC ^= rk9;
x0 ^= rkC;
rkD ^= rkA;
x1 ^= rkD;
rkE ^= rkB;
x2 ^= rkE;
rkF ^= rkC;
x3 ^= rkF;
AES_ROUND_NOKEY(x0, x1, x2, x3);
p4 ^= x0;
p5 ^= x1;
p6 ^= x2;
p7 ^= x3;
/* round 4 */
KEY_EXPAND_ELT(rk0, rk1, rk2, rk3);
rk0 ^= rkC;
rk1 ^= rkD;
rk2 ^= rkE;
rk3 ^= rkF;
x0 = p4 ^ rk0;
x1 = p5 ^ rk1;
x2 = p6 ^ rk2;
x3 = p7 ^ rk3;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk4, rk5, rk6, rk7);
rk4 ^= rk0;
rk5 ^= rk1;
rk6 ^= rk2;
rk7 ^= rk3;
x0 ^= rk4;
x1 ^= rk5;
x2 ^= rk6;
x3 ^= rk7;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk8, rk9, rkA, rkB);
rk8 ^= rk4;
rk9 ^= rk5 ^ sc->count1;
rkA ^= rk6 ^ SPH_T32(~sc->count0);
rkB ^= rk7;
x0 ^= rk8;
x1 ^= rk9;
x2 ^= rkA;
x3 ^= rkB;
AES_ROUND_NOKEY(x0, x1, x2, x3);
p0 ^= x0;
p1 ^= x1;
p2 ^= x2;
p3 ^= x3;
/* round 5 */
KEY_EXPAND_ELT(rkC, rkD, rkE, rkF);
rkC ^= rk8;
rkD ^= rk9;
rkE ^= rkA;
rkF ^= rkB;
x0 = p0 ^ rkC;
x1 = p1 ^ rkD;
x2 = p2 ^ rkE;
x3 = p3 ^ rkF;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk0 ^= rkD;
x0 ^= rk0;
rk1 ^= rkE;
x1 ^= rk1;
rk2 ^= rkF;
x2 ^= rk2;
rk3 ^= rk0;
x3 ^= rk3;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk4 ^= rk1;
x0 ^= rk4;
rk5 ^= rk2;
x1 ^= rk5;
rk6 ^= rk3;
x2 ^= rk6;
rk7 ^= rk4;
x3 ^= rk7;
AES_ROUND_NOKEY(x0, x1, x2, x3);
p4 ^= x0;
p5 ^= x1;
p6 ^= x2;
p7 ^= x3;
/* round 6 */
rk8 ^= rk5;
x0 = p4 ^ rk8;
rk9 ^= rk6;
x1 = p5 ^ rk9;
rkA ^= rk7;
x2 = p6 ^ rkA;
rkB ^= rk8;
x3 = p7 ^ rkB;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rkC ^= rk9;
x0 ^= rkC;
rkD ^= rkA;
x1 ^= rkD;
rkE ^= rkB;
x2 ^= rkE;
rkF ^= rkC;
x3 ^= rkF;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk0, rk1, rk2, rk3);
rk0 ^= rkC;
rk1 ^= rkD;
rk2 ^= rkE;
rk3 ^= rkF;
x0 ^= rk0;
x1 ^= rk1;
x2 ^= rk2;
x3 ^= rk3;
AES_ROUND_NOKEY(x0, x1, x2, x3);
p0 ^= x0;
p1 ^= x1;
p2 ^= x2;
p3 ^= x3;
/* round 7 */
KEY_EXPAND_ELT(rk4, rk5, rk6, rk7);
rk4 ^= rk0;
rk5 ^= rk1;
rk6 ^= rk2 ^ sc->count1;
rk7 ^= rk3 ^ SPH_T32(~sc->count0);
x0 = p0 ^ rk4;
x1 = p1 ^ rk5;
x2 = p2 ^ rk6;
x3 = p3 ^ rk7;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk8, rk9, rkA, rkB);
rk8 ^= rk4;
rk9 ^= rk5;
rkA ^= rk6;
rkB ^= rk7;
x0 ^= rk8;
x1 ^= rk9;
x2 ^= rkA;
x3 ^= rkB;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rkC, rkD, rkE, rkF);
rkC ^= rk8;
rkD ^= rk9;
rkE ^= rkA;
rkF ^= rkB;
x0 ^= rkC;
x1 ^= rkD;
x2 ^= rkE;
x3 ^= rkF;
AES_ROUND_NOKEY(x0, x1, x2, x3);
p4 ^= x0;
p5 ^= x1;
p6 ^= x2;
p7 ^= x3;
/* round 8 */
rk0 ^= rkD;
x0 = p4 ^ rk0;
rk1 ^= rkE;
x1 = p5 ^ rk1;
rk2 ^= rkF;
x2 = p6 ^ rk2;
rk3 ^= rk0;
x3 = p7 ^ rk3;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk4 ^= rk1;
x0 ^= rk4;
rk5 ^= rk2;
x1 ^= rk5;
rk6 ^= rk3;
x2 ^= rk6;
rk7 ^= rk4;
x3 ^= rk7;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk8 ^= rk5;
x0 ^= rk8;
rk9 ^= rk6;
x1 ^= rk9;
rkA ^= rk7;
x2 ^= rkA;
rkB ^= rk8;
x3 ^= rkB;
AES_ROUND_NOKEY(x0, x1, x2, x3);
p0 ^= x0;
p1 ^= x1;
p2 ^= x2;
p3 ^= x3;
/* round 9 */
rkC ^= rk9;
x0 = p0 ^ rkC;
rkD ^= rkA;
x1 = p1 ^ rkD;
rkE ^= rkB;
x2 = p2 ^ rkE;
rkF ^= rkC;
x3 = p3 ^ rkF;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk0, rk1, rk2, rk3);
rk0 ^= rkC;
rk1 ^= rkD;
rk2 ^= rkE;
rk3 ^= rkF;
x0 ^= rk0;
x1 ^= rk1;
x2 ^= rk2;
x3 ^= rk3;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rk4, rk5, rk6, rk7);
rk4 ^= rk0;
rk5 ^= rk1;
rk6 ^= rk2;
rk7 ^= rk3;
x0 ^= rk4;
x1 ^= rk5;
x2 ^= rk6;
x3 ^= rk7;
AES_ROUND_NOKEY(x0, x1, x2, x3);
p4 ^= x0;
p5 ^= x1;
p6 ^= x2;
p7 ^= x3;
/* round 10 */
KEY_EXPAND_ELT(rk8, rk9, rkA, rkB);
rk8 ^= rk4;
rk9 ^= rk5;
rkA ^= rk6;
rkB ^= rk7;
x0 = p4 ^ rk8;
x1 = p5 ^ rk9;
x2 = p6 ^ rkA;
x3 = p7 ^ rkB;
AES_ROUND_NOKEY(x0, x1, x2, x3);
KEY_EXPAND_ELT(rkC, rkD, rkE, rkF);
rkC ^= rk8 ^ sc->count0;
rkD ^= rk9;
rkE ^= rkA;
rkF ^= rkB ^ SPH_T32(~sc->count1);
x0 ^= rkC;
x1 ^= rkD;
x2 ^= rkE;
x3 ^= rkF;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk0 ^= rkD;
x0 ^= rk0;
rk1 ^= rkE;
x1 ^= rk1;
rk2 ^= rkF;
x2 ^= rk2;
rk3 ^= rk0;
x3 ^= rk3;
AES_ROUND_NOKEY(x0, x1, x2, x3);
p0 ^= x0;
p1 ^= x1;
p2 ^= x2;
p3 ^= x3;
/* round 11 */
rk4 ^= rk1;
x0 = p0 ^ rk4;
rk5 ^= rk2;
x1 = p1 ^ rk5;
rk6 ^= rk3;
x2 = p2 ^ rk6;
rk7 ^= rk4;
x3 = p3 ^ rk7;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk8 ^= rk5;
x0 ^= rk8;
rk9 ^= rk6;
x1 ^= rk9;
rkA ^= rk7;
x2 ^= rkA;
rkB ^= rk8;
x3 ^= rkB;
AES_ROUND_NOKEY(x0, x1, x2, x3);
rkC ^= rk9;
x0 ^= rkC;
rkD ^= rkA;
x1 ^= rkD;
rkE ^= rkB;
x2 ^= rkE;
rkF ^= rkC;
x3 ^= rkF;
AES_ROUND_NOKEY(x0, x1, x2, x3);
p4 ^= x0;
p5 ^= x1;
p6 ^= x2;
p7 ^= x3;
sc->h[0x0] ^= p0;
sc->h[0x1] ^= p1;
sc->h[0x2] ^= p2;
sc->h[0x3] ^= p3;
sc->h[0x4] ^= p4;
sc->h[0x5] ^= p5;
sc->h[0x6] ^= p6;
sc->h[0x7] ^= p7;
}
/*
* This function assumes that "msg" is aligned for 32-bit access.
*/
static void
c256(sph_shavite_small_context *sc, const void *msg)
{
sph_u32 p0, p1, p2, p3, p4, p5, p6, p7;
sph_u32 rk[144];
size_t u;
int r, s;
#if SPH_LITTLE_ENDIAN
memcpy(rk, msg, 64);
#else
for (u = 0; u < 16; u += 4) {
rk[u + 0] = sph_dec32le_aligned(
(const unsigned char *)msg + (u << 2) + 0);
rk[u + 1] = sph_dec32le_aligned(
(const unsigned char *)msg + (u << 2) + 4);
rk[u + 2] = sph_dec32le_aligned(
(const unsigned char *)msg + (u << 2) + 8);
rk[u + 3] = sph_dec32le_aligned(
(const unsigned char *)msg + (u << 2) + 12);
}
#endif
u = 16;
for (r = 0; r < 4; r ++) {
for (s = 0; s < 2; s ++) {
sph_u32 x0, x1, x2, x3;
x0 = rk[u - 15];
x1 = rk[u - 14];
x2 = rk[u - 13];
x3 = rk[u - 16];
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk[u + 0] = x0 ^ rk[u - 4];
rk[u + 1] = x1 ^ rk[u - 3];
rk[u + 2] = x2 ^ rk[u - 2];
rk[u + 3] = x3 ^ rk[u - 1];
if (u == 16) {
rk[ 16] ^= sc->count0;
rk[ 17] ^= SPH_T32(~sc->count1);
} else if (u == 56) {
rk[ 57] ^= sc->count1;
rk[ 58] ^= SPH_T32(~sc->count0);
}
u += 4;
x0 = rk[u - 15];
x1 = rk[u - 14];
x2 = rk[u - 13];
x3 = rk[u - 16];
AES_ROUND_NOKEY(x0, x1, x2, x3);
rk[u + 0] = x0 ^ rk[u - 4];
rk[u + 1] = x1 ^ rk[u - 3];
rk[u + 2] = x2 ^ rk[u - 2];
rk[u + 3] = x3 ^ rk[u - 1];
if (u == 84) {
rk[ 86] ^= sc->count1;
rk[ 87] ^= SPH_T32(~sc->count0);
} else if (u == 124) {
rk[124] ^= sc->count0;
rk[127] ^= SPH_T32(~sc->count1);
}
u += 4;
}
for (s = 0; s < 4; s ++) {
rk[u + 0] = rk[u - 16] ^ rk[u - 3];
rk[u + 1] = rk[u - 15] ^ rk[u - 2];
rk[u + 2] = rk[u - 14] ^ rk[u - 1];
rk[u + 3] = rk[u - 13] ^ rk[u - 0];
u += 4;
}
}
p0 = sc->h[0x0];
p1 = sc->h[0x1];
p2 = sc->h[0x2];
p3 = sc->h[0x3];
p4 = sc->h[0x4];
p5 = sc->h[0x5];
p6 = sc->h[0x6];
p7 = sc->h[0x7];
u = 0;
for (r = 0; r < 6; r ++) {
sph_u32 x0, x1, x2, x3;
x0 = p4 ^ rk[u ++];
x1 = p5 ^ rk[u ++];
x2 = p6 ^ rk[u ++];
x3 = p7 ^ rk[u ++];
AES_ROUND_NOKEY(x0, x1, x2, x3);
x0 ^= rk[u ++];
x1 ^= rk[u ++];
x2 ^= rk[u ++];
x3 ^= rk[u ++];
AES_ROUND_NOKEY(x0, x1, x2, x3);
x0 ^= rk[u ++];
x1 ^= rk[u ++];
x2 ^= rk[u ++];
x3 ^= rk[u ++];
AES_ROUND_NOKEY(x0, x1, x2, x3);
p0 ^= x0;
p1 ^= x1;
p2 ^= x2;
p3 ^= x3;
x0 = p0 ^ rk[u ++];
x1 = p1 ^ rk[u ++];
x2 = p2 ^ rk[u ++];
x3 = p3 ^ rk[u ++];
AES_ROUND_NOKEY(x0, x1, x2, x3);
x0 ^= rk[u ++];
x1 ^= rk[u ++];
x2 ^= rk[u ++];
x3 ^= rk[u ++];
AES_ROUND_NOKEY(x0, x1, x2, x3);
x0 ^= rk[u ++];
x1 ^= rk[u ++];
x2 ^= rk[u ++];
x3 ^= rk[u ++];
AES_ROUND_NOKEY(x0, x1, x2, x3);
p4 ^= x0;
p5 ^= x1;
p6 ^= x2;
p7 ^= x3;
}
sc->h[0x0] ^= p0;
sc->h[0x1] ^= p1;
sc->h[0x2] ^= p2;
sc->h[0x3] ^= p3;
sc->h[0x4] ^= p4;
sc->h[0x5] ^= p5;
sc->h[0x6] ^= p6;
sc->h[0x7] ^= p7;
}
/*
* Write out HAVAL output. The output length is tailored to the requested
* length.
*/
static void
haval_out(sph_haval_context *sc, void *dst)
{
DSTATE;
unsigned char *buf;
buf = dst;
RSTATE;
switch (sc->olen) {
case 4:
sph_enc32le(buf, SPH_T32(s0 + mix128(s7, s4, s5, s6, 24)));
sph_enc32le(buf + 4, SPH_T32(s1 + mix128(s6, s7, s4, s5, 16)));
sph_enc32le(buf + 8, SPH_T32(s2 + mix128(s5, s6, s7, s4, 8)));
sph_enc32le(buf + 12, SPH_T32(s3 + mix128(s4, s5, s6, s7, 0)));
break;
case 5:
sph_enc32le(buf, SPH_T32(s0 + mix160_0(s5, s6, s7)));
sph_enc32le(buf + 4, SPH_T32(s1 + mix160_1(s5, s6, s7)));
sph_enc32le(buf + 8, SPH_T32(s2 + mix160_2(s5, s6, s7)));
sph_enc32le(buf + 12, SPH_T32(s3 + mix160_3(s5, s6, s7)));
sph_enc32le(buf + 16, SPH_T32(s4 + mix160_4(s5, s6, s7)));
break;
case 6:
sph_enc32le(buf, SPH_T32(s0 + mix192_0(s6, s7)));
sph_enc32le(buf + 4, SPH_T32(s1 + mix192_1(s6, s7)));
sph_enc32le(buf + 8, SPH_T32(s2 + mix192_2(s6, s7)));
sph_enc32le(buf + 12, SPH_T32(s3 + mix192_3(s6, s7)));
sph_enc32le(buf + 16, SPH_T32(s4 + mix192_4(s6, s7)));
sph_enc32le(buf + 20, SPH_T32(s5 + mix192_5(s6, s7)));
break;
case 7:
sph_enc32le(buf, SPH_T32(s0 + ((s7 >> 27) & 0x1F)));
sph_enc32le(buf + 4, SPH_T32(s1 + ((s7 >> 22) & 0x1F)));
sph_enc32le(buf + 8, SPH_T32(s2 + ((s7 >> 18) & 0x0F)));
sph_enc32le(buf + 12, SPH_T32(s3 + ((s7 >> 13) & 0x1F)));
sph_enc32le(buf + 16, SPH_T32(s4 + ((s7 >> 9) & 0x0F)));
sph_enc32le(buf + 20, SPH_T32(s5 + ((s7 >> 4) & 0x1F)));
sph_enc32le(buf + 24, SPH_T32(s6 + ((s7 ) & 0x0F)));
break;
case 8:
sph_enc32le(buf, s0);
sph_enc32le(buf + 4, s1);
sph_enc32le(buf + 8, s2);
sph_enc32le(buf + 12, s3);
sph_enc32le(buf + 16, s4);
sph_enc32le(buf + 20, s5);
sph_enc32le(buf + 24, s6);
sph_enc32le(buf + 28, s7);
break;
}
}