void blake28_compress( state *S, const u8 *block ) {
u32 v[16], m[16], i;
#define ROT(x,n) (((x)<<(32-n))|( (x)>>(n)))
#define G(a,b,c,d,e) \
v[a] += (m[sigma[i][e]] ^ cst[sigma[i][e+1]]) + v[b]; \
v[d] = ROT( v[d] ^ v[a],16); \
v[c] += v[d]; \
v[b] = ROT( v[b] ^ v[c],12); \
v[a] += (m[sigma[i][e+1]] ^ cst[sigma[i][e]])+v[b]; \
v[d] = ROT( v[d] ^ v[a], 8); \
v[c] += v[d]; \
v[b] = ROT( v[b] ^ v[c], 7);
for(i=0; i<16;++i) m[i] = U8TO32(block + i*4);
for(i=0; i< 8;++i) v[i] = S->h[i];
v[ 8] = S->s[0] ^ 0x243F6A88;
v[ 9] = S->s[1] ^ 0x85A308D3;
v[10] = S->s[2] ^ 0x13198A2E;
v[11] = S->s[3] ^ 0x03707344;
v[12] = 0xA4093822;
v[13] = 0x299F31D0;
v[14] = 0x082EFA98;
v[15] = 0xEC4E6C89;
if (S->nullt == 0) {
v[12] ^= S->t[0];
v[13] ^= S->t[0];
v[14] ^= S->t[1];
v[15] ^= S->t[1];
}
for(i=0; i<10; ++i) {
G( 0, 4, 8,12, 0);
G( 1, 5, 9,13, 2);
G( 2, 6,10,14, 4);
G( 3, 7,11,15, 6);
G( 3, 4, 9,14,14);
G( 2, 7, 8,13,12);
G( 0, 5,10,15, 8);
G( 1, 6,11,12,10);
}
for(i=0; i<16;++i) S->h[i%8] ^= v[i];
for(i=0; i<8 ;++i) S->h[i] ^= S->s[i%4];
}
void blake256_compress( state *S, const u8 *block )
{
u32 m0;
u32 m1;
u32 m2;
u32 m3;
u32 m4;
u32 m5;
u32 m6;
u32 m7;
u32 m8;
u32 m9;
u32 m10;
u32 m11;
u32 m12;
u32 m13;
u32 m14;
u32 m15;
u32 v0;
u32 v1;
u32 v2;
u32 v3;
u32 v4;
u32 v5;
u32 v6;
u32 v7;
u32 v8;
u32 v9;
u32 v10;
u32 v11;
u32 v12;
u32 v13;
u32 v14;
u32 v15;
m0 = U8TO32(block + 0);
m1 = U8TO32(block + 4);
m2 = U8TO32(block + 8);
m3 = U8TO32(block + 12);
m4 = U8TO32(block + 16);
m5 = U8TO32(block + 20);
m6 = U8TO32(block + 24);
m7 = U8TO32(block + 28);
m8 = U8TO32(block + 32);
m9 = U8TO32(block + 36);
m10 = U8TO32(block + 40);
m11 = U8TO32(block + 44);
m12 = U8TO32(block + 48);
m13 = U8TO32(block + 52);
m14 = U8TO32(block + 56);
m15 = U8TO32(block + 60);
v0 = S->h[0];
v1 = S->h[1];
v2 = S->h[2];
v3 = S->h[3];
v4 = S->h[4];
v5 = S->h[5];
v6 = S->h[6];
v7 = S->h[7];
v8 = S->s[0] ^ 0x243F6A88;
v9 = S->s[1] ^ 0x85A308D3;
v10 = S->s[2] ^ 0x13198A2E;
v11 = S->s[3] ^ 0x03707344;
v12 = 0xA4093822;
v13 = 0x299F31D0;
v14 = 0x082EFA98;
v15 = 0xEC4E6C89;
if (S->nullt == 0) {
v12 ^= S->t[0];
v13 ^= S->t[0];
v14 ^= S->t[1];
v15 ^= S->t[1];
}
#define ROUND(m0,c0,m1,c1,m2,c2,m3,c3,m4,c4,m5,c5,m6,c6,m7,c7,m8,c8,m9,c9,m10,c10,m11,c11,m12,c12,m13,c13,m14,c14,m15,c15) \
v0 += m0 ^ c0; \
v0 += v4; \
v12 ^= v0; \
v12 = ROT( v12,16); \
v8 += v12; \
v4 ^= v8; \
v4 = ROT( v4,12); \
v1 += m2 ^ c2; \
v1 += v5; \
v13 ^= v1; \
v13 = ROT( v13,16); \
v9 += v13; \
v5 ^= v9; \
v5 = ROT( v5,12); \
v2 += m4 ^ c4; \
v2 += v6; \
v14 ^= v2; \
v14 = ROT( v14,16); \
v10 += v14; \
v6 ^= v10; \
v6 = ROT( v6,12); \
v3 += m6 ^ c6; \
v3 += v7; \
v15 ^= v3; \
v15 = ROT( v15,16); \
v11 += v15; \
v7 ^= v11; \
v7 = ROT( v7,12); \
v2 += m5 ^ c5; \
v2 += v6; \
v14 ^= v2; \
v14 = ROT( v14, 8); \
v10 += v14; \
v6 ^= v10; \
v6 = ROT( v6, 7); \
v3 += m7 ^ c7; \
v3 += v7; \
v15 ^= v3; \
v15 = ROT( v15, 8); \
v11 += v15; \
v7 ^= v11; \
v7 = ROT( v7, 7); \
v1 += m3 ^ c3; \
v1 += v5; \
v13 ^= v1; \
v13 = ROT( v13, 8); \
v9 += v13; \
v5 ^= v9; \
v5 = ROT( v5, 7); \
v0 += m1 ^ c1; \
v0 += v4; \
v12 ^= v0; \
v12 = ROT( v12, 8); \
v8 += v12; \
v4 ^= v8; \
v4 = ROT( v4, 7); \
v0 += m8 ^ c8; \
v0 += v5; \
v15 ^= v0; \
v15 = ROT( v15,16); \
v10 += v15; \
v5 ^= v10; \
v5 = ROT( v5,12); \
v1 += m10 ^ c10; \
v1 += v6; \
v12 ^= v1; \
v12 = ROT( v12,16); \
v11 += v12; \
v6 ^= v11; \
v6 = ROT( v6,12); \
v2 += m12 ^ c12; \
v2 += v7; \
v13 ^= v2; \
v13 = ROT( v13,16); \
v8 += v13; \
v7 ^= v8; \
v7 = ROT( v7,12); \
v3 += m14 ^ c14; \
v3 += v4; \
v14 ^= v3; \
v14 = ROT( v14,16); \
v9 += v14; \
v4 ^= v9; \
v4 = ROT( v4,12); \
v2 += m13 ^ c13; \
v2 += v7; \
v13 ^= v2; \
v13 = ROT( v13, 8); \
v8 += v13; \
v7 ^= v8; \
v7 = ROT( v7, 7); \
v3 += m15 ^ c15; \
v3 += v4; \
v14 ^= v3; \
v14 = ROT( v14, 8); \
v9 += v14; \
v4 ^= v9; \
v4 = ROT( v4, 7); \
v1 += m11 ^ c11; \
v1 += v6; \
v12 ^= v1; \
v12 = ROT( v12, 8); \
v11 += v12; \
v6 ^= v11; \
v6 = ROT( v6, 7); \
v0 += m9 ^ c9; \
v0 += v5; \
v15 ^= v0; \
v15 = ROT( v15, 8); \
v10 += v15; \
v5 ^= v10; \
v5 = ROT( v5, 7); \
ROUND(m0,cst[1],m1,cst[0],m2,cst[3],m3,cst[2],m4,cst[5],m5,cst[4],m6,cst[7],m7,cst[6],m8,cst[9],m9,cst[8],m10,cst[11],m11,cst[10],m12,cst[13],m13,cst[12],m14,cst[15],m15,cst[14])
ROUND(m14,cst[10],m10,cst[14],m4,cst[8],m8,cst[4],m9,cst[15],m15,cst[9],m13,cst[6],m6,cst[13],m1,cst[12],m12,cst[1],m0,cst[2],m2,cst[0],m11,cst[7],m7,cst[11],m5,cst[3],m3,cst[5])
ROUND(m11,cst[8],m8,cst[11],m12,cst[0],m0,cst[12],m5,cst[2],m2,cst[5],m15,cst[13],m13,cst[15],m10,cst[14],m14,cst[10],m3,cst[6],m6,cst[3],m7,cst[1],m1,cst[7],m9,cst[4],m4,cst[9])
ROUND(m7,cst[9],m9,cst[7],m3,cst[1],m1,cst[3],m13,cst[12],m12,cst[13],m11,cst[14],m14,cst[11],m2,cst[6],m6,cst[2],m5,cst[10],m10,cst[5],m4,cst[0],m0,cst[4],m15,cst[8],m8,cst[15])
ROUND(m9,cst[0],m0,cst[9],m5,cst[7],m7,cst[5],m2,cst[4],m4,cst[2],m10,cst[15],m15,cst[10],m14,cst[1],m1,cst[14],m11,cst[12],m12,cst[11],m6,cst[8],m8,cst[6],m3,cst[13],m13,cst[3])
ROUND(m2,cst[12],m12,cst[2],m6,cst[10],m10,cst[6],m0,cst[11],m11,cst[0],m8,cst[3],m3,cst[8],m4,cst[13],m13,cst[4],m7,cst[5],m5,cst[7],m15,cst[14],m14,cst[15],m1,cst[9],m9,cst[1])
ROUND(m12,cst[5],m5,cst[12],m1,cst[15],m15,cst[1],m14,cst[13],m13,cst[14],m4,cst[10],m10,cst[4],m0,cst[7],m7,cst[0],m6,cst[3],m3,cst[6],m9,cst[2],m2,cst[9],m8,cst[11],m11,cst[8])
ROUND(m13,cst[11],m11,cst[13],m7,cst[14],m14,cst[7],m12,cst[1],m1,cst[12],m3,cst[9],m9,cst[3],m5,cst[0],m0,cst[5],m15,cst[4],m4,cst[15],m8,cst[6],m6,cst[8],m2,cst[10],m10,cst[2])
ROUND(m6,cst[15],m15,cst[6],m14,cst[9],m9,cst[14],m11,cst[3],m3,cst[11],m0,cst[8],m8,cst[0],m12,cst[2],m2,cst[12],m13,cst[7],m7,cst[13],m1,cst[4],m4,cst[1],m10,cst[5],m5,cst[10])
ROUND(m10,cst[2],m2,cst[10],m8,cst[4],m4,cst[8],m7,cst[6],m6,cst[7],m1,cst[5],m5,cst[1],m15,cst[11],m11,cst[15],m9,cst[14],m14,cst[9],m3,cst[12],m12,cst[3],m13,cst[0],m0,cst[13])
ROUND(m0,cst[1],m1,cst[0],m2,cst[3],m3,cst[2],m4,cst[5],m5,cst[4],m6,cst[7],m7,cst[6],m8,cst[9],m9,cst[8],m10,cst[11],m11,cst[10],m12,cst[13],m13,cst[12],m14,cst[15],m15,cst[14])
ROUND(m14,cst[10],m10,cst[14],m4,cst[8],m8,cst[4],m9,cst[15],m15,cst[9],m13,cst[6],m6,cst[13],m1,cst[12],m12,cst[1],m0,cst[2],m2,cst[0],m11,cst[7],m7,cst[11],m5,cst[3],m3,cst[5])
ROUND(m11,cst[8],m8,cst[11],m12,cst[0],m0,cst[12],m5,cst[2],m2,cst[5],m15,cst[13],m13,cst[15],m10,cst[14],m14,cst[10],m3,cst[6],m6,cst[3],m7,cst[1],m1,cst[7],m9,cst[4],m4,cst[9])
ROUND(m7,cst[9],m9,cst[7],m3,cst[1],m1,cst[3],m13,cst[12],m12,cst[13],m11,cst[14],m14,cst[11],m2,cst[6],m6,cst[2],m5,cst[10],m10,cst[5],m4,cst[0],m0,cst[4],m15,cst[8],m8,cst[15])
v0 ^= v8;
v1 ^= v9;
v2 ^= v10;
v3 ^= v11;
v4 ^= v12;
v5 ^= v13;
v6 ^= v14;
v7 ^= v15;
v0 ^= S->s[0];
v1 ^= S->s[1];
v2 ^= S->s[2];
v3 ^= S->s[3];
v4 ^= S->s[0];
v5 ^= S->s[1];
v6 ^= S->s[2];
v7 ^= S->s[3];
S->h[0] ^= v0;
S->h[1] ^= v1;
S->h[2] ^= v2;
S->h[3] ^= v3;
S->h[4] ^= v4;
S->h[5] ^= v5;
S->h[6] ^= v6;
S->h[7] ^= v7;
}
