/*internal static void XorLittleEndian32(byte[] buf, int offset, UInt32 value)
* {
* buf[offset + 0] ^= (byte)value;
* buf[offset + 1] ^= (byte)(value >> 8);
* buf[offset + 2] ^= (byte)(value >> 16);
* buf[offset + 3] ^= (byte)(value >> 24);
* }*/
/*internal static void XorLittleEndian32(byte[] output, int outputOffset, byte[] input, int inputOffset, UInt32 value)
* {
* output[outputOffset + 0] = (byte)(input[inputOffset + 0] ^ value);
* output[outputOffset + 1] = (byte)(input[inputOffset + 1] ^ (value >> 8));
* output[outputOffset + 2] = (byte)(input[inputOffset + 2] ^ (value >> 16));
* output[outputOffset + 3] = (byte)(input[inputOffset + 3] ^ (value >> 24));
* }*/
#endregion
#region Array8
internal static void Array8LoadLittleEndian32(out Array8 <UInt32> output, byte[] input, int inputOffset)
{
output.x0 = LoadLittleEndian32(input, inputOffset + 0);
output.x1 = LoadLittleEndian32(input, inputOffset + 4);
output.x2 = LoadLittleEndian32(input, inputOffset + 8);
output.x3 = LoadLittleEndian32(input, inputOffset + 12);
output.x4 = LoadLittleEndian32(input, inputOffset + 16);
output.x5 = LoadLittleEndian32(input, inputOffset + 20);
output.x6 = LoadLittleEndian32(input, inputOffset + 24);
output.x7 = LoadLittleEndian32(input, inputOffset + 28);
}
internal static void Sha512Init(out Array8 <UInt64> state)
{
state.x0 = 0x6a09e667f3bcc908;
state.x1 = 0xbb67ae8584caa73b;
state.x2 = 0x3c6ef372fe94f82b;
state.x3 = 0xa54ff53a5f1d36f1;
state.x4 = 0x510e527fade682d1;
state.x5 = 0x9b05688c2b3e6c1f;
state.x6 = 0x1f83d9abfb41bd6b;
state.x7 = 0x5be0cd19137e2179;
}
// written by floodyberry (Andrew M.)
// original license: MIT or PUBLIC DOMAIN
// https://github.com/floodyberry/poly1305-donna/blob/master/poly1305-donna-unrolled.c
internal static void poly1305_auth(byte[] output, int outputOffset, byte[] m, int mStart, int mLength, ref Array8 <UInt32> key)
{
UInt32 t0, t1, t2, t3;
UInt32 h0, h1, h2, h3, h4;
UInt32 r0, r1, r2, r3, r4;
UInt32 s1, s2, s3, s4;
UInt32 b, nb;
int j;
UInt64 tt0, tt1, tt2, tt3, tt4;
UInt64 f0, f1, f2, f3;
UInt32 g0, g1, g2, g3, g4;
UInt64 c;
/* clamp key */
t0 = key.x0;
t1 = key.x1;
t2 = key.x2;
t3 = key.x3;
/* precompute multipliers */
r0 = t0 & 0x3ffffff; t0 >>= 26; t0 |= t1 << 6;
r1 = t0 & 0x3ffff03; t1 >>= 20; t1 |= t2 << 12;
r2 = t1 & 0x3ffc0ff; t2 >>= 14; t2 |= t3 << 18;
r3 = t2 & 0x3f03fff; t3 >>= 8;
r4 = t3 & 0x00fffff;
s1 = r1 * 5;
s2 = r2 * 5;
s3 = r3 * 5;
s4 = r4 * 5;
/* init state */
h0 = 0;
h1 = 0;
h2 = 0;
h3 = 0;
h4 = 0;
/* full blocks */
if (mLength < 16)
{
goto poly1305_donna_atmost15bytes;
}
poly1305_donna_16bytes:
mStart += 16;
mLength -= 16;
t0 = ByteIntegerConverter.LoadLittleEndian32(m, mStart - 16);
t1 = ByteIntegerConverter.LoadLittleEndian32(m, mStart - 12);
t2 = ByteIntegerConverter.LoadLittleEndian32(m, mStart - 8);
t3 = ByteIntegerConverter.LoadLittleEndian32(m, mStart - 4);
//todo: looks like these can be simplified a bit
h0 += t0 & 0x3ffffff;
h1 += (uint)(((((UInt64)t1 << 32) | t0) >> 26) & 0x3ffffff);
h2 += (uint)(((((UInt64)t2 << 32) | t1) >> 20) & 0x3ffffff);
h3 += (uint)(((((UInt64)t3 << 32) | t2) >> 14) & 0x3ffffff);
h4 += (t3 >> 8) | (1 << 24);
poly1305_donna_mul:
tt0 = (ulong)h0 * r0 + (ulong)h1 * s4 + (ulong)h2 * s3 + (ulong)h3 * s2 + (ulong)h4 * s1;
tt1 = (ulong)h0 * r1 + (ulong)h1 * r0 + (ulong)h2 * s4 + (ulong)h3 * s3 + (ulong)h4 * s2;
tt2 = (ulong)h0 * r2 + (ulong)h1 * r1 + (ulong)h2 * r0 + (ulong)h3 * s4 + (ulong)h4 * s3;
tt3 = (ulong)h0 * r3 + (ulong)h1 * r2 + (ulong)h2 * r1 + (ulong)h3 * r0 + (ulong)h4 * s4;
tt4 = (ulong)h0 * r4 + (ulong)h1 * r3 + (ulong)h2 * r2 + (ulong)h3 * r1 + (ulong)h4 * r0;
unchecked
{
h0 = (UInt32)tt0 & 0x3ffffff; c = (tt0 >> 26);
tt1 += c; h1 = (UInt32)tt1 & 0x3ffffff; b = (UInt32)(tt1 >> 26);
tt2 += b; h2 = (UInt32)tt2 & 0x3ffffff; b = (UInt32)(tt2 >> 26);
tt3 += b; h3 = (UInt32)tt3 & 0x3ffffff; b = (UInt32)(tt3 >> 26);
tt4 += b; h4 = (UInt32)tt4 & 0x3ffffff; b = (UInt32)(tt4 >> 26);
}
h0 += b * 5;
if (mLength >= 16)
{
goto poly1305_donna_16bytes;
}
/* final bytes */
poly1305_donna_atmost15bytes:
if (mLength == 0)
{
goto poly1305_donna_finish;
}
byte[] mp = new byte[16];//todo remove allocation
for (j = 0; j < mLength; j++)
{
mp[j] = m[mStart + j];
}
mp[j++] = 1;
for (; j < 16; j++)
{
mp[j] = 0;
}
mLength = 0;
t0 = ByteIntegerConverter.LoadLittleEndian32(mp, 0);
t1 = ByteIntegerConverter.LoadLittleEndian32(mp, 4);
t2 = ByteIntegerConverter.LoadLittleEndian32(mp, 8);
t3 = ByteIntegerConverter.LoadLittleEndian32(mp, 12);
CryptoBytes.Wipe(mp);
h0 += t0 & 0x3ffffff;
h1 += (uint)(((((UInt64)t1 << 32) | t0) >> 26) & 0x3ffffff);
h2 += (uint)(((((UInt64)t2 << 32) | t1) >> 20) & 0x3ffffff);
h3 += (uint)(((((UInt64)t3 << 32) | t2) >> 14) & 0x3ffffff);
h4 += t3 >> 8;
goto poly1305_donna_mul;
poly1305_donna_finish:
b = h0 >> 26; h0 = h0 & 0x3ffffff;
h1 += b; b = h1 >> 26; h1 = h1 & 0x3ffffff;
h2 += b; b = h2 >> 26; h2 = h2 & 0x3ffffff;
h3 += b; b = h3 >> 26; h3 = h3 & 0x3ffffff;
h4 += b; b = h4 >> 26; h4 = h4 & 0x3ffffff;
h0 += b * 5;
g0 = h0 + 5; b = g0 >> 26; g0 &= 0x3ffffff;
g1 = h1 + b; b = g1 >> 26; g1 &= 0x3ffffff;
g2 = h2 + b; b = g2 >> 26; g2 &= 0x3ffffff;
g3 = h3 + b; b = g3 >> 26; g3 &= 0x3ffffff;
g4 = unchecked (h4 + b - (1 << 26));
b = (g4 >> 31) - 1;
nb = ~b;
h0 = (h0 & nb) | (g0 & b);
h1 = (h1 & nb) | (g1 & b);
h2 = (h2 & nb) | (g2 & b);
h3 = (h3 & nb) | (g3 & b);
h4 = (h4 & nb) | (g4 & b);
f0 = ((h0) | (h1 << 26)) + (UInt64)key.x4;
f1 = ((h1 >> 6) | (h2 << 20)) + (UInt64)key.x5;
f2 = ((h2 >> 12) | (h3 << 14)) + (UInt64)key.x6;
f3 = ((h3 >> 18) | (h4 << 8)) + (UInt64)key.x7;
unchecked
{
ByteIntegerConverter.StoreLittleEndian32(output, outputOffset + 0, (uint)f0); f1 += (f0 >> 32);
ByteIntegerConverter.StoreLittleEndian32(output, outputOffset + 4, (uint)f1); f2 += (f1 >> 32);
ByteIntegerConverter.StoreLittleEndian32(output, outputOffset + 8, (uint)f2); f3 += (f2 >> 32);
ByteIntegerConverter.StoreLittleEndian32(output, outputOffset + 12, (uint)f3);
}
}
internal static void Core(out Array8 <UInt64> outputState, ref Array8 <UInt64> inputState, ref Array16 <UInt64> input)
{
unchecked
{
UInt64 a = inputState.x0;
UInt64 b = inputState.x1;
UInt64 c = inputState.x2;
UInt64 d = inputState.x3;
UInt64 e = inputState.x4;
UInt64 f = inputState.x5;
UInt64 g = inputState.x6;
UInt64 h = inputState.x7;
UInt64 w0 = input.x0;
UInt64 w1 = input.x1;
UInt64 w2 = input.x2;
UInt64 w3 = input.x3;
UInt64 w4 = input.x4;
UInt64 w5 = input.x5;
UInt64 w6 = input.x6;
UInt64 w7 = input.x7;
UInt64 w8 = input.x8;
UInt64 w9 = input.x9;
UInt64 w10 = input.x10;
UInt64 w11 = input.x11;
UInt64 w12 = input.x12;
UInt64 w13 = input.x13;
UInt64 w14 = input.x14;
UInt64 w15 = input.x15;
int t = 0;
while (true)
{
ulong t1, t2;
{//0
t1 = h +
((e >> 14) ^ (e << (64 - 14)) ^ (e >> 18) ^ (e << (64 - 18)) ^ (e >> 41) ^ (e << (64 - 41))) +
//Sigma1(e)
((e & f) ^ (~e & g)) + //Ch(e,f,g)
K[t] + w0;
t2 = ((a >> 28) ^ (a << (64 - 28)) ^ (a >> 34) ^ (a << (64 - 34)) ^ (a >> 39) ^ (a << (64 - 39))) +
//Sigma0(a)
((a & b) ^ (a & c) ^ (b & c)); //Maj(a,b,c)
h = g;
g = f;
f = e;
e = d + t1;
d = c;
c = b;
b = a;
a = t1 + t2;
t++;
}
{//1
t1 = h +
((e >> 14) ^ (e << (64 - 14)) ^ (e >> 18) ^ (e << (64 - 18)) ^ (e >> 41) ^ (e << (64 - 41))) +
//Sigma1(e)
((e & f) ^ (~e & g)) + //Ch(e,f,g)
K[t] + w1;
t2 = ((a >> 28) ^ (a << (64 - 28)) ^ (a >> 34) ^ (a << (64 - 34)) ^ (a >> 39) ^ (a << (64 - 39))) +
//Sigma0(a)
((a & b) ^ (a & c) ^ (b & c)); //Maj(a,b,c)
h = g;
g = f;
f = e;
e = d + t1;
d = c;
c = b;
b = a;
a = t1 + t2;
t++;
}
{//2
t1 = h +
((e >> 14) ^ (e << (64 - 14)) ^ (e >> 18) ^ (e << (64 - 18)) ^ (e >> 41) ^ (e << (64 - 41))) +
//Sigma1(e)
((e & f) ^ (~e & g)) + //Ch(e,f,g)
K[t] + w2;
t2 = ((a >> 28) ^ (a << (64 - 28)) ^ (a >> 34) ^ (a << (64 - 34)) ^ (a >> 39) ^ (a << (64 - 39))) +
//Sigma0(a)
((a & b) ^ (a & c) ^ (b & c)); //Maj(a,b,c)
h = g;
g = f;
f = e;
e = d + t1;
d = c;
c = b;
b = a;
a = t1 + t2;
t++;
}
{//3
t1 = h +
((e >> 14) ^ (e << (64 - 14)) ^ (e >> 18) ^ (e << (64 - 18)) ^ (e >> 41) ^ (e << (64 - 41))) +
//Sigma1(e)
((e & f) ^ (~e & g)) + //Ch(e,f,g)
K[t] + w3;
t2 = ((a >> 28) ^ (a << (64 - 28)) ^ (a >> 34) ^ (a << (64 - 34)) ^ (a >> 39) ^ (a << (64 - 39))) +
//Sigma0(a)
((a & b) ^ (a & c) ^ (b & c)); //Maj(a,b,c)
h = g;
g = f;
f = e;
e = d + t1;
d = c;
c = b;
b = a;
a = t1 + t2;
t++;
}
{//4
t1 = h +
((e >> 14) ^ (e << (64 - 14)) ^ (e >> 18) ^ (e << (64 - 18)) ^ (e >> 41) ^ (e << (64 - 41))) +
//Sigma1(e)
((e & f) ^ (~e & g)) + //Ch(e,f,g)
K[t] + w4;
t2 = ((a >> 28) ^ (a << (64 - 28)) ^ (a >> 34) ^ (a << (64 - 34)) ^ (a >> 39) ^ (a << (64 - 39))) +
//Sigma0(a)
((a & b) ^ (a & c) ^ (b & c)); //Maj(a,b,c)
h = g;
g = f;
f = e;
e = d + t1;
d = c;
c = b;
b = a;
a = t1 + t2;
t++;
}
{//5
t1 = h +
((e >> 14) ^ (e << (64 - 14)) ^ (e >> 18) ^ (e << (64 - 18)) ^ (e >> 41) ^ (e << (64 - 41))) +
//Sigma1(e)
((e & f) ^ (~e & g)) + //Ch(e,f,g)
K[t] + w5;
t2 = ((a >> 28) ^ (a << (64 - 28)) ^ (a >> 34) ^ (a << (64 - 34)) ^ (a >> 39) ^ (a << (64 - 39))) +
//Sigma0(a)
((a & b) ^ (a & c) ^ (b & c)); //Maj(a,b,c)
h = g;
g = f;
f = e;
e = d + t1;
d = c;
c = b;
b = a;
a = t1 + t2;
t++;
}
{//6
t1 = h +
((e >> 14) ^ (e << (64 - 14)) ^ (e >> 18) ^ (e << (64 - 18)) ^ (e >> 41) ^ (e << (64 - 41))) +
//Sigma1(e)
((e & f) ^ (~e & g)) + //Ch(e,f,g)
K[t] + w6;
t2 = ((a >> 28) ^ (a << (64 - 28)) ^ (a >> 34) ^ (a << (64 - 34)) ^ (a >> 39) ^ (a << (64 - 39))) +
//Sigma0(a)
((a & b) ^ (a & c) ^ (b & c)); //Maj(a,b,c)
h = g;
g = f;
f = e;
e = d + t1;
d = c;
c = b;
b = a;
a = t1 + t2;
t++;
}
{//7
t1 = h +
((e >> 14) ^ (e << (64 - 14)) ^ (e >> 18) ^ (e << (64 - 18)) ^ (e >> 41) ^ (e << (64 - 41))) +
//Sigma1(e)
((e & f) ^ (~e & g)) + //Ch(e,f,g)
K[t] + w7;
t2 = ((a >> 28) ^ (a << (64 - 28)) ^ (a >> 34) ^ (a << (64 - 34)) ^ (a >> 39) ^ (a << (64 - 39))) +
//Sigma0(a)
((a & b) ^ (a & c) ^ (b & c)); //Maj(a,b,c)
h = g;
g = f;
f = e;
e = d + t1;
d = c;
c = b;
b = a;
a = t1 + t2;
t++;
}
{//8
t1 = h +
((e >> 14) ^ (e << (64 - 14)) ^ (e >> 18) ^ (e << (64 - 18)) ^ (e >> 41) ^ (e << (64 - 41))) +
//Sigma1(e)
((e & f) ^ (~e & g)) + //Ch(e,f,g)
K[t] + w8;
t2 = ((a >> 28) ^ (a << (64 - 28)) ^ (a >> 34) ^ (a << (64 - 34)) ^ (a >> 39) ^ (a << (64 - 39))) +
//Sigma0(a)
((a & b) ^ (a & c) ^ (b & c)); //Maj(a,b,c)
h = g;
g = f;
f = e;
e = d + t1;
d = c;
c = b;
b = a;
a = t1 + t2;
t++;
}
{//9
t1 = h +
((e >> 14) ^ (e << (64 - 14)) ^ (e >> 18) ^ (e << (64 - 18)) ^ (e >> 41) ^ (e << (64 - 41))) +
//Sigma1(e)
((e & f) ^ (~e & g)) + //Ch(e,f,g)
K[t] + w9;
t2 = ((a >> 28) ^ (a << (64 - 28)) ^ (a >> 34) ^ (a << (64 - 34)) ^ (a >> 39) ^ (a << (64 - 39))) +
//Sigma0(a)
((a & b) ^ (a & c) ^ (b & c)); //Maj(a,b,c)
h = g;
g = f;
f = e;
e = d + t1;
d = c;
c = b;
b = a;
a = t1 + t2;
t++;
}
{//10
t1 = h +
((e >> 14) ^ (e << (64 - 14)) ^ (e >> 18) ^ (e << (64 - 18)) ^ (e >> 41) ^ (e << (64 - 41))) +
//Sigma1(e)
((e & f) ^ (~e & g)) + //Ch(e,f,g)
K[t] + w10;
t2 = ((a >> 28) ^ (a << (64 - 28)) ^ (a >> 34) ^ (a << (64 - 34)) ^ (a >> 39) ^ (a << (64 - 39))) +
//Sigma0(a)
((a & b) ^ (a & c) ^ (b & c)); //Maj(a,b,c)
h = g;
g = f;
f = e;
e = d + t1;
d = c;
c = b;
b = a;
a = t1 + t2;
t++;
}
{//11
t1 = h +
((e >> 14) ^ (e << (64 - 14)) ^ (e >> 18) ^ (e << (64 - 18)) ^ (e >> 41) ^ (e << (64 - 41))) +
//Sigma1(e)
((e & f) ^ (~e & g)) + //Ch(e,f,g)
K[t] + w11;
t2 = ((a >> 28) ^ (a << (64 - 28)) ^ (a >> 34) ^ (a << (64 - 34)) ^ (a >> 39) ^ (a << (64 - 39))) +
//Sigma0(a)
((a & b) ^ (a & c) ^ (b & c)); //Maj(a,b,c)
h = g;
g = f;
f = e;
e = d + t1;
d = c;
c = b;
b = a;
a = t1 + t2;
t++;
}
{//12
t1 = h +
((e >> 14) ^ (e << (64 - 14)) ^ (e >> 18) ^ (e << (64 - 18)) ^ (e >> 41) ^ (e << (64 - 41))) +
//Sigma1(e)
((e & f) ^ (~e & g)) + //Ch(e,f,g)
K[t] + w12;
t2 = ((a >> 28) ^ (a << (64 - 28)) ^ (a >> 34) ^ (a << (64 - 34)) ^ (a >> 39) ^ (a << (64 - 39))) +
//Sigma0(a)
((a & b) ^ (a & c) ^ (b & c)); //Maj(a,b,c)
h = g;
g = f;
f = e;
e = d + t1;
d = c;
c = b;
b = a;
a = t1 + t2;
t++;
}
{//13
t1 = h +
((e >> 14) ^ (e << (64 - 14)) ^ (e >> 18) ^ (e << (64 - 18)) ^ (e >> 41) ^ (e << (64 - 41))) +
//Sigma1(e)
((e & f) ^ (~e & g)) + //Ch(e,f,g)
K[t] + w13;
t2 = ((a >> 28) ^ (a << (64 - 28)) ^ (a >> 34) ^ (a << (64 - 34)) ^ (a >> 39) ^ (a << (64 - 39))) +
//Sigma0(a)
((a & b) ^ (a & c) ^ (b & c)); //Maj(a,b,c)
h = g;
g = f;
f = e;
e = d + t1;
d = c;
c = b;
b = a;
a = t1 + t2;
t++;
}
{//14
t1 = h +
((e >> 14) ^ (e << (64 - 14)) ^ (e >> 18) ^ (e << (64 - 18)) ^ (e >> 41) ^ (e << (64 - 41))) +
//Sigma1(e)
((e & f) ^ (~e & g)) + //Ch(e,f,g)
K[t] + w14;
t2 = ((a >> 28) ^ (a << (64 - 28)) ^ (a >> 34) ^ (a << (64 - 34)) ^ (a >> 39) ^ (a << (64 - 39))) +
//Sigma0(a)
((a & b) ^ (a & c) ^ (b & c)); //Maj(a,b,c)
h = g;
g = f;
f = e;
e = d + t1;
d = c;
c = b;
b = a;
a = t1 + t2;
t++;
}
{//15
t1 = h +
((e >> 14) ^ (e << (64 - 14)) ^ (e >> 18) ^ (e << (64 - 18)) ^ (e >> 41) ^ (e << (64 - 41))) +
//Sigma1(e)
((e & f) ^ (~e & g)) + //Ch(e,f,g)
K[t] + w15;
t2 = ((a >> 28) ^ (a << (64 - 28)) ^ (a >> 34) ^ (a << (64 - 34)) ^ (a >> 39) ^ (a << (64 - 39))) +
//Sigma0(a)
((a & b) ^ (a & c) ^ (b & c)); //Maj(a,b,c)
h = g;
g = f;
f = e;
e = d + t1;
d = c;
c = b;
b = a;
a = t1 + t2;
t++;
}
if (t == 80)
{
break;
}
w0 += ((w14 >> 19) ^ (w14 << (64 - 19)) ^ (w14 >> 61) ^ (w14 << (64 - 61)) ^ (w14 >> 6)) +
w9 +
((w1 >> 1) ^ (w1 << (64 - 1)) ^ (w1 >> 8) ^ (w1 << (64 - 8)) ^ (w1 >> 7));
w1 += ((w15 >> 19) ^ (w15 << (64 - 19)) ^ (w15 >> 61) ^ (w15 << (64 - 61)) ^ (w15 >> 6)) +
w10 +
((w2 >> 1) ^ (w2 << (64 - 1)) ^ (w2 >> 8) ^ (w2 << (64 - 8)) ^ (w2 >> 7));
w2 += ((w0 >> 19) ^ (w0 << (64 - 19)) ^ (w0 >> 61) ^ (w0 << (64 - 61)) ^ (w0 >> 6)) +
w11 +
((w3 >> 1) ^ (w3 << (64 - 1)) ^ (w3 >> 8) ^ (w3 << (64 - 8)) ^ (w3 >> 7));
w3 += ((w1 >> 19) ^ (w1 << (64 - 19)) ^ (w1 >> 61) ^ (w1 << (64 - 61)) ^ (w1 >> 6)) +
w12 +
((w4 >> 1) ^ (w4 << (64 - 1)) ^ (w4 >> 8) ^ (w4 << (64 - 8)) ^ (w4 >> 7));
w4 += ((w2 >> 19) ^ (w2 << (64 - 19)) ^ (w2 >> 61) ^ (w2 << (64 - 61)) ^ (w2 >> 6)) +
w13 +
((w5 >> 1) ^ (w5 << (64 - 1)) ^ (w5 >> 8) ^ (w5 << (64 - 8)) ^ (w5 >> 7));
w5 += ((w3 >> 19) ^ (w3 << (64 - 19)) ^ (w3 >> 61) ^ (w3 << (64 - 61)) ^ (w3 >> 6)) +
w14 +
((w6 >> 1) ^ (w6 << (64 - 1)) ^ (w6 >> 8) ^ (w6 << (64 - 8)) ^ (w6 >> 7));
w6 += ((w4 >> 19) ^ (w4 << (64 - 19)) ^ (w4 >> 61) ^ (w4 << (64 - 61)) ^ (w4 >> 6)) +
w15 +
((w7 >> 1) ^ (w7 << (64 - 1)) ^ (w7 >> 8) ^ (w7 << (64 - 8)) ^ (w7 >> 7));
w7 += ((w5 >> 19) ^ (w5 << (64 - 19)) ^ (w5 >> 61) ^ (w5 << (64 - 61)) ^ (w5 >> 6)) +
w0 +
((w8 >> 1) ^ (w8 << (64 - 1)) ^ (w8 >> 8) ^ (w8 << (64 - 8)) ^ (w8 >> 7));
w8 += ((w6 >> 19) ^ (w6 << (64 - 19)) ^ (w6 >> 61) ^ (w6 << (64 - 61)) ^ (w6 >> 6)) +
w1 +
((w9 >> 1) ^ (w9 << (64 - 1)) ^ (w9 >> 8) ^ (w9 << (64 - 8)) ^ (w9 >> 7));
w9 += ((w7 >> 19) ^ (w7 << (64 - 19)) ^ (w7 >> 61) ^ (w7 << (64 - 61)) ^ (w7 >> 6)) +
w2 +
((w10 >> 1) ^ (w10 << (64 - 1)) ^ (w10 >> 8) ^ (w10 << (64 - 8)) ^ (w10 >> 7));
w10 += ((w8 >> 19) ^ (w8 << (64 - 19)) ^ (w8 >> 61) ^ (w8 << (64 - 61)) ^ (w8 >> 6)) +
w3 +
((w11 >> 1) ^ (w11 << (64 - 1)) ^ (w11 >> 8) ^ (w11 << (64 - 8)) ^ (w11 >> 7));
w11 += ((w9 >> 19) ^ (w9 << (64 - 19)) ^ (w9 >> 61) ^ (w9 << (64 - 61)) ^ (w9 >> 6)) +
w4 +
((w12 >> 1) ^ (w12 << (64 - 1)) ^ (w12 >> 8) ^ (w12 << (64 - 8)) ^ (w12 >> 7));
w12 += ((w10 >> 19) ^ (w10 << (64 - 19)) ^ (w10 >> 61) ^ (w10 << (64 - 61)) ^ (w10 >> 6)) +
w5 +
((w13 >> 1) ^ (w13 << (64 - 1)) ^ (w13 >> 8) ^ (w13 << (64 - 8)) ^ (w13 >> 7));
w13 += ((w11 >> 19) ^ (w11 << (64 - 19)) ^ (w11 >> 61) ^ (w11 << (64 - 61)) ^ (w11 >> 6)) +
w6 +
((w14 >> 1) ^ (w14 << (64 - 1)) ^ (w14 >> 8) ^ (w14 << (64 - 8)) ^ (w14 >> 7));
w14 += ((w12 >> 19) ^ (w12 << (64 - 19)) ^ (w12 >> 61) ^ (w12 << (64 - 61)) ^ (w12 >> 6)) +
w7 +
((w15 >> 1) ^ (w15 << (64 - 1)) ^ (w15 >> 8) ^ (w15 << (64 - 8)) ^ (w15 >> 7));
w15 += ((w13 >> 19) ^ (w13 << (64 - 19)) ^ (w13 >> 61) ^ (w13 << (64 - 61)) ^ (w13 >> 6)) +
w8 +
((w0 >> 1) ^ (w0 << (64 - 1)) ^ (w0 >> 8) ^ (w0 << (64 - 8)) ^ (w0 >> 7));
}
outputState.x0 = inputState.x0 + a;
outputState.x1 = inputState.x1 + b;
outputState.x2 = inputState.x2 + c;
outputState.x3 = inputState.x3 + d;
outputState.x4 = inputState.x4 + e;
outputState.x5 = inputState.x5 + f;
outputState.x6 = inputState.x6 + g;
outputState.x7 = inputState.x7 + h;
}
}