#version 300 es

precision highp int;
precision highp float;

uniform uint in_w5;
uniform uint in_step;

out uvec4 out_v;


// SHA256 utils
#define Ch(x, y, z)   ((x & (y ^ z)) ^ z)
#define Maj(x, y, z)  ((x & (y | z)) | (y & z))
#define SHR(x, n)     (x >> n)
#define ROTR(x, n)    ((x >> n) | (x << (32 - n)))

#define S0(x)         (ROTR(x,  2) ^ ROTR(x, 13) ^ ROTR(x, 22))
#define S1(x)         (ROTR(x,  6) ^ ROTR(x, 11) ^ ROTR(x, 25))
#define s0(x)         (ROTR(x,  7) ^ ROTR(x, 18) ^ SHR(x, 3))
#define s1(x)         (ROTR(x, 17) ^ ROTR(x, 19) ^ SHR(x, 10))

/* SHA256 round function */
#define RND(a, b, c, d, e, f, g, h, k)  \
  t0 = h + S1(e) + Ch(e, f, g) + k;     \
  t1 = S0(a) + Maj(a, b, c);            \
  d += t0;                              \
  h  = t0 + t1;


const uint K[64] = uint[64] (
  0x428A2F98u, 0x71374491u, 0xB5C0FBCFu, 0xE9B5DBA5u,
  0x3956C25Bu, 0x59F111F1u, 0x923F82A4u, 0xAB1C5ED5u,
  0xD807AA98u, 0x12835B01u, 0x243185BEu, 0x550C7DC3u,
  0x72BE5D74u, 0x80DEB1FEu, 0x9BDC06A7u, 0xC19BF174u,
  0xE49B69C1u, 0xEFBE4786u, 0x0FC19DC6u, 0x240CA1CCu,
  0x2DE92C6Fu, 0x4A7484AAu, 0x5CB0A9DCu, 0x76F988DAu,
  0x983E5152u, 0xA831C66Du, 0xB00327C8u, 0xBF597FC7u,
  0xC6E00BF3u, 0xD5A79147u, 0x06CA6351u, 0x14292967u,
  0x27B70A85u, 0x2E1B2138u, 0x4D2C6DFCu, 0x53380D13u,
  0x650A7354u, 0x766A0ABBu, 0x81C2C92Eu, 0x92722C85u,
  0xA2BFE8A1u, 0xA81A664Bu, 0xC24B8B70u, 0xC76C51A3u,
  0xD192E819u, 0xD6990624u, 0xF40E3585u, 0x106AA070u,
  0x19A4C116u, 0x1E376C08u, 0x2748774Cu, 0x34B0BCB5u,
  0x391C0CB3u, 0x4ED8AA4Au, 0x5B9CCA4Fu, 0x682E6FF3u,
  0x748F82EEu, 0x78A5636Fu, 0x84C87814u, 0x8CC70208u,
  0x90BEFFFAu, 0xA4506CEBu, 0xBEF9A3F7u, 0xC67178F2u
);

void main() {
  uint x = uint(gl_FragCoord.x);
  uint y = uint(gl_FragCoord.y);
  uint thread_id = y * __TEX_W__u + x;

  uint W[16], S[8];
  uint t0, t1;

  for (uint i = in_step; i > 0u; i--) {
    S[0] = 0x6A09E667u;
    S[1] = 0xBB67AE85u;
    S[2] = 0x3C6EF372u;
    S[3] = 0xA54FF53Au;
    S[4] = 0x510E527Fu;
    S[5] = 0x9B05688Cu;
    S[6] = 0x1F83D9ABu;
    S[7] = 0x5BE0CD19u;

    // challenge
    W[0] = __W0__u;
    W[1] = __W1__u;
    W[2] = __W2__u;
    W[3] = __W3__u;

    // nonce
    W[4] = __W4__u;
    W[5] = in_w5;
    W[6] = thread_id;
    W[7] = i;

    // padding
    W[8] = 0x80000000u; // 1 << 31
    W[9] = 0u;
    W[10] = 0u;
    W[11] = 0u;
    W[12] = 0u;
    W[13] = 0u;
    W[14] = 0u;
    W[15] = 256u;       // input bit len

    RND(S[0], S[1], S[2], S[3], S[4], S[5], S[6], S[7], W[ 0] + K[0])
    RND(S[7], S[0], S[1], S[2], S[3], S[4], S[5], S[6], W[ 1] + K[1])
    RND(S[6], S[7], S[0], S[1], S[2], S[3], S[4], S[5], W[ 2] + K[2])
    RND(S[5], S[6], S[7], S[0], S[1], S[2], S[3], S[4], W[ 3] + K[3])
    RND(S[4], S[5], S[6], S[7], S[0], S[1], S[2], S[3], W[ 4] + K[4])
    RND(S[3], S[4], S[5], S[6], S[7], S[0], S[1], S[2], W[ 5] + K[5])
    RND(S[2], S[3], S[4], S[5], S[6], S[7], S[0], S[1], W[ 6] + K[6])
    RND(S[1], S[2], S[3], S[4], S[5], S[6], S[7], S[0], W[ 7] + K[7])
    RND(S[0], S[1], S[2], S[3], S[4], S[5], S[6], S[7], W[ 8] + K[8])
    RND(S[7], S[0], S[1], S[2], S[3], S[4], S[5], S[6], W[ 9] + K[9])
    RND(S[6], S[7], S[0], S[1], S[2], S[3], S[4], S[5], W[10] + K[10])
    RND(S[5], S[6], S[7], S[0], S[1], S[2], S[3], S[4], W[11] + K[11])
    RND(S[4], S[5], S[6], S[7], S[0], S[1], S[2], S[3], W[12] + K[12])
    RND(S[3], S[4], S[5], S[6], S[7], S[0], S[1], S[2], W[13] + K[13])
    RND(S[2], S[3], S[4], S[5], S[6], S[7], S[0], S[1], W[14] + K[14])
    RND(S[1], S[2], S[3], S[4], S[5], S[6], S[7], S[0], W[15] + K[15])

    W[ 0] += s1(W[14]) + W[ 9] + s0(W[ 1]);
    W[ 1] += s1(W[15]) + W[10] + s0(W[ 2]);
    W[ 2] += s1(W[ 0]) + W[11] + s0(W[ 3]);
    W[ 3] += s1(W[ 1]) + W[12] + s0(W[ 4]);
    W[ 4] += s1(W[ 2]) + W[13] + s0(W[ 5]);
    W[ 5] += s1(W[ 3]) + W[14] + s0(W[ 6]);
    W[ 6] += s1(W[ 4]) + W[15] + s0(W[ 7]);
    W[ 7] += s1(W[ 5]) + W[ 0] + s0(W[ 8]);
    W[ 8] += s1(W[ 6]) + W[ 1] + s0(W[ 9]);
    W[ 9] += s1(W[ 7]) + W[ 2] + s0(W[10]);
    W[10] += s1(W[ 8]) + W[ 3] + s0(W[11]);
    W[11] += s1(W[ 9]) + W[ 4] + s0(W[12]);
    W[12] += s1(W[10]) + W[ 5] + s0(W[13]);
    W[13] += s1(W[11]) + W[ 6] + s0(W[14]);
    W[14] += s1(W[12]) + W[ 7] + s0(W[15]);
    W[15] += s1(W[13]) + W[ 8] + s0(W[ 0]);

    RND(S[0], S[1], S[2], S[3], S[4], S[5], S[6], S[7], W[ 0] + K[16])
    RND(S[7], S[0], S[1], S[2], S[3], S[4], S[5], S[6], W[ 1] + K[17])
    RND(S[6], S[7], S[0], S[1], S[2], S[3], S[4], S[5], W[ 2] + K[18])
    RND(S[5], S[6], S[7], S[0], S[1], S[2], S[3], S[4], W[ 3] + K[19])
    RND(S[4], S[5], S[6], S[7], S[0], S[1], S[2], S[3], W[ 4] + K[20])
    RND(S[3], S[4], S[5], S[6], S[7], S[0], S[1], S[2], W[ 5] + K[21])
    RND(S[2], S[3], S[4], S[5], S[6], S[7], S[0], S[1], W[ 6] + K[22])
    RND(S[1], S[2], S[3], S[4], S[5], S[6], S[7], S[0], W[ 7] + K[23])
    RND(S[0], S[1], S[2], S[3], S[4], S[5], S[6], S[7], W[ 8] + K[24])
    RND(S[7], S[0], S[1], S[2], S[3], S[4], S[5], S[6], W[ 9] + K[25])
    RND(S[6], S[7], S[0], S[1], S[2], S[3], S[4], S[5], W[10] + K[26])
    RND(S[5], S[6], S[7], S[0], S[1], S[2], S[3], S[4], W[11] + K[27])
    RND(S[4], S[5], S[6], S[7], S[0], S[1], S[2], S[3], W[12] + K[28])
    RND(S[3], S[4], S[5], S[6], S[7], S[0], S[1], S[2], W[13] + K[29])
    RND(S[2], S[3], S[4], S[5], S[6], S[7], S[0], S[1], W[14] + K[30])
    RND(S[1], S[2], S[3], S[4], S[5], S[6], S[7], S[0], W[15] + K[31])

    W[ 0] += s1(W[14]) + W[ 9] + s0(W[ 1]);
    W[ 1] += s1(W[15]) + W[10] + s0(W[ 2]);
    W[ 2] += s1(W[ 0]) + W[11] + s0(W[ 3]);
    W[ 3] += s1(W[ 1]) + W[12] + s0(W[ 4]);
    W[ 4] += s1(W[ 2]) + W[13] + s0(W[ 5]);
    W[ 5] += s1(W[ 3]) + W[14] + s0(W[ 6]);
    W[ 6] += s1(W[ 4]) + W[15] + s0(W[ 7]);
    W[ 7] += s1(W[ 5]) + W[ 0] + s0(W[ 8]);
    W[ 8] += s1(W[ 6]) + W[ 1] + s0(W[ 9]);
    W[ 9] += s1(W[ 7]) + W[ 2] + s0(W[10]);
    W[10] += s1(W[ 8]) + W[ 3] + s0(W[11]);
    W[11] += s1(W[ 9]) + W[ 4] + s0(W[12]);
    W[12] += s1(W[10]) + W[ 5] + s0(W[13]);
    W[13] += s1(W[11]) + W[ 6] + s0(W[14]);
    W[14] += s1(W[12]) + W[ 7] + s0(W[15]);
    W[15] += s1(W[13]) + W[ 8] + s0(W[ 0]);

    RND(S[0], S[1], S[2], S[3], S[4], S[5], S[6], S[7], W[ 0] + K[32])
    RND(S[7], S[0], S[1], S[2], S[3], S[4], S[5], S[6], W[ 1] + K[33])
    RND(S[6], S[7], S[0], S[1], S[2], S[3], S[4], S[5], W[ 2] + K[34])
    RND(S[5], S[6], S[7], S[0], S[1], S[2], S[3], S[4], W[ 3] + K[35])
    RND(S[4], S[5], S[6], S[7], S[0], S[1], S[2], S[3], W[ 4] + K[36])
    RND(S[3], S[4], S[5], S[6], S[7], S[0], S[1], S[2], W[ 5] + K[37])
    RND(S[2], S[3], S[4], S[5], S[6], S[7], S[0], S[1], W[ 6] + K[38])
    RND(S[1], S[2], S[3], S[4], S[5], S[6], S[7], S[0], W[ 7] + K[39])
    RND(S[0], S[1], S[2], S[3], S[4], S[5], S[6], S[7], W[ 8] + K[40])
    RND(S[7], S[0], S[1], S[2], S[3], S[4], S[5], S[6], W[ 9] + K[41])
    RND(S[6], S[7], S[0], S[1], S[2], S[3], S[4], S[5], W[10] + K[42])
    RND(S[5], S[6], S[7], S[0], S[1], S[2], S[3], S[4], W[11] + K[43])
    RND(S[4], S[5], S[6], S[7], S[0], S[1], S[2], S[3], W[12] + K[44])
    RND(S[3], S[4], S[5], S[6], S[7], S[0], S[1], S[2], W[13] + K[45])
    RND(S[2], S[3], S[4], S[5], S[6], S[7], S[0], S[1], W[14] + K[46])
    RND(S[1], S[2], S[3], S[4], S[5], S[6], S[7], S[0], W[15] + K[47])

    W[ 0] += s1(W[14]) + W[ 9] + s0(W[ 1]);
    W[ 1] += s1(W[15]) + W[10] + s0(W[ 2]);
    W[ 2] += s1(W[ 0]) + W[11] + s0(W[ 3]);
    W[ 3] += s1(W[ 1]) + W[12] + s0(W[ 4]);
    W[ 4] += s1(W[ 2]) + W[13] + s0(W[ 5]);
    W[ 5] += s1(W[ 3]) + W[14] + s0(W[ 6]);
    W[ 6] += s1(W[ 4]) + W[15] + s0(W[ 7]);
    W[ 7] += s1(W[ 5]) + W[ 0] + s0(W[ 8]);
    W[ 8] += s1(W[ 6]) + W[ 1] + s0(W[ 9]);
    W[ 9] += s1(W[ 7]) + W[ 2] + s0(W[10]);
    W[10] += s1(W[ 8]) + W[ 3] + s0(W[11]);
    W[11] += s1(W[ 9]) + W[ 4] + s0(W[12]);
    W[12] += s1(W[10]) + W[ 5] + s0(W[13]);
    W[13] += s1(W[11]) + W[ 6] + s0(W[14]);
    W[14] += s1(W[12]) + W[ 7] + s0(W[15]);
    W[15] += s1(W[13]) + W[ 8] + s0(W[ 0]);

    RND(S[0], S[1], S[2], S[3], S[4], S[5], S[6], S[7], W[ 0] + K[48])
    RND(S[7], S[0], S[1], S[2], S[3], S[4], S[5], S[6], W[ 1] + K[49])
    RND(S[6], S[7], S[0], S[1], S[2], S[3], S[4], S[5], W[ 2] + K[50])
    RND(S[5], S[6], S[7], S[0], S[1], S[2], S[3], S[4], W[ 3] + K[51])
    RND(S[4], S[5], S[6], S[7], S[0], S[1], S[2], S[3], W[ 4] + K[52])
    RND(S[3], S[4], S[5], S[6], S[7], S[0], S[1], S[2], W[ 5] + K[53])
    RND(S[2], S[3], S[4], S[5], S[6], S[7], S[0], S[1], W[ 6] + K[54])
    RND(S[1], S[2], S[3], S[4], S[5], S[6], S[7], S[0], W[ 7] + K[55])
    RND(S[0], S[1], S[2], S[3], S[4], S[5], S[6], S[7], W[ 8] + K[56])
    RND(S[7], S[0], S[1], S[2], S[3], S[4], S[5], S[6], W[ 9] + K[57])
    RND(S[6], S[7], S[0], S[1], S[2], S[3], S[4], S[5], W[10] + K[58])
    RND(S[5], S[6], S[7], S[0], S[1], S[2], S[3], S[4], W[11] + K[59])
    RND(S[4], S[5], S[6], S[7], S[0], S[1], S[2], S[3], W[12] + K[60])
    RND(S[3], S[4], S[5], S[6], S[7], S[0], S[1], S[2], W[13] + K[61])
    RND(S[2], S[3], S[4], S[5], S[6], S[7], S[0], S[1], W[14] + K[62])
    RND(S[1], S[2], S[3], S[4], S[5], S[6], S[7], S[0], W[15] + K[63])

    if (0u == ((S[0] + 0x6A09E667u) & __MASK0__u) &&
        0u == ((S[1] + 0xBB67AE85u) & __MASK1__u)
    ) {
      out_v = uvec4(0u, __MAGIC_CODE__u, thread_id, i);
      return;
    }
  }
  discard;
}