v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
o0 = v4 + m8;
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_67.x = v6;
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_23.x = v2;
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;
v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_67.x = v6;
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
o0 = v2 + m0;
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_23.x = v2;
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
o0 = v2 + m4;
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;
v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
o0 = v4 + m4;
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_67.x = v6;
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
o0 = v2 + m6;
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_23.x = v2;
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
o0 = v4 + m2;
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;
v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_67.x = v6;
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_23.x = v2;
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
o0 = v4 + m8;
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
o0 = v4 + m0;
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;
v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
o0 = v4 + m4;
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
o0 = v4 + m8;
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_67.x = v6;
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_23.x = v2;
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;
v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
o0 = v4 + m0;
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_67.x = v6;
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
o0 = v6 + m6;
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_23.x = v2;
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;
v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
o0 = v6 + m8;
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_67.x = v6;
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_23.x = v2;
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
o0 = v2 + m6;
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
o0 = v4 + m4;
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;
v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
o0 = v4 + m2;
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
o0 = v6 + m6;
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_67.x = v6;
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_23.x = v2;
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
o0 = v2 + m8;
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;
v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
o0 = v4 + m6;
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
o0 = v6 + m0;
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_67.x = v6;
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_23.x = v2;
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
o0 = v4 + m2;
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
o0 = v4 + m8;
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;
v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
o0 = v6 + m2;
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_67.x = v6;
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_23.x = v2;
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
o0 = v4 + m6;
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;
v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
o0 = v4 + m8;
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_67.x = v6;
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_23.x = v2;
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;
v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2425.x = v24;
v_2425.y = v25;
v_1617 = v_1617 + v_2425 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2425.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v24 = v_2425.x;
v25 = v_2425.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_2627.x = v26;
v_2627.y = v27;
v_1819 = v_1819 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2627.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1213.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1213 ^ v_2021;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_67.x = v6;
v_67 = v_67 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_67.x + v_1415.x < v_67.x);
v6 = v_67.x;
v7 = v_67.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
xor = v_1415 ^ v_2223;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+0]]
o0 = v2 + m0;
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + b
v_23.x = v2;
v_23 = v_23 + v_1213 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_23.x + v_1213.x < v_23.x);
v2 = v_23.x;
v3 = v_23.y;
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
// a = a + m[sigma[r][2*i+1]]
o0 = v2 + m4;
xor = v_1213 ^ v_2223;
v_1213 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v12 = v_1213.x;
- v13 = v_1213.y;
+
+
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+0]]
// skip since adding 0u does nothing
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x >> 24u) | (xor.y << 8u), (xor.y >> 24u) | (xor.x << 8u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + b
v_45.x = v4;
v_45 = v_45 + v_1415 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_45.x + v_1415.x < v_45.x);
v4 = v_45.x;
v5 = v_45.y;
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
// a = a + m[sigma[r][2*i+1]]
// skip since adding 0u does nothing
v_2627.x = v26;
v_2627.y = v27;
v_1617 = v_1617 + v_2627 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1617.x + v_2627.x < v_1617.x);
- v16 = v_1617.x;
- v17 = v_1617.y;
+
+
v26 = v_2627.x;
v27 = v_2627.y;
xor = v_1415 ^ v_1617;
v_1415 = vec2((xor.x << 1u) | (xor.y >> 31u), (xor.y << 1u) | (xor.x >> 31u));
- v14 = v_1415.x;
- v15 = v_1415.y;
+
+
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;
v_2829.x = v28;
v_2829.y = v29;
v_1819 = v_1819 + v_2829 + select(vec2<u32>(0u), vec2<u32>(0u, 1u), v_1819.x + v_2829.x < v_1819.x);
- v18 = v_1819.x;
- v19 = v_1819.y;
+
+
v28 = v_2829.x;
v29 = v_2829.y;