diff --git a/src/air/layouts/dex/public_input.cairo b/src/air/layouts/dex/public_input.cairo
index 616ab8437..772fe53e9 100644
--- a/src/air/layouts/dex/public_input.cairo
+++ b/src/air/layouts/dex/public_input.cairo
@@ -97,7 +97,7 @@ impl DexPublicInputImpl of PublicInputTrait {
                 ref memory_index
             );
 
-        // 3. Output segment 
+        // 3. Output segment
         let output_len = output_stop - output_start;
         let output = memory
             .extract_range(
diff --git a/src/air/layouts/recursive/public_input.cairo b/src/air/layouts/recursive/public_input.cairo
index ae1d0634a..35eb87e24 100644
--- a/src/air/layouts/recursive/public_input.cairo
+++ b/src/air/layouts/recursive/public_input.cairo
@@ -111,7 +111,7 @@ impl RecursivePublicInputImpl of PublicInputTrait {
                 ref memory_index
             );
 
-        // 3. Output segment 
+        // 3. Output segment
         let output_len = output_stop - output_start;
         let output = memory
             .extract_range(
diff --git a/src/air/layouts/recursive_with_poseidon/public_input.cairo b/src/air/layouts/recursive_with_poseidon/public_input.cairo
index 3c9c6d268..35aab85ee 100644
--- a/src/air/layouts/recursive_with_poseidon/public_input.cairo
+++ b/src/air/layouts/recursive_with_poseidon/public_input.cairo
@@ -110,7 +110,7 @@ impl RecursiveWithPoseidonPublicInputImpl of PublicInputTrait {
                 ref memory_index
             );
 
-        // 3. Output segment 
+        // 3. Output segment
         let output_len = output_stop - output_start;
         let output = memory
             .extract_range(
diff --git a/src/air/layouts/small/public_input.cairo b/src/air/layouts/small/public_input.cairo
index 9f36acff3..98d74fa1f 100644
--- a/src/air/layouts/small/public_input.cairo
+++ b/src/air/layouts/small/public_input.cairo
@@ -109,7 +109,7 @@ impl SmallPublicInputImpl of PublicInputTrait {
                 ref memory_index
             );
 
-        // 3. Output segment 
+        // 3. Output segment
         let output_len = output_stop - output_start;
         let output = memory
             .extract_range(
diff --git a/src/air/layouts/starknet/public_input.cairo b/src/air/layouts/starknet/public_input.cairo
index ae0fccbc9..f9cb56e26 100644
--- a/src/air/layouts/starknet/public_input.cairo
+++ b/src/air/layouts/starknet/public_input.cairo
@@ -110,7 +110,7 @@ impl StarknetPublicInputImpl of PublicInputTrait {
                 ref memory_index
             );
 
-        // 3. Output segment 
+        // 3. Output segment
         let output_len = output_stop - output_start;
         let output = memory
             .extract_range(
diff --git a/src/air/layouts/starknet_with_keccak/public_input.cairo b/src/air/layouts/starknet_with_keccak/public_input.cairo
index e0ac4d3b4..ca1932266 100644
--- a/src/air/layouts/starknet_with_keccak/public_input.cairo
+++ b/src/air/layouts/starknet_with_keccak/public_input.cairo
@@ -110,7 +110,7 @@ impl StarknetWithKeccakPublicInputImpl of PublicInputTrait {
                 ref memory_index
             );
 
-        // 3. Output segment 
+        // 3. Output segment
         let output_len = output_stop - output_start;
         let output = memory
             .extract_range(
diff --git a/src/air/public_input.cairo b/src/air/public_input.cairo
index a7f98dfdc..0834f73de 100644
--- a/src/air/public_input.cairo
+++ b/src/air/public_input.cairo
@@ -193,22 +193,21 @@ fn verify_cairo1_public_input(public_input: @PublicInput) -> (felt252, felt252)
 
     // 1. Program segment
     assert(initial_pc == INITIAL_PC, 'Invalid initial_pc');
-    let program = memory
-        .extract_range_unchecked(initial_pc.try_into().unwrap(), memory.len() - output_len);
+    let program = memory.extract_range_unchecked(0, memory.len() - output_len);
     let program_hash = poseidon_hash_span(program);
 
-    // 2. Output segment 
+    // 2. Output segment
     let output = memory.extract_range_unchecked(memory.len() - output_len, output_len);
     let output_hash = poseidon_hash_span(output);
     (program_hash, output_hash)
 }
 
+// === RECURSIVE BEGIN ===
 #[cfg(test)]
 mod tests {
     use core::debug::PrintTrait;
     use super::get_public_input_hash;
     use cairo_verifier::tests::stone_proof_fibonacci_keccak::public_input::get;
-    // test data from cairo0-verifier run on stone-prover generated proof
     #[test]
     #[available_gas(9999999999)]
     fn test_get_public_input_hash() {
@@ -220,3 +219,6 @@ mod tests {
         )
     }
 }
+// === RECURSIVE END ===
+
+
diff --git a/src/air/public_memory.cairo b/src/air/public_memory.cairo
index 3d84bbbfd..8284b22b8 100644
--- a/src/air/public_memory.cairo
+++ b/src/air/public_memory.cairo
@@ -58,7 +58,7 @@ impl PageImpl of PageTrait {
 
             let current = *self.at(offset);
 
-            // TODO is this needed? If not we can just use slice directly 
+            // TODO is this needed? If not we can just use slice directly
             assert(current.address == (addr + i).into(), 'Invalid address');
             arr.append(current.value);
             i += 1;
diff --git a/src/common/blake2s.cairo b/src/common/blake2s.cairo
index d4174f236..063e40f09 100644
--- a/src/common/blake2s.cairo
+++ b/src/common/blake2s.cairo
@@ -18,52 +18,28 @@ fn truncated_blake2s(data: Array<u32>) -> felt252 {
 
 // internals:
 
-fn load32(p0: u8, p1: u8, p2: u8, p3: u8) -> u32 {
-    let mut x: u32 = p3.into();
-    x = x * 256 + p2.into();
-    x = x * 256 + p1.into();
-    x = x * 256 + p0.into();
-    x
-}
-
-fn get_sigma(r: u32) -> Array<u32> {
-    if r == 0 {
-        array![0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15]
-    } else if r == 1 {
-        array![14, 10, 4, 8, 9, 15, 13, 6, 1, 12, 0, 2, 11, 7, 5, 3]
-    } else if r == 2 {
-        array![11, 8, 12, 0, 5, 2, 15, 13, 10, 14, 3, 6, 7, 1, 9, 4]
-    } else if r == 3 {
-        array![7, 9, 3, 1, 13, 12, 11, 14, 2, 6, 5, 10, 4, 0, 15, 8]
-    } else if r == 4 {
-        array![9, 0, 5, 7, 2, 4, 10, 15, 14, 1, 11, 12, 6, 8, 3, 13]
-    } else if r == 5 {
-        array![2, 12, 6, 10, 0, 11, 8, 3, 4, 13, 7, 5, 15, 14, 1, 9]
-    } else if r == 6 {
-        array![12, 5, 1, 15, 14, 13, 4, 10, 0, 7, 6, 3, 9, 2, 8, 11]
-    } else if r == 7 {
-        array![13, 11, 7, 14, 12, 1, 3, 9, 5, 0, 15, 4, 8, 6, 2, 10]
-    } else if r == 8 {
-        array![6, 15, 14, 9, 11, 3, 0, 8, 12, 2, 13, 7, 1, 4, 10, 5]
-    } else { // r == 9
-        array![10, 2, 8, 4, 7, 6, 1, 5, 15, 11, 9, 14, 3, 12, 13, 0]
-    }
-}
-
+#[inline(always)]
 fn rotr16(n: u32) -> u32 {
-    n / 65536 + (n % 65536) * 65536
+    let (high, low) = DivRem::div_rem(n, 65536);
+    TryInto::<felt252, u32>::try_into(high.into() + low.into() * 65536).unwrap()
 }
 
+#[inline(always)]
 fn rotr12(n: u32) -> u32 {
-    n / 4096 + (n % 4096) * 1048576
+    let (high, low) = DivRem::div_rem(n, 4096);
+    TryInto::<felt252, u32>::try_into(high.into() + low.into() * 1048576).unwrap()
 }
 
+#[inline(always)]
 fn rotr8(n: u32) -> u32 {
-    n / 256 + (n % 256) * 16777216
+    let (high, low) = DivRem::div_rem(n, 256);
+    TryInto::<felt252, u32>::try_into(high.into() + low.into() * 16777216).unwrap()
 }
 
+#[inline(always)]
 fn rotr7(n: u32) -> u32 {
-    n / 128 + (n % 128) * 33554432
+    let (high, low) = DivRem::div_rem(n, 128);
+    TryInto::<felt252, u32>::try_into(high.into() + low.into() * 33554432).unwrap()
 }
 
 #[derive(Drop, Clone)]
@@ -113,98 +89,256 @@ fn blake2s_compress(mut s: blake2s_state, m: Array<u32>) -> blake2s_state {
     let mut v15: u32 = 0x5BE0CD19; // f1 is always 0
 
     let m_span = m.span();
+    let mut sigma = array![
+        0,
+        1,
+        2,
+        3,
+        4,
+        5,
+        6,
+        7,
+        8,
+        9,
+        10,
+        11,
+        12,
+        13,
+        14,
+        15,
+        14,
+        10,
+        4,
+        8,
+        9,
+        15,
+        13,
+        6,
+        1,
+        12,
+        0,
+        2,
+        11,
+        7,
+        5,
+        3,
+        11,
+        8,
+        12,
+        0,
+        5,
+        2,
+        15,
+        13,
+        10,
+        14,
+        3,
+        6,
+        7,
+        1,
+        9,
+        4,
+        7,
+        9,
+        3,
+        1,
+        13,
+        12,
+        11,
+        14,
+        2,
+        6,
+        5,
+        10,
+        4,
+        0,
+        15,
+        8,
+        9,
+        0,
+        5,
+        7,
+        2,
+        4,
+        10,
+        15,
+        14,
+        1,
+        11,
+        12,
+        6,
+        8,
+        3,
+        13,
+        2,
+        12,
+        6,
+        10,
+        0,
+        11,
+        8,
+        3,
+        4,
+        13,
+        7,
+        5,
+        15,
+        14,
+        1,
+        9,
+        12,
+        5,
+        1,
+        15,
+        14,
+        13,
+        4,
+        10,
+        0,
+        7,
+        6,
+        3,
+        9,
+        2,
+        8,
+        11,
+        13,
+        11,
+        7,
+        14,
+        12,
+        1,
+        3,
+        9,
+        5,
+        0,
+        15,
+        4,
+        8,
+        6,
+        2,
+        10,
+        6,
+        15,
+        14,
+        9,
+        11,
+        3,
+        0,
+        8,
+        12,
+        2,
+        13,
+        7,
+        1,
+        4,
+        10,
+        5,
+        10,
+        2,
+        8,
+        4,
+        7,
+        6,
+        1,
+        5,
+        15,
+        11,
+        9,
+        14,
+        3,
+        12,
+        13,
+        0,
+    ]
+        .span();
 
-    let mut r = 0;
     loop {
-        if r == 10 {
+        if sigma.is_empty() {
             break;
         }
 
-        let sigma = get_sigma(r);
-
         // ROUND function begin
 
         // 0 - 0,4,8,12
-        v0 = u32_wrapping_add(u32_wrapping_add(v0, v4), *m_span.at(*sigma[0]));
+        v0 = u32_wrapping_add(u32_wrapping_add(v0, v4), *m_span.at(*sigma.pop_front().unwrap()));
         v12 = rotr16(v12 ^ v0);
         v8 = u32_wrapping_add(v8, v12);
         v4 = rotr12(v4 ^ v8);
-        v0 = u32_wrapping_add(u32_wrapping_add(v0, v4), *m_span.at(*sigma[1]));
+        v0 = u32_wrapping_add(u32_wrapping_add(v0, v4), *m_span.at(*sigma.pop_front().unwrap()));
         v12 = rotr8(v12 ^ v0);
         v8 = u32_wrapping_add(v8, v12);
         v4 = rotr7(v4 ^ v8);
 
         // 1 - 1,5,9,13
-        v1 = u32_wrapping_add(u32_wrapping_add(v1, v5), *m_span.at(*sigma[2]));
+        v1 = u32_wrapping_add(u32_wrapping_add(v1, v5), *m_span.at(*sigma.pop_front().unwrap()));
         v13 = rotr16(v13 ^ v1);
         v9 = u32_wrapping_add(v9, v13);
         v5 = rotr12(v5 ^ v9);
-        v1 = u32_wrapping_add(u32_wrapping_add(v1, v5), *m_span.at(*sigma[3]));
+        v1 = u32_wrapping_add(u32_wrapping_add(v1, v5), *m_span.at(*sigma.pop_front().unwrap()));
         v13 = rotr8(v13 ^ v1);
         v9 = u32_wrapping_add(v9, v13);
         v5 = rotr7(v5 ^ v9);
 
         // 2 - 2,6,10,14
-        v2 = u32_wrapping_add(u32_wrapping_add(v2, v6), *m_span.at(*sigma[4]));
+        v2 = u32_wrapping_add(u32_wrapping_add(v2, v6), *m_span.at(*sigma.pop_front().unwrap()));
         v14 = rotr16(v14 ^ v2);
         v10 = u32_wrapping_add(v10, v14);
         v6 = rotr12(v6 ^ v10);
-        v2 = u32_wrapping_add(u32_wrapping_add(v2, v6), *m_span.at(*sigma[5]));
+        v2 = u32_wrapping_add(u32_wrapping_add(v2, v6), *m_span.at(*sigma.pop_front().unwrap()));
         v14 = rotr8(v14 ^ v2);
         v10 = u32_wrapping_add(v10, v14);
         v6 = rotr7(v6 ^ v10);
 
         // 3 - 3,7,11,15
-        v3 = u32_wrapping_add(u32_wrapping_add(v3, v7), *m_span.at(*sigma[6]));
+        v3 = u32_wrapping_add(u32_wrapping_add(v3, v7), *m_span.at(*sigma.pop_front().unwrap()));
         v15 = rotr16(v15 ^ v3);
         v11 = u32_wrapping_add(v11, v15);
         v7 = rotr12(v7 ^ v11);
-        v3 = u32_wrapping_add(u32_wrapping_add(v3, v7), *m_span.at(*sigma[7]));
+        v3 = u32_wrapping_add(u32_wrapping_add(v3, v7), *m_span.at(*sigma.pop_front().unwrap()));
         v15 = rotr8(v15 ^ v3);
         v11 = u32_wrapping_add(v11, v15);
         v7 = rotr7(v7 ^ v11);
 
         // 4 - 0,5,10,15
-        v0 = u32_wrapping_add(u32_wrapping_add(v0, v5), *m_span.at(*sigma[8]));
+        v0 = u32_wrapping_add(u32_wrapping_add(v0, v5), *m_span.at(*sigma.pop_front().unwrap()));
         v15 = rotr16(v15 ^ v0);
         v10 = u32_wrapping_add(v10, v15);
         v5 = rotr12(v5 ^ v10);
-        v0 = u32_wrapping_add(u32_wrapping_add(v0, v5), *m_span.at(*sigma[9]));
+        v0 = u32_wrapping_add(u32_wrapping_add(v0, v5), *m_span.at(*sigma.pop_front().unwrap()));
         v15 = rotr8(v15 ^ v0);
         v10 = u32_wrapping_add(v10, v15);
         v5 = rotr7(v5 ^ v10);
 
         // 5 - 1,6,11,12
-        v1 = u32_wrapping_add(u32_wrapping_add(v1, v6), *m_span.at(*sigma[10]));
+        v1 = u32_wrapping_add(u32_wrapping_add(v1, v6), *m_span.at(*sigma.pop_front().unwrap()));
         v12 = rotr16(v12 ^ v1);
         v11 = u32_wrapping_add(v11, v12);
         v6 = rotr12(v6 ^ v11);
-        v1 = u32_wrapping_add(u32_wrapping_add(v1, v6), *m_span.at(*sigma[11]));
+        v1 = u32_wrapping_add(u32_wrapping_add(v1, v6), *m_span.at(*sigma.pop_front().unwrap()));
         v12 = rotr8(v12 ^ v1);
         v11 = u32_wrapping_add(v11, v12);
         v6 = rotr7(v6 ^ v11);
 
         // 6 - 2,7,8,13
-        v2 = u32_wrapping_add(u32_wrapping_add(v2, v7), *m_span.at(*sigma[12]));
+        v2 = u32_wrapping_add(u32_wrapping_add(v2, v7), *m_span.at(*sigma.pop_front().unwrap()));
         v13 = rotr16(v13 ^ v2);
         v8 = u32_wrapping_add(v8, v13);
         v7 = rotr12(v7 ^ v8);
-        v2 = u32_wrapping_add(u32_wrapping_add(v2, v7), *m_span.at(*sigma[13]));
+        v2 = u32_wrapping_add(u32_wrapping_add(v2, v7), *m_span.at(*sigma.pop_front().unwrap()));
         v13 = rotr8(v13 ^ v2);
         v8 = u32_wrapping_add(v8, v13);
         v7 = rotr7(v7 ^ v8);
 
         // 7 - 3,4,9,14
-        v3 = u32_wrapping_add(u32_wrapping_add(v3, v4), *m_span.at(*sigma[14]));
+        v3 = u32_wrapping_add(u32_wrapping_add(v3, v4), *m_span.at(*sigma.pop_front().unwrap()));
         v14 = rotr16(v14 ^ v3);
         v9 = u32_wrapping_add(v9, v14);
         v4 = rotr12(v4 ^ v9);
-        v3 = u32_wrapping_add(u32_wrapping_add(v3, v4), *m_span.at(*sigma[15]));
+        v3 = u32_wrapping_add(u32_wrapping_add(v3, v4), *m_span.at(*sigma.pop_front().unwrap()));
         v14 = rotr8(v14 ^ v3);
         v9 = u32_wrapping_add(v9, v14);
         v4 = rotr7(v4 ^ v9);
-
-        r += 1;
     };
 
     let mut new_h = ArrayTrait::new();
@@ -324,7 +458,7 @@ fn blake2s_update(mut s: blake2s_state, in: Array<u32>) -> blake2s_state {
 fn blake2s_final(mut s: blake2s_state) -> u256 {
     assert(s.f0 == 0, 'blake2s_is_lastblock');
 
-    // blake2s_increment_counter 
+    // blake2s_increment_counter
     s.t0 = u32_wrapping_add(s.t0, s.buflen * 4);
     if s.t0 < s.buflen {
         s.t1 = u32_wrapping_add(s.t1, 1);
diff --git a/src/common/blake2s_u8.cairo b/src/common/blake2s_u8.cairo
index 63a0f765e..2f2e890ea 100644
--- a/src/common/blake2s_u8.cairo
+++ b/src/common/blake2s_u8.cairo
@@ -10,53 +10,34 @@ fn blake2s(data: Array<u8>) -> u256 {
 // internals:
 
 fn load32(p0: u8, p1: u8, p2: u8, p3: u8) -> u32 {
-    let mut x: u32 = p3.into();
+    let mut x: felt252 = p3.into();
     x = x * 256 + p2.into();
     x = x * 256 + p1.into();
     x = x * 256 + p0.into();
-    x
-}
-
-fn get_sigma(r: u32) -> Array<u32> {
-    if r == 0 {
-        array![0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15]
-    } else if r == 1 {
-        array![14, 10, 4, 8, 9, 15, 13, 6, 1, 12, 0, 2, 11, 7, 5, 3]
-    } else if r == 2 {
-        array![11, 8, 12, 0, 5, 2, 15, 13, 10, 14, 3, 6, 7, 1, 9, 4]
-    } else if r == 3 {
-        array![7, 9, 3, 1, 13, 12, 11, 14, 2, 6, 5, 10, 4, 0, 15, 8]
-    } else if r == 4 {
-        array![9, 0, 5, 7, 2, 4, 10, 15, 14, 1, 11, 12, 6, 8, 3, 13]
-    } else if r == 5 {
-        array![2, 12, 6, 10, 0, 11, 8, 3, 4, 13, 7, 5, 15, 14, 1, 9]
-    } else if r == 6 {
-        array![12, 5, 1, 15, 14, 13, 4, 10, 0, 7, 6, 3, 9, 2, 8, 11]
-    } else if r == 7 {
-        array![13, 11, 7, 14, 12, 1, 3, 9, 5, 0, 15, 4, 8, 6, 2, 10]
-    } else if r == 8 {
-        array![6, 15, 14, 9, 11, 3, 0, 8, 12, 2, 13, 7, 1, 4, 10, 5]
-    } else { // r == 9
-        array![10, 2, 8, 4, 7, 6, 1, 5, 15, 11, 9, 14, 3, 12, 13, 0]
-    }
+    x.try_into().unwrap()
 }
 
 fn rotr16(n: u32) -> u32 {
-    n / 65536 + (n % 65536) * 65536
+    let (high, low) = DivRem::div_rem(n, 65536);
+    TryInto::<felt252, u32>::try_into(high.into() + low.into() * 65536).unwrap()
 }
 
 fn rotr12(n: u32) -> u32 {
-    n / 4096 + (n % 4096) * 1048576
+    let (high, low) = DivRem::div_rem(n, 4096);
+    TryInto::<felt252, u32>::try_into(high.into() + low.into() * 1048576).unwrap()
 }
 
 fn rotr8(n: u32) -> u32 {
-    n / 256 + (n % 256) * 16777216
+    let (high, low) = DivRem::div_rem(n, 256);
+    TryInto::<felt252, u32>::try_into(high.into() + low.into() * 16777216).unwrap()
 }
 
 fn rotr7(n: u32) -> u32 {
-    n / 128 + (n % 128) * 33554432
+    let (high, low) = DivRem::div_rem(n, 128);
+    TryInto::<felt252, u32>::try_into(high.into() + low.into() * 33554432).unwrap()
 }
 
+
 #[derive(Drop, Clone)]
 struct blake2s_state {
     h: Array<u32>, // length: 8
@@ -115,100 +96,257 @@ fn blake2s_compress(mut s: blake2s_state, in: Array<u8>) -> blake2s_state {
     let mut v15: u32 = 0x5BE0CD19; // f1 is always 0
 
     let m_span = m.span();
+    let mut sigma = array![
+        0,
+        1,
+        2,
+        3,
+        4,
+        5,
+        6,
+        7,
+        8,
+        9,
+        10,
+        11,
+        12,
+        13,
+        14,
+        15,
+        14,
+        10,
+        4,
+        8,
+        9,
+        15,
+        13,
+        6,
+        1,
+        12,
+        0,
+        2,
+        11,
+        7,
+        5,
+        3,
+        11,
+        8,
+        12,
+        0,
+        5,
+        2,
+        15,
+        13,
+        10,
+        14,
+        3,
+        6,
+        7,
+        1,
+        9,
+        4,
+        7,
+        9,
+        3,
+        1,
+        13,
+        12,
+        11,
+        14,
+        2,
+        6,
+        5,
+        10,
+        4,
+        0,
+        15,
+        8,
+        9,
+        0,
+        5,
+        7,
+        2,
+        4,
+        10,
+        15,
+        14,
+        1,
+        11,
+        12,
+        6,
+        8,
+        3,
+        13,
+        2,
+        12,
+        6,
+        10,
+        0,
+        11,
+        8,
+        3,
+        4,
+        13,
+        7,
+        5,
+        15,
+        14,
+        1,
+        9,
+        12,
+        5,
+        1,
+        15,
+        14,
+        13,
+        4,
+        10,
+        0,
+        7,
+        6,
+        3,
+        9,
+        2,
+        8,
+        11,
+        13,
+        11,
+        7,
+        14,
+        12,
+        1,
+        3,
+        9,
+        5,
+        0,
+        15,
+        4,
+        8,
+        6,
+        2,
+        10,
+        6,
+        15,
+        14,
+        9,
+        11,
+        3,
+        0,
+        8,
+        12,
+        2,
+        13,
+        7,
+        1,
+        4,
+        10,
+        5,
+        10,
+        2,
+        8,
+        4,
+        7,
+        6,
+        1,
+        5,
+        15,
+        11,
+        9,
+        14,
+        3,
+        12,
+        13,
+        0,
+    ]
+        .span();
 
-    let mut r = 0;
     loop {
-        if r == 10 {
+        if sigma.is_empty() {
             break;
         }
 
-        let sigma = get_sigma(r);
-
         // ROUND function begin
 
         // 0 - 0,4,8,12
-        v0 = u32_wrapping_add(u32_wrapping_add(v0, v4), *m_span.at(*sigma[0]));
+        v0 = u32_wrapping_add(u32_wrapping_add(v0, v4), *m_span.at(*sigma.pop_front().unwrap()));
         v12 = rotr16(v12 ^ v0);
         v8 = u32_wrapping_add(v8, v12);
         v4 = rotr12(v4 ^ v8);
-        v0 = u32_wrapping_add(u32_wrapping_add(v0, v4), *m_span.at(*sigma[1]));
+        v0 = u32_wrapping_add(u32_wrapping_add(v0, v4), *m_span.at(*sigma.pop_front().unwrap()));
         v12 = rotr8(v12 ^ v0);
         v8 = u32_wrapping_add(v8, v12);
         v4 = rotr7(v4 ^ v8);
 
         // 1 - 1,5,9,13
-        v1 = u32_wrapping_add(u32_wrapping_add(v1, v5), *m_span.at(*sigma[2]));
+        v1 = u32_wrapping_add(u32_wrapping_add(v1, v5), *m_span.at(*sigma.pop_front().unwrap()));
         v13 = rotr16(v13 ^ v1);
         v9 = u32_wrapping_add(v9, v13);
         v5 = rotr12(v5 ^ v9);
-        v1 = u32_wrapping_add(u32_wrapping_add(v1, v5), *m_span.at(*sigma[3]));
+        v1 = u32_wrapping_add(u32_wrapping_add(v1, v5), *m_span.at(*sigma.pop_front().unwrap()));
         v13 = rotr8(v13 ^ v1);
         v9 = u32_wrapping_add(v9, v13);
         v5 = rotr7(v5 ^ v9);
 
         // 2 - 2,6,10,14
-        v2 = u32_wrapping_add(u32_wrapping_add(v2, v6), *m_span.at(*sigma[4]));
+        v2 = u32_wrapping_add(u32_wrapping_add(v2, v6), *m_span.at(*sigma.pop_front().unwrap()));
         v14 = rotr16(v14 ^ v2);
         v10 = u32_wrapping_add(v10, v14);
         v6 = rotr12(v6 ^ v10);
-        v2 = u32_wrapping_add(u32_wrapping_add(v2, v6), *m_span.at(*sigma[5]));
+        v2 = u32_wrapping_add(u32_wrapping_add(v2, v6), *m_span.at(*sigma.pop_front().unwrap()));
         v14 = rotr8(v14 ^ v2);
         v10 = u32_wrapping_add(v10, v14);
         v6 = rotr7(v6 ^ v10);
 
         // 3 - 3,7,11,15
-        v3 = u32_wrapping_add(u32_wrapping_add(v3, v7), *m_span.at(*sigma[6]));
+        v3 = u32_wrapping_add(u32_wrapping_add(v3, v7), *m_span.at(*sigma.pop_front().unwrap()));
         v15 = rotr16(v15 ^ v3);
         v11 = u32_wrapping_add(v11, v15);
         v7 = rotr12(v7 ^ v11);
-        v3 = u32_wrapping_add(u32_wrapping_add(v3, v7), *m_span.at(*sigma[7]));
+        v3 = u32_wrapping_add(u32_wrapping_add(v3, v7), *m_span.at(*sigma.pop_front().unwrap()));
         v15 = rotr8(v15 ^ v3);
         v11 = u32_wrapping_add(v11, v15);
         v7 = rotr7(v7 ^ v11);
 
         // 4 - 0,5,10,15
-        v0 = u32_wrapping_add(u32_wrapping_add(v0, v5), *m_span.at(*sigma[8]));
+        v0 = u32_wrapping_add(u32_wrapping_add(v0, v5), *m_span.at(*sigma.pop_front().unwrap()));
         v15 = rotr16(v15 ^ v0);
         v10 = u32_wrapping_add(v10, v15);
         v5 = rotr12(v5 ^ v10);
-        v0 = u32_wrapping_add(u32_wrapping_add(v0, v5), *m_span.at(*sigma[9]));
+        v0 = u32_wrapping_add(u32_wrapping_add(v0, v5), *m_span.at(*sigma.pop_front().unwrap()));
         v15 = rotr8(v15 ^ v0);
         v10 = u32_wrapping_add(v10, v15);
         v5 = rotr7(v5 ^ v10);
 
         // 5 - 1,6,11,12
-        v1 = u32_wrapping_add(u32_wrapping_add(v1, v6), *m_span.at(*sigma[10]));
+        v1 = u32_wrapping_add(u32_wrapping_add(v1, v6), *m_span.at(*sigma.pop_front().unwrap()));
         v12 = rotr16(v12 ^ v1);
         v11 = u32_wrapping_add(v11, v12);
         v6 = rotr12(v6 ^ v11);
-        v1 = u32_wrapping_add(u32_wrapping_add(v1, v6), *m_span.at(*sigma[11]));
+        v1 = u32_wrapping_add(u32_wrapping_add(v1, v6), *m_span.at(*sigma.pop_front().unwrap()));
         v12 = rotr8(v12 ^ v1);
         v11 = u32_wrapping_add(v11, v12);
         v6 = rotr7(v6 ^ v11);
 
         // 6 - 2,7,8,13
-        v2 = u32_wrapping_add(u32_wrapping_add(v2, v7), *m_span.at(*sigma[12]));
+        v2 = u32_wrapping_add(u32_wrapping_add(v2, v7), *m_span.at(*sigma.pop_front().unwrap()));
         v13 = rotr16(v13 ^ v2);
         v8 = u32_wrapping_add(v8, v13);
         v7 = rotr12(v7 ^ v8);
-        v2 = u32_wrapping_add(u32_wrapping_add(v2, v7), *m_span.at(*sigma[13]));
+        v2 = u32_wrapping_add(u32_wrapping_add(v2, v7), *m_span.at(*sigma.pop_front().unwrap()));
         v13 = rotr8(v13 ^ v2);
         v8 = u32_wrapping_add(v8, v13);
         v7 = rotr7(v7 ^ v8);
 
         // 7 - 3,4,9,14
-        v3 = u32_wrapping_add(u32_wrapping_add(v3, v4), *m_span.at(*sigma[14]));
+        v3 = u32_wrapping_add(u32_wrapping_add(v3, v4), *m_span.at(*sigma.pop_front().unwrap()));
         v14 = rotr16(v14 ^ v3);
         v9 = u32_wrapping_add(v9, v14);
         v4 = rotr12(v4 ^ v9);
-        v3 = u32_wrapping_add(u32_wrapping_add(v3, v4), *m_span.at(*sigma[15]));
+        v3 = u32_wrapping_add(u32_wrapping_add(v3, v4), *m_span.at(*sigma.pop_front().unwrap()));
         v14 = rotr8(v14 ^ v3);
         v9 = u32_wrapping_add(v9, v14);
         v4 = rotr7(v4 ^ v9);
-
-        // ROUND function end
-
-        r += 1;
+    // ROUND function end
     };
 
     let mut new_h = ArrayTrait::new();
@@ -328,7 +466,7 @@ fn blake2s_update(mut s: blake2s_state, in: Array<u8>) -> blake2s_state {
 fn blake2s_final(mut s: blake2s_state) -> u256 {
     assert(s.f0 == 0, 'blake2s_is_lastblock');
 
-    // blake2s_increment_counter 
+    // blake2s_increment_counter
     s.t0 = u32_wrapping_add(s.t0, s.buflen);
     if s.t0 < s.buflen {
         s.t1 = u32_wrapping_add(s.t1, 1);
diff --git a/src/common/tests/test_blake2s.cairo b/src/common/tests/test_blake2s.cairo
index 84f51f4ab..18f9cbac6 100644
--- a/src/common/tests/test_blake2s.cairo
+++ b/src/common/tests/test_blake2s.cairo
@@ -1,5 +1,5 @@
 use cairo_verifier::common::{
-    array_append::ArrayAppendTrait, blake2s::{blake2s, truncated_blake2s, load32}
+    array_append::ArrayAppendTrait, blake2s::{blake2s, truncated_blake2s}, blake2s_u8::load32,
 };
 
 fn get_arr_v1(n: u32) -> Array<u32> {
diff --git a/src/proof_of_work/tests/test_proof_of_work.cairo b/src/proof_of_work/tests/test_proof_of_work.cairo
index 809f5009a..e2b6dd6de 100644
--- a/src/proof_of_work/tests/test_proof_of_work.cairo
+++ b/src/proof_of_work/tests/test_proof_of_work.cairo
@@ -9,7 +9,7 @@ use cairo_verifier::proof_of_work::proof_of_work::verify_proof_of_work;
 //     let n_bits: u8 = 20;
 //     verify_proof_of_work(digest, n_bits, nonce);
 // }
-// 
+//
 // #[test]
 // #[should_panic]
 // #[available_gas(9999999999)]
diff --git a/src/stark/tests/test_stark_commit.cairo b/src/stark/tests/test_stark_commit.cairo
index 620694c6a..c4949bfd0 100644
--- a/src/stark/tests/test_stark_commit.cairo
+++ b/src/stark/tests/test_stark_commit.cairo
@@ -10,19 +10,19 @@ use cairo_verifier::{
 //     let mut channel = ChannelTrait::new_with_counter(
 //         0xaf91f2c71f4a594b1575d258ce82464475c82d8fb244142d0db450491c1b52, 0x0
 //     );
-// 
+//
 //     let public_input = stone_proof_fibonacci::public_input::get();
 //     let unsent_commitment = stone_proof_fibonacci::stark::unsent_commitment::get();
 //     let config = stone_proof_fibonacci::stark::config::get();
 //     let stark_domains = stone_proof_fibonacci::stark::domains::get();
-// 
+//
 //     assert(
 //         stark_commit(
 //             ref channel, @public_input, @unsent_commitment, @config, @stark_domains
 //         ) == stone_proof_fibonacci::stark::commitment::get(),
 //         'Invalid value'
 //     );
-// 
+//
 //     assert(
 //         channel.digest == 0x9c769c7e0797cf043b06b980072a798b141f2bc41b14e85ad93ba178b13de7,
 //         'Invalid value'
diff --git a/src/stark/tests/test_stark_verify.cairo b/src/stark/tests/test_stark_verify.cairo
index 79ed05e3e..a8136800c 100644
--- a/src/stark/tests/test_stark_verify.cairo
+++ b/src/stark/tests/test_stark_verify.cairo
@@ -12,7 +12,7 @@ use cairo_verifier::{
 //     let commitment = stone_proof_fibonacci::stark::commitment::get();
 //     let witness = stone_proof_fibonacci::stark::witness::get();
 //     let stark_domains = stone_proof_fibonacci::stark::domains::get();
-// 
+//
 //     stark_verify(
 //         NUM_COLUMNS_FIRST, NUM_COLUMNS_SECOND, queries, commitment, witness, stark_domains,
 //     )
diff --git a/src/table_commitment/tests/test_table_commitment_decommit.cairo b/src/table_commitment/tests/test_table_commitment_decommit.cairo
index 6fecdb096..5b845691c 100644
--- a/src/table_commitment/tests/test_table_commitment_decommit.cairo
+++ b/src/table_commitment/tests/test_table_commitment_decommit.cairo
@@ -17,7 +17,7 @@ use cairo_verifier::{
 //     let queries = stone_proof_fibonacci::queries::get().span();
 //     let decommitment = stone_proof_fibonacci::traces::decommitment::get().original;
 //     let witness = stone_proof_fibonacci::traces::witness::get().original;
-// 
+//
 //     table_decommit(commitment, queries, decommitment, witness);
 // }
 // === BLAKE2S END ===