diff --git a/hashes/src/ripemd160/crypto.rs b/hashes/src/ripemd160/crypto.rs
new file mode 100644
index 000000000..4c6226892
--- /dev/null
+++ b/hashes/src/ripemd160/crypto.rs
@@ -0,0 +1,321 @@
+// SPDX-License-Identifier: CC0-1.0
+
+use super::{HashEngine, BLOCK_SIZE};
+
+#[cfg(feature = "small-hash")]
+#[macro_use]
+mod small_hash {
+    #[rustfmt::skip]
+    pub(super) fn round(a: u32, _b: u32, c: u32, _d: u32, e: u32,
+                        x: u32, bits: u32, add: u32, round: u32,
+    ) -> (u32, u32) {
+        let a = a.wrapping_add(round).wrapping_add(x).wrapping_add(add);
+        let a = a.rotate_left(bits).wrapping_add(e);
+        let c = c.rotate_left(10);
+
+        (a, c)
+    }
+
+    macro_rules! round(
+        ($a:expr, $b:expr, $c:expr, $d:expr, $e:expr,
+         $x:expr, $bits:expr, $add:expr, $round:expr) => ({
+            let updates = small_hash::round($a, $b, $c, $d, $e, $x, $bits, $add, $round);
+            $a = updates.0;
+            $c = updates.1;
+        });
+    );
+}
+
+#[cfg(not(feature = "small-hash"))]
+macro_rules! round(
+    ($a:expr, $b:expr, $c:expr, $d:expr, $e:expr,
+     $x:expr, $bits:expr, $add:expr, $round:expr) => ({
+        $a = $a.wrapping_add($round).wrapping_add($x).wrapping_add($add);
+        $a = $a.rotate_left($bits).wrapping_add($e);
+        $c = $c.rotate_left(10);
+    });
+);
+
+macro_rules! process_block(
+    ($h:expr, $data:expr,
+     $( round1: h_ordering $f0:expr, $f1:expr, $f2:expr, $f3:expr, $f4:expr;
+                data_index $data_index1:expr; roll_shift $bits1:expr; )*
+     $( round2: h_ordering $g0:expr, $g1:expr, $g2:expr, $g3:expr, $g4:expr;
+                data_index $data_index2:expr; roll_shift $bits2:expr; )*
+     $( round3: h_ordering $h0:expr, $h1:expr, $h2:expr, $h3:expr, $h4:expr;
+                data_index $data_index3:expr; roll_shift $bits3:expr; )*
+     $( round4: h_ordering $i0:expr, $i1:expr, $i2:expr, $i3:expr, $i4:expr;
+                data_index $data_index4:expr; roll_shift $bits4:expr; )*
+     $( round5: h_ordering $j0:expr, $j1:expr, $j2:expr, $j3:expr, $j4:expr;
+                data_index $data_index5:expr; roll_shift $bits5:expr; )*
+     $( par_round1: h_ordering $pj0:expr, $pj1:expr, $pj2:expr, $pj3:expr, $pj4:expr;
+                    data_index $pdata_index1:expr; roll_shift $pbits1:expr; )*
+     $( par_round2: h_ordering $pi0:expr, $pi1:expr, $pi2:expr, $pi3:expr, $pi4:expr;
+                    data_index $pdata_index2:expr; roll_shift $pbits2:expr; )*
+     $( par_round3: h_ordering $ph0:expr, $ph1:expr, $ph2:expr, $ph3:expr, $ph4:expr;
+                    data_index $pdata_index3:expr; roll_shift $pbits3:expr; )*
+     $( par_round4: h_ordering $pg0:expr, $pg1:expr, $pg2:expr, $pg3:expr, $pg4:expr;
+                    data_index $pdata_index4:expr; roll_shift $pbits4:expr; )*
+     $( par_round5: h_ordering $pf0:expr, $pf1:expr, $pf2:expr, $pf3:expr, $pf4:expr;
+                    data_index $pdata_index5:expr; roll_shift $pbits5:expr; )*
+    ) => ({
+        let mut bb = $h;
+        let mut bbb = $h;
+
+        // Round 1
+        $( round!(bb[$f0], bb[$f1], bb[$f2], bb[$f3], bb[$f4],
+                  $data[$data_index1], $bits1, 0x00000000,
+                  bb[$f1] ^ bb[$f2] ^ bb[$f3]); )*
+
+        // Round 2
+        $( round!(bb[$g0], bb[$g1], bb[$g2], bb[$g3], bb[$g4],
+                  $data[$data_index2], $bits2, 0x5a827999,
+                  (bb[$g1] & bb[$g2]) | (!bb[$g1] & bb[$g3])); )*
+
+        // Round 3
+        $( round!(bb[$h0], bb[$h1], bb[$h2], bb[$h3], bb[$h4],
+                  $data[$data_index3], $bits3, 0x6ed9eba1,
+                  (bb[$h1] | !bb[$h2]) ^ bb[$h3]); )*
+
+        // Round 4
+        $( round!(bb[$i0], bb[$i1], bb[$i2], bb[$i3], bb[$i4],
+                  $data[$data_index4], $bits4, 0x8f1bbcdc,
+                  (bb[$i1] & bb[$i3]) | (bb[$i2] & !bb[$i3])); )*
+
+        // Round 5
+        $( round!(bb[$j0], bb[$j1], bb[$j2], bb[$j3], bb[$j4],
+                  $data[$data_index5], $bits5, 0xa953fd4e,
+                  bb[$j1] ^ (bb[$j2] | !bb[$j3])); )*
+
+        // Parallel rounds: these are the same as the previous five
+        // rounds except that the constants have changed, we work
+        // with the other buffer, and they are applied in reverse
+        // order.
+
+        // Parallel Round 1
+        $( round!(bbb[$pj0], bbb[$pj1], bbb[$pj2], bbb[$pj3], bbb[$pj4],
+                  $data[$pdata_index1], $pbits1, 0x50a28be6,
+                  bbb[$pj1] ^ (bbb[$pj2] | !bbb[$pj3])); )*
+
+        // Porallel Round 2
+        $( round!(bbb[$pi0], bbb[$pi1], bbb[$pi2], bbb[$pi3], bbb[$pi4],
+                  $data[$pdata_index2], $pbits2, 0x5c4dd124,
+                  (bbb[$pi1] & bbb[$pi3]) | (bbb[$pi2] & !bbb[$pi3])); )*
+
+        // Parallel Round 3
+        $( round!(bbb[$ph0], bbb[$ph1], bbb[$ph2], bbb[$ph3], bbb[$ph4],
+                  $data[$pdata_index3], $pbits3, 0x6d703ef3,
+                  (bbb[$ph1] | !bbb[$ph2]) ^ bbb[$ph3]); )*
+
+        // Parallel Round 4
+        $( round!(bbb[$pg0], bbb[$pg1], bbb[$pg2], bbb[$pg3], bbb[$pg4],
+                  $data[$pdata_index4], $pbits4, 0x7a6d76e9,
+                  (bbb[$pg1] & bbb[$pg2]) | (!bbb[$pg1] & bbb[$pg3])); )*
+
+        // Parallel Round 5
+        $( round!(bbb[$pf0], bbb[$pf1], bbb[$pf2], bbb[$pf3], bbb[$pf4],
+                  $data[$pdata_index5], $pbits5, 0x00000000,
+                  bbb[$pf1] ^ bbb[$pf2] ^ bbb[$pf3]); )*
+
+        // Combine results
+        bbb[3] = bbb[3].wrapping_add($h[1]).wrapping_add(bb[2]);
+        $h[1]  =  $h[2].wrapping_add(bb[3]).wrapping_add(bbb[4]);
+        $h[2]  =  $h[3].wrapping_add(bb[4]).wrapping_add(bbb[0]);
+        $h[3]  =  $h[4].wrapping_add(bb[0]).wrapping_add(bbb[1]);
+        $h[4]  =  $h[0].wrapping_add(bb[1]).wrapping_add(bbb[2]);
+        $h[0]  =                                         bbb[3];
+    });
+);
+
+impl HashEngine {
+    pub(super) fn process_block(&mut self) {
+        debug_assert_eq!(self.buffer.len(), BLOCK_SIZE);
+
+        let mut w = [0u32; 16];
+        for (w_val, buff_bytes) in w.iter_mut().zip(self.buffer.chunks_exact(4)) {
+            *w_val = u32::from_le_bytes(buff_bytes.try_into().expect("4 byte slice"))
+        }
+
+        process_block!(self.h, w,
+            // Round 1
+            round1: h_ordering 0, 1, 2, 3, 4; data_index  0; roll_shift 11;
+            round1: h_ordering 4, 0, 1, 2, 3; data_index  1; roll_shift 14;
+            round1: h_ordering 3, 4, 0, 1, 2; data_index  2; roll_shift 15;
+            round1: h_ordering 2, 3, 4, 0, 1; data_index  3; roll_shift 12;
+            round1: h_ordering 1, 2, 3, 4, 0; data_index  4; roll_shift  5;
+            round1: h_ordering 0, 1, 2, 3, 4; data_index  5; roll_shift  8;
+            round1: h_ordering 4, 0, 1, 2, 3; data_index  6; roll_shift  7;
+            round1: h_ordering 3, 4, 0, 1, 2; data_index  7; roll_shift  9;
+            round1: h_ordering 2, 3, 4, 0, 1; data_index  8; roll_shift 11;
+            round1: h_ordering 1, 2, 3, 4, 0; data_index  9; roll_shift 13;
+            round1: h_ordering 0, 1, 2, 3, 4; data_index 10; roll_shift 14;
+            round1: h_ordering 4, 0, 1, 2, 3; data_index 11; roll_shift 15;
+            round1: h_ordering 3, 4, 0, 1, 2; data_index 12; roll_shift  6;
+            round1: h_ordering 2, 3, 4, 0, 1; data_index 13; roll_shift  7;
+            round1: h_ordering 1, 2, 3, 4, 0; data_index 14; roll_shift  9;
+            round1: h_ordering 0, 1, 2, 3, 4; data_index 15; roll_shift  8;
+
+            // Round 2
+            round2: h_ordering 4, 0, 1, 2, 3; data_index  7; roll_shift  7;
+            round2: h_ordering 3, 4, 0, 1, 2; data_index  4; roll_shift  6;
+            round2: h_ordering 2, 3, 4, 0, 1; data_index 13; roll_shift  8;
+            round2: h_ordering 1, 2, 3, 4, 0; data_index  1; roll_shift 13;
+            round2: h_ordering 0, 1, 2, 3, 4; data_index 10; roll_shift 11;
+            round2: h_ordering 4, 0, 1, 2, 3; data_index  6; roll_shift  9;
+            round2: h_ordering 3, 4, 0, 1, 2; data_index 15; roll_shift  7;
+            round2: h_ordering 2, 3, 4, 0, 1; data_index  3; roll_shift 15;
+            round2: h_ordering 1, 2, 3, 4, 0; data_index 12; roll_shift  7;
+            round2: h_ordering 0, 1, 2, 3, 4; data_index  0; roll_shift 12;
+            round2: h_ordering 4, 0, 1, 2, 3; data_index  9; roll_shift 15;
+            round2: h_ordering 3, 4, 0, 1, 2; data_index  5; roll_shift  9;
+            round2: h_ordering 2, 3, 4, 0, 1; data_index  2; roll_shift 11;
+            round2: h_ordering 1, 2, 3, 4, 0; data_index 14; roll_shift  7;
+            round2: h_ordering 0, 1, 2, 3, 4; data_index 11; roll_shift 13;
+            round2: h_ordering 4, 0, 1, 2, 3; data_index  8; roll_shift 12;
+
+            // Round 3
+            round3: h_ordering 3, 4, 0, 1, 2; data_index  3; roll_shift 11;
+            round3: h_ordering 2, 3, 4, 0, 1; data_index 10; roll_shift 13;
+            round3: h_ordering 1, 2, 3, 4, 0; data_index 14; roll_shift  6;
+            round3: h_ordering 0, 1, 2, 3, 4; data_index  4; roll_shift  7;
+            round3: h_ordering 4, 0, 1, 2, 3; data_index  9; roll_shift 14;
+            round3: h_ordering 3, 4, 0, 1, 2; data_index 15; roll_shift  9;
+            round3: h_ordering 2, 3, 4, 0, 1; data_index  8; roll_shift 13;
+            round3: h_ordering 1, 2, 3, 4, 0; data_index  1; roll_shift 15;
+            round3: h_ordering 0, 1, 2, 3, 4; data_index  2; roll_shift 14;
+            round3: h_ordering 4, 0, 1, 2, 3; data_index  7; roll_shift  8;
+            round3: h_ordering 3, 4, 0, 1, 2; data_index  0; roll_shift 13;
+            round3: h_ordering 2, 3, 4, 0, 1; data_index  6; roll_shift  6;
+            round3: h_ordering 1, 2, 3, 4, 0; data_index 13; roll_shift  5;
+            round3: h_ordering 0, 1, 2, 3, 4; data_index 11; roll_shift 12;
+            round3: h_ordering 4, 0, 1, 2, 3; data_index  5; roll_shift  7;
+            round3: h_ordering 3, 4, 0, 1, 2; data_index 12; roll_shift  5;
+
+            // Round 4
+            round4: h_ordering 2, 3, 4, 0, 1; data_index  1; roll_shift 11;
+            round4: h_ordering 1, 2, 3, 4, 0; data_index  9; roll_shift 12;
+            round4: h_ordering 0, 1, 2, 3, 4; data_index 11; roll_shift 14;
+            round4: h_ordering 4, 0, 1, 2, 3; data_index 10; roll_shift 15;
+            round4: h_ordering 3, 4, 0, 1, 2; data_index  0; roll_shift 14;
+            round4: h_ordering 2, 3, 4, 0, 1; data_index  8; roll_shift 15;
+            round4: h_ordering 1, 2, 3, 4, 0; data_index 12; roll_shift  9;
+            round4: h_ordering 0, 1, 2, 3, 4; data_index  4; roll_shift  8;
+            round4: h_ordering 4, 0, 1, 2, 3; data_index 13; roll_shift  9;
+            round4: h_ordering 3, 4, 0, 1, 2; data_index  3; roll_shift 14;
+            round4: h_ordering 2, 3, 4, 0, 1; data_index  7; roll_shift  5;
+            round4: h_ordering 1, 2, 3, 4, 0; data_index 15; roll_shift  6;
+            round4: h_ordering 0, 1, 2, 3, 4; data_index 14; roll_shift  8;
+            round4: h_ordering 4, 0, 1, 2, 3; data_index  5; roll_shift  6;
+            round4: h_ordering 3, 4, 0, 1, 2; data_index  6; roll_shift  5;
+            round4: h_ordering 2, 3, 4, 0, 1; data_index  2; roll_shift 12;
+
+            // Round 5
+            round5: h_ordering 1, 2, 3, 4, 0; data_index  4; roll_shift  9;
+            round5: h_ordering 0, 1, 2, 3, 4; data_index  0; roll_shift 15;
+            round5: h_ordering 4, 0, 1, 2, 3; data_index  5; roll_shift  5;
+            round5: h_ordering 3, 4, 0, 1, 2; data_index  9; roll_shift 11;
+            round5: h_ordering 2, 3, 4, 0, 1; data_index  7; roll_shift  6;
+            round5: h_ordering 1, 2, 3, 4, 0; data_index 12; roll_shift  8;
+            round5: h_ordering 0, 1, 2, 3, 4; data_index  2; roll_shift 13;
+            round5: h_ordering 4, 0, 1, 2, 3; data_index 10; roll_shift 12;
+            round5: h_ordering 3, 4, 0, 1, 2; data_index 14; roll_shift  5;
+            round5: h_ordering 2, 3, 4, 0, 1; data_index  1; roll_shift 12;
+            round5: h_ordering 1, 2, 3, 4, 0; data_index  3; roll_shift 13;
+            round5: h_ordering 0, 1, 2, 3, 4; data_index  8; roll_shift 14;
+            round5: h_ordering 4, 0, 1, 2, 3; data_index 11; roll_shift 11;
+            round5: h_ordering 3, 4, 0, 1, 2; data_index  6; roll_shift  8;
+            round5: h_ordering 2, 3, 4, 0, 1; data_index 15; roll_shift  5;
+            round5: h_ordering 1, 2, 3, 4, 0; data_index 13; roll_shift  6;
+
+            // Porallel Round 1;
+            par_round1: h_ordering 0, 1, 2, 3, 4; data_index  5; roll_shift  8;
+            par_round1: h_ordering 4, 0, 1, 2, 3; data_index 14; roll_shift  9;
+            par_round1: h_ordering 3, 4, 0, 1, 2; data_index  7; roll_shift  9;
+            par_round1: h_ordering 2, 3, 4, 0, 1; data_index  0; roll_shift 11;
+            par_round1: h_ordering 1, 2, 3, 4, 0; data_index  9; roll_shift 13;
+            par_round1: h_ordering 0, 1, 2, 3, 4; data_index  2; roll_shift 15;
+            par_round1: h_ordering 4, 0, 1, 2, 3; data_index 11; roll_shift 15;
+            par_round1: h_ordering 3, 4, 0, 1, 2; data_index  4; roll_shift  5;
+            par_round1: h_ordering 2, 3, 4, 0, 1; data_index 13; roll_shift  7;
+            par_round1: h_ordering 1, 2, 3, 4, 0; data_index  6; roll_shift  7;
+            par_round1: h_ordering 0, 1, 2, 3, 4; data_index 15; roll_shift  8;
+            par_round1: h_ordering 4, 0, 1, 2, 3; data_index  8; roll_shift 11;
+            par_round1: h_ordering 3, 4, 0, 1, 2; data_index  1; roll_shift 14;
+            par_round1: h_ordering 2, 3, 4, 0, 1; data_index 10; roll_shift 14;
+            par_round1: h_ordering 1, 2, 3, 4, 0; data_index  3; roll_shift 12;
+            par_round1: h_ordering 0, 1, 2, 3, 4; data_index 12; roll_shift  6;
+
+            // Parallel Round 2
+            par_round2: h_ordering 4, 0, 1, 2, 3; data_index  6; roll_shift  9;
+            par_round2: h_ordering 3, 4, 0, 1, 2; data_index 11; roll_shift 13;
+            par_round2: h_ordering 2, 3, 4, 0, 1; data_index  3; roll_shift 15;
+            par_round2: h_ordering 1, 2, 3, 4, 0; data_index  7; roll_shift  7;
+            par_round2: h_ordering 0, 1, 2, 3, 4; data_index  0; roll_shift 12;
+            par_round2: h_ordering 4, 0, 1, 2, 3; data_index 13; roll_shift  8;
+            par_round2: h_ordering 3, 4, 0, 1, 2; data_index  5; roll_shift  9;
+            par_round2: h_ordering 2, 3, 4, 0, 1; data_index 10; roll_shift 11;
+            par_round2: h_ordering 1, 2, 3, 4, 0; data_index 14; roll_shift  7;
+            par_round2: h_ordering 0, 1, 2, 3, 4; data_index 15; roll_shift  7;
+            par_round2: h_ordering 4, 0, 1, 2, 3; data_index  8; roll_shift 12;
+            par_round2: h_ordering 3, 4, 0, 1, 2; data_index 12; roll_shift  7;
+            par_round2: h_ordering 2, 3, 4, 0, 1; data_index  4; roll_shift  6;
+            par_round2: h_ordering 1, 2, 3, 4, 0; data_index  9; roll_shift 15;
+            par_round2: h_ordering 0, 1, 2, 3, 4; data_index  1; roll_shift 13;
+            par_round2: h_ordering 4, 0, 1, 2, 3; data_index  2; roll_shift 11;
+
+            // Parallel Round 3
+            par_round3: h_ordering 3, 4, 0, 1, 2; data_index 15; roll_shift  9;
+            par_round3: h_ordering 2, 3, 4, 0, 1; data_index  5; roll_shift  7;
+            par_round3: h_ordering 1, 2, 3, 4, 0; data_index  1; roll_shift 15;
+            par_round3: h_ordering 0, 1, 2, 3, 4; data_index  3; roll_shift 11;
+            par_round3: h_ordering 4, 0, 1, 2, 3; data_index  7; roll_shift  8;
+            par_round3: h_ordering 3, 4, 0, 1, 2; data_index 14; roll_shift  6;
+            par_round3: h_ordering 2, 3, 4, 0, 1; data_index  6; roll_shift  6;
+            par_round3: h_ordering 1, 2, 3, 4, 0; data_index  9; roll_shift 14;
+            par_round3: h_ordering 0, 1, 2, 3, 4; data_index 11; roll_shift 12;
+            par_round3: h_ordering 4, 0, 1, 2, 3; data_index  8; roll_shift 13;
+            par_round3: h_ordering 3, 4, 0, 1, 2; data_index 12; roll_shift  5;
+            par_round3: h_ordering 2, 3, 4, 0, 1; data_index  2; roll_shift 14;
+            par_round3: h_ordering 1, 2, 3, 4, 0; data_index 10; roll_shift 13;
+            par_round3: h_ordering 0, 1, 2, 3, 4; data_index  0; roll_shift 13;
+            par_round3: h_ordering 4, 0, 1, 2, 3; data_index  4; roll_shift  7;
+            par_round3: h_ordering 3, 4, 0, 1, 2; data_index 13; roll_shift  5;
+
+            // Parallel Round 4
+            par_round4: h_ordering 2, 3, 4, 0, 1; data_index  8; roll_shift 15;
+            par_round4: h_ordering 1, 2, 3, 4, 0; data_index  6; roll_shift  5;
+            par_round4: h_ordering 0, 1, 2, 3, 4; data_index  4; roll_shift  8;
+            par_round4: h_ordering 4, 0, 1, 2, 3; data_index  1; roll_shift 11;
+            par_round4: h_ordering 3, 4, 0, 1, 2; data_index  3; roll_shift 14;
+            par_round4: h_ordering 2, 3, 4, 0, 1; data_index 11; roll_shift 14;
+            par_round4: h_ordering 1, 2, 3, 4, 0; data_index 15; roll_shift  6;
+            par_round4: h_ordering 0, 1, 2, 3, 4; data_index  0; roll_shift 14;
+            par_round4: h_ordering 4, 0, 1, 2, 3; data_index  5; roll_shift  6;
+            par_round4: h_ordering 3, 4, 0, 1, 2; data_index 12; roll_shift  9;
+            par_round4: h_ordering 2, 3, 4, 0, 1; data_index  2; roll_shift 12;
+            par_round4: h_ordering 1, 2, 3, 4, 0; data_index 13; roll_shift  9;
+            par_round4: h_ordering 0, 1, 2, 3, 4; data_index  9; roll_shift 12;
+            par_round4: h_ordering 4, 0, 1, 2, 3; data_index  7; roll_shift  5;
+            par_round4: h_ordering 3, 4, 0, 1, 2; data_index 10; roll_shift 15;
+            par_round4: h_ordering 2, 3, 4, 0, 1; data_index 14; roll_shift  8;
+
+            // Parallel Round 5
+            par_round5: h_ordering 1, 2, 3, 4, 0; data_index 12; roll_shift  8;
+            par_round5: h_ordering 0, 1, 2, 3, 4; data_index 15; roll_shift  5;
+            par_round5: h_ordering 4, 0, 1, 2, 3; data_index 10; roll_shift 12;
+            par_round5: h_ordering 3, 4, 0, 1, 2; data_index  4; roll_shift  9;
+            par_round5: h_ordering 2, 3, 4, 0, 1; data_index  1; roll_shift 12;
+            par_round5: h_ordering 1, 2, 3, 4, 0; data_index  5; roll_shift  5;
+            par_round5: h_ordering 0, 1, 2, 3, 4; data_index  8; roll_shift 14;
+            par_round5: h_ordering 4, 0, 1, 2, 3; data_index  7; roll_shift  6;
+            par_round5: h_ordering 3, 4, 0, 1, 2; data_index  6; roll_shift  8;
+            par_round5: h_ordering 2, 3, 4, 0, 1; data_index  2; roll_shift 13;
+            par_round5: h_ordering 1, 2, 3, 4, 0; data_index 13; roll_shift  6;
+            par_round5: h_ordering 0, 1, 2, 3, 4; data_index 14; roll_shift  5;
+            par_round5: h_ordering 4, 0, 1, 2, 3; data_index  0; roll_shift 15;
+            par_round5: h_ordering 3, 4, 0, 1, 2; data_index  3; roll_shift 13;
+            par_round5: h_ordering 2, 3, 4, 0, 1; data_index  9; roll_shift 11;
+            par_round5: h_ordering 1, 2, 3, 4, 0; data_index 11; roll_shift 11;
+        );
+    }
+}
diff --git a/hashes/src/ripemd160/mod.rs b/hashes/src/ripemd160/mod.rs
index fa0893d7a..8ef659770 100644
--- a/hashes/src/ripemd160/mod.rs
+++ b/hashes/src/ripemd160/mod.rs
@@ -2,6 +2,8 @@
 
 //! RIPEMD160 implementation.
 
+mod crypto;
+
 use core::cmp;
 
 use crate::{incomplete_block_len, HashEngine as _};
@@ -88,324 +90,6 @@ impl crate::HashEngine for HashEngine {
     crate::internal_macros::engine_input_impl!();
 }
 
-#[cfg(feature = "small-hash")]
-#[macro_use]
-mod small_hash {
-    #[rustfmt::skip]
-    pub(super) fn round(a: u32, _b: u32, c: u32, _d: u32, e: u32,
-                        x: u32, bits: u32, add: u32, round: u32,
-    ) -> (u32, u32) {
-        let a = a.wrapping_add(round).wrapping_add(x).wrapping_add(add);
-        let a = a.rotate_left(bits).wrapping_add(e);
-        let c = c.rotate_left(10);
-
-        (a, c)
-    }
-
-    macro_rules! round(
-        ($a:expr, $b:expr, $c:expr, $d:expr, $e:expr,
-         $x:expr, $bits:expr, $add:expr, $round:expr) => ({
-            let updates = small_hash::round($a, $b, $c, $d, $e, $x, $bits, $add, $round);
-            $a = updates.0;
-            $c = updates.1;
-        });
-    );
-}
-
-#[cfg(not(feature = "small-hash"))]
-macro_rules! round(
-    ($a:expr, $b:expr, $c:expr, $d:expr, $e:expr,
-     $x:expr, $bits:expr, $add:expr, $round:expr) => ({
-        $a = $a.wrapping_add($round).wrapping_add($x).wrapping_add($add);
-        $a = $a.rotate_left($bits).wrapping_add($e);
-        $c = $c.rotate_left(10);
-    });
-);
-
-macro_rules! process_block(
-    ($h:expr, $data:expr,
-     $( round1: h_ordering $f0:expr, $f1:expr, $f2:expr, $f3:expr, $f4:expr;
-                data_index $data_index1:expr; roll_shift $bits1:expr; )*
-     $( round2: h_ordering $g0:expr, $g1:expr, $g2:expr, $g3:expr, $g4:expr;
-                data_index $data_index2:expr; roll_shift $bits2:expr; )*
-     $( round3: h_ordering $h0:expr, $h1:expr, $h2:expr, $h3:expr, $h4:expr;
-                data_index $data_index3:expr; roll_shift $bits3:expr; )*
-     $( round4: h_ordering $i0:expr, $i1:expr, $i2:expr, $i3:expr, $i4:expr;
-                data_index $data_index4:expr; roll_shift $bits4:expr; )*
-     $( round5: h_ordering $j0:expr, $j1:expr, $j2:expr, $j3:expr, $j4:expr;
-                data_index $data_index5:expr; roll_shift $bits5:expr; )*
-     $( par_round1: h_ordering $pj0:expr, $pj1:expr, $pj2:expr, $pj3:expr, $pj4:expr;
-                    data_index $pdata_index1:expr; roll_shift $pbits1:expr; )*
-     $( par_round2: h_ordering $pi0:expr, $pi1:expr, $pi2:expr, $pi3:expr, $pi4:expr;
-                    data_index $pdata_index2:expr; roll_shift $pbits2:expr; )*
-     $( par_round3: h_ordering $ph0:expr, $ph1:expr, $ph2:expr, $ph3:expr, $ph4:expr;
-                    data_index $pdata_index3:expr; roll_shift $pbits3:expr; )*
-     $( par_round4: h_ordering $pg0:expr, $pg1:expr, $pg2:expr, $pg3:expr, $pg4:expr;
-                    data_index $pdata_index4:expr; roll_shift $pbits4:expr; )*
-     $( par_round5: h_ordering $pf0:expr, $pf1:expr, $pf2:expr, $pf3:expr, $pf4:expr;
-                    data_index $pdata_index5:expr; roll_shift $pbits5:expr; )*
-    ) => ({
-        let mut bb = $h;
-        let mut bbb = $h;
-
-        // Round 1
-        $( round!(bb[$f0], bb[$f1], bb[$f2], bb[$f3], bb[$f4],
-                  $data[$data_index1], $bits1, 0x00000000,
-                  bb[$f1] ^ bb[$f2] ^ bb[$f3]); )*
-
-        // Round 2
-        $( round!(bb[$g0], bb[$g1], bb[$g2], bb[$g3], bb[$g4],
-                  $data[$data_index2], $bits2, 0x5a827999,
-                  (bb[$g1] & bb[$g2]) | (!bb[$g1] & bb[$g3])); )*
-
-        // Round 3
-        $( round!(bb[$h0], bb[$h1], bb[$h2], bb[$h3], bb[$h4],
-                  $data[$data_index3], $bits3, 0x6ed9eba1,
-                  (bb[$h1] | !bb[$h2]) ^ bb[$h3]); )*
-
-        // Round 4
-        $( round!(bb[$i0], bb[$i1], bb[$i2], bb[$i3], bb[$i4],
-                  $data[$data_index4], $bits4, 0x8f1bbcdc,
-                  (bb[$i1] & bb[$i3]) | (bb[$i2] & !bb[$i3])); )*
-
-        // Round 5
-        $( round!(bb[$j0], bb[$j1], bb[$j2], bb[$j3], bb[$j4],
-                  $data[$data_index5], $bits5, 0xa953fd4e,
-                  bb[$j1] ^ (bb[$j2] | !bb[$j3])); )*
-
-        // Parallel rounds: these are the same as the previous five
-        // rounds except that the constants have changed, we work
-        // with the other buffer, and they are applied in reverse
-        // order.
-
-        // Parallel Round 1
-        $( round!(bbb[$pj0], bbb[$pj1], bbb[$pj2], bbb[$pj3], bbb[$pj4],
-                  $data[$pdata_index1], $pbits1, 0x50a28be6,
-                  bbb[$pj1] ^ (bbb[$pj2] | !bbb[$pj3])); )*
-
-        // Porallel Round 2
-        $( round!(bbb[$pi0], bbb[$pi1], bbb[$pi2], bbb[$pi3], bbb[$pi4],
-                  $data[$pdata_index2], $pbits2, 0x5c4dd124,
-                  (bbb[$pi1] & bbb[$pi3]) | (bbb[$pi2] & !bbb[$pi3])); )*
-
-        // Parallel Round 3
-        $( round!(bbb[$ph0], bbb[$ph1], bbb[$ph2], bbb[$ph3], bbb[$ph4],
-                  $data[$pdata_index3], $pbits3, 0x6d703ef3,
-                  (bbb[$ph1] | !bbb[$ph2]) ^ bbb[$ph3]); )*
-
-        // Parallel Round 4
-        $( round!(bbb[$pg0], bbb[$pg1], bbb[$pg2], bbb[$pg3], bbb[$pg4],
-                  $data[$pdata_index4], $pbits4, 0x7a6d76e9,
-                  (bbb[$pg1] & bbb[$pg2]) | (!bbb[$pg1] & bbb[$pg3])); )*
-
-        // Parallel Round 5
-        $( round!(bbb[$pf0], bbb[$pf1], bbb[$pf2], bbb[$pf3], bbb[$pf4],
-                  $data[$pdata_index5], $pbits5, 0x00000000,
-                  bbb[$pf1] ^ bbb[$pf2] ^ bbb[$pf3]); )*
-
-        // Combine results
-        bbb[3] = bbb[3].wrapping_add($h[1]).wrapping_add(bb[2]);
-        $h[1]  =  $h[2].wrapping_add(bb[3]).wrapping_add(bbb[4]);
-        $h[2]  =  $h[3].wrapping_add(bb[4]).wrapping_add(bbb[0]);
-        $h[3]  =  $h[4].wrapping_add(bb[0]).wrapping_add(bbb[1]);
-        $h[4]  =  $h[0].wrapping_add(bb[1]).wrapping_add(bbb[2]);
-        $h[0]  =                                         bbb[3];
-    });
-);
-
-impl HashEngine {
-    fn process_block(&mut self) {
-        debug_assert_eq!(self.buffer.len(), BLOCK_SIZE);
-
-        let mut w = [0u32; 16];
-        for (w_val, buff_bytes) in w.iter_mut().zip(self.buffer.chunks_exact(4)) {
-            *w_val = u32::from_le_bytes(buff_bytes.try_into().expect("4 byte slice"))
-        }
-
-        process_block!(self.h, w,
-            // Round 1
-            round1: h_ordering 0, 1, 2, 3, 4; data_index  0; roll_shift 11;
-            round1: h_ordering 4, 0, 1, 2, 3; data_index  1; roll_shift 14;
-            round1: h_ordering 3, 4, 0, 1, 2; data_index  2; roll_shift 15;
-            round1: h_ordering 2, 3, 4, 0, 1; data_index  3; roll_shift 12;
-            round1: h_ordering 1, 2, 3, 4, 0; data_index  4; roll_shift  5;
-            round1: h_ordering 0, 1, 2, 3, 4; data_index  5; roll_shift  8;
-            round1: h_ordering 4, 0, 1, 2, 3; data_index  6; roll_shift  7;
-            round1: h_ordering 3, 4, 0, 1, 2; data_index  7; roll_shift  9;
-            round1: h_ordering 2, 3, 4, 0, 1; data_index  8; roll_shift 11;
-            round1: h_ordering 1, 2, 3, 4, 0; data_index  9; roll_shift 13;
-            round1: h_ordering 0, 1, 2, 3, 4; data_index 10; roll_shift 14;
-            round1: h_ordering 4, 0, 1, 2, 3; data_index 11; roll_shift 15;
-            round1: h_ordering 3, 4, 0, 1, 2; data_index 12; roll_shift  6;
-            round1: h_ordering 2, 3, 4, 0, 1; data_index 13; roll_shift  7;
-            round1: h_ordering 1, 2, 3, 4, 0; data_index 14; roll_shift  9;
-            round1: h_ordering 0, 1, 2, 3, 4; data_index 15; roll_shift  8;
-
-            // Round 2
-            round2: h_ordering 4, 0, 1, 2, 3; data_index  7; roll_shift  7;
-            round2: h_ordering 3, 4, 0, 1, 2; data_index  4; roll_shift  6;
-            round2: h_ordering 2, 3, 4, 0, 1; data_index 13; roll_shift  8;
-            round2: h_ordering 1, 2, 3, 4, 0; data_index  1; roll_shift 13;
-            round2: h_ordering 0, 1, 2, 3, 4; data_index 10; roll_shift 11;
-            round2: h_ordering 4, 0, 1, 2, 3; data_index  6; roll_shift  9;
-            round2: h_ordering 3, 4, 0, 1, 2; data_index 15; roll_shift  7;
-            round2: h_ordering 2, 3, 4, 0, 1; data_index  3; roll_shift 15;
-            round2: h_ordering 1, 2, 3, 4, 0; data_index 12; roll_shift  7;
-            round2: h_ordering 0, 1, 2, 3, 4; data_index  0; roll_shift 12;
-            round2: h_ordering 4, 0, 1, 2, 3; data_index  9; roll_shift 15;
-            round2: h_ordering 3, 4, 0, 1, 2; data_index  5; roll_shift  9;
-            round2: h_ordering 2, 3, 4, 0, 1; data_index  2; roll_shift 11;
-            round2: h_ordering 1, 2, 3, 4, 0; data_index 14; roll_shift  7;
-            round2: h_ordering 0, 1, 2, 3, 4; data_index 11; roll_shift 13;
-            round2: h_ordering 4, 0, 1, 2, 3; data_index  8; roll_shift 12;
-
-            // Round 3
-            round3: h_ordering 3, 4, 0, 1, 2; data_index  3; roll_shift 11;
-            round3: h_ordering 2, 3, 4, 0, 1; data_index 10; roll_shift 13;
-            round3: h_ordering 1, 2, 3, 4, 0; data_index 14; roll_shift  6;
-            round3: h_ordering 0, 1, 2, 3, 4; data_index  4; roll_shift  7;
-            round3: h_ordering 4, 0, 1, 2, 3; data_index  9; roll_shift 14;
-            round3: h_ordering 3, 4, 0, 1, 2; data_index 15; roll_shift  9;
-            round3: h_ordering 2, 3, 4, 0, 1; data_index  8; roll_shift 13;
-            round3: h_ordering 1, 2, 3, 4, 0; data_index  1; roll_shift 15;
-            round3: h_ordering 0, 1, 2, 3, 4; data_index  2; roll_shift 14;
-            round3: h_ordering 4, 0, 1, 2, 3; data_index  7; roll_shift  8;
-            round3: h_ordering 3, 4, 0, 1, 2; data_index  0; roll_shift 13;
-            round3: h_ordering 2, 3, 4, 0, 1; data_index  6; roll_shift  6;
-            round3: h_ordering 1, 2, 3, 4, 0; data_index 13; roll_shift  5;
-            round3: h_ordering 0, 1, 2, 3, 4; data_index 11; roll_shift 12;
-            round3: h_ordering 4, 0, 1, 2, 3; data_index  5; roll_shift  7;
-            round3: h_ordering 3, 4, 0, 1, 2; data_index 12; roll_shift  5;
-
-            // Round 4
-            round4: h_ordering 2, 3, 4, 0, 1; data_index  1; roll_shift 11;
-            round4: h_ordering 1, 2, 3, 4, 0; data_index  9; roll_shift 12;
-            round4: h_ordering 0, 1, 2, 3, 4; data_index 11; roll_shift 14;
-            round4: h_ordering 4, 0, 1, 2, 3; data_index 10; roll_shift 15;
-            round4: h_ordering 3, 4, 0, 1, 2; data_index  0; roll_shift 14;
-            round4: h_ordering 2, 3, 4, 0, 1; data_index  8; roll_shift 15;
-            round4: h_ordering 1, 2, 3, 4, 0; data_index 12; roll_shift  9;
-            round4: h_ordering 0, 1, 2, 3, 4; data_index  4; roll_shift  8;
-            round4: h_ordering 4, 0, 1, 2, 3; data_index 13; roll_shift  9;
-            round4: h_ordering 3, 4, 0, 1, 2; data_index  3; roll_shift 14;
-            round4: h_ordering 2, 3, 4, 0, 1; data_index  7; roll_shift  5;
-            round4: h_ordering 1, 2, 3, 4, 0; data_index 15; roll_shift  6;
-            round4: h_ordering 0, 1, 2, 3, 4; data_index 14; roll_shift  8;
-            round4: h_ordering 4, 0, 1, 2, 3; data_index  5; roll_shift  6;
-            round4: h_ordering 3, 4, 0, 1, 2; data_index  6; roll_shift  5;
-            round4: h_ordering 2, 3, 4, 0, 1; data_index  2; roll_shift 12;
-
-            // Round 5
-            round5: h_ordering 1, 2, 3, 4, 0; data_index  4; roll_shift  9;
-            round5: h_ordering 0, 1, 2, 3, 4; data_index  0; roll_shift 15;
-            round5: h_ordering 4, 0, 1, 2, 3; data_index  5; roll_shift  5;
-            round5: h_ordering 3, 4, 0, 1, 2; data_index  9; roll_shift 11;
-            round5: h_ordering 2, 3, 4, 0, 1; data_index  7; roll_shift  6;
-            round5: h_ordering 1, 2, 3, 4, 0; data_index 12; roll_shift  8;
-            round5: h_ordering 0, 1, 2, 3, 4; data_index  2; roll_shift 13;
-            round5: h_ordering 4, 0, 1, 2, 3; data_index 10; roll_shift 12;
-            round5: h_ordering 3, 4, 0, 1, 2; data_index 14; roll_shift  5;
-            round5: h_ordering 2, 3, 4, 0, 1; data_index  1; roll_shift 12;
-            round5: h_ordering 1, 2, 3, 4, 0; data_index  3; roll_shift 13;
-            round5: h_ordering 0, 1, 2, 3, 4; data_index  8; roll_shift 14;
-            round5: h_ordering 4, 0, 1, 2, 3; data_index 11; roll_shift 11;
-            round5: h_ordering 3, 4, 0, 1, 2; data_index  6; roll_shift  8;
-            round5: h_ordering 2, 3, 4, 0, 1; data_index 15; roll_shift  5;
-            round5: h_ordering 1, 2, 3, 4, 0; data_index 13; roll_shift  6;
-
-            // Porallel Round 1;
-            par_round1: h_ordering 0, 1, 2, 3, 4; data_index  5; roll_shift  8;
-            par_round1: h_ordering 4, 0, 1, 2, 3; data_index 14; roll_shift  9;
-            par_round1: h_ordering 3, 4, 0, 1, 2; data_index  7; roll_shift  9;
-            par_round1: h_ordering 2, 3, 4, 0, 1; data_index  0; roll_shift 11;
-            par_round1: h_ordering 1, 2, 3, 4, 0; data_index  9; roll_shift 13;
-            par_round1: h_ordering 0, 1, 2, 3, 4; data_index  2; roll_shift 15;
-            par_round1: h_ordering 4, 0, 1, 2, 3; data_index 11; roll_shift 15;
-            par_round1: h_ordering 3, 4, 0, 1, 2; data_index  4; roll_shift  5;
-            par_round1: h_ordering 2, 3, 4, 0, 1; data_index 13; roll_shift  7;
-            par_round1: h_ordering 1, 2, 3, 4, 0; data_index  6; roll_shift  7;
-            par_round1: h_ordering 0, 1, 2, 3, 4; data_index 15; roll_shift  8;
-            par_round1: h_ordering 4, 0, 1, 2, 3; data_index  8; roll_shift 11;
-            par_round1: h_ordering 3, 4, 0, 1, 2; data_index  1; roll_shift 14;
-            par_round1: h_ordering 2, 3, 4, 0, 1; data_index 10; roll_shift 14;
-            par_round1: h_ordering 1, 2, 3, 4, 0; data_index  3; roll_shift 12;
-            par_round1: h_ordering 0, 1, 2, 3, 4; data_index 12; roll_shift  6;
-
-            // Parallel Round 2
-            par_round2: h_ordering 4, 0, 1, 2, 3; data_index  6; roll_shift  9;
-            par_round2: h_ordering 3, 4, 0, 1, 2; data_index 11; roll_shift 13;
-            par_round2: h_ordering 2, 3, 4, 0, 1; data_index  3; roll_shift 15;
-            par_round2: h_ordering 1, 2, 3, 4, 0; data_index  7; roll_shift  7;
-            par_round2: h_ordering 0, 1, 2, 3, 4; data_index  0; roll_shift 12;
-            par_round2: h_ordering 4, 0, 1, 2, 3; data_index 13; roll_shift  8;
-            par_round2: h_ordering 3, 4, 0, 1, 2; data_index  5; roll_shift  9;
-            par_round2: h_ordering 2, 3, 4, 0, 1; data_index 10; roll_shift 11;
-            par_round2: h_ordering 1, 2, 3, 4, 0; data_index 14; roll_shift  7;
-            par_round2: h_ordering 0, 1, 2, 3, 4; data_index 15; roll_shift  7;
-            par_round2: h_ordering 4, 0, 1, 2, 3; data_index  8; roll_shift 12;
-            par_round2: h_ordering 3, 4, 0, 1, 2; data_index 12; roll_shift  7;
-            par_round2: h_ordering 2, 3, 4, 0, 1; data_index  4; roll_shift  6;
-            par_round2: h_ordering 1, 2, 3, 4, 0; data_index  9; roll_shift 15;
-            par_round2: h_ordering 0, 1, 2, 3, 4; data_index  1; roll_shift 13;
-            par_round2: h_ordering 4, 0, 1, 2, 3; data_index  2; roll_shift 11;
-
-            // Parallel Round 3
-            par_round3: h_ordering 3, 4, 0, 1, 2; data_index 15; roll_shift  9;
-            par_round3: h_ordering 2, 3, 4, 0, 1; data_index  5; roll_shift  7;
-            par_round3: h_ordering 1, 2, 3, 4, 0; data_index  1; roll_shift 15;
-            par_round3: h_ordering 0, 1, 2, 3, 4; data_index  3; roll_shift 11;
-            par_round3: h_ordering 4, 0, 1, 2, 3; data_index  7; roll_shift  8;
-            par_round3: h_ordering 3, 4, 0, 1, 2; data_index 14; roll_shift  6;
-            par_round3: h_ordering 2, 3, 4, 0, 1; data_index  6; roll_shift  6;
-            par_round3: h_ordering 1, 2, 3, 4, 0; data_index  9; roll_shift 14;
-            par_round3: h_ordering 0, 1, 2, 3, 4; data_index 11; roll_shift 12;
-            par_round3: h_ordering 4, 0, 1, 2, 3; data_index  8; roll_shift 13;
-            par_round3: h_ordering 3, 4, 0, 1, 2; data_index 12; roll_shift  5;
-            par_round3: h_ordering 2, 3, 4, 0, 1; data_index  2; roll_shift 14;
-            par_round3: h_ordering 1, 2, 3, 4, 0; data_index 10; roll_shift 13;
-            par_round3: h_ordering 0, 1, 2, 3, 4; data_index  0; roll_shift 13;
-            par_round3: h_ordering 4, 0, 1, 2, 3; data_index  4; roll_shift  7;
-            par_round3: h_ordering 3, 4, 0, 1, 2; data_index 13; roll_shift  5;
-
-            // Parallel Round 4
-            par_round4: h_ordering 2, 3, 4, 0, 1; data_index  8; roll_shift 15;
-            par_round4: h_ordering 1, 2, 3, 4, 0; data_index  6; roll_shift  5;
-            par_round4: h_ordering 0, 1, 2, 3, 4; data_index  4; roll_shift  8;
-            par_round4: h_ordering 4, 0, 1, 2, 3; data_index  1; roll_shift 11;
-            par_round4: h_ordering 3, 4, 0, 1, 2; data_index  3; roll_shift 14;
-            par_round4: h_ordering 2, 3, 4, 0, 1; data_index 11; roll_shift 14;
-            par_round4: h_ordering 1, 2, 3, 4, 0; data_index 15; roll_shift  6;
-            par_round4: h_ordering 0, 1, 2, 3, 4; data_index  0; roll_shift 14;
-            par_round4: h_ordering 4, 0, 1, 2, 3; data_index  5; roll_shift  6;
-            par_round4: h_ordering 3, 4, 0, 1, 2; data_index 12; roll_shift  9;
-            par_round4: h_ordering 2, 3, 4, 0, 1; data_index  2; roll_shift 12;
-            par_round4: h_ordering 1, 2, 3, 4, 0; data_index 13; roll_shift  9;
-            par_round4: h_ordering 0, 1, 2, 3, 4; data_index  9; roll_shift 12;
-            par_round4: h_ordering 4, 0, 1, 2, 3; data_index  7; roll_shift  5;
-            par_round4: h_ordering 3, 4, 0, 1, 2; data_index 10; roll_shift 15;
-            par_round4: h_ordering 2, 3, 4, 0, 1; data_index 14; roll_shift  8;
-
-            // Parallel Round 5
-            par_round5: h_ordering 1, 2, 3, 4, 0; data_index 12; roll_shift  8;
-            par_round5: h_ordering 0, 1, 2, 3, 4; data_index 15; roll_shift  5;
-            par_round5: h_ordering 4, 0, 1, 2, 3; data_index 10; roll_shift 12;
-            par_round5: h_ordering 3, 4, 0, 1, 2; data_index  4; roll_shift  9;
-            par_round5: h_ordering 2, 3, 4, 0, 1; data_index  1; roll_shift 12;
-            par_round5: h_ordering 1, 2, 3, 4, 0; data_index  5; roll_shift  5;
-            par_round5: h_ordering 0, 1, 2, 3, 4; data_index  8; roll_shift 14;
-            par_round5: h_ordering 4, 0, 1, 2, 3; data_index  7; roll_shift  6;
-            par_round5: h_ordering 3, 4, 0, 1, 2; data_index  6; roll_shift  8;
-            par_round5: h_ordering 2, 3, 4, 0, 1; data_index  2; roll_shift 13;
-            par_round5: h_ordering 1, 2, 3, 4, 0; data_index 13; roll_shift  6;
-            par_round5: h_ordering 0, 1, 2, 3, 4; data_index 14; roll_shift  5;
-            par_round5: h_ordering 4, 0, 1, 2, 3; data_index  0; roll_shift 15;
-            par_round5: h_ordering 3, 4, 0, 1, 2; data_index  3; roll_shift 13;
-            par_round5: h_ordering 2, 3, 4, 0, 1; data_index  9; roll_shift 11;
-            par_round5: h_ordering 1, 2, 3, 4, 0; data_index 11; roll_shift 11;
-        );
-    }
-}
-
 #[cfg(test)]
 mod tests {
     #[test]
diff --git a/hashes/src/sha1/crypto.rs b/hashes/src/sha1/crypto.rs
new file mode 100644
index 000000000..6f8a3fd8a
--- /dev/null
+++ b/hashes/src/sha1/crypto.rs
@@ -0,0 +1,48 @@
+// SPDX-License-Identifier: CC0-1.0
+
+use super::{HashEngine, BLOCK_SIZE};
+
+impl HashEngine {
+    // Basic unoptimized algorithm from Wikipedia
+    pub(super) fn process_block(&mut self) {
+        debug_assert_eq!(self.buffer.len(), BLOCK_SIZE);
+
+        let mut w = [0u32; 80];
+        for (w_val, buff_bytes) in w.iter_mut().zip(self.buffer.chunks_exact(4)) {
+            *w_val = u32::from_be_bytes(buff_bytes.try_into().expect("4 bytes slice"))
+        }
+        for i in 16..80 {
+            w[i] = (w[i - 3] ^ w[i - 8] ^ w[i - 14] ^ w[i - 16]).rotate_left(1);
+        }
+
+        let mut a = self.h[0];
+        let mut b = self.h[1];
+        let mut c = self.h[2];
+        let mut d = self.h[3];
+        let mut e = self.h[4];
+
+        for (i, &wi) in w.iter().enumerate() {
+            let (f, k) = match i {
+                0..=19 => ((b & c) | (!b & d), 0x5a827999),
+                20..=39 => (b ^ c ^ d, 0x6ed9eba1),
+                40..=59 => ((b & c) | (b & d) | (c & d), 0x8f1bbcdc),
+                60..=79 => (b ^ c ^ d, 0xca62c1d6),
+                _ => unreachable!(),
+            };
+
+            let new_a =
+                a.rotate_left(5).wrapping_add(f).wrapping_add(e).wrapping_add(k).wrapping_add(wi);
+            e = d;
+            d = c;
+            c = b.rotate_left(30);
+            b = a;
+            a = new_a;
+        }
+
+        self.h[0] = self.h[0].wrapping_add(a);
+        self.h[1] = self.h[1].wrapping_add(b);
+        self.h[2] = self.h[2].wrapping_add(c);
+        self.h[3] = self.h[3].wrapping_add(d);
+        self.h[4] = self.h[4].wrapping_add(e);
+    }
+}
diff --git a/hashes/src/sha1/mod.rs b/hashes/src/sha1/mod.rs
index 152a8f348..f0748c33b 100644
--- a/hashes/src/sha1/mod.rs
+++ b/hashes/src/sha1/mod.rs
@@ -2,6 +2,8 @@
 
 //! SHA1 implementation.
 
+mod crypto;
+
 use core::cmp;
 
 use crate::{incomplete_block_len, HashEngine as _};
@@ -80,51 +82,6 @@ impl crate::HashEngine for HashEngine {
     crate::internal_macros::engine_input_impl!();
 }
 
-impl HashEngine {
-    // Basic unoptimized algorithm from Wikipedia
-    fn process_block(&mut self) {
-        debug_assert_eq!(self.buffer.len(), BLOCK_SIZE);
-
-        let mut w = [0u32; 80];
-        for (w_val, buff_bytes) in w.iter_mut().zip(self.buffer.chunks_exact(4)) {
-            *w_val = u32::from_be_bytes(buff_bytes.try_into().expect("4 bytes slice"))
-        }
-        for i in 16..80 {
-            w[i] = (w[i - 3] ^ w[i - 8] ^ w[i - 14] ^ w[i - 16]).rotate_left(1);
-        }
-
-        let mut a = self.h[0];
-        let mut b = self.h[1];
-        let mut c = self.h[2];
-        let mut d = self.h[3];
-        let mut e = self.h[4];
-
-        for (i, &wi) in w.iter().enumerate() {
-            let (f, k) = match i {
-                0..=19 => ((b & c) | (!b & d), 0x5a827999),
-                20..=39 => (b ^ c ^ d, 0x6ed9eba1),
-                40..=59 => ((b & c) | (b & d) | (c & d), 0x8f1bbcdc),
-                60..=79 => (b ^ c ^ d, 0xca62c1d6),
-                _ => unreachable!(),
-            };
-
-            let new_a =
-                a.rotate_left(5).wrapping_add(f).wrapping_add(e).wrapping_add(k).wrapping_add(wi);
-            e = d;
-            d = c;
-            c = b.rotate_left(30);
-            b = a;
-            a = new_a;
-        }
-
-        self.h[0] = self.h[0].wrapping_add(a);
-        self.h[1] = self.h[1].wrapping_add(b);
-        self.h[2] = self.h[2].wrapping_add(c);
-        self.h[3] = self.h[3].wrapping_add(d);
-        self.h[4] = self.h[4].wrapping_add(e);
-    }
-}
-
 #[cfg(test)]
 mod tests {
     #[test]
diff --git a/hashes/src/sha256/crypto.rs b/hashes/src/sha256/crypto.rs
new file mode 100644
index 000000000..210eb6e6e
--- /dev/null
+++ b/hashes/src/sha256/crypto.rs
@@ -0,0 +1,619 @@
+// SPDX-License-Identifier: CC0-1.0
+
+#[cfg(all(feature = "std", target_arch = "x86"))]
+use core::arch::x86::*;
+#[cfg(all(feature = "std", target_arch = "x86_64"))]
+use core::arch::x86_64::*;
+
+use super::{HashEngine, Midstate, BLOCK_SIZE};
+
+#[allow(non_snake_case)]
+const fn Ch(x: u32, y: u32, z: u32) -> u32 { z ^ (x & (y ^ z)) }
+#[allow(non_snake_case)]
+const fn Maj(x: u32, y: u32, z: u32) -> u32 { (x & y) | (z & (x | y)) }
+#[allow(non_snake_case)]
+const fn Sigma0(x: u32) -> u32 { x.rotate_left(30) ^ x.rotate_left(19) ^ x.rotate_left(10) }
+#[allow(non_snake_case)]
+const fn Sigma1(x: u32) -> u32 { x.rotate_left(26) ^ x.rotate_left(21) ^ x.rotate_left(7) }
+const fn sigma0(x: u32) -> u32 { x.rotate_left(25) ^ x.rotate_left(14) ^ (x >> 3) }
+const fn sigma1(x: u32) -> u32 { x.rotate_left(15) ^ x.rotate_left(13) ^ (x >> 10) }
+
+#[cfg(feature = "small-hash")]
+#[macro_use]
+mod small_hash {
+    use super::*;
+
+    #[rustfmt::skip]
+    pub(super) const fn round(a: u32, b: u32, c: u32, d: u32, e: u32,
+                              f: u32, g: u32, h: u32, k: u32, w: u32) -> (u32, u32) {
+        let t1 =
+            h.wrapping_add(Sigma1(e)).wrapping_add(Ch(e, f, g)).wrapping_add(k).wrapping_add(w);
+        let t2 = Sigma0(a).wrapping_add(Maj(a, b, c));
+        (d.wrapping_add(t1), t1.wrapping_add(t2))
+    }
+    #[rustfmt::skip]
+    pub(super) const fn later_round(a: u32, b: u32, c: u32, d: u32, e: u32,
+                                    f: u32, g: u32, h: u32, k: u32, w: u32,
+                                    w1: u32, w2: u32, w3: u32,
+    ) -> (u32, u32, u32) {
+        let w = w.wrapping_add(sigma1(w1)).wrapping_add(w2).wrapping_add(sigma0(w3));
+        let (d, h) = round(a, b, c, d, e, f, g, h, k, w);
+        (d, h, w)
+    }
+
+    macro_rules! round(
+        // first round
+        ($a:expr, $b:expr, $c:expr, $d:expr, $e:expr, $f:expr, $g:expr, $h:expr, $k:expr, $w:expr) => (
+            let updates = small_hash::round($a, $b, $c, $d, $e, $f, $g, $h, $k, $w);
+            $d = updates.0;
+            $h = updates.1;
+        );
+        // later rounds we reassign $w before doing the first-round computation
+        ($a:expr, $b:expr, $c:expr, $d:expr, $e:expr, $f:expr, $g:expr, $h:expr, $k:expr, $w:expr, $w1:expr, $w2:expr, $w3:expr) => (
+            let updates = small_hash::later_round($a, $b, $c, $d, $e, $f, $g, $h, $k, $w, $w1, $w2, $w3);
+            $d = updates.0;
+            $h = updates.1;
+            $w = updates.2;
+        )
+    );
+}
+
+#[cfg(not(feature = "small-hash"))]
+#[macro_use]
+mod fast_hash {
+    macro_rules! round(
+        // first round
+        ($a:expr, $b:expr, $c:expr, $d:expr, $e:expr, $f:expr, $g:expr, $h:expr, $k:expr, $w:expr) => (
+            let t1 = $h.wrapping_add(Sigma1($e)).wrapping_add(Ch($e, $f, $g)).wrapping_add($k).wrapping_add($w);
+            let t2 = Sigma0($a).wrapping_add(Maj($a, $b, $c));
+            $d = $d.wrapping_add(t1);
+            $h = t1.wrapping_add(t2);
+        );
+        // later rounds we reassign $w before doing the first-round computation
+        ($a:expr, $b:expr, $c:expr, $d:expr, $e:expr, $f:expr, $g:expr, $h:expr, $k:expr, $w:expr, $w1:expr, $w2:expr, $w3:expr) => (
+            $w = $w.wrapping_add(sigma1($w1)).wrapping_add($w2).wrapping_add(sigma0($w3));
+            round!($a, $b, $c, $d, $e, $f, $g, $h, $k, $w);
+        )
+    );
+}
+
+impl Midstate {
+    #[allow(clippy::identity_op)] // more readble
+    const fn read_u32(bytes: &[u8], index: usize) -> u32 {
+        ((bytes[index + 0] as u32) << 24)
+            | ((bytes[index + 1] as u32) << 16)
+            | ((bytes[index + 2] as u32) << 8)
+            | ((bytes[index + 3] as u32) << 0)
+    }
+
+    const fn copy_w(bytes: &[u8], index: usize) -> [u32; 16] {
+        let mut w = [0u32; 16];
+        let mut i = 0;
+        while i < 16 {
+            w[i] = Self::read_u32(bytes, index + i * 4);
+            i += 1;
+        }
+        w
+    }
+
+    pub(super) const fn compute_midstate_unoptimized(bytes: &[u8], finalize: bool) -> Self {
+        let mut state = [
+            0x6a09e667u32,
+            0xbb67ae85,
+            0x3c6ef372,
+            0xa54ff53a,
+            0x510e527f,
+            0x9b05688c,
+            0x1f83d9ab,
+            0x5be0cd19,
+        ];
+
+        let num_chunks = (bytes.len() + 9 + 63) / 64;
+        let mut chunk = 0;
+        #[allow(clippy::precedence)]
+        while chunk < num_chunks {
+            if !finalize && chunk + 1 == num_chunks {
+                break;
+            }
+            let mut w = if chunk * 64 + 64 <= bytes.len() {
+                Self::copy_w(bytes, chunk * 64)
+            } else {
+                let mut buf = [0; 64];
+                let mut i = 0;
+                let offset = chunk * 64;
+                while offset + i < bytes.len() {
+                    buf[i] = bytes[offset + i];
+                    i += 1;
+                }
+                if (bytes.len() % 64 <= 64 - 9) || (chunk + 2 == num_chunks) {
+                    buf[i] = 0x80;
+                }
+                #[allow(clippy::identity_op)] // more readble
+                #[allow(clippy::erasing_op)]
+                if chunk + 1 == num_chunks {
+                    let bit_len = bytes.len() as u64 * 8;
+                    buf[64 - 8] = ((bit_len >> 8 * 7) & 0xFF) as u8;
+                    buf[64 - 7] = ((bit_len >> 8 * 6) & 0xFF) as u8;
+                    buf[64 - 6] = ((bit_len >> 8 * 5) & 0xFF) as u8;
+                    buf[64 - 5] = ((bit_len >> 8 * 4) & 0xFF) as u8;
+                    buf[64 - 4] = ((bit_len >> 8 * 3) & 0xFF) as u8;
+                    buf[64 - 3] = ((bit_len >> 8 * 2) & 0xFF) as u8;
+                    buf[64 - 2] = ((bit_len >> 8 * 1) & 0xFF) as u8;
+                    buf[64 - 1] = ((bit_len >> 8 * 0) & 0xFF) as u8;
+                }
+                Self::copy_w(&buf, 0)
+            };
+            chunk += 1;
+
+            let mut a = state[0];
+            let mut b = state[1];
+            let mut c = state[2];
+            let mut d = state[3];
+            let mut e = state[4];
+            let mut f = state[5];
+            let mut g = state[6];
+            let mut h = state[7];
+
+            round!(a, b, c, d, e, f, g, h, 0x428a2f98, w[0]);
+            round!(h, a, b, c, d, e, f, g, 0x71374491, w[1]);
+            round!(g, h, a, b, c, d, e, f, 0xb5c0fbcf, w[2]);
+            round!(f, g, h, a, b, c, d, e, 0xe9b5dba5, w[3]);
+            round!(e, f, g, h, a, b, c, d, 0x3956c25b, w[4]);
+            round!(d, e, f, g, h, a, b, c, 0x59f111f1, w[5]);
+            round!(c, d, e, f, g, h, a, b, 0x923f82a4, w[6]);
+            round!(b, c, d, e, f, g, h, a, 0xab1c5ed5, w[7]);
+            round!(a, b, c, d, e, f, g, h, 0xd807aa98, w[8]);
+            round!(h, a, b, c, d, e, f, g, 0x12835b01, w[9]);
+            round!(g, h, a, b, c, d, e, f, 0x243185be, w[10]);
+            round!(f, g, h, a, b, c, d, e, 0x550c7dc3, w[11]);
+            round!(e, f, g, h, a, b, c, d, 0x72be5d74, w[12]);
+            round!(d, e, f, g, h, a, b, c, 0x80deb1fe, w[13]);
+            round!(c, d, e, f, g, h, a, b, 0x9bdc06a7, w[14]);
+            round!(b, c, d, e, f, g, h, a, 0xc19bf174, w[15]);
+
+            round!(a, b, c, d, e, f, g, h, 0xe49b69c1, w[0], w[14], w[9], w[1]);
+            round!(h, a, b, c, d, e, f, g, 0xefbe4786, w[1], w[15], w[10], w[2]);
+            round!(g, h, a, b, c, d, e, f, 0x0fc19dc6, w[2], w[0], w[11], w[3]);
+            round!(f, g, h, a, b, c, d, e, 0x240ca1cc, w[3], w[1], w[12], w[4]);
+            round!(e, f, g, h, a, b, c, d, 0x2de92c6f, w[4], w[2], w[13], w[5]);
+            round!(d, e, f, g, h, a, b, c, 0x4a7484aa, w[5], w[3], w[14], w[6]);
+            round!(c, d, e, f, g, h, a, b, 0x5cb0a9dc, w[6], w[4], w[15], w[7]);
+            round!(b, c, d, e, f, g, h, a, 0x76f988da, w[7], w[5], w[0], w[8]);
+            round!(a, b, c, d, e, f, g, h, 0x983e5152, w[8], w[6], w[1], w[9]);
+            round!(h, a, b, c, d, e, f, g, 0xa831c66d, w[9], w[7], w[2], w[10]);
+            round!(g, h, a, b, c, d, e, f, 0xb00327c8, w[10], w[8], w[3], w[11]);
+            round!(f, g, h, a, b, c, d, e, 0xbf597fc7, w[11], w[9], w[4], w[12]);
+            round!(e, f, g, h, a, b, c, d, 0xc6e00bf3, w[12], w[10], w[5], w[13]);
+            round!(d, e, f, g, h, a, b, c, 0xd5a79147, w[13], w[11], w[6], w[14]);
+            round!(c, d, e, f, g, h, a, b, 0x06ca6351, w[14], w[12], w[7], w[15]);
+            round!(b, c, d, e, f, g, h, a, 0x14292967, w[15], w[13], w[8], w[0]);
+
+            round!(a, b, c, d, e, f, g, h, 0x27b70a85, w[0], w[14], w[9], w[1]);
+            round!(h, a, b, c, d, e, f, g, 0x2e1b2138, w[1], w[15], w[10], w[2]);
+            round!(g, h, a, b, c, d, e, f, 0x4d2c6dfc, w[2], w[0], w[11], w[3]);
+            round!(f, g, h, a, b, c, d, e, 0x53380d13, w[3], w[1], w[12], w[4]);
+            round!(e, f, g, h, a, b, c, d, 0x650a7354, w[4], w[2], w[13], w[5]);
+            round!(d, e, f, g, h, a, b, c, 0x766a0abb, w[5], w[3], w[14], w[6]);
+            round!(c, d, e, f, g, h, a, b, 0x81c2c92e, w[6], w[4], w[15], w[7]);
+            round!(b, c, d, e, f, g, h, a, 0x92722c85, w[7], w[5], w[0], w[8]);
+            round!(a, b, c, d, e, f, g, h, 0xa2bfe8a1, w[8], w[6], w[1], w[9]);
+            round!(h, a, b, c, d, e, f, g, 0xa81a664b, w[9], w[7], w[2], w[10]);
+            round!(g, h, a, b, c, d, e, f, 0xc24b8b70, w[10], w[8], w[3], w[11]);
+            round!(f, g, h, a, b, c, d, e, 0xc76c51a3, w[11], w[9], w[4], w[12]);
+            round!(e, f, g, h, a, b, c, d, 0xd192e819, w[12], w[10], w[5], w[13]);
+            round!(d, e, f, g, h, a, b, c, 0xd6990624, w[13], w[11], w[6], w[14]);
+            round!(c, d, e, f, g, h, a, b, 0xf40e3585, w[14], w[12], w[7], w[15]);
+            round!(b, c, d, e, f, g, h, a, 0x106aa070, w[15], w[13], w[8], w[0]);
+
+            round!(a, b, c, d, e, f, g, h, 0x19a4c116, w[0], w[14], w[9], w[1]);
+            round!(h, a, b, c, d, e, f, g, 0x1e376c08, w[1], w[15], w[10], w[2]);
+            round!(g, h, a, b, c, d, e, f, 0x2748774c, w[2], w[0], w[11], w[3]);
+            round!(f, g, h, a, b, c, d, e, 0x34b0bcb5, w[3], w[1], w[12], w[4]);
+            round!(e, f, g, h, a, b, c, d, 0x391c0cb3, w[4], w[2], w[13], w[5]);
+            round!(d, e, f, g, h, a, b, c, 0x4ed8aa4a, w[5], w[3], w[14], w[6]);
+            round!(c, d, e, f, g, h, a, b, 0x5b9cca4f, w[6], w[4], w[15], w[7]);
+            round!(b, c, d, e, f, g, h, a, 0x682e6ff3, w[7], w[5], w[0], w[8]);
+            round!(a, b, c, d, e, f, g, h, 0x748f82ee, w[8], w[6], w[1], w[9]);
+            round!(h, a, b, c, d, e, f, g, 0x78a5636f, w[9], w[7], w[2], w[10]);
+            round!(g, h, a, b, c, d, e, f, 0x84c87814, w[10], w[8], w[3], w[11]);
+            round!(f, g, h, a, b, c, d, e, 0x8cc70208, w[11], w[9], w[4], w[12]);
+            round!(e, f, g, h, a, b, c, d, 0x90befffa, w[12], w[10], w[5], w[13]);
+            round!(d, e, f, g, h, a, b, c, 0xa4506ceb, w[13], w[11], w[6], w[14]);
+            round!(c, d, e, f, g, h, a, b, 0xbef9a3f7, w[14], w[12], w[7], w[15]);
+            round!(b, c, d, e, f, g, h, a, 0xc67178f2, w[15], w[13], w[8], w[0]);
+
+            state[0] = state[0].wrapping_add(a);
+            state[1] = state[1].wrapping_add(b);
+            state[2] = state[2].wrapping_add(c);
+            state[3] = state[3].wrapping_add(d);
+            state[4] = state[4].wrapping_add(e);
+            state[5] = state[5].wrapping_add(f);
+            state[6] = state[6].wrapping_add(g);
+            state[7] = state[7].wrapping_add(h);
+        }
+        let mut output = [0u8; 32];
+        let mut i = 0;
+        #[allow(clippy::identity_op)] // more readble
+        while i < 8 {
+            output[i * 4 + 0] = (state[i + 0] >> 24) as u8;
+            output[i * 4 + 1] = (state[i + 0] >> 16) as u8;
+            output[i * 4 + 2] = (state[i + 0] >> 8) as u8;
+            output[i * 4 + 3] = (state[i + 0] >> 0) as u8;
+            i += 1;
+        }
+        Midstate { bytes: output, bytes_hashed: bytes.len() as u64 }
+    }
+}
+
+impl HashEngine {
+    pub(super) fn process_block(&mut self) {
+        #[cfg(all(feature = "std", any(target_arch = "x86", target_arch = "x86_64")))]
+        {
+            if std::is_x86_feature_detected!("sse4.1")
+                && std::is_x86_feature_detected!("sha")
+                && std::is_x86_feature_detected!("sse2")
+                && std::is_x86_feature_detected!("ssse3")
+            {
+                return unsafe { self.process_block_simd_x86_intrinsics() };
+            }
+        }
+
+        // fallback implementation without using any intrinsics
+        self.software_process_block()
+    }
+
+    #[cfg(all(feature = "std", any(target_arch = "x86", target_arch = "x86_64")))]
+    #[target_feature(enable = "sha,sse2,ssse3,sse4.1")]
+    unsafe fn process_block_simd_x86_intrinsics(&mut self) {
+        // Code translated and based on from
+        // https://github.com/noloader/SHA-Intrinsics/blob/4899efc81d1af159c1fd955936c673139f35aea9/sha256-x86.c
+
+        /* sha256-x86.c - Intel SHA extensions using C intrinsics  */
+        /*   Written and place in public domain by Jeffrey Walton  */
+        /*   Based on code from Intel, and by Sean Gulley for      */
+        /*   the miTLS project.                                    */
+
+        // Variable names are also kept the same as in the original C code for easier comparison.
+        let (mut state0, mut state1);
+        let (mut msg, mut tmp);
+
+        let (mut msg0, mut msg1, mut msg2, mut msg3);
+
+        let (abef_save, cdgh_save);
+
+        #[allow(non_snake_case)]
+        let MASK: __m128i =
+            _mm_set_epi64x(0x0c0d_0e0f_0809_0a0bu64 as i64, 0x0405_0607_0001_0203u64 as i64);
+
+        let block_offset = 0;
+
+        // Load initial values
+        // CAST SAFETY: loadu_si128 documentation states that mem_addr does not
+        // need to be aligned on any particular boundary.
+        tmp = _mm_loadu_si128(self.h.as_ptr().add(0) as *const __m128i);
+        state1 = _mm_loadu_si128(self.h.as_ptr().add(4) as *const __m128i);
+
+        tmp = _mm_shuffle_epi32(tmp, 0xB1); // CDAB
+        state1 = _mm_shuffle_epi32(state1, 0x1B); // EFGH
+        state0 = _mm_alignr_epi8(tmp, state1, 8); // ABEF
+        state1 = _mm_blend_epi16(state1, tmp, 0xF0); // CDGH
+
+        // Process a single block
+        {
+            // Save current state
+            abef_save = state0;
+            cdgh_save = state1;
+
+            // Rounds 0-3
+            msg = _mm_loadu_si128(self.buffer.as_ptr().add(block_offset) as *const __m128i);
+            msg0 = _mm_shuffle_epi8(msg, MASK);
+            msg = _mm_add_epi32(
+                msg0,
+                _mm_set_epi64x(0xE9B5DBA5B5C0FBCFu64 as i64, 0x71374491428A2F98u64 as i64),
+            );
+            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
+            msg = _mm_shuffle_epi32(msg, 0x0E);
+            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
+
+            // Rounds 4-7
+            msg1 = _mm_loadu_si128(self.buffer.as_ptr().add(block_offset + 16) as *const __m128i);
+            msg1 = _mm_shuffle_epi8(msg1, MASK);
+            msg = _mm_add_epi32(
+                msg1,
+                _mm_set_epi64x(0xAB1C5ED5923F82A4u64 as i64, 0x59F111F13956C25Bu64 as i64),
+            );
+            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
+            msg = _mm_shuffle_epi32(msg, 0x0E);
+            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
+            msg0 = _mm_sha256msg1_epu32(msg0, msg1);
+
+            // Rounds 8-11
+            msg2 = _mm_loadu_si128(self.buffer.as_ptr().add(block_offset + 32) as *const __m128i);
+            msg2 = _mm_shuffle_epi8(msg2, MASK);
+            msg = _mm_add_epi32(
+                msg2,
+                _mm_set_epi64x(0x550C7DC3243185BEu64 as i64, 0x12835B01D807AA98u64 as i64),
+            );
+            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
+            msg = _mm_shuffle_epi32(msg, 0x0E);
+            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
+            msg1 = _mm_sha256msg1_epu32(msg1, msg2);
+
+            // Rounds 12-15
+            msg3 = _mm_loadu_si128(self.buffer.as_ptr().add(block_offset + 48) as *const __m128i);
+            msg3 = _mm_shuffle_epi8(msg3, MASK);
+            msg = _mm_add_epi32(
+                msg3,
+                _mm_set_epi64x(0xC19BF1749BDC06A7u64 as i64, 0x80DEB1FE72BE5D74u64 as i64),
+            );
+            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
+            tmp = _mm_alignr_epi8(msg3, msg2, 4);
+            msg0 = _mm_add_epi32(msg0, tmp);
+            msg0 = _mm_sha256msg2_epu32(msg0, msg3);
+            msg = _mm_shuffle_epi32(msg, 0x0E);
+            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
+            msg2 = _mm_sha256msg1_epu32(msg2, msg3);
+
+            // Rounds 16-19
+            msg = _mm_add_epi32(
+                msg0,
+                _mm_set_epi64x(0x240CA1CC0FC19DC6u64 as i64, 0xEFBE4786E49B69C1u64 as i64),
+            );
+            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
+            tmp = _mm_alignr_epi8(msg0, msg3, 4);
+            msg1 = _mm_add_epi32(msg1, tmp);
+            msg1 = _mm_sha256msg2_epu32(msg1, msg0);
+            msg = _mm_shuffle_epi32(msg, 0x0E);
+            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
+            msg3 = _mm_sha256msg1_epu32(msg3, msg0);
+
+            // Rounds 20-23
+            msg = _mm_add_epi32(
+                msg1,
+                _mm_set_epi64x(0x76F988DA5CB0A9DCu64 as i64, 0x4A7484AA2DE92C6Fu64 as i64),
+            );
+            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
+            tmp = _mm_alignr_epi8(msg1, msg0, 4);
+            msg2 = _mm_add_epi32(msg2, tmp);
+            msg2 = _mm_sha256msg2_epu32(msg2, msg1);
+            msg = _mm_shuffle_epi32(msg, 0x0E);
+            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
+            msg0 = _mm_sha256msg1_epu32(msg0, msg1);
+
+            // Rounds 24-27
+            msg = _mm_add_epi32(
+                msg2,
+                _mm_set_epi64x(0xBF597FC7B00327C8u64 as i64, 0xA831C66D983E5152u64 as i64),
+            );
+            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
+            tmp = _mm_alignr_epi8(msg2, msg1, 4);
+            msg3 = _mm_add_epi32(msg3, tmp);
+            msg3 = _mm_sha256msg2_epu32(msg3, msg2);
+            msg = _mm_shuffle_epi32(msg, 0x0E);
+            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
+            msg1 = _mm_sha256msg1_epu32(msg1, msg2);
+
+            // Rounds 28-31
+            msg = _mm_add_epi32(
+                msg3,
+                _mm_set_epi64x(0x1429296706CA6351u64 as i64, 0xD5A79147C6E00BF3u64 as i64),
+            );
+            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
+            tmp = _mm_alignr_epi8(msg3, msg2, 4);
+            msg0 = _mm_add_epi32(msg0, tmp);
+            msg0 = _mm_sha256msg2_epu32(msg0, msg3);
+            msg = _mm_shuffle_epi32(msg, 0x0E);
+            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
+            msg2 = _mm_sha256msg1_epu32(msg2, msg3);
+
+            // Rounds 32-35
+            msg = _mm_add_epi32(
+                msg0,
+                _mm_set_epi64x(0x53380D134D2C6DFCu64 as i64, 0x2E1B213827B70A85u64 as i64),
+            );
+            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
+            tmp = _mm_alignr_epi8(msg0, msg3, 4);
+            msg1 = _mm_add_epi32(msg1, tmp);
+            msg1 = _mm_sha256msg2_epu32(msg1, msg0);
+            msg = _mm_shuffle_epi32(msg, 0x0E);
+            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
+            msg3 = _mm_sha256msg1_epu32(msg3, msg0);
+
+            // Rounds 36-39
+            msg = _mm_add_epi32(
+                msg1,
+                _mm_set_epi64x(0x92722C8581C2C92Eu64 as i64, 0x766A0ABB650A7354u64 as i64),
+            );
+            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
+            tmp = _mm_alignr_epi8(msg1, msg0, 4);
+            msg2 = _mm_add_epi32(msg2, tmp);
+            msg2 = _mm_sha256msg2_epu32(msg2, msg1);
+            msg = _mm_shuffle_epi32(msg, 0x0E);
+            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
+            msg0 = _mm_sha256msg1_epu32(msg0, msg1);
+
+            // Rounds 40-43
+            msg = _mm_add_epi32(
+                msg2,
+                _mm_set_epi64x(0xC76C51A3C24B8B70u64 as i64, 0xA81A664BA2BFE8A1u64 as i64),
+            );
+            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
+            tmp = _mm_alignr_epi8(msg2, msg1, 4);
+            msg3 = _mm_add_epi32(msg3, tmp);
+            msg3 = _mm_sha256msg2_epu32(msg3, msg2);
+            msg = _mm_shuffle_epi32(msg, 0x0E);
+            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
+            msg1 = _mm_sha256msg1_epu32(msg1, msg2);
+
+            // Rounds 44-47
+            msg = _mm_add_epi32(
+                msg3,
+                _mm_set_epi64x(0x106AA070F40E3585u64 as i64, 0xD6990624D192E819u64 as i64),
+            );
+            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
+            tmp = _mm_alignr_epi8(msg3, msg2, 4);
+            msg0 = _mm_add_epi32(msg0, tmp);
+            msg0 = _mm_sha256msg2_epu32(msg0, msg3);
+            msg = _mm_shuffle_epi32(msg, 0x0E);
+            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
+            msg2 = _mm_sha256msg1_epu32(msg2, msg3);
+
+            // Rounds 48-51
+            msg = _mm_add_epi32(
+                msg0,
+                _mm_set_epi64x(0x34B0BCB52748774Cu64 as i64, 0x1E376C0819A4C116u64 as i64),
+            );
+            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
+            tmp = _mm_alignr_epi8(msg0, msg3, 4);
+            msg1 = _mm_add_epi32(msg1, tmp);
+            msg1 = _mm_sha256msg2_epu32(msg1, msg0);
+            msg = _mm_shuffle_epi32(msg, 0x0E);
+            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
+            msg3 = _mm_sha256msg1_epu32(msg3, msg0);
+
+            // Rounds 52-55
+            msg = _mm_add_epi32(
+                msg1,
+                _mm_set_epi64x(0x682E6FF35B9CCA4Fu64 as i64, 0x4ED8AA4A391C0CB3u64 as i64),
+            );
+            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
+            tmp = _mm_alignr_epi8(msg1, msg0, 4);
+            msg2 = _mm_add_epi32(msg2, tmp);
+            msg2 = _mm_sha256msg2_epu32(msg2, msg1);
+            msg = _mm_shuffle_epi32(msg, 0x0E);
+            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
+
+            // Rounds 56-59
+            msg = _mm_add_epi32(
+                msg2,
+                _mm_set_epi64x(0x8CC7020884C87814u64 as i64, 0x78A5636F748F82EEu64 as i64),
+            );
+            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
+            tmp = _mm_alignr_epi8(msg2, msg1, 4);
+            msg3 = _mm_add_epi32(msg3, tmp);
+            msg3 = _mm_sha256msg2_epu32(msg3, msg2);
+            msg = _mm_shuffle_epi32(msg, 0x0E);
+            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
+
+            // Rounds 60-63
+            msg = _mm_add_epi32(
+                msg3,
+                _mm_set_epi64x(0xC67178F2BEF9A3F7u64 as i64, 0xA4506CEB90BEFFFAu64 as i64),
+            );
+            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
+            msg = _mm_shuffle_epi32(msg, 0x0E);
+            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
+
+            // Combine state
+            state0 = _mm_add_epi32(state0, abef_save);
+            state1 = _mm_add_epi32(state1, cdgh_save);
+        }
+
+        tmp = _mm_shuffle_epi32(state0, 0x1B); // FEBA
+        state1 = _mm_shuffle_epi32(state1, 0xB1); // DCHG
+        state0 = _mm_blend_epi16(tmp, state1, 0xF0); // DCBA
+        state1 = _mm_alignr_epi8(state1, tmp, 8); // ABEF
+
+        // Save state
+        // CAST SAFETY: storeu_si128 documentation states that mem_addr does not
+        // need to be aligned on any particular boundary.
+        _mm_storeu_si128(self.h.as_mut_ptr().add(0) as *mut __m128i, state0);
+        _mm_storeu_si128(self.h.as_mut_ptr().add(4) as *mut __m128i, state1);
+    }
+
+    // Algorithm copied from libsecp256k1
+    fn software_process_block(&mut self) {
+        debug_assert_eq!(self.buffer.len(), BLOCK_SIZE);
+
+        let mut w = [0u32; 16];
+        for (w_val, buff_bytes) in w.iter_mut().zip(self.buffer.chunks_exact(4)) {
+            *w_val = u32::from_be_bytes(buff_bytes.try_into().expect("4 byte slice"));
+        }
+
+        let mut a = self.h[0];
+        let mut b = self.h[1];
+        let mut c = self.h[2];
+        let mut d = self.h[3];
+        let mut e = self.h[4];
+        let mut f = self.h[5];
+        let mut g = self.h[6];
+        let mut h = self.h[7];
+
+        round!(a, b, c, d, e, f, g, h, 0x428a2f98, w[0]);
+        round!(h, a, b, c, d, e, f, g, 0x71374491, w[1]);
+        round!(g, h, a, b, c, d, e, f, 0xb5c0fbcf, w[2]);
+        round!(f, g, h, a, b, c, d, e, 0xe9b5dba5, w[3]);
+        round!(e, f, g, h, a, b, c, d, 0x3956c25b, w[4]);
+        round!(d, e, f, g, h, a, b, c, 0x59f111f1, w[5]);
+        round!(c, d, e, f, g, h, a, b, 0x923f82a4, w[6]);
+        round!(b, c, d, e, f, g, h, a, 0xab1c5ed5, w[7]);
+        round!(a, b, c, d, e, f, g, h, 0xd807aa98, w[8]);
+        round!(h, a, b, c, d, e, f, g, 0x12835b01, w[9]);
+        round!(g, h, a, b, c, d, e, f, 0x243185be, w[10]);
+        round!(f, g, h, a, b, c, d, e, 0x550c7dc3, w[11]);
+        round!(e, f, g, h, a, b, c, d, 0x72be5d74, w[12]);
+        round!(d, e, f, g, h, a, b, c, 0x80deb1fe, w[13]);
+        round!(c, d, e, f, g, h, a, b, 0x9bdc06a7, w[14]);
+        round!(b, c, d, e, f, g, h, a, 0xc19bf174, w[15]);
+
+        round!(a, b, c, d, e, f, g, h, 0xe49b69c1, w[0], w[14], w[9], w[1]);
+        round!(h, a, b, c, d, e, f, g, 0xefbe4786, w[1], w[15], w[10], w[2]);
+        round!(g, h, a, b, c, d, e, f, 0x0fc19dc6, w[2], w[0], w[11], w[3]);
+        round!(f, g, h, a, b, c, d, e, 0x240ca1cc, w[3], w[1], w[12], w[4]);
+        round!(e, f, g, h, a, b, c, d, 0x2de92c6f, w[4], w[2], w[13], w[5]);
+        round!(d, e, f, g, h, a, b, c, 0x4a7484aa, w[5], w[3], w[14], w[6]);
+        round!(c, d, e, f, g, h, a, b, 0x5cb0a9dc, w[6], w[4], w[15], w[7]);
+        round!(b, c, d, e, f, g, h, a, 0x76f988da, w[7], w[5], w[0], w[8]);
+        round!(a, b, c, d, e, f, g, h, 0x983e5152, w[8], w[6], w[1], w[9]);
+        round!(h, a, b, c, d, e, f, g, 0xa831c66d, w[9], w[7], w[2], w[10]);
+        round!(g, h, a, b, c, d, e, f, 0xb00327c8, w[10], w[8], w[3], w[11]);
+        round!(f, g, h, a, b, c, d, e, 0xbf597fc7, w[11], w[9], w[4], w[12]);
+        round!(e, f, g, h, a, b, c, d, 0xc6e00bf3, w[12], w[10], w[5], w[13]);
+        round!(d, e, f, g, h, a, b, c, 0xd5a79147, w[13], w[11], w[6], w[14]);
+        round!(c, d, e, f, g, h, a, b, 0x06ca6351, w[14], w[12], w[7], w[15]);
+        round!(b, c, d, e, f, g, h, a, 0x14292967, w[15], w[13], w[8], w[0]);
+
+        round!(a, b, c, d, e, f, g, h, 0x27b70a85, w[0], w[14], w[9], w[1]);
+        round!(h, a, b, c, d, e, f, g, 0x2e1b2138, w[1], w[15], w[10], w[2]);
+        round!(g, h, a, b, c, d, e, f, 0x4d2c6dfc, w[2], w[0], w[11], w[3]);
+        round!(f, g, h, a, b, c, d, e, 0x53380d13, w[3], w[1], w[12], w[4]);
+        round!(e, f, g, h, a, b, c, d, 0x650a7354, w[4], w[2], w[13], w[5]);
+        round!(d, e, f, g, h, a, b, c, 0x766a0abb, w[5], w[3], w[14], w[6]);
+        round!(c, d, e, f, g, h, a, b, 0x81c2c92e, w[6], w[4], w[15], w[7]);
+        round!(b, c, d, e, f, g, h, a, 0x92722c85, w[7], w[5], w[0], w[8]);
+        round!(a, b, c, d, e, f, g, h, 0xa2bfe8a1, w[8], w[6], w[1], w[9]);
+        round!(h, a, b, c, d, e, f, g, 0xa81a664b, w[9], w[7], w[2], w[10]);
+        round!(g, h, a, b, c, d, e, f, 0xc24b8b70, w[10], w[8], w[3], w[11]);
+        round!(f, g, h, a, b, c, d, e, 0xc76c51a3, w[11], w[9], w[4], w[12]);
+        round!(e, f, g, h, a, b, c, d, 0xd192e819, w[12], w[10], w[5], w[13]);
+        round!(d, e, f, g, h, a, b, c, 0xd6990624, w[13], w[11], w[6], w[14]);
+        round!(c, d, e, f, g, h, a, b, 0xf40e3585, w[14], w[12], w[7], w[15]);
+        round!(b, c, d, e, f, g, h, a, 0x106aa070, w[15], w[13], w[8], w[0]);
+
+        round!(a, b, c, d, e, f, g, h, 0x19a4c116, w[0], w[14], w[9], w[1]);
+        round!(h, a, b, c, d, e, f, g, 0x1e376c08, w[1], w[15], w[10], w[2]);
+        round!(g, h, a, b, c, d, e, f, 0x2748774c, w[2], w[0], w[11], w[3]);
+        round!(f, g, h, a, b, c, d, e, 0x34b0bcb5, w[3], w[1], w[12], w[4]);
+        round!(e, f, g, h, a, b, c, d, 0x391c0cb3, w[4], w[2], w[13], w[5]);
+        round!(d, e, f, g, h, a, b, c, 0x4ed8aa4a, w[5], w[3], w[14], w[6]);
+        round!(c, d, e, f, g, h, a, b, 0x5b9cca4f, w[6], w[4], w[15], w[7]);
+        round!(b, c, d, e, f, g, h, a, 0x682e6ff3, w[7], w[5], w[0], w[8]);
+        round!(a, b, c, d, e, f, g, h, 0x748f82ee, w[8], w[6], w[1], w[9]);
+        round!(h, a, b, c, d, e, f, g, 0x78a5636f, w[9], w[7], w[2], w[10]);
+        round!(g, h, a, b, c, d, e, f, 0x84c87814, w[10], w[8], w[3], w[11]);
+        round!(f, g, h, a, b, c, d, e, 0x8cc70208, w[11], w[9], w[4], w[12]);
+        round!(e, f, g, h, a, b, c, d, 0x90befffa, w[12], w[10], w[5], w[13]);
+        round!(d, e, f, g, h, a, b, c, 0xa4506ceb, w[13], w[11], w[6], w[14]);
+        round!(c, d, e, f, g, h, a, b, 0xbef9a3f7, w[14], w[12], w[7], w[15]);
+        round!(b, c, d, e, f, g, h, a, 0xc67178f2, w[15], w[13], w[8], w[0]);
+
+        self.h[0] = self.h[0].wrapping_add(a);
+        self.h[1] = self.h[1].wrapping_add(b);
+        self.h[2] = self.h[2].wrapping_add(c);
+        self.h[3] = self.h[3].wrapping_add(d);
+        self.h[4] = self.h[4].wrapping_add(e);
+        self.h[5] = self.h[5].wrapping_add(f);
+        self.h[6] = self.h[6].wrapping_add(g);
+        self.h[7] = self.h[7].wrapping_add(h);
+    }
+}
diff --git a/hashes/src/sha256/mod.rs b/hashes/src/sha256/mod.rs
index 8770b332b..cd6beb24f 100644
--- a/hashes/src/sha256/mod.rs
+++ b/hashes/src/sha256/mod.rs
@@ -2,10 +2,8 @@
 
 //! SHA256 implementation.
 
-#[cfg(all(feature = "std", target_arch = "x86"))]
-use core::arch::x86::*;
-#[cfg(all(feature = "std", target_arch = "x86_64"))]
-use core::arch::x86_64::*;
+mod crypto;
+
 use core::{cmp, convert, fmt};
 
 use crate::{incomplete_block_len, sha256d, HashEngine as _};
@@ -256,617 +254,6 @@ impl fmt::Display for MidstateError {
 #[cfg(feature = "std")]
 impl std::error::Error for MidstateError {}
 
-#[allow(non_snake_case)]
-const fn Ch(x: u32, y: u32, z: u32) -> u32 { z ^ (x & (y ^ z)) }
-#[allow(non_snake_case)]
-const fn Maj(x: u32, y: u32, z: u32) -> u32 { (x & y) | (z & (x | y)) }
-#[allow(non_snake_case)]
-const fn Sigma0(x: u32) -> u32 { x.rotate_left(30) ^ x.rotate_left(19) ^ x.rotate_left(10) }
-#[allow(non_snake_case)]
-const fn Sigma1(x: u32) -> u32 { x.rotate_left(26) ^ x.rotate_left(21) ^ x.rotate_left(7) }
-const fn sigma0(x: u32) -> u32 { x.rotate_left(25) ^ x.rotate_left(14) ^ (x >> 3) }
-const fn sigma1(x: u32) -> u32 { x.rotate_left(15) ^ x.rotate_left(13) ^ (x >> 10) }
-
-#[cfg(feature = "small-hash")]
-#[macro_use]
-mod small_hash {
-    use super::*;
-
-    #[rustfmt::skip]
-    pub(super) const fn round(a: u32, b: u32, c: u32, d: u32, e: u32,
-                              f: u32, g: u32, h: u32, k: u32, w: u32) -> (u32, u32) {
-        let t1 =
-            h.wrapping_add(Sigma1(e)).wrapping_add(Ch(e, f, g)).wrapping_add(k).wrapping_add(w);
-        let t2 = Sigma0(a).wrapping_add(Maj(a, b, c));
-        (d.wrapping_add(t1), t1.wrapping_add(t2))
-    }
-    #[rustfmt::skip]
-    pub(super) const fn later_round(a: u32, b: u32, c: u32, d: u32, e: u32,
-                                    f: u32, g: u32, h: u32, k: u32, w: u32,
-                                    w1: u32, w2: u32, w3: u32,
-    ) -> (u32, u32, u32) {
-        let w = w.wrapping_add(sigma1(w1)).wrapping_add(w2).wrapping_add(sigma0(w3));
-        let (d, h) = round(a, b, c, d, e, f, g, h, k, w);
-        (d, h, w)
-    }
-
-    macro_rules! round(
-        // first round
-        ($a:expr, $b:expr, $c:expr, $d:expr, $e:expr, $f:expr, $g:expr, $h:expr, $k:expr, $w:expr) => (
-            let updates = small_hash::round($a, $b, $c, $d, $e, $f, $g, $h, $k, $w);
-            $d = updates.0;
-            $h = updates.1;
-        );
-        // later rounds we reassign $w before doing the first-round computation
-        ($a:expr, $b:expr, $c:expr, $d:expr, $e:expr, $f:expr, $g:expr, $h:expr, $k:expr, $w:expr, $w1:expr, $w2:expr, $w3:expr) => (
-            let updates = small_hash::later_round($a, $b, $c, $d, $e, $f, $g, $h, $k, $w, $w1, $w2, $w3);
-            $d = updates.0;
-            $h = updates.1;
-            $w = updates.2;
-        )
-    );
-}
-
-#[cfg(not(feature = "small-hash"))]
-#[macro_use]
-mod fast_hash {
-    macro_rules! round(
-        // first round
-        ($a:expr, $b:expr, $c:expr, $d:expr, $e:expr, $f:expr, $g:expr, $h:expr, $k:expr, $w:expr) => (
-            let t1 = $h.wrapping_add(Sigma1($e)).wrapping_add(Ch($e, $f, $g)).wrapping_add($k).wrapping_add($w);
-            let t2 = Sigma0($a).wrapping_add(Maj($a, $b, $c));
-            $d = $d.wrapping_add(t1);
-            $h = t1.wrapping_add(t2);
-        );
-        // later rounds we reassign $w before doing the first-round computation
-        ($a:expr, $b:expr, $c:expr, $d:expr, $e:expr, $f:expr, $g:expr, $h:expr, $k:expr, $w:expr, $w1:expr, $w2:expr, $w3:expr) => (
-            $w = $w.wrapping_add(sigma1($w1)).wrapping_add($w2).wrapping_add(sigma0($w3));
-            round!($a, $b, $c, $d, $e, $f, $g, $h, $k, $w);
-        )
-    );
-}
-
-impl Midstate {
-    #[allow(clippy::identity_op)] // more readble
-    const fn read_u32(bytes: &[u8], index: usize) -> u32 {
-        ((bytes[index + 0] as u32) << 24)
-            | ((bytes[index + 1] as u32) << 16)
-            | ((bytes[index + 2] as u32) << 8)
-            | ((bytes[index + 3] as u32) << 0)
-    }
-
-    const fn copy_w(bytes: &[u8], index: usize) -> [u32; 16] {
-        let mut w = [0u32; 16];
-        let mut i = 0;
-        while i < 16 {
-            w[i] = Self::read_u32(bytes, index + i * 4);
-            i += 1;
-        }
-        w
-    }
-
-    const fn compute_midstate_unoptimized(bytes: &[u8], finalize: bool) -> Self {
-        let mut state = [
-            0x6a09e667u32,
-            0xbb67ae85,
-            0x3c6ef372,
-            0xa54ff53a,
-            0x510e527f,
-            0x9b05688c,
-            0x1f83d9ab,
-            0x5be0cd19,
-        ];
-
-        let num_chunks = (bytes.len() + 9 + 63) / 64;
-        let mut chunk = 0;
-        #[allow(clippy::precedence)]
-        while chunk < num_chunks {
-            if !finalize && chunk + 1 == num_chunks {
-                break;
-            }
-            let mut w = if chunk * 64 + 64 <= bytes.len() {
-                Self::copy_w(bytes, chunk * 64)
-            } else {
-                let mut buf = [0; 64];
-                let mut i = 0;
-                let offset = chunk * 64;
-                while offset + i < bytes.len() {
-                    buf[i] = bytes[offset + i];
-                    i += 1;
-                }
-                if (bytes.len() % 64 <= 64 - 9) || (chunk + 2 == num_chunks) {
-                    buf[i] = 0x80;
-                }
-                #[allow(clippy::identity_op)] // more readble
-                #[allow(clippy::erasing_op)]
-                if chunk + 1 == num_chunks {
-                    let bit_len = bytes.len() as u64 * 8;
-                    buf[64 - 8] = ((bit_len >> 8 * 7) & 0xFF) as u8;
-                    buf[64 - 7] = ((bit_len >> 8 * 6) & 0xFF) as u8;
-                    buf[64 - 6] = ((bit_len >> 8 * 5) & 0xFF) as u8;
-                    buf[64 - 5] = ((bit_len >> 8 * 4) & 0xFF) as u8;
-                    buf[64 - 4] = ((bit_len >> 8 * 3) & 0xFF) as u8;
-                    buf[64 - 3] = ((bit_len >> 8 * 2) & 0xFF) as u8;
-                    buf[64 - 2] = ((bit_len >> 8 * 1) & 0xFF) as u8;
-                    buf[64 - 1] = ((bit_len >> 8 * 0) & 0xFF) as u8;
-                }
-                Self::copy_w(&buf, 0)
-            };
-            chunk += 1;
-
-            let mut a = state[0];
-            let mut b = state[1];
-            let mut c = state[2];
-            let mut d = state[3];
-            let mut e = state[4];
-            let mut f = state[5];
-            let mut g = state[6];
-            let mut h = state[7];
-
-            round!(a, b, c, d, e, f, g, h, 0x428a2f98, w[0]);
-            round!(h, a, b, c, d, e, f, g, 0x71374491, w[1]);
-            round!(g, h, a, b, c, d, e, f, 0xb5c0fbcf, w[2]);
-            round!(f, g, h, a, b, c, d, e, 0xe9b5dba5, w[3]);
-            round!(e, f, g, h, a, b, c, d, 0x3956c25b, w[4]);
-            round!(d, e, f, g, h, a, b, c, 0x59f111f1, w[5]);
-            round!(c, d, e, f, g, h, a, b, 0x923f82a4, w[6]);
-            round!(b, c, d, e, f, g, h, a, 0xab1c5ed5, w[7]);
-            round!(a, b, c, d, e, f, g, h, 0xd807aa98, w[8]);
-            round!(h, a, b, c, d, e, f, g, 0x12835b01, w[9]);
-            round!(g, h, a, b, c, d, e, f, 0x243185be, w[10]);
-            round!(f, g, h, a, b, c, d, e, 0x550c7dc3, w[11]);
-            round!(e, f, g, h, a, b, c, d, 0x72be5d74, w[12]);
-            round!(d, e, f, g, h, a, b, c, 0x80deb1fe, w[13]);
-            round!(c, d, e, f, g, h, a, b, 0x9bdc06a7, w[14]);
-            round!(b, c, d, e, f, g, h, a, 0xc19bf174, w[15]);
-
-            round!(a, b, c, d, e, f, g, h, 0xe49b69c1, w[0], w[14], w[9], w[1]);
-            round!(h, a, b, c, d, e, f, g, 0xefbe4786, w[1], w[15], w[10], w[2]);
-            round!(g, h, a, b, c, d, e, f, 0x0fc19dc6, w[2], w[0], w[11], w[3]);
-            round!(f, g, h, a, b, c, d, e, 0x240ca1cc, w[3], w[1], w[12], w[4]);
-            round!(e, f, g, h, a, b, c, d, 0x2de92c6f, w[4], w[2], w[13], w[5]);
-            round!(d, e, f, g, h, a, b, c, 0x4a7484aa, w[5], w[3], w[14], w[6]);
-            round!(c, d, e, f, g, h, a, b, 0x5cb0a9dc, w[6], w[4], w[15], w[7]);
-            round!(b, c, d, e, f, g, h, a, 0x76f988da, w[7], w[5], w[0], w[8]);
-            round!(a, b, c, d, e, f, g, h, 0x983e5152, w[8], w[6], w[1], w[9]);
-            round!(h, a, b, c, d, e, f, g, 0xa831c66d, w[9], w[7], w[2], w[10]);
-            round!(g, h, a, b, c, d, e, f, 0xb00327c8, w[10], w[8], w[3], w[11]);
-            round!(f, g, h, a, b, c, d, e, 0xbf597fc7, w[11], w[9], w[4], w[12]);
-            round!(e, f, g, h, a, b, c, d, 0xc6e00bf3, w[12], w[10], w[5], w[13]);
-            round!(d, e, f, g, h, a, b, c, 0xd5a79147, w[13], w[11], w[6], w[14]);
-            round!(c, d, e, f, g, h, a, b, 0x06ca6351, w[14], w[12], w[7], w[15]);
-            round!(b, c, d, e, f, g, h, a, 0x14292967, w[15], w[13], w[8], w[0]);
-
-            round!(a, b, c, d, e, f, g, h, 0x27b70a85, w[0], w[14], w[9], w[1]);
-            round!(h, a, b, c, d, e, f, g, 0x2e1b2138, w[1], w[15], w[10], w[2]);
-            round!(g, h, a, b, c, d, e, f, 0x4d2c6dfc, w[2], w[0], w[11], w[3]);
-            round!(f, g, h, a, b, c, d, e, 0x53380d13, w[3], w[1], w[12], w[4]);
-            round!(e, f, g, h, a, b, c, d, 0x650a7354, w[4], w[2], w[13], w[5]);
-            round!(d, e, f, g, h, a, b, c, 0x766a0abb, w[5], w[3], w[14], w[6]);
-            round!(c, d, e, f, g, h, a, b, 0x81c2c92e, w[6], w[4], w[15], w[7]);
-            round!(b, c, d, e, f, g, h, a, 0x92722c85, w[7], w[5], w[0], w[8]);
-            round!(a, b, c, d, e, f, g, h, 0xa2bfe8a1, w[8], w[6], w[1], w[9]);
-            round!(h, a, b, c, d, e, f, g, 0xa81a664b, w[9], w[7], w[2], w[10]);
-            round!(g, h, a, b, c, d, e, f, 0xc24b8b70, w[10], w[8], w[3], w[11]);
-            round!(f, g, h, a, b, c, d, e, 0xc76c51a3, w[11], w[9], w[4], w[12]);
-            round!(e, f, g, h, a, b, c, d, 0xd192e819, w[12], w[10], w[5], w[13]);
-            round!(d, e, f, g, h, a, b, c, 0xd6990624, w[13], w[11], w[6], w[14]);
-            round!(c, d, e, f, g, h, a, b, 0xf40e3585, w[14], w[12], w[7], w[15]);
-            round!(b, c, d, e, f, g, h, a, 0x106aa070, w[15], w[13], w[8], w[0]);
-
-            round!(a, b, c, d, e, f, g, h, 0x19a4c116, w[0], w[14], w[9], w[1]);
-            round!(h, a, b, c, d, e, f, g, 0x1e376c08, w[1], w[15], w[10], w[2]);
-            round!(g, h, a, b, c, d, e, f, 0x2748774c, w[2], w[0], w[11], w[3]);
-            round!(f, g, h, a, b, c, d, e, 0x34b0bcb5, w[3], w[1], w[12], w[4]);
-            round!(e, f, g, h, a, b, c, d, 0x391c0cb3, w[4], w[2], w[13], w[5]);
-            round!(d, e, f, g, h, a, b, c, 0x4ed8aa4a, w[5], w[3], w[14], w[6]);
-            round!(c, d, e, f, g, h, a, b, 0x5b9cca4f, w[6], w[4], w[15], w[7]);
-            round!(b, c, d, e, f, g, h, a, 0x682e6ff3, w[7], w[5], w[0], w[8]);
-            round!(a, b, c, d, e, f, g, h, 0x748f82ee, w[8], w[6], w[1], w[9]);
-            round!(h, a, b, c, d, e, f, g, 0x78a5636f, w[9], w[7], w[2], w[10]);
-            round!(g, h, a, b, c, d, e, f, 0x84c87814, w[10], w[8], w[3], w[11]);
-            round!(f, g, h, a, b, c, d, e, 0x8cc70208, w[11], w[9], w[4], w[12]);
-            round!(e, f, g, h, a, b, c, d, 0x90befffa, w[12], w[10], w[5], w[13]);
-            round!(d, e, f, g, h, a, b, c, 0xa4506ceb, w[13], w[11], w[6], w[14]);
-            round!(c, d, e, f, g, h, a, b, 0xbef9a3f7, w[14], w[12], w[7], w[15]);
-            round!(b, c, d, e, f, g, h, a, 0xc67178f2, w[15], w[13], w[8], w[0]);
-
-            state[0] = state[0].wrapping_add(a);
-            state[1] = state[1].wrapping_add(b);
-            state[2] = state[2].wrapping_add(c);
-            state[3] = state[3].wrapping_add(d);
-            state[4] = state[4].wrapping_add(e);
-            state[5] = state[5].wrapping_add(f);
-            state[6] = state[6].wrapping_add(g);
-            state[7] = state[7].wrapping_add(h);
-        }
-        let mut output = [0u8; 32];
-        let mut i = 0;
-        #[allow(clippy::identity_op)] // more readble
-        while i < 8 {
-            output[i * 4 + 0] = (state[i + 0] >> 24) as u8;
-            output[i * 4 + 1] = (state[i + 0] >> 16) as u8;
-            output[i * 4 + 2] = (state[i + 0] >> 8) as u8;
-            output[i * 4 + 3] = (state[i + 0] >> 0) as u8;
-            i += 1;
-        }
-        Midstate { bytes: output, bytes_hashed: bytes.len() as u64 }
-    }
-}
-
-impl HashEngine {
-    fn process_block(&mut self) {
-        #[cfg(all(feature = "std", any(target_arch = "x86", target_arch = "x86_64")))]
-        {
-            if std::is_x86_feature_detected!("sse4.1")
-                && std::is_x86_feature_detected!("sha")
-                && std::is_x86_feature_detected!("sse2")
-                && std::is_x86_feature_detected!("ssse3")
-            {
-                return unsafe { self.process_block_simd_x86_intrinsics() };
-            }
-        }
-
-        // fallback implementation without using any intrinsics
-        self.software_process_block()
-    }
-
-    #[cfg(all(feature = "std", any(target_arch = "x86", target_arch = "x86_64")))]
-    #[target_feature(enable = "sha,sse2,ssse3,sse4.1")]
-    unsafe fn process_block_simd_x86_intrinsics(&mut self) {
-        // Code translated and based on from
-        // https://github.com/noloader/SHA-Intrinsics/blob/4899efc81d1af159c1fd955936c673139f35aea9/sha256-x86.c
-
-        /* sha256-x86.c - Intel SHA extensions using C intrinsics  */
-        /*   Written and place in public domain by Jeffrey Walton  */
-        /*   Based on code from Intel, and by Sean Gulley for      */
-        /*   the miTLS project.                                    */
-
-        // Variable names are also kept the same as in the original C code for easier comparison.
-        let (mut state0, mut state1);
-        let (mut msg, mut tmp);
-
-        let (mut msg0, mut msg1, mut msg2, mut msg3);
-
-        let (abef_save, cdgh_save);
-
-        #[allow(non_snake_case)]
-        let MASK: __m128i =
-            _mm_set_epi64x(0x0c0d_0e0f_0809_0a0bu64 as i64, 0x0405_0607_0001_0203u64 as i64);
-
-        let block_offset = 0;
-
-        // Load initial values
-        // CAST SAFETY: loadu_si128 documentation states that mem_addr does not
-        // need to be aligned on any particular boundary.
-        tmp = _mm_loadu_si128(self.h.as_ptr().add(0) as *const __m128i);
-        state1 = _mm_loadu_si128(self.h.as_ptr().add(4) as *const __m128i);
-
-        tmp = _mm_shuffle_epi32(tmp, 0xB1); // CDAB
-        state1 = _mm_shuffle_epi32(state1, 0x1B); // EFGH
-        state0 = _mm_alignr_epi8(tmp, state1, 8); // ABEF
-        state1 = _mm_blend_epi16(state1, tmp, 0xF0); // CDGH
-
-        // Process a single block
-        {
-            // Save current state
-            abef_save = state0;
-            cdgh_save = state1;
-
-            // Rounds 0-3
-            msg = _mm_loadu_si128(self.buffer.as_ptr().add(block_offset) as *const __m128i);
-            msg0 = _mm_shuffle_epi8(msg, MASK);
-            msg = _mm_add_epi32(
-                msg0,
-                _mm_set_epi64x(0xE9B5DBA5B5C0FBCFu64 as i64, 0x71374491428A2F98u64 as i64),
-            );
-            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
-            msg = _mm_shuffle_epi32(msg, 0x0E);
-            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
-
-            // Rounds 4-7
-            msg1 = _mm_loadu_si128(self.buffer.as_ptr().add(block_offset + 16) as *const __m128i);
-            msg1 = _mm_shuffle_epi8(msg1, MASK);
-            msg = _mm_add_epi32(
-                msg1,
-                _mm_set_epi64x(0xAB1C5ED5923F82A4u64 as i64, 0x59F111F13956C25Bu64 as i64),
-            );
-            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
-            msg = _mm_shuffle_epi32(msg, 0x0E);
-            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
-            msg0 = _mm_sha256msg1_epu32(msg0, msg1);
-
-            // Rounds 8-11
-            msg2 = _mm_loadu_si128(self.buffer.as_ptr().add(block_offset + 32) as *const __m128i);
-            msg2 = _mm_shuffle_epi8(msg2, MASK);
-            msg = _mm_add_epi32(
-                msg2,
-                _mm_set_epi64x(0x550C7DC3243185BEu64 as i64, 0x12835B01D807AA98u64 as i64),
-            );
-            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
-            msg = _mm_shuffle_epi32(msg, 0x0E);
-            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
-            msg1 = _mm_sha256msg1_epu32(msg1, msg2);
-
-            // Rounds 12-15
-            msg3 = _mm_loadu_si128(self.buffer.as_ptr().add(block_offset + 48) as *const __m128i);
-            msg3 = _mm_shuffle_epi8(msg3, MASK);
-            msg = _mm_add_epi32(
-                msg3,
-                _mm_set_epi64x(0xC19BF1749BDC06A7u64 as i64, 0x80DEB1FE72BE5D74u64 as i64),
-            );
-            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
-            tmp = _mm_alignr_epi8(msg3, msg2, 4);
-            msg0 = _mm_add_epi32(msg0, tmp);
-            msg0 = _mm_sha256msg2_epu32(msg0, msg3);
-            msg = _mm_shuffle_epi32(msg, 0x0E);
-            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
-            msg2 = _mm_sha256msg1_epu32(msg2, msg3);
-
-            // Rounds 16-19
-            msg = _mm_add_epi32(
-                msg0,
-                _mm_set_epi64x(0x240CA1CC0FC19DC6u64 as i64, 0xEFBE4786E49B69C1u64 as i64),
-            );
-            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
-            tmp = _mm_alignr_epi8(msg0, msg3, 4);
-            msg1 = _mm_add_epi32(msg1, tmp);
-            msg1 = _mm_sha256msg2_epu32(msg1, msg0);
-            msg = _mm_shuffle_epi32(msg, 0x0E);
-            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
-            msg3 = _mm_sha256msg1_epu32(msg3, msg0);
-
-            // Rounds 20-23
-            msg = _mm_add_epi32(
-                msg1,
-                _mm_set_epi64x(0x76F988DA5CB0A9DCu64 as i64, 0x4A7484AA2DE92C6Fu64 as i64),
-            );
-            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
-            tmp = _mm_alignr_epi8(msg1, msg0, 4);
-            msg2 = _mm_add_epi32(msg2, tmp);
-            msg2 = _mm_sha256msg2_epu32(msg2, msg1);
-            msg = _mm_shuffle_epi32(msg, 0x0E);
-            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
-            msg0 = _mm_sha256msg1_epu32(msg0, msg1);
-
-            // Rounds 24-27
-            msg = _mm_add_epi32(
-                msg2,
-                _mm_set_epi64x(0xBF597FC7B00327C8u64 as i64, 0xA831C66D983E5152u64 as i64),
-            );
-            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
-            tmp = _mm_alignr_epi8(msg2, msg1, 4);
-            msg3 = _mm_add_epi32(msg3, tmp);
-            msg3 = _mm_sha256msg2_epu32(msg3, msg2);
-            msg = _mm_shuffle_epi32(msg, 0x0E);
-            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
-            msg1 = _mm_sha256msg1_epu32(msg1, msg2);
-
-            // Rounds 28-31
-            msg = _mm_add_epi32(
-                msg3,
-                _mm_set_epi64x(0x1429296706CA6351u64 as i64, 0xD5A79147C6E00BF3u64 as i64),
-            );
-            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
-            tmp = _mm_alignr_epi8(msg3, msg2, 4);
-            msg0 = _mm_add_epi32(msg0, tmp);
-            msg0 = _mm_sha256msg2_epu32(msg0, msg3);
-            msg = _mm_shuffle_epi32(msg, 0x0E);
-            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
-            msg2 = _mm_sha256msg1_epu32(msg2, msg3);
-
-            // Rounds 32-35
-            msg = _mm_add_epi32(
-                msg0,
-                _mm_set_epi64x(0x53380D134D2C6DFCu64 as i64, 0x2E1B213827B70A85u64 as i64),
-            );
-            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
-            tmp = _mm_alignr_epi8(msg0, msg3, 4);
-            msg1 = _mm_add_epi32(msg1, tmp);
-            msg1 = _mm_sha256msg2_epu32(msg1, msg0);
-            msg = _mm_shuffle_epi32(msg, 0x0E);
-            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
-            msg3 = _mm_sha256msg1_epu32(msg3, msg0);
-
-            // Rounds 36-39
-            msg = _mm_add_epi32(
-                msg1,
-                _mm_set_epi64x(0x92722C8581C2C92Eu64 as i64, 0x766A0ABB650A7354u64 as i64),
-            );
-            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
-            tmp = _mm_alignr_epi8(msg1, msg0, 4);
-            msg2 = _mm_add_epi32(msg2, tmp);
-            msg2 = _mm_sha256msg2_epu32(msg2, msg1);
-            msg = _mm_shuffle_epi32(msg, 0x0E);
-            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
-            msg0 = _mm_sha256msg1_epu32(msg0, msg1);
-
-            // Rounds 40-43
-            msg = _mm_add_epi32(
-                msg2,
-                _mm_set_epi64x(0xC76C51A3C24B8B70u64 as i64, 0xA81A664BA2BFE8A1u64 as i64),
-            );
-            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
-            tmp = _mm_alignr_epi8(msg2, msg1, 4);
-            msg3 = _mm_add_epi32(msg3, tmp);
-            msg3 = _mm_sha256msg2_epu32(msg3, msg2);
-            msg = _mm_shuffle_epi32(msg, 0x0E);
-            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
-            msg1 = _mm_sha256msg1_epu32(msg1, msg2);
-
-            // Rounds 44-47
-            msg = _mm_add_epi32(
-                msg3,
-                _mm_set_epi64x(0x106AA070F40E3585u64 as i64, 0xD6990624D192E819u64 as i64),
-            );
-            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
-            tmp = _mm_alignr_epi8(msg3, msg2, 4);
-            msg0 = _mm_add_epi32(msg0, tmp);
-            msg0 = _mm_sha256msg2_epu32(msg0, msg3);
-            msg = _mm_shuffle_epi32(msg, 0x0E);
-            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
-            msg2 = _mm_sha256msg1_epu32(msg2, msg3);
-
-            // Rounds 48-51
-            msg = _mm_add_epi32(
-                msg0,
-                _mm_set_epi64x(0x34B0BCB52748774Cu64 as i64, 0x1E376C0819A4C116u64 as i64),
-            );
-            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
-            tmp = _mm_alignr_epi8(msg0, msg3, 4);
-            msg1 = _mm_add_epi32(msg1, tmp);
-            msg1 = _mm_sha256msg2_epu32(msg1, msg0);
-            msg = _mm_shuffle_epi32(msg, 0x0E);
-            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
-            msg3 = _mm_sha256msg1_epu32(msg3, msg0);
-
-            // Rounds 52-55
-            msg = _mm_add_epi32(
-                msg1,
-                _mm_set_epi64x(0x682E6FF35B9CCA4Fu64 as i64, 0x4ED8AA4A391C0CB3u64 as i64),
-            );
-            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
-            tmp = _mm_alignr_epi8(msg1, msg0, 4);
-            msg2 = _mm_add_epi32(msg2, tmp);
-            msg2 = _mm_sha256msg2_epu32(msg2, msg1);
-            msg = _mm_shuffle_epi32(msg, 0x0E);
-            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
-
-            // Rounds 56-59
-            msg = _mm_add_epi32(
-                msg2,
-                _mm_set_epi64x(0x8CC7020884C87814u64 as i64, 0x78A5636F748F82EEu64 as i64),
-            );
-            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
-            tmp = _mm_alignr_epi8(msg2, msg1, 4);
-            msg3 = _mm_add_epi32(msg3, tmp);
-            msg3 = _mm_sha256msg2_epu32(msg3, msg2);
-            msg = _mm_shuffle_epi32(msg, 0x0E);
-            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
-
-            // Rounds 60-63
-            msg = _mm_add_epi32(
-                msg3,
-                _mm_set_epi64x(0xC67178F2BEF9A3F7u64 as i64, 0xA4506CEB90BEFFFAu64 as i64),
-            );
-            state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
-            msg = _mm_shuffle_epi32(msg, 0x0E);
-            state0 = _mm_sha256rnds2_epu32(state0, state1, msg);
-
-            // Combine state
-            state0 = _mm_add_epi32(state0, abef_save);
-            state1 = _mm_add_epi32(state1, cdgh_save);
-        }
-
-        tmp = _mm_shuffle_epi32(state0, 0x1B); // FEBA
-        state1 = _mm_shuffle_epi32(state1, 0xB1); // DCHG
-        state0 = _mm_blend_epi16(tmp, state1, 0xF0); // DCBA
-        state1 = _mm_alignr_epi8(state1, tmp, 8); // ABEF
-
-        // Save state
-        // CAST SAFETY: storeu_si128 documentation states that mem_addr does not
-        // need to be aligned on any particular boundary.
-        _mm_storeu_si128(self.h.as_mut_ptr().add(0) as *mut __m128i, state0);
-        _mm_storeu_si128(self.h.as_mut_ptr().add(4) as *mut __m128i, state1);
-    }
-
-    // Algorithm copied from libsecp256k1
-    fn software_process_block(&mut self) {
-        debug_assert_eq!(self.buffer.len(), BLOCK_SIZE);
-
-        let mut w = [0u32; 16];
-        for (w_val, buff_bytes) in w.iter_mut().zip(self.buffer.chunks_exact(4)) {
-            *w_val = u32::from_be_bytes(buff_bytes.try_into().expect("4 byte slice"));
-        }
-
-        let mut a = self.h[0];
-        let mut b = self.h[1];
-        let mut c = self.h[2];
-        let mut d = self.h[3];
-        let mut e = self.h[4];
-        let mut f = self.h[5];
-        let mut g = self.h[6];
-        let mut h = self.h[7];
-
-        round!(a, b, c, d, e, f, g, h, 0x428a2f98, w[0]);
-        round!(h, a, b, c, d, e, f, g, 0x71374491, w[1]);
-        round!(g, h, a, b, c, d, e, f, 0xb5c0fbcf, w[2]);
-        round!(f, g, h, a, b, c, d, e, 0xe9b5dba5, w[3]);
-        round!(e, f, g, h, a, b, c, d, 0x3956c25b, w[4]);
-        round!(d, e, f, g, h, a, b, c, 0x59f111f1, w[5]);
-        round!(c, d, e, f, g, h, a, b, 0x923f82a4, w[6]);
-        round!(b, c, d, e, f, g, h, a, 0xab1c5ed5, w[7]);
-        round!(a, b, c, d, e, f, g, h, 0xd807aa98, w[8]);
-        round!(h, a, b, c, d, e, f, g, 0x12835b01, w[9]);
-        round!(g, h, a, b, c, d, e, f, 0x243185be, w[10]);
-        round!(f, g, h, a, b, c, d, e, 0x550c7dc3, w[11]);
-        round!(e, f, g, h, a, b, c, d, 0x72be5d74, w[12]);
-        round!(d, e, f, g, h, a, b, c, 0x80deb1fe, w[13]);
-        round!(c, d, e, f, g, h, a, b, 0x9bdc06a7, w[14]);
-        round!(b, c, d, e, f, g, h, a, 0xc19bf174, w[15]);
-
-        round!(a, b, c, d, e, f, g, h, 0xe49b69c1, w[0], w[14], w[9], w[1]);
-        round!(h, a, b, c, d, e, f, g, 0xefbe4786, w[1], w[15], w[10], w[2]);
-        round!(g, h, a, b, c, d, e, f, 0x0fc19dc6, w[2], w[0], w[11], w[3]);
-        round!(f, g, h, a, b, c, d, e, 0x240ca1cc, w[3], w[1], w[12], w[4]);
-        round!(e, f, g, h, a, b, c, d, 0x2de92c6f, w[4], w[2], w[13], w[5]);
-        round!(d, e, f, g, h, a, b, c, 0x4a7484aa, w[5], w[3], w[14], w[6]);
-        round!(c, d, e, f, g, h, a, b, 0x5cb0a9dc, w[6], w[4], w[15], w[7]);
-        round!(b, c, d, e, f, g, h, a, 0x76f988da, w[7], w[5], w[0], w[8]);
-        round!(a, b, c, d, e, f, g, h, 0x983e5152, w[8], w[6], w[1], w[9]);
-        round!(h, a, b, c, d, e, f, g, 0xa831c66d, w[9], w[7], w[2], w[10]);
-        round!(g, h, a, b, c, d, e, f, 0xb00327c8, w[10], w[8], w[3], w[11]);
-        round!(f, g, h, a, b, c, d, e, 0xbf597fc7, w[11], w[9], w[4], w[12]);
-        round!(e, f, g, h, a, b, c, d, 0xc6e00bf3, w[12], w[10], w[5], w[13]);
-        round!(d, e, f, g, h, a, b, c, 0xd5a79147, w[13], w[11], w[6], w[14]);
-        round!(c, d, e, f, g, h, a, b, 0x06ca6351, w[14], w[12], w[7], w[15]);
-        round!(b, c, d, e, f, g, h, a, 0x14292967, w[15], w[13], w[8], w[0]);
-
-        round!(a, b, c, d, e, f, g, h, 0x27b70a85, w[0], w[14], w[9], w[1]);
-        round!(h, a, b, c, d, e, f, g, 0x2e1b2138, w[1], w[15], w[10], w[2]);
-        round!(g, h, a, b, c, d, e, f, 0x4d2c6dfc, w[2], w[0], w[11], w[3]);
-        round!(f, g, h, a, b, c, d, e, 0x53380d13, w[3], w[1], w[12], w[4]);
-        round!(e, f, g, h, a, b, c, d, 0x650a7354, w[4], w[2], w[13], w[5]);
-        round!(d, e, f, g, h, a, b, c, 0x766a0abb, w[5], w[3], w[14], w[6]);
-        round!(c, d, e, f, g, h, a, b, 0x81c2c92e, w[6], w[4], w[15], w[7]);
-        round!(b, c, d, e, f, g, h, a, 0x92722c85, w[7], w[5], w[0], w[8]);
-        round!(a, b, c, d, e, f, g, h, 0xa2bfe8a1, w[8], w[6], w[1], w[9]);
-        round!(h, a, b, c, d, e, f, g, 0xa81a664b, w[9], w[7], w[2], w[10]);
-        round!(g, h, a, b, c, d, e, f, 0xc24b8b70, w[10], w[8], w[3], w[11]);
-        round!(f, g, h, a, b, c, d, e, 0xc76c51a3, w[11], w[9], w[4], w[12]);
-        round!(e, f, g, h, a, b, c, d, 0xd192e819, w[12], w[10], w[5], w[13]);
-        round!(d, e, f, g, h, a, b, c, 0xd6990624, w[13], w[11], w[6], w[14]);
-        round!(c, d, e, f, g, h, a, b, 0xf40e3585, w[14], w[12], w[7], w[15]);
-        round!(b, c, d, e, f, g, h, a, 0x106aa070, w[15], w[13], w[8], w[0]);
-
-        round!(a, b, c, d, e, f, g, h, 0x19a4c116, w[0], w[14], w[9], w[1]);
-        round!(h, a, b, c, d, e, f, g, 0x1e376c08, w[1], w[15], w[10], w[2]);
-        round!(g, h, a, b, c, d, e, f, 0x2748774c, w[2], w[0], w[11], w[3]);
-        round!(f, g, h, a, b, c, d, e, 0x34b0bcb5, w[3], w[1], w[12], w[4]);
-        round!(e, f, g, h, a, b, c, d, 0x391c0cb3, w[4], w[2], w[13], w[5]);
-        round!(d, e, f, g, h, a, b, c, 0x4ed8aa4a, w[5], w[3], w[14], w[6]);
-        round!(c, d, e, f, g, h, a, b, 0x5b9cca4f, w[6], w[4], w[15], w[7]);
-        round!(b, c, d, e, f, g, h, a, 0x682e6ff3, w[7], w[5], w[0], w[8]);
-        round!(a, b, c, d, e, f, g, h, 0x748f82ee, w[8], w[6], w[1], w[9]);
-        round!(h, a, b, c, d, e, f, g, 0x78a5636f, w[9], w[7], w[2], w[10]);
-        round!(g, h, a, b, c, d, e, f, 0x84c87814, w[10], w[8], w[3], w[11]);
-        round!(f, g, h, a, b, c, d, e, 0x8cc70208, w[11], w[9], w[4], w[12]);
-        round!(e, f, g, h, a, b, c, d, 0x90befffa, w[12], w[10], w[5], w[13]);
-        round!(d, e, f, g, h, a, b, c, 0xa4506ceb, w[13], w[11], w[6], w[14]);
-        round!(c, d, e, f, g, h, a, b, 0xbef9a3f7, w[14], w[12], w[7], w[15]);
-        round!(b, c, d, e, f, g, h, a, 0xc67178f2, w[15], w[13], w[8], w[0]);
-
-        self.h[0] = self.h[0].wrapping_add(a);
-        self.h[1] = self.h[1].wrapping_add(b);
-        self.h[2] = self.h[2].wrapping_add(c);
-        self.h[3] = self.h[3].wrapping_add(d);
-        self.h[4] = self.h[4].wrapping_add(e);
-        self.h[5] = self.h[5].wrapping_add(f);
-        self.h[6] = self.h[6].wrapping_add(g);
-        self.h[7] = self.h[7].wrapping_add(h);
-    }
-}
-
 #[cfg(test)]
 mod tests {
     use core::array;
diff --git a/hashes/src/sha512/crypto.rs b/hashes/src/sha512/crypto.rs
new file mode 100644
index 000000000..cbb1c82ab
--- /dev/null
+++ b/hashes/src/sha512/crypto.rs
@@ -0,0 +1,185 @@
+// SPDX-License-Identifier: CC0-1.0
+
+use super::{HashEngine, BLOCK_SIZE};
+
+#[allow(non_snake_case)]
+fn Ch(x: u64, y: u64, z: u64) -> u64 { z ^ (x & (y ^ z)) }
+#[allow(non_snake_case)]
+fn Maj(x: u64, y: u64, z: u64) -> u64 { (x & y) | (z & (x | y)) }
+#[allow(non_snake_case)]
+fn Sigma0(x: u64) -> u64 { x.rotate_left(36) ^ x.rotate_left(30) ^ x.rotate_left(25) }
+#[allow(non_snake_case)]
+fn Sigma1(x: u64) -> u64 { x.rotate_left(50) ^ x.rotate_left(46) ^ x.rotate_left(23) }
+fn sigma0(x: u64) -> u64 { x.rotate_left(63) ^ x.rotate_left(56) ^ (x >> 7) }
+fn sigma1(x: u64) -> u64 { x.rotate_left(45) ^ x.rotate_left(3) ^ (x >> 6) }
+
+#[cfg(feature = "small-hash")]
+#[macro_use]
+mod small_hash {
+    use super::*;
+
+    #[rustfmt::skip]
+    pub(super) fn round(a: u64, b: u64, c: u64, d: &mut u64, e: u64,
+                        f: u64, g: u64, h: &mut u64, k: u64, w: u64,
+    ) {
+        let t1 =
+            h.wrapping_add(Sigma1(e)).wrapping_add(Ch(e, f, g)).wrapping_add(k).wrapping_add(w);
+        let t2 = Sigma0(a).wrapping_add(Maj(a, b, c));
+        *d = d.wrapping_add(t1);
+        *h = t1.wrapping_add(t2);
+    }
+    #[rustfmt::skip]
+    pub(super) fn later_round(a: u64, b: u64, c: u64, d: &mut u64, e: u64,
+                              f: u64, g: u64, h: &mut u64, k: u64, w: u64,
+                              w1: u64, w2: u64, w3: u64,
+    ) -> u64 {
+        let w = w.wrapping_add(sigma1(w1)).wrapping_add(w2).wrapping_add(sigma0(w3));
+        round(a, b, c, d, e, f, g, h, k, w);
+        w
+    }
+
+    macro_rules! round(
+        // first round
+        ($a:expr, $b:expr, $c:expr, $d:expr, $e:expr, $f:expr, $g:expr, $h:expr, $k:expr, $w:expr) => (
+            small_hash::round($a, $b, $c, &mut $d, $e, $f, $g, &mut $h, $k, $w)
+        );
+        // later rounds we reassign $w before doing the first-round computation
+        ($a:expr, $b:expr, $c:expr, $d:expr, $e:expr, $f:expr, $g:expr, $h:expr, $k:expr, $w:expr, $w1:expr, $w2:expr, $w3:expr) => (
+            $w = small_hash::later_round($a, $b, $c, &mut $d, $e, $f, $g, &mut $h, $k, $w, $w1, $w2, $w3)
+        )
+    );
+}
+
+#[cfg(not(feature = "small-hash"))]
+#[macro_use]
+mod fast_hash {
+    macro_rules! round(
+        // first round
+        ($a:expr, $b:expr, $c:expr, $d:expr, $e:expr, $f:expr, $g:expr, $h:expr, $k:expr, $w:expr) => (
+            let t1 = $h.wrapping_add(Sigma1($e)).wrapping_add(Ch($e, $f, $g)).wrapping_add($k).wrapping_add($w);
+            let t2 = Sigma0($a).wrapping_add(Maj($a, $b, $c));
+            $d = $d.wrapping_add(t1);
+            $h = t1.wrapping_add(t2);
+        );
+        // later rounds we reassign $w before doing the first-round computation
+        ($a:expr, $b:expr, $c:expr, $d:expr, $e:expr, $f:expr, $g:expr, $h:expr, $k:expr, $w:expr, $w1:expr, $w2:expr, $w3:expr) => (
+            $w = $w.wrapping_add(sigma1($w1)).wrapping_add($w2).wrapping_add(sigma0($w3));
+            round!($a, $b, $c, $d, $e, $f, $g, $h, $k, $w);
+        )
+    );
+}
+
+impl HashEngine {
+    // Algorithm copied from libsecp256k1
+    pub(crate) fn process_block(&mut self) {
+        debug_assert_eq!(self.buffer.len(), BLOCK_SIZE);
+
+        let mut w = [0u64; 16];
+        for (w_val, buff_bytes) in w.iter_mut().zip(self.buffer.chunks_exact(8)) {
+            *w_val = u64::from_be_bytes(buff_bytes.try_into().expect("8 byte slice"));
+        }
+
+        let mut a = self.h[0];
+        let mut b = self.h[1];
+        let mut c = self.h[2];
+        let mut d = self.h[3];
+        let mut e = self.h[4];
+        let mut f = self.h[5];
+        let mut g = self.h[6];
+        let mut h = self.h[7];
+
+        round!(a, b, c, d, e, f, g, h, 0x428a2f98d728ae22, w[0]);
+        round!(h, a, b, c, d, e, f, g, 0x7137449123ef65cd, w[1]);
+        round!(g, h, a, b, c, d, e, f, 0xb5c0fbcfec4d3b2f, w[2]);
+        round!(f, g, h, a, b, c, d, e, 0xe9b5dba58189dbbc, w[3]);
+        round!(e, f, g, h, a, b, c, d, 0x3956c25bf348b538, w[4]);
+        round!(d, e, f, g, h, a, b, c, 0x59f111f1b605d019, w[5]);
+        round!(c, d, e, f, g, h, a, b, 0x923f82a4af194f9b, w[6]);
+        round!(b, c, d, e, f, g, h, a, 0xab1c5ed5da6d8118, w[7]);
+        round!(a, b, c, d, e, f, g, h, 0xd807aa98a3030242, w[8]);
+        round!(h, a, b, c, d, e, f, g, 0x12835b0145706fbe, w[9]);
+        round!(g, h, a, b, c, d, e, f, 0x243185be4ee4b28c, w[10]);
+        round!(f, g, h, a, b, c, d, e, 0x550c7dc3d5ffb4e2, w[11]);
+        round!(e, f, g, h, a, b, c, d, 0x72be5d74f27b896f, w[12]);
+        round!(d, e, f, g, h, a, b, c, 0x80deb1fe3b1696b1, w[13]);
+        round!(c, d, e, f, g, h, a, b, 0x9bdc06a725c71235, w[14]);
+        round!(b, c, d, e, f, g, h, a, 0xc19bf174cf692694, w[15]);
+
+        round!(a, b, c, d, e, f, g, h, 0xe49b69c19ef14ad2, w[0], w[14], w[9], w[1]);
+        round!(h, a, b, c, d, e, f, g, 0xefbe4786384f25e3, w[1], w[15], w[10], w[2]);
+        round!(g, h, a, b, c, d, e, f, 0x0fc19dc68b8cd5b5, w[2], w[0], w[11], w[3]);
+        round!(f, g, h, a, b, c, d, e, 0x240ca1cc77ac9c65, w[3], w[1], w[12], w[4]);
+        round!(e, f, g, h, a, b, c, d, 0x2de92c6f592b0275, w[4], w[2], w[13], w[5]);
+        round!(d, e, f, g, h, a, b, c, 0x4a7484aa6ea6e483, w[5], w[3], w[14], w[6]);
+        round!(c, d, e, f, g, h, a, b, 0x5cb0a9dcbd41fbd4, w[6], w[4], w[15], w[7]);
+        round!(b, c, d, e, f, g, h, a, 0x76f988da831153b5, w[7], w[5], w[0], w[8]);
+        round!(a, b, c, d, e, f, g, h, 0x983e5152ee66dfab, w[8], w[6], w[1], w[9]);
+        round!(h, a, b, c, d, e, f, g, 0xa831c66d2db43210, w[9], w[7], w[2], w[10]);
+        round!(g, h, a, b, c, d, e, f, 0xb00327c898fb213f, w[10], w[8], w[3], w[11]);
+        round!(f, g, h, a, b, c, d, e, 0xbf597fc7beef0ee4, w[11], w[9], w[4], w[12]);
+        round!(e, f, g, h, a, b, c, d, 0xc6e00bf33da88fc2, w[12], w[10], w[5], w[13]);
+        round!(d, e, f, g, h, a, b, c, 0xd5a79147930aa725, w[13], w[11], w[6], w[14]);
+        round!(c, d, e, f, g, h, a, b, 0x06ca6351e003826f, w[14], w[12], w[7], w[15]);
+        round!(b, c, d, e, f, g, h, a, 0x142929670a0e6e70, w[15], w[13], w[8], w[0]);
+
+        round!(a, b, c, d, e, f, g, h, 0x27b70a8546d22ffc, w[0], w[14], w[9], w[1]);
+        round!(h, a, b, c, d, e, f, g, 0x2e1b21385c26c926, w[1], w[15], w[10], w[2]);
+        round!(g, h, a, b, c, d, e, f, 0x4d2c6dfc5ac42aed, w[2], w[0], w[11], w[3]);
+        round!(f, g, h, a, b, c, d, e, 0x53380d139d95b3df, w[3], w[1], w[12], w[4]);
+        round!(e, f, g, h, a, b, c, d, 0x650a73548baf63de, w[4], w[2], w[13], w[5]);
+        round!(d, e, f, g, h, a, b, c, 0x766a0abb3c77b2a8, w[5], w[3], w[14], w[6]);
+        round!(c, d, e, f, g, h, a, b, 0x81c2c92e47edaee6, w[6], w[4], w[15], w[7]);
+        round!(b, c, d, e, f, g, h, a, 0x92722c851482353b, w[7], w[5], w[0], w[8]);
+        round!(a, b, c, d, e, f, g, h, 0xa2bfe8a14cf10364, w[8], w[6], w[1], w[9]);
+        round!(h, a, b, c, d, e, f, g, 0xa81a664bbc423001, w[9], w[7], w[2], w[10]);
+        round!(g, h, a, b, c, d, e, f, 0xc24b8b70d0f89791, w[10], w[8], w[3], w[11]);
+        round!(f, g, h, a, b, c, d, e, 0xc76c51a30654be30, w[11], w[9], w[4], w[12]);
+        round!(e, f, g, h, a, b, c, d, 0xd192e819d6ef5218, w[12], w[10], w[5], w[13]);
+        round!(d, e, f, g, h, a, b, c, 0xd69906245565a910, w[13], w[11], w[6], w[14]);
+        round!(c, d, e, f, g, h, a, b, 0xf40e35855771202a, w[14], w[12], w[7], w[15]);
+        round!(b, c, d, e, f, g, h, a, 0x106aa07032bbd1b8, w[15], w[13], w[8], w[0]);
+
+        round!(a, b, c, d, e, f, g, h, 0x19a4c116b8d2d0c8, w[0], w[14], w[9], w[1]);
+        round!(h, a, b, c, d, e, f, g, 0x1e376c085141ab53, w[1], w[15], w[10], w[2]);
+        round!(g, h, a, b, c, d, e, f, 0x2748774cdf8eeb99, w[2], w[0], w[11], w[3]);
+        round!(f, g, h, a, b, c, d, e, 0x34b0bcb5e19b48a8, w[3], w[1], w[12], w[4]);
+        round!(e, f, g, h, a, b, c, d, 0x391c0cb3c5c95a63, w[4], w[2], w[13], w[5]);
+        round!(d, e, f, g, h, a, b, c, 0x4ed8aa4ae3418acb, w[5], w[3], w[14], w[6]);
+        round!(c, d, e, f, g, h, a, b, 0x5b9cca4f7763e373, w[6], w[4], w[15], w[7]);
+        round!(b, c, d, e, f, g, h, a, 0x682e6ff3d6b2b8a3, w[7], w[5], w[0], w[8]);
+        round!(a, b, c, d, e, f, g, h, 0x748f82ee5defb2fc, w[8], w[6], w[1], w[9]);
+        round!(h, a, b, c, d, e, f, g, 0x78a5636f43172f60, w[9], w[7], w[2], w[10]);
+        round!(g, h, a, b, c, d, e, f, 0x84c87814a1f0ab72, w[10], w[8], w[3], w[11]);
+        round!(f, g, h, a, b, c, d, e, 0x8cc702081a6439ec, w[11], w[9], w[4], w[12]);
+        round!(e, f, g, h, a, b, c, d, 0x90befffa23631e28, w[12], w[10], w[5], w[13]);
+        round!(d, e, f, g, h, a, b, c, 0xa4506cebde82bde9, w[13], w[11], w[6], w[14]);
+        round!(c, d, e, f, g, h, a, b, 0xbef9a3f7b2c67915, w[14], w[12], w[7], w[15]);
+        round!(b, c, d, e, f, g, h, a, 0xc67178f2e372532b, w[15], w[13], w[8], w[0]);
+
+        round!(a, b, c, d, e, f, g, h, 0xca273eceea26619c, w[0], w[14], w[9], w[1]);
+        round!(h, a, b, c, d, e, f, g, 0xd186b8c721c0c207, w[1], w[15], w[10], w[2]);
+        round!(g, h, a, b, c, d, e, f, 0xeada7dd6cde0eb1e, w[2], w[0], w[11], w[3]);
+        round!(f, g, h, a, b, c, d, e, 0xf57d4f7fee6ed178, w[3], w[1], w[12], w[4]);
+        round!(e, f, g, h, a, b, c, d, 0x06f067aa72176fba, w[4], w[2], w[13], w[5]);
+        round!(d, e, f, g, h, a, b, c, 0x0a637dc5a2c898a6, w[5], w[3], w[14], w[6]);
+        round!(c, d, e, f, g, h, a, b, 0x113f9804bef90dae, w[6], w[4], w[15], w[7]);
+        round!(b, c, d, e, f, g, h, a, 0x1b710b35131c471b, w[7], w[5], w[0], w[8]);
+        round!(a, b, c, d, e, f, g, h, 0x28db77f523047d84, w[8], w[6], w[1], w[9]);
+        round!(h, a, b, c, d, e, f, g, 0x32caab7b40c72493, w[9], w[7], w[2], w[10]);
+        round!(g, h, a, b, c, d, e, f, 0x3c9ebe0a15c9bebc, w[10], w[8], w[3], w[11]);
+        round!(f, g, h, a, b, c, d, e, 0x431d67c49c100d4c, w[11], w[9], w[4], w[12]);
+        round!(e, f, g, h, a, b, c, d, 0x4cc5d4becb3e42b6, w[12], w[10], w[5], w[13]);
+        round!(d, e, f, g, h, a, b, c, 0x597f299cfc657e2a, w[13], w[11], w[6], w[14]);
+        round!(c, d, e, f, g, h, a, b, 0x5fcb6fab3ad6faec, w[14], w[12], w[7], w[15]);
+        round!(b, c, d, e, f, g, h, a, 0x6c44198c4a475817, w[15], w[13], w[8], w[0]);
+
+        self.h[0] = self.h[0].wrapping_add(a);
+        self.h[1] = self.h[1].wrapping_add(b);
+        self.h[2] = self.h[2].wrapping_add(c);
+        self.h[3] = self.h[3].wrapping_add(d);
+        self.h[4] = self.h[4].wrapping_add(e);
+        self.h[5] = self.h[5].wrapping_add(f);
+        self.h[6] = self.h[6].wrapping_add(g);
+        self.h[7] = self.h[7].wrapping_add(h);
+    }
+}
diff --git a/hashes/src/sha512/mod.rs b/hashes/src/sha512/mod.rs
index 1ecd563df..4db0bec54 100644
--- a/hashes/src/sha512/mod.rs
+++ b/hashes/src/sha512/mod.rs
@@ -2,6 +2,8 @@
 
 //! SHA512 implementation.
 
+mod crypto;
+
 use core::cmp;
 
 use crate::{incomplete_block_len, HashEngine as _};
@@ -121,188 +123,6 @@ impl crate::HashEngine for HashEngine {
     crate::internal_macros::engine_input_impl!();
 }
 
-#[allow(non_snake_case)]
-fn Ch(x: u64, y: u64, z: u64) -> u64 { z ^ (x & (y ^ z)) }
-#[allow(non_snake_case)]
-fn Maj(x: u64, y: u64, z: u64) -> u64 { (x & y) | (z & (x | y)) }
-#[allow(non_snake_case)]
-fn Sigma0(x: u64) -> u64 { x.rotate_left(36) ^ x.rotate_left(30) ^ x.rotate_left(25) }
-#[allow(non_snake_case)]
-fn Sigma1(x: u64) -> u64 { x.rotate_left(50) ^ x.rotate_left(46) ^ x.rotate_left(23) }
-fn sigma0(x: u64) -> u64 { x.rotate_left(63) ^ x.rotate_left(56) ^ (x >> 7) }
-fn sigma1(x: u64) -> u64 { x.rotate_left(45) ^ x.rotate_left(3) ^ (x >> 6) }
-
-#[cfg(feature = "small-hash")]
-#[macro_use]
-mod small_hash {
-    use super::*;
-
-    #[rustfmt::skip]
-    pub(super) fn round(a: u64, b: u64, c: u64, d: &mut u64, e: u64,
-                        f: u64, g: u64, h: &mut u64, k: u64, w: u64,
-    ) {
-        let t1 =
-            h.wrapping_add(Sigma1(e)).wrapping_add(Ch(e, f, g)).wrapping_add(k).wrapping_add(w);
-        let t2 = Sigma0(a).wrapping_add(Maj(a, b, c));
-        *d = d.wrapping_add(t1);
-        *h = t1.wrapping_add(t2);
-    }
-    #[rustfmt::skip]
-    pub(super) fn later_round(a: u64, b: u64, c: u64, d: &mut u64, e: u64,
-                              f: u64, g: u64, h: &mut u64, k: u64, w: u64,
-                              w1: u64, w2: u64, w3: u64,
-    ) -> u64 {
-        let w = w.wrapping_add(sigma1(w1)).wrapping_add(w2).wrapping_add(sigma0(w3));
-        round(a, b, c, d, e, f, g, h, k, w);
-        w
-    }
-
-    macro_rules! round(
-        // first round
-        ($a:expr, $b:expr, $c:expr, $d:expr, $e:expr, $f:expr, $g:expr, $h:expr, $k:expr, $w:expr) => (
-            small_hash::round($a, $b, $c, &mut $d, $e, $f, $g, &mut $h, $k, $w)
-        );
-        // later rounds we reassign $w before doing the first-round computation
-        ($a:expr, $b:expr, $c:expr, $d:expr, $e:expr, $f:expr, $g:expr, $h:expr, $k:expr, $w:expr, $w1:expr, $w2:expr, $w3:expr) => (
-            $w = small_hash::later_round($a, $b, $c, &mut $d, $e, $f, $g, &mut $h, $k, $w, $w1, $w2, $w3)
-        )
-    );
-}
-
-#[cfg(not(feature = "small-hash"))]
-#[macro_use]
-mod fast_hash {
-    macro_rules! round(
-        // first round
-        ($a:expr, $b:expr, $c:expr, $d:expr, $e:expr, $f:expr, $g:expr, $h:expr, $k:expr, $w:expr) => (
-            let t1 = $h.wrapping_add(Sigma1($e)).wrapping_add(Ch($e, $f, $g)).wrapping_add($k).wrapping_add($w);
-            let t2 = Sigma0($a).wrapping_add(Maj($a, $b, $c));
-            $d = $d.wrapping_add(t1);
-            $h = t1.wrapping_add(t2);
-        );
-        // later rounds we reassign $w before doing the first-round computation
-        ($a:expr, $b:expr, $c:expr, $d:expr, $e:expr, $f:expr, $g:expr, $h:expr, $k:expr, $w:expr, $w1:expr, $w2:expr, $w3:expr) => (
-            $w = $w.wrapping_add(sigma1($w1)).wrapping_add($w2).wrapping_add(sigma0($w3));
-            round!($a, $b, $c, $d, $e, $f, $g, $h, $k, $w);
-        )
-    );
-}
-
-impl HashEngine {
-    // Algorithm copied from libsecp256k1
-    pub(crate) fn process_block(&mut self) {
-        debug_assert_eq!(self.buffer.len(), BLOCK_SIZE);
-
-        let mut w = [0u64; 16];
-        for (w_val, buff_bytes) in w.iter_mut().zip(self.buffer.chunks_exact(8)) {
-            *w_val = u64::from_be_bytes(buff_bytes.try_into().expect("8 byte slice"));
-        }
-
-        let mut a = self.h[0];
-        let mut b = self.h[1];
-        let mut c = self.h[2];
-        let mut d = self.h[3];
-        let mut e = self.h[4];
-        let mut f = self.h[5];
-        let mut g = self.h[6];
-        let mut h = self.h[7];
-
-        round!(a, b, c, d, e, f, g, h, 0x428a2f98d728ae22, w[0]);
-        round!(h, a, b, c, d, e, f, g, 0x7137449123ef65cd, w[1]);
-        round!(g, h, a, b, c, d, e, f, 0xb5c0fbcfec4d3b2f, w[2]);
-        round!(f, g, h, a, b, c, d, e, 0xe9b5dba58189dbbc, w[3]);
-        round!(e, f, g, h, a, b, c, d, 0x3956c25bf348b538, w[4]);
-        round!(d, e, f, g, h, a, b, c, 0x59f111f1b605d019, w[5]);
-        round!(c, d, e, f, g, h, a, b, 0x923f82a4af194f9b, w[6]);
-        round!(b, c, d, e, f, g, h, a, 0xab1c5ed5da6d8118, w[7]);
-        round!(a, b, c, d, e, f, g, h, 0xd807aa98a3030242, w[8]);
-        round!(h, a, b, c, d, e, f, g, 0x12835b0145706fbe, w[9]);
-        round!(g, h, a, b, c, d, e, f, 0x243185be4ee4b28c, w[10]);
-        round!(f, g, h, a, b, c, d, e, 0x550c7dc3d5ffb4e2, w[11]);
-        round!(e, f, g, h, a, b, c, d, 0x72be5d74f27b896f, w[12]);
-        round!(d, e, f, g, h, a, b, c, 0x80deb1fe3b1696b1, w[13]);
-        round!(c, d, e, f, g, h, a, b, 0x9bdc06a725c71235, w[14]);
-        round!(b, c, d, e, f, g, h, a, 0xc19bf174cf692694, w[15]);
-
-        round!(a, b, c, d, e, f, g, h, 0xe49b69c19ef14ad2, w[0], w[14], w[9], w[1]);
-        round!(h, a, b, c, d, e, f, g, 0xefbe4786384f25e3, w[1], w[15], w[10], w[2]);
-        round!(g, h, a, b, c, d, e, f, 0x0fc19dc68b8cd5b5, w[2], w[0], w[11], w[3]);
-        round!(f, g, h, a, b, c, d, e, 0x240ca1cc77ac9c65, w[3], w[1], w[12], w[4]);
-        round!(e, f, g, h, a, b, c, d, 0x2de92c6f592b0275, w[4], w[2], w[13], w[5]);
-        round!(d, e, f, g, h, a, b, c, 0x4a7484aa6ea6e483, w[5], w[3], w[14], w[6]);
-        round!(c, d, e, f, g, h, a, b, 0x5cb0a9dcbd41fbd4, w[6], w[4], w[15], w[7]);
-        round!(b, c, d, e, f, g, h, a, 0x76f988da831153b5, w[7], w[5], w[0], w[8]);
-        round!(a, b, c, d, e, f, g, h, 0x983e5152ee66dfab, w[8], w[6], w[1], w[9]);
-        round!(h, a, b, c, d, e, f, g, 0xa831c66d2db43210, w[9], w[7], w[2], w[10]);
-        round!(g, h, a, b, c, d, e, f, 0xb00327c898fb213f, w[10], w[8], w[3], w[11]);
-        round!(f, g, h, a, b, c, d, e, 0xbf597fc7beef0ee4, w[11], w[9], w[4], w[12]);
-        round!(e, f, g, h, a, b, c, d, 0xc6e00bf33da88fc2, w[12], w[10], w[5], w[13]);
-        round!(d, e, f, g, h, a, b, c, 0xd5a79147930aa725, w[13], w[11], w[6], w[14]);
-        round!(c, d, e, f, g, h, a, b, 0x06ca6351e003826f, w[14], w[12], w[7], w[15]);
-        round!(b, c, d, e, f, g, h, a, 0x142929670a0e6e70, w[15], w[13], w[8], w[0]);
-
-        round!(a, b, c, d, e, f, g, h, 0x27b70a8546d22ffc, w[0], w[14], w[9], w[1]);
-        round!(h, a, b, c, d, e, f, g, 0x2e1b21385c26c926, w[1], w[15], w[10], w[2]);
-        round!(g, h, a, b, c, d, e, f, 0x4d2c6dfc5ac42aed, w[2], w[0], w[11], w[3]);
-        round!(f, g, h, a, b, c, d, e, 0x53380d139d95b3df, w[3], w[1], w[12], w[4]);
-        round!(e, f, g, h, a, b, c, d, 0x650a73548baf63de, w[4], w[2], w[13], w[5]);
-        round!(d, e, f, g, h, a, b, c, 0x766a0abb3c77b2a8, w[5], w[3], w[14], w[6]);
-        round!(c, d, e, f, g, h, a, b, 0x81c2c92e47edaee6, w[6], w[4], w[15], w[7]);
-        round!(b, c, d, e, f, g, h, a, 0x92722c851482353b, w[7], w[5], w[0], w[8]);
-        round!(a, b, c, d, e, f, g, h, 0xa2bfe8a14cf10364, w[8], w[6], w[1], w[9]);
-        round!(h, a, b, c, d, e, f, g, 0xa81a664bbc423001, w[9], w[7], w[2], w[10]);
-        round!(g, h, a, b, c, d, e, f, 0xc24b8b70d0f89791, w[10], w[8], w[3], w[11]);
-        round!(f, g, h, a, b, c, d, e, 0xc76c51a30654be30, w[11], w[9], w[4], w[12]);
-        round!(e, f, g, h, a, b, c, d, 0xd192e819d6ef5218, w[12], w[10], w[5], w[13]);
-        round!(d, e, f, g, h, a, b, c, 0xd69906245565a910, w[13], w[11], w[6], w[14]);
-        round!(c, d, e, f, g, h, a, b, 0xf40e35855771202a, w[14], w[12], w[7], w[15]);
-        round!(b, c, d, e, f, g, h, a, 0x106aa07032bbd1b8, w[15], w[13], w[8], w[0]);
-
-        round!(a, b, c, d, e, f, g, h, 0x19a4c116b8d2d0c8, w[0], w[14], w[9], w[1]);
-        round!(h, a, b, c, d, e, f, g, 0x1e376c085141ab53, w[1], w[15], w[10], w[2]);
-        round!(g, h, a, b, c, d, e, f, 0x2748774cdf8eeb99, w[2], w[0], w[11], w[3]);
-        round!(f, g, h, a, b, c, d, e, 0x34b0bcb5e19b48a8, w[3], w[1], w[12], w[4]);
-        round!(e, f, g, h, a, b, c, d, 0x391c0cb3c5c95a63, w[4], w[2], w[13], w[5]);
-        round!(d, e, f, g, h, a, b, c, 0x4ed8aa4ae3418acb, w[5], w[3], w[14], w[6]);
-        round!(c, d, e, f, g, h, a, b, 0x5b9cca4f7763e373, w[6], w[4], w[15], w[7]);
-        round!(b, c, d, e, f, g, h, a, 0x682e6ff3d6b2b8a3, w[7], w[5], w[0], w[8]);
-        round!(a, b, c, d, e, f, g, h, 0x748f82ee5defb2fc, w[8], w[6], w[1], w[9]);
-        round!(h, a, b, c, d, e, f, g, 0x78a5636f43172f60, w[9], w[7], w[2], w[10]);
-        round!(g, h, a, b, c, d, e, f, 0x84c87814a1f0ab72, w[10], w[8], w[3], w[11]);
-        round!(f, g, h, a, b, c, d, e, 0x8cc702081a6439ec, w[11], w[9], w[4], w[12]);
-        round!(e, f, g, h, a, b, c, d, 0x90befffa23631e28, w[12], w[10], w[5], w[13]);
-        round!(d, e, f, g, h, a, b, c, 0xa4506cebde82bde9, w[13], w[11], w[6], w[14]);
-        round!(c, d, e, f, g, h, a, b, 0xbef9a3f7b2c67915, w[14], w[12], w[7], w[15]);
-        round!(b, c, d, e, f, g, h, a, 0xc67178f2e372532b, w[15], w[13], w[8], w[0]);
-
-        round!(a, b, c, d, e, f, g, h, 0xca273eceea26619c, w[0], w[14], w[9], w[1]);
-        round!(h, a, b, c, d, e, f, g, 0xd186b8c721c0c207, w[1], w[15], w[10], w[2]);
-        round!(g, h, a, b, c, d, e, f, 0xeada7dd6cde0eb1e, w[2], w[0], w[11], w[3]);
-        round!(f, g, h, a, b, c, d, e, 0xf57d4f7fee6ed178, w[3], w[1], w[12], w[4]);
-        round!(e, f, g, h, a, b, c, d, 0x06f067aa72176fba, w[4], w[2], w[13], w[5]);
-        round!(d, e, f, g, h, a, b, c, 0x0a637dc5a2c898a6, w[5], w[3], w[14], w[6]);
-        round!(c, d, e, f, g, h, a, b, 0x113f9804bef90dae, w[6], w[4], w[15], w[7]);
-        round!(b, c, d, e, f, g, h, a, 0x1b710b35131c471b, w[7], w[5], w[0], w[8]);
-        round!(a, b, c, d, e, f, g, h, 0x28db77f523047d84, w[8], w[6], w[1], w[9]);
-        round!(h, a, b, c, d, e, f, g, 0x32caab7b40c72493, w[9], w[7], w[2], w[10]);
-        round!(g, h, a, b, c, d, e, f, 0x3c9ebe0a15c9bebc, w[10], w[8], w[3], w[11]);
-        round!(f, g, h, a, b, c, d, e, 0x431d67c49c100d4c, w[11], w[9], w[4], w[12]);
-        round!(e, f, g, h, a, b, c, d, 0x4cc5d4becb3e42b6, w[12], w[10], w[5], w[13]);
-        round!(d, e, f, g, h, a, b, c, 0x597f299cfc657e2a, w[13], w[11], w[6], w[14]);
-        round!(c, d, e, f, g, h, a, b, 0x5fcb6fab3ad6faec, w[14], w[12], w[7], w[15]);
-        round!(b, c, d, e, f, g, h, a, 0x6c44198c4a475817, w[15], w[13], w[8], w[0]);
-
-        self.h[0] = self.h[0].wrapping_add(a);
-        self.h[1] = self.h[1].wrapping_add(b);
-        self.h[2] = self.h[2].wrapping_add(c);
-        self.h[3] = self.h[3].wrapping_add(d);
-        self.h[4] = self.h[4].wrapping_add(e);
-        self.h[5] = self.h[5].wrapping_add(f);
-        self.h[6] = self.h[6].wrapping_add(g);
-        self.h[7] = self.h[7].wrapping_add(h);
-    }
-}
-
 #[cfg(test)]
 mod tests {
     #[test]