Fix zerocopy validation logic and add detailed explanations

Author: zebreus

src/lib.rs

@@ -1102,8 +1102,16 @@ mod impl_zerocopy {
             let my_candidate =
                 unsafe { candidate.assume_validity::<zerocopy::pointer::invariant::Valid>() };
             {
-                (my_candidate.read_unaligned::<zerocopy::pointer::BecauseImmutable>() ^ T::ALL_BITS)
-                    == T::EMPTY
+                // ALL_BITS has all valid bits set to 1. If we invert it we get a mask with all invalid bits.
+                let invalid_bits = !T::ALL_BITS;
+                // TODO: Currently this assumes that the candidate is aligned. We actually need to check this beforehand
+                // Dereference the pointer to the candidate
+                let candidate =
+                    my_candidate.read_unaligned::<zerocopy::pointer::BecauseImmutable>();
+                // By applying the invalid_bits mask to the candidate, only invalid bits will remain 1. So if there are any 1s left in this value we know that the candidate is invalid.
+                let invalid_bits_in_candidate = candidate & invalid_bits;
+                // Verify that there are no 1s left.
+                return invalid_bits_in_candidate == T::EMPTY;
             }
         }
     }