diff --git a/src/int/mod.rs b/src/int/mod.rs
index 1ab49f5a..5814f6e1 100644
--- a/src/int/mod.rs
+++ b/src/int/mod.rs
@@ -4,9 +4,9 @@ pub mod addsub;
 pub mod mul;
 pub mod shift;
 
+pub mod sdiv;
 mod specialized_div_rem;
 pub mod udiv;
-pub mod sdiv;
 
 /// Trait for some basic operations on integers
 pub trait Int:
diff --git a/src/int/sdiv.rs b/src/int/sdiv.rs
index f75f2ca9..af75cbdb 100644
--- a/src/int/sdiv.rs
+++ b/src/int/sdiv.rs
@@ -38,7 +38,7 @@ intrinsics! {
     pub extern "C" fn __moddi3(a: i64, b: i64) -> i64 {
         i64_div_rem(a, b).1
     }
-    
+
     #[aapcs_on_arm]
     /// Returns `n / d` and sets `*rem = n % d`
     pub extern "C" fn __divmoddi4(a: i64, b: i64, rem: &mut i64) -> i64 {
@@ -52,7 +52,7 @@ intrinsics! {
     pub extern "C" fn __divti3(a: i128, b: i128) -> i128 {
         i128_div_rem(a, b).0
     }
-    
+
     #[win64_128bit_abi_hack]
     /// Returns `n % d`
     pub extern "C" fn __modti3(a: i128, b: i128) -> i128 {
diff --git a/src/int/specialized_div_rem/asymmetric.rs b/src/int/specialized_div_rem/asymmetric.rs
index 133d8249..3a663dcf 100644
--- a/src/int/specialized_div_rem/asymmetric.rs
+++ b/src/int/specialized_div_rem/asymmetric.rs
@@ -14,7 +14,7 @@ macro_rules! impl_asymmetric {
     ) => {
         /// Computes the quotient and remainder of `duo` divided by `div` and returns them as a
         /// tuple.
-        /// 
+        ///
         /// This is optimized for dividing integers with the same bitwidth as the largest operand in
         /// an asymmetrically sized division. For example, the x86-64 `divq` assembly instruction
         /// can divide a 128 bit integer by a 64 bit integer if the quotient fits in 64 bits.
@@ -129,7 +129,7 @@ macro_rules! impl_asymmetric {
                 }
                 // Note that this is a large $uD multiplication being used here
                 let mut rem = duo - ((quo as $uD) * div);
-    
+
                 if rem >= div {
                     quo += 1;
                     rem -= div;
@@ -140,7 +140,7 @@ macro_rules! impl_asymmetric {
 
         /// Computes the quotient and remainder of `duo` divided by `div` and returns them as a
         /// tuple.
-        /// 
+        ///
         /// This is optimized for dividing integers with the same bitwidth as the largest operand in
         /// an asymmetrically sized division. For example, the x86-64 `divq` assembly instruction
         /// can divide a 128 bit integer by a 64 bit integer if the quotient fits in 64 bits.
diff --git a/src/int/specialized_div_rem/binary_long.rs b/src/int/specialized_div_rem/binary_long.rs
index dc80b75f..dc1b261e 100644
--- a/src/int/specialized_div_rem/binary_long.rs
+++ b/src/int/specialized_div_rem/binary_long.rs
@@ -11,7 +11,7 @@ macro_rules! impl_binary_long {
 
         /// Computes the quotient and remainder of `duo` divided by `div` and returns them as a
         /// tuple.
-        /// 
+        ///
         /// This uses binary shift long division only, and is designed for CPUs without fast
         /// multiplication or division hardware.
         ///
@@ -79,7 +79,7 @@ macro_rules! impl_binary_long {
 
         /// Computes the quotient and remainder of `duo` divided by `div` and returns them as a
         /// tuple.
-        /// 
+        ///
         /// This uses binary shift long division only, and is designed for CPUs without fast
         /// multiplication or division hardware.
         ///
diff --git a/src/int/specialized_div_rem/delegate.rs b/src/int/specialized_div_rem/delegate.rs
index bd6fc117..86bb2d6a 100644
--- a/src/int/specialized_div_rem/delegate.rs
+++ b/src/int/specialized_div_rem/delegate.rs
@@ -14,7 +14,7 @@ macro_rules! impl_delegate {
 
         /// Computes the quotient and remainder of `duo` divided by `div` and returns them as a
         /// tuple.
-        /// 
+        ///
         /// This uses binary shift long division, but if it can delegates work to a smaller
         /// division. This function is used for CPUs with a register size smaller than the division
         /// size, and that do not have fast multiplication or division hardware. For CPUs with a
@@ -168,7 +168,7 @@ macro_rules! impl_delegate {
 
         /// Computes the quotient and remainder of `duo` divided by `div` and returns them as a
         /// tuple.
-        /// 
+        ///
         /// This uses binary shift long division, but if it can delegates work to a smaller
         /// division. This function is used for CPUs with a register size smaller than the division
         /// size, and that do not have fast multiplication or division hardware. For CPUs with a
diff --git a/src/int/specialized_div_rem/mod.rs b/src/int/specialized_div_rem/mod.rs
index 9e55cb26..f628ca9d 100644
--- a/src/int/specialized_div_rem/mod.rs
+++ b/src/int/specialized_div_rem/mod.rs
@@ -19,7 +19,7 @@
 //    be used which calls `u32_div_rem_binary_long`. The `impl_binary_long!` macro uses
 //    no half division, so the chain of calls ends there.
 //  - LLVM supplies 16 bit divisions and smaller
-//  
+//
 // I could replace all `u64_by_u64_div_rem` in the macros by `u64_div_rem` and likewise
 // for `u32_by_u32_div_rem`, but that would mean that hardware divisions would never be
 // used.
@@ -79,7 +79,7 @@ unsafe fn u128_by_u64_div_rem(duo: u128, div: u64) -> (u64, u64) {
         : "{rax}"(duo_lo), "{rdx}"(duo_hi), "r"(div)
         : "rax", "rdx"
     );
-    return (quo, rem)
+    return (quo, rem);
 }
 
 // use `_asymmetric` instead of `_trifecta`, because x86_64 supplies the `divq` instruction
diff --git a/src/int/specialized_div_rem/trifecta.rs b/src/int/specialized_div_rem/trifecta.rs
index 12652b6f..1e1db3d0 100644
--- a/src/int/specialized_div_rem/trifecta.rs
+++ b/src/int/specialized_div_rem/trifecta.rs
@@ -13,13 +13,13 @@ macro_rules! impl_trifecta {
     ) => {
         /// Computes the quotient and remainder of `duo` divided by `div` and returns them as a
         /// tuple.
-        /// 
+        ///
         /// This is optimized for division of two integers with bit widths twice as large
         /// as the largest hardware integer division supported. Note that some architectures supply
         /// a division of an integer larger than register size by a regular sized integer (e.x.
         /// x86_64 has a `divq` instruction which can divide a 128 bit integer by a 64 bit integer).
         /// In that case, the `_asymmetric` algorithm should be used instead of this one.
-        /// 
+        ///
         /// This is called the trifecta algorithm because it uses three main algorithms: short
         /// division for small divisors, the "mul or mul - 1" algorithm for when the divisor is
         /// large enough for the quotient to be determined to be one of two values via only one
@@ -92,10 +92,10 @@ macro_rules! impl_trifecta {
                     tmp.1 as $uD
                 )
             }
-            
+
             // relative leading significant bits, cannot overflow because of above branches
             let rel_leading_sb = div_lz.wrapping_sub(duo_lz);
-            
+
             // `{2^n, 2^div_sb} <= duo < 2^n_d`
             // `1 <= div < {2^duo_sb, 2^(n_d - 1)}`
             // short division branch
@@ -122,7 +122,7 @@ macro_rules! impl_trifecta {
                     rem_1 as $uD
                 )
             }
-            
+
             // `{2^n, 2^div_sb} <= duo < 2^n_d`
             // `2^n_h <= div < {2^duo_sb, 2^(n_d - 1)}`
             // `mul` or `mul - 1` branch
@@ -266,11 +266,11 @@ macro_rules! impl_trifecta {
             // correct it. This long division algorithm has been carefully constructed to always
             // underguess the quotient by slim margins. This allows different subalgorithms
             // to be blindly jumped to without needing an extra correction step.
-            // 
+            //
             // The only problem is that it will not work
             // for many ranges of `duo` and `div`. Fortunately, the short division,
             // mul or mul - 1 algorithms, and simple divisions happen to exactly fill these gaps.
-            // 
+            //
             // For an example, consider the division of 76543210 by 213 and assume that `n_h`
             // is equal to two decimal digits (note: we are working with base 10 here for
             // readability).
@@ -399,13 +399,13 @@ macro_rules! impl_trifecta {
 
         /// Computes the quotient and remainder of `duo` divided by `div` and returns them as a
         /// tuple.
-        /// 
+        ///
         /// This is optimized for division of two integers with bit widths twice as large
         /// as the largest hardware integer division supported. Note that some architectures supply
         /// a division of an integer larger than register size by a regular sized integer (e.x.
         /// x86_64 has a `divq` instruction which can divide a 128 bit integer by a 64 bit integer).
         /// In that case, the `_asymmetric` algorithm should be used instead of this one.
-        /// 
+        ///
         /// This is called the trifecta algorithm because it uses three main algorithms: short
         /// division for small divisors, the "mul or mul - 1" algorithm for when the divisor is
         /// large enough for the quotient to be determined to be one of two values via only one
diff --git a/src/int/udiv.rs b/src/int/udiv.rs
index 1fd02899..198ad004 100644
--- a/src/int/udiv.rs
+++ b/src/int/udiv.rs
@@ -12,13 +12,13 @@ intrinsics! {
     pub extern "C" fn __udivsi3(n: u32, d: u32) -> u32 {
         u32_div_rem(n, d).0
     }
-    
+
     #[maybe_use_optimized_c_shim]
     /// Returns `n % d`
     pub extern "C" fn __umodsi3(n: u32, d: u32) -> u32 {
         u32_div_rem(n, d).1
     }
-    
+
     #[maybe_use_optimized_c_shim]
     /// Returns `n / d` and sets `*rem = n % d`
     pub extern "C" fn __udivmodsi4(n: u32, d: u32, rem: Option<&mut u32>) -> u32 {
@@ -28,19 +28,19 @@ intrinsics! {
         }
         quo_rem.0
     }
-    
+
     #[maybe_use_optimized_c_shim]
     /// Returns `n / d`
     pub extern "C" fn __udivdi3(n: u64, d: u64) -> u64 {
         u64_div_rem(n, d).0
     }
-    
+
     #[maybe_use_optimized_c_shim]
     /// Returns `n % d`
     pub extern "C" fn __umoddi3(n: u64, d: u64) -> u64 {
         u64_div_rem(n, d).1
     }
-    
+
     /// Returns `n / d` and sets `*rem = n % d`
     pub extern "C" fn __udivmoddi4(n: u64, d: u64, rem: Option<&mut u64>) -> u64 {
         let quo_rem = u64_div_rem(n, d);
@@ -49,7 +49,7 @@ intrinsics! {
         }
         quo_rem.0
     }
-    
+
     #[win64_128bit_abi_hack]
     /// Returns `n / d`
     pub extern "C" fn __udivti3(n: u128, d: u128) -> u128 {