diff --git a/benches/numparse.rs b/benches/numparse.rs
index fbb48f7..f2aa676 100644
--- a/benches/numparse.rs
+++ b/benches/numparse.rs
@@ -54,7 +54,7 @@ impl<'a, S: Simd> NumberParserSimd<'a, S> {
         Self {
             // Transmuting here is safe because we're converting from u8 to i8
             // This is necessary to make loading easier below
-            string: unsafe { core::mem::transmute(string.as_bytes()) },
+            string: unsafe { core::mem::transmute::<&[u8], &[i8]>(string.as_bytes()) },
 
             _simd: std::marker::PhantomData,
         }
diff --git a/examples/string_parse_sum.rs b/examples/string_parse_sum.rs
index cb19b21..ccb4c0a 100644
--- a/examples/string_parse_sum.rs
+++ b/examples/string_parse_sum.rs
@@ -75,7 +75,7 @@ impl<'a, S: Simd> NumberParserSimd<'a, S> {
         Self {
             // Transmuting here is safe because we're converting from u8 to i8
             // This is necessary to make loading easier below
-            string: unsafe { core::mem::transmute(string.as_bytes()) },
+            string: unsafe { core::mem::transmute::<&[u8], &[i8]>(string.as_bytes()) },
             _simd: std::marker::PhantomData,
         }
     }
diff --git a/src/base/iters.rs b/src/base/iters.rs
index 2cc6c9c..e0902b5 100644
--- a/src/base/iters.rs
+++ b/src/base/iters.rs
@@ -25,7 +25,7 @@ pub struct SimdArrayIterator<'a, S: SimdBaseIo> {
     index: usize,
 }
 
-impl<'a, S: SimdBaseIo> Iterator for SimdArrayIterator<'a, S> {
+impl<S: SimdBaseIo> Iterator for SimdArrayIterator<'_, S> {
     type Item = S::Scalar;
 
     #[inline(always)]
diff --git a/src/lib.rs b/src/lib.rs
index 2078dbe..81e47d7 100644
--- a/src/lib.rs
+++ b/src/lib.rs
@@ -147,15 +147,11 @@
 //!
 //! * `distance<S:Simd>`      the generic version of your function
 //! * `distance_compiletime`  the fastest instruction set availble for the given compile time
-//! feature set
+//!   feature set
 //!
 //! You may also forgo the macros if you know what you are doing, just keep in mind there are lots
 //! of arcane subtleties with inlining and target_features that must be managed. See how the macros
 //! expand for more detail.
-#![cfg_attr(
-    all(target_arch = "wasm32", not(feature = "stable")),
-    feature(core_intrinsics)
-)]
 #![allow(clippy::missing_safety_doc)] // TODO: Work on the safety of functions
 #![cfg_attr(all(feature = "no_std", not(test)), no_std)]
 #[macro_use]
diff --git a/src/libm_ext/mod.rs b/src/libm_ext/mod.rs
index cecece7..952f6a8 100644
--- a/src/libm_ext/mod.rs
+++ b/src/libm_ext/mod.rs
@@ -1,3 +1,4 @@
+#![allow(dead_code)]
 #[cfg(feature = "no_std")]
 mod no_std;
 
diff --git a/src/ops/f64.rs b/src/ops/f64.rs
index 0ed63d7..1995046 100644
--- a/src/ops/f64.rs
+++ b/src/ops/f64.rs
@@ -612,7 +612,7 @@ impl_op! {
 impl_op! {
     fn cast_i64<f64> {
         for Avx2(a: __m256d) -> __m256i {
-            let nums_arr = core::mem::transmute::<_, [f64; 4]>(a);
+            let nums_arr = core::mem::transmute::<__m256d, [f64; 4]>(a);
             let ceil = [
                 nums_arr[0].m_round() as i64,
                 nums_arr[1].m_round() as i64,
@@ -622,7 +622,7 @@ impl_op! {
             core::mem::transmute::<_, __m256i>(ceil)
         }
         for Sse41(a: __m128d) -> __m128i {
-            let nums_arr = core::mem::transmute::<_, [f64; 2]>(a);
+            let nums_arr = core::mem::transmute::<__m128d, [f64; 2]>(a);
             let ceil = [
                 nums_arr[0].m_round() as i64,
                 nums_arr[1].m_round() as i64,
@@ -630,7 +630,7 @@ impl_op! {
             core::mem::transmute::<_, __m128i>(ceil)
         }
         for Sse2(a: __m128d) -> __m128i {
-            let nums_arr = core::mem::transmute::<_, [f64; 2]>(a);
+            let nums_arr = core::mem::transmute::<__m128d, [f64; 2]>(a);
             let ceil = [
                 nums_arr[0].m_round() as i64,
                 nums_arr[1].m_round() as i64,
@@ -641,7 +641,7 @@ impl_op! {
             a.m_round() as i64
         }
         for Neon(a: float64x2_t) -> int64x2_t {
-            let nums_arr = core::mem::transmute::<_, [f64; 2]>(a);
+            let nums_arr = core::mem::transmute::<int64x2_t, [f64; 2]>(a);
             let ceil = [
                 nums_arr[0].m_round() as i64,
                 nums_arr[1].m_round() as i64,
diff --git a/src/ops/i16.rs b/src/ops/i16.rs
index a94433f..e51edef 100644
--- a/src/ops/i16.rs
+++ b/src/ops/i16.rs
@@ -415,7 +415,7 @@ impl_op! {
                 arr[6] as i32,
                 arr[7] as i32,
             ];
-            (core::mem::transmute(a), core::mem::transmute(b))
+            (core::mem::transmute::<[i32; 4], __m128i>(a), core::mem::transmute::<[i32; 4], __m128i>(b))
         }
         for Scalar(val: i16) -> (i32, i32) {
             (val as i32, 0)
@@ -454,7 +454,7 @@ impl_op! {
                 arr[6] as u16 as u32 as i32,
                 arr[7] as u16 as u32 as i32,
             ];
-            (core::mem::transmute(a), core::mem::transmute(b))
+            (core::mem::transmute::<[i32; 4], __m128i>(a), core::mem::transmute::<[i32; 4], __m128i>(b))
         }
         for Scalar(val: i16) -> (i32, i32) {
             (val as u16 as u32 as i32, 0)
diff --git a/src/ops/i32.rs b/src/ops/i32.rs
index e823462..2e5b087 100644
--- a/src/ops/i32.rs
+++ b/src/ops/i32.rs
@@ -49,8 +49,8 @@ impl_op! {
             _mm_mullo_epi32(a, b)
         }
         for Sse2(a: __m128i, b: __m128i) -> __m128i {
-            let a_arr = core::mem::transmute::<_, [i32; 4]>(a);
-            let b_arr = core::mem::transmute::<_, [i32; 4]>(b);
+            let a_arr = core::mem::transmute::<__m128i, [i32; 4]>(a);
+            let b_arr = core::mem::transmute::<__m128i, [i32; 4]>(b);
             let c_arr = [
                 a_arr[0].wrapping_mul(b_arr[0]),
                 a_arr[1].wrapping_mul(b_arr[1]),
@@ -450,10 +450,13 @@ impl_op! {
             Ops::<Sse2, i32>::extend_i64(val)
         }
         for Sse2(val: __m128i) -> (__m128i, __m128i) {
-            let arr = core::mem::transmute::<_, [i32; 4]>(val);
+            let arr = core::mem::transmute::<__m128i, [i32; 4]>(val);
             let a = [arr[0] as i64, arr[1] as i64];
             let b = [arr[2] as i64, arr[3] as i64];
-            (core::mem::transmute(a), core::mem::transmute(b))
+            (
+                core::mem::transmute::<[i64; 2], __m128i>(a),
+                core::mem::transmute::<[i64; 2], __m128i>(b)
+            )
         }
         for Scalar(val: i32) -> (i64, i64) {
             (val as i64, 0)
@@ -477,10 +480,10 @@ impl_op! {
             Ops::<Sse2, i32>::unsigned_extend_i64(val)
         }
         for Sse2(val: __m128i) -> (__m128i, __m128i) {
-            let arr = core::mem::transmute::<_, [i32; 4]>(val);
+            let arr = core::mem::transmute::<__m128i, [i32; 4]>(val);
             let a = [arr[0] as u32 as u64 as i64, arr[1] as u32 as u64 as i64];
             let b = [arr[2] as u32 as u64 as i64, arr[3] as u32 as u64 as i64];
-            (core::mem::transmute(a), core::mem::transmute(b))
+            (core::mem::transmute::<[i64; 2], __m128i>(a), core::mem::transmute::<[i64; 2], __m128i>(b))
         }
         for Scalar(val: i32) -> (i64, i64) {
             (val as u32 as u64 as i64, 0)
diff --git a/src/ops/i64.rs b/src/ops/i64.rs
index 84dc11f..532f62e 100644
--- a/src/ops/i64.rs
+++ b/src/ops/i64.rs
@@ -43,8 +43,8 @@ impl_op! {
 impl_op! {
     fn mul<i64> {
         for Avx2(a: __m256i, b: __m256i) -> __m256i {
-            let a_arr = core::mem::transmute::<_, [i64; 4]>(a);
-            let b_arr = core::mem::transmute::<_, [i64; 4]>(b);
+            let a_arr = core::mem::transmute::<__m256i, [i64; 4]>(a);
+            let b_arr = core::mem::transmute::<__m256i, [i64; 4]>(b);
             let c_arr = [
                 a_arr[0].wrapping_mul(b_arr[0]),
                 a_arr[1].wrapping_mul(b_arr[1]),
@@ -54,8 +54,8 @@ impl_op! {
             core::mem::transmute::<_, __m256i>(c_arr)
         }
         for Sse41(a: __m128i, b: __m128i) -> __m128i {
-            let a_arr = core::mem::transmute::<_, [i64; 2]>(a);
-            let b_arr = core::mem::transmute::<_, [i64; 2]>(b);
+            let a_arr = core::mem::transmute::<__m128i, [i64; 2]>(a);
+            let b_arr = core::mem::transmute::<__m128i, [i64; 2]>(b);
             let c_arr = [
                 a_arr[0].wrapping_mul(b_arr[0]),
                 a_arr[1].wrapping_mul(b_arr[1]),
@@ -63,8 +63,8 @@ impl_op! {
             core::mem::transmute::<_, __m128i>(c_arr)
         }
         for Sse2(a: __m128i, b: __m128i) -> __m128i {
-            let a_arr = core::mem::transmute::<_, [i64; 2]>(a);
-            let b_arr = core::mem::transmute::<_, [i64; 2]>(b);
+            let a_arr = core::mem::transmute::<__m128i, [i64; 2]>(a);
+            let b_arr = core::mem::transmute::<__m128i, [i64; 2]>(b);
             let c_arr = [
                 a_arr[0].wrapping_mul(b_arr[0]),
                 a_arr[1].wrapping_mul(b_arr[1]),
@@ -75,8 +75,8 @@ impl_op! {
             a.wrapping_mul(b)
         }
         for Neon(a: int64x2_t, b: int64x2_t) -> int64x2_t {
-            let a_arr = core::mem::transmute::<_, [i64; 2]>(a);
-            let b_arr = core::mem::transmute::<_, [i64; 2]>(b);
+            let a_arr = core::mem::transmute::<int64x2_t, [i64; 2]>(a);
+            let b_arr = core::mem::transmute::<int64x2_t, [i64; 2]>(b);
             let c_arr = [
                 a_arr[0].wrapping_mul(b_arr[0]),
                 a_arr[1].wrapping_mul(b_arr[1]),
@@ -430,7 +430,7 @@ impl_imm8_op! {
 impl_op! {
     fn cast_f64<i64> {
         for Avx2(a: __m256i) -> __m256d {
-            let arr = core::mem::transmute::<_, [i64; 4]>(a);
+            let arr = core::mem::transmute::<__m256i, [i64; 4]>(a);
             let result = [
                 arr[0] as f64,
                 arr[1] as f64,
@@ -440,7 +440,7 @@ impl_op! {
             core::mem::transmute::<_, __m256d>(result)
         }
         for Sse41(a: __m128i) -> __m128d {
-            let arr = core::mem::transmute::<_, [i64; 2]>(a);
+            let arr = core::mem::transmute::<__m128i, [i64; 2]>(a);
             let result = [
                 arr[0] as f64,
                 arr[1] as f64,
@@ -448,7 +448,7 @@ impl_op! {
             core::mem::transmute::<_, __m128d>(result)
         }
         for Sse2(a: __m128i) -> __m128d {
-            let arr = core::mem::transmute::<_, [i64; 2]>(a);
+            let arr = core::mem::transmute::<__m128i, [i64; 2]>(a);
             let result = [
                 arr[0] as f64,
                 arr[1] as f64,
diff --git a/src/ops/i8.rs b/src/ops/i8.rs
index 27bf42d..a50d5a0 100644
--- a/src/ops/i8.rs
+++ b/src/ops/i8.rs
@@ -43,24 +43,24 @@ impl_op! {
 impl_op! {
     fn mul<i8> {
         for Avx2(a: __m256i, b: __m256i) -> __m256i {
-            let mut arr1 = core::mem::transmute::<_, [i8; 32]>(a);
-            let arr2 = core::mem::transmute::<_, [i8; 32]>(b);
+            let mut arr1 = core::mem::transmute::<__m256i, [i8; 32]>(a);
+            let arr2 = core::mem::transmute::<__m256i, [i8; 32]>(b);
             for i in 0..32 {
                 arr1[i] = arr1[i].wrapping_mul(arr2[i]);
             }
             core::mem::transmute::<_, _>(arr1)
         }
         for Sse41(a: __m128i, b: __m128i) -> __m128i {
-            let mut arr1 = core::mem::transmute::<_, [i8; 16]>(a);
-            let arr2 = core::mem::transmute::<_, [i8; 16]>(b);
+            let mut arr1 = core::mem::transmute::<__m128i, [i8; 16]>(a);
+            let arr2 = core::mem::transmute::<__m128i, [i8; 16]>(b);
             for i in 0..16 {
                 arr1[i] = arr1[i].wrapping_mul(arr2[i]);
             }
             core::mem::transmute::<_, _>(arr1)
         }
         for Sse2(a: __m128i, b: __m128i) -> __m128i {
-            let mut arr1 = core::mem::transmute::<_, [i8; 16]>(a);
-            let arr2 = core::mem::transmute::<_, [i8; 16]>(b);
+            let mut arr1 = core::mem::transmute::<__m128i, [i8; 16]>(a);
+            let arr2 = core::mem::transmute::<__m128i, [i8; 16]>(b);
             for i in 0..16 {
                 arr1[i] = arr1[i].wrapping_mul(arr2[i]);
             }
@@ -464,7 +464,7 @@ impl_op! {
                 arr[14] as i16,
                 arr[15] as i16,
             ];
-            (core::mem::transmute(a), core::mem::transmute(b))
+            (core::mem::transmute::<[i16; 8], __m128i>(a), core::mem::transmute::<[i16; 8], __m128i>(b))
         }
         for Scalar(val: i8) -> (i16, i16) {
             (val as i16, 0)
@@ -511,7 +511,7 @@ impl_op! {
                 arr[14] as u8 as u16 as i16,
                 arr[15] as u8 as u16 as i16,
             ];
-            (core::mem::transmute(a), core::mem::transmute(b))
+            (core::mem::transmute::<[i16; 8], __m128i>(a), core::mem::transmute::<[i16; 8], __m128i>(b))
         }
         for Scalar(val: i8) -> (i16, i16) {
             (val as u8 as u16 as i16, 0)
diff --git a/src/tests/lib/arbitrary.rs b/src/tests/lib/arbitrary.rs
index 7eb50eb..21121f1 100644
--- a/src/tests/lib/arbitrary.rs
+++ b/src/tests/lib/arbitrary.rs
@@ -21,8 +21,8 @@ const IMPORTANT_F32: [f32; 10] = [
     -0.5,
     1.5,
     -1.5,
-    std::f32::MAX,
-    std::f32::MIN,
+    f32::MAX,
+    f32::MIN,
     f32::NAN,
 ];
 
@@ -34,15 +34,15 @@ const IMPORTANT_F64: [f64; 10] = [
     -0.5,
     1.5,
     -1.5,
-    std::f64::MAX,
-    std::f64::MIN,
+    f64::MAX,
+    f64::MIN,
     f64::NAN,
 ];
 
-const IMPORTANT_I8: [i8; 7] = [0, 1, -1, 2, -2, std::i8::MAX, std::i8::MIN];
-const IMPORTANT_I16: [i16; 7] = [0, 1, -1, 2, -2, std::i16::MAX, std::i16::MIN];
-const IMPORTANT_I32: [i32; 7] = [0, 1, -1, 2, -2, std::i32::MAX, std::i32::MIN];
-const IMPORTANT_I64: [i64; 7] = [0, 1, -1, 2, -2, std::i64::MAX, std::i64::MIN];
+const IMPORTANT_I8: [i8; 7] = [0, 1, -1, 2, -2, i8::MAX, i8::MIN];
+const IMPORTANT_I16: [i16; 7] = [0, 1, -1, 2, -2, i16::MAX, i16::MIN];
+const IMPORTANT_I32: [i32; 7] = [0, 1, -1, 2, -2, i32::MAX, i32::MIN];
+const IMPORTANT_I64: [i64; 7] = [0, 1, -1, 2, -2, i64::MAX, i64::MIN];
 
 fn iter_arbitrary_f32(interval: usize) -> impl Iterator<Item = f32> {
     assert!(interval > IMPORTANT_F32.len());
@@ -101,19 +101,19 @@ fn iter_arbitrary_ints<T: SampleUniform + Clone>(
 }
 
 fn iter_arbitrary_i8(interval: usize) -> impl Iterator<Item = i8> {
-    iter_arbitrary_ints(interval, &IMPORTANT_I8, std::i8::MIN..=std::i8::MAX)
+    iter_arbitrary_ints(interval, &IMPORTANT_I8, i8::MIN..=i8::MAX)
 }
 
 fn iter_arbitrary_i16(interval: usize) -> impl Iterator<Item = i16> {
-    iter_arbitrary_ints(interval, &IMPORTANT_I16, std::i16::MIN..=std::i16::MAX)
+    iter_arbitrary_ints(interval, &IMPORTANT_I16, i16::MIN..=i16::MAX)
 }
 
 fn iter_arbitrary_i32(interval: usize) -> impl Iterator<Item = i32> {
-    iter_arbitrary_ints(interval, &IMPORTANT_I32, std::i32::MIN..=std::i32::MAX)
+    iter_arbitrary_ints(interval, &IMPORTANT_I32, i32::MIN..=i32::MAX)
 }
 
 fn iter_arbitrary_i64(interval: usize) -> impl Iterator<Item = i64> {
-    iter_arbitrary_ints(interval, &IMPORTANT_I64, std::i64::MIN..=std::i64::MAX)
+    iter_arbitrary_ints(interval, &IMPORTANT_I64, i64::MIN..=i64::MAX)
 }
 
 fn iter_arbitrary_blendv_i8() -> impl Iterator<Item = i8> {
diff --git a/src/tests/lib/fn_tuple.rs b/src/tests/lib/fn_tuple.rs
index fb99586..5ce49ab 100644
--- a/src/tests/lib/fn_tuple.rs
+++ b/src/tests/lib/fn_tuple.rs
@@ -1,3 +1,5 @@
+#![allow(dead_code)]
+#![allow(non_snake_case)]
 //! Some traits for using tuples as function arguments. Mostly copied from warp, with minor modifications.
 //! https://github.com/seanmonstar/warp/blob/4e9c4fd6ce238197fd1088061bbc07fa2852cb0f/src/generic.rs
 //!
diff --git a/src/tests/lib/mod.rs b/src/tests/lib/mod.rs
index 46216f3..723ec11 100644
--- a/src/tests/lib/mod.rs
+++ b/src/tests/lib/mod.rs
@@ -4,7 +4,6 @@ mod arbitrary;
 pub use arbitrary::*;
 
 mod fn_tuple;
-pub use fn_tuple::*;
 
 mod tester;
 pub use tester::*;
@@ -13,4 +12,3 @@ mod numbers;
 pub use numbers::*;
 
 mod constify;
-pub use constify::*;