diff --git a/lax/src/tridiagonal.rs b/lax/src/tridiagonal.rs
index f6cbfc7e..2b1d98e2 100644
--- a/lax/src/tridiagonal.rs
+++ b/lax/src/tridiagonal.rs
@@ -15,7 +15,7 @@ use std::ops::{Index, IndexMut};
 ///  ...           ...,  u{n-1},
 ///   0,  ...,  l{n-1},  d{n-1},]
 /// ```
-#[derive(Clone, PartialEq)]
+#[derive(Clone, PartialEq, Eq)]
 pub struct Tridiagonal<A: Scalar> {
     /// layout of raw matrix
     pub l: MatrixLayout,
diff --git a/ndarray-linalg/examples/eig.rs b/ndarray-linalg/examples/eig.rs
index 3e41556a..746ec2e4 100644
--- a/ndarray-linalg/examples/eig.rs
+++ b/ndarray-linalg/examples/eig.rs
@@ -3,7 +3,7 @@ use ndarray_linalg::*;
 
 fn main() {
     let a = arr2(&[[2.0, 1.0, 2.0], [-2.0, 2.0, 1.0], [1.0, 2.0, -2.0]]);
-    let (e, vecs) = a.clone().eig().unwrap();
+    let (e, vecs) = a.eig().unwrap();
     println!("eigenvalues = \n{:?}", e);
     println!("V = \n{:?}", vecs);
     let a_c: Array2<c64> = a.map(|f| c64::new(*f, 0.0));
diff --git a/ndarray-linalg/examples/eigh.rs b/ndarray-linalg/examples/eigh.rs
index c9bcc941..f4f36841 100644
--- a/ndarray-linalg/examples/eigh.rs
+++ b/ndarray-linalg/examples/eigh.rs
@@ -6,7 +6,7 @@ use ndarray_linalg::*;
 
 fn main() {
     let a = arr2(&[[3.0, 1.0, 1.0], [1.0, 3.0, 1.0], [1.0, 1.0, 3.0]]);
-    let (e, vecs) = a.clone().eigh(UPLO::Upper).unwrap();
+    let (e, vecs) = a.eigh(UPLO::Upper).unwrap();
     println!("eigenvalues = \n{:?}", e);
     println!("V = \n{:?}", vecs);
     let av = a.dot(&vecs);
diff --git a/ndarray-linalg/src/krylov/householder.rs b/ndarray-linalg/src/krylov/householder.rs
index 844d0e7e..c04d9049 100644
--- a/ndarray-linalg/src/krylov/householder.rs
+++ b/ndarray-linalg/src/krylov/householder.rs
@@ -34,7 +34,7 @@ where
 {
     assert_eq!(w.len(), a.len());
     let n = a.len();
-    let c = A::from(2.0).unwrap() * w.inner(&a);
+    let c = A::from(2.0).unwrap() * w.inner(a);
     for l in 0..n {
         a[l] -= c * w[l];
     }
diff --git a/ndarray-linalg/src/krylov/mgs.rs b/ndarray-linalg/src/krylov/mgs.rs
index dc0dfba6..67806a2c 100644
--- a/ndarray-linalg/src/krylov/mgs.rs
+++ b/ndarray-linalg/src/krylov/mgs.rs
@@ -50,7 +50,7 @@ impl<A: Scalar + Lapack> Orthogonalizer for MGS<A> {
         let mut coef = Array1::zeros(self.len() + 1);
         for i in 0..self.len() {
             let q = &self.q[i];
-            let c = q.inner(&a);
+            let c = q.inner(a);
             azip!((a in &mut *a, &q in q) *a -= c * q);
             coef[i] = c;
         }
@@ -77,12 +77,12 @@ impl<A: Scalar + Lapack> Orthogonalizer for MGS<A> {
         self.div_append(&mut a)
     }
 
-    fn div_append<S>(&mut self, mut a: &mut ArrayBase<S, Ix1>) -> AppendResult<A>
+    fn div_append<S>(&mut self, a: &mut ArrayBase<S, Ix1>) -> AppendResult<A>
     where
         A: Lapack,
         S: DataMut<Elem = A>,
     {
-        let coef = self.decompose(&mut a);
+        let coef = self.decompose(a);
         let nrm = coef[coef.len() - 1].re();
         if nrm < self.tol {
             // Linearly dependent
diff --git a/ndarray-linalg/src/layout.rs b/ndarray-linalg/src/layout.rs
index 9ca772bb..d0d13585 100644
--- a/ndarray-linalg/src/layout.rs
+++ b/ndarray-linalg/src/layout.rs
@@ -67,9 +67,8 @@ where
     }
 
     fn as_allocated(&self) -> Result<&[A]> {
-        Ok(self
-            .as_slice_memory_order()
-            .ok_or_else(|| LinalgError::MemoryNotCont)?)
+        self.as_slice_memory_order()
+            .ok_or(LinalgError::MemoryNotCont)
     }
 }
 
@@ -78,8 +77,7 @@ where
     S: DataMut<Elem = A>,
 {
     fn as_allocated_mut(&mut self) -> Result<&mut [A]> {
-        Ok(self
-            .as_slice_memory_order_mut()
-            .ok_or_else(|| LinalgError::MemoryNotCont)?)
+        self.as_slice_memory_order_mut()
+            .ok_or(LinalgError::MemoryNotCont)
     }
 }
diff --git a/ndarray-linalg/src/least_squares.rs b/ndarray-linalg/src/least_squares.rs
index 25e700e4..86ca17f3 100644
--- a/ndarray-linalg/src/least_squares.rs
+++ b/ndarray-linalg/src/least_squares.rs
@@ -304,12 +304,12 @@ where
         a.layout()?,
         a.as_allocated_mut()?,
         rhs.as_slice_memory_order_mut()
-            .ok_or_else(|| LinalgError::MemoryNotCont)?,
+            .ok_or(LinalgError::MemoryNotCont)?,
     )?;
 
     let (m, n) = (a.shape()[0], a.shape()[1]);
     let solution = rhs.slice(s![0..n]).to_owned();
-    let residual_sum_of_squares = compute_residual_scalar(m, n, rank, &rhs);
+    let residual_sum_of_squares = compute_residual_scalar(m, n, rank, rhs);
     Ok(LeastSquaresResult {
         solution,
         singular_values: Array::from_shape_vec((singular_values.len(),), singular_values)?,
@@ -399,7 +399,7 @@ where
     let solution: Array2<E> = rhs.slice(s![..a.shape()[1], ..]).to_owned();
     let singular_values = Array::from_shape_vec((singular_values.len(),), singular_values)?;
     let (m, n) = (a.shape()[0], a.shape()[1]);
-    let residual_sum_of_squares = compute_residual_array1(m, n, rank, &rhs);
+    let residual_sum_of_squares = compute_residual_array1(m, n, rank, rhs);
     Ok(LeastSquaresResult {
         solution,
         singular_values,
diff --git a/ndarray-linalg/src/lobpcg/lobpcg.rs b/ndarray-linalg/src/lobpcg/lobpcg.rs
index 1bc4f2ad..8b7ab616 100644
--- a/ndarray-linalg/src/lobpcg/lobpcg.rs
+++ b/ndarray-linalg/src/lobpcg/lobpcg.rs
@@ -83,12 +83,11 @@ fn apply_constraints<A: Scalar + Lapack>(
     let u = gram_yv
         .columns()
         .into_iter()
-        .map(|x| {
+        .flat_map(|x| {
             let res = cholesky_yy.solvec(&x).unwrap();
 
             res.to_vec()
         })
-        .flatten()
         .collect::<Vec<A>>();
 
     let rows = gram_yv.len_of(Axis(0));
diff --git a/ndarray-linalg/src/lobpcg/svd.rs b/ndarray-linalg/src/lobpcg/svd.rs
index 7a326293..f46857cc 100644
--- a/ndarray-linalg/src/lobpcg/svd.rs
+++ b/ndarray-linalg/src/lobpcg/svd.rs
@@ -30,7 +30,7 @@ impl<A: Float + PartialOrd + DivAssign<A> + 'static + MagnitudeCorrection> Trunc
         let mut a = self.eigvals.iter().enumerate().collect::<Vec<_>>();
 
         // sort by magnitude
-        a.sort_by(|(_, x), (_, y)| x.partial_cmp(&y).unwrap().reverse());
+        a.sort_by(|(_, x), (_, y)| x.partial_cmp(y).unwrap().reverse());
 
         // calculate cut-off magnitude (borrowed from scipy)
         let cutoff = A::epsilon() * // float precision
diff --git a/ndarray-linalg/src/tridiagonal.rs b/ndarray-linalg/src/tridiagonal.rs
index b603a5ee..b5aebbf2 100644
--- a/ndarray-linalg/src/tridiagonal.rs
+++ b/ndarray-linalg/src/tridiagonal.rs
@@ -272,7 +272,7 @@ where
         Sb: DataMut<Elem = A>,
     {
         A::solve_tridiagonal(
-            &self,
+            self,
             rhs.layout()?,
             Transpose::No,
             rhs.as_slice_mut().unwrap(),
@@ -287,7 +287,7 @@ where
         Sb: DataMut<Elem = A>,
     {
         A::solve_tridiagonal(
-            &self,
+            self,
             rhs.layout()?,
             Transpose::Transpose,
             rhs.as_slice_mut().unwrap(),
@@ -302,7 +302,7 @@ where
         Sb: DataMut<Elem = A>,
     {
         A::solve_tridiagonal(
-            &self,
+            self,
             rhs.layout()?,
             Transpose::Hermite,
             rhs.as_slice_mut().unwrap(),
@@ -622,7 +622,7 @@ where
 {
     fn det_tridiagonal(&self) -> Result<A> {
         let n = self.d.len();
-        Ok(rec_rel(&self)[n])
+        Ok(rec_rel(self)[n])
     }
 }
 
@@ -671,7 +671,7 @@ where
     A: Scalar + Lapack,
 {
     fn rcond_tridiagonal(&self) -> Result<A::Real> {
-        Ok(A::rcond_tridiagonal(&self)?)
+        Ok(A::rcond_tridiagonal(self)?)
     }
 }