diff --git a/benchmarks/static_map/insert_or_apply_bench.cu b/benchmarks/static_map/insert_or_apply_bench.cu
index 60cd91caf..ad668b6ac 100644
--- a/benchmarks/static_map/insert_or_apply_bench.cu
+++ b/benchmarks/static_map/insert_or_apply_bench.cu
@@ -21,11 +21,9 @@
 #include <cuco/utility/key_generator.cuh>
 #include <cuco/utility/reduction_functors.cuh>
 
-#include <cuda/std/cmath>
-
-
 #include <nvbench/nvbench.cuh>
 
+#include <cuda/std/cmath>
 #include <thrust/device_vector.h>
 #include <thrust/transform.h>
 
@@ -45,7 +43,9 @@ std::enable_if_t<(sizeof(Key) == sizeof(Value)), void> static_map_insert_or_appl
   auto const occupancy    = state.get_float64("Occupancy");
   auto const multiplicity = state.get_int64("Multiplicity");
 
-  std::size_t const size = static_cast<size_t>(cuda::std::ceil(static_cast<double>(num_keys) / static_cast<double>(multiplicity))) / occupancy;
+  std::size_t const size = static_cast<size_t>(cuda::std::ceil(static_cast<double>(num_keys) /
+                                                               static_cast<double>(multiplicity))) /
+                           occupancy;
 
   thrust::device_vector<Key> keys(num_keys);
 
diff --git a/include/cuco/detail/extent/extent.inl b/include/cuco/detail/extent/extent.inl
index 88261fd4b..6ef69a705 100644
--- a/include/cuco/detail/extent/extent.inl
+++ b/include/cuco/detail/extent/extent.inl
@@ -23,10 +23,8 @@
 #include <cuco/storage.cuh>
 #include <cuco/utility/fast_int.cuh>
 
-#include <cuda/std/type_traits>
-
 #include <cuda/std/cmath>
-
+#include <cuda/std/type_traits>
 
 namespace cuco {
 
@@ -93,8 +91,9 @@ template <int32_t CGSize, int32_t BucketSize, typename SizeType, std::size_t N>
     (static_cast<uint64_t>(std::numeric_limits<SizeType>::max()) < max_prime)
       ? std::numeric_limits<SizeType>::max()
       : static_cast<SizeType>(max_prime);
-  auto const size = static_cast<SizeType>(cuda::std::ceil(static_cast<double>(
-    std::max(static_cast<SizeType>(ext), static_cast<SizeType>(1))) / static_cast<double>(CGSize * BucketSize)));
+  auto const size = static_cast<SizeType>(cuda::std::ceil(
+    static_cast<double>(std::max(static_cast<SizeType>(ext), static_cast<SizeType>(1))) /
+    static_cast<double>(CGSize * BucketSize)));
   if (size > max_value) { CUCO_FAIL("Invalid input extent"); }
 
   if constexpr (N == dynamic_extent) {
diff --git a/include/cuco/detail/prime.hpp b/include/cuco/detail/prime.hpp
index f28db0e8d..26e876ab3 100644
--- a/include/cuco/detail/prime.hpp
+++ b/include/cuco/detail/prime.hpp
@@ -16,13 +16,12 @@
 
 #pragma once
 
+#include <cuda/std/cmath>
 
 #include <algorithm>
 #include <array>
 #include <cstdint>
 #include <limits>
-#include <cuda/std/cmath>
-
 
 namespace cuco {
 namespace detail {
@@ -20155,7 +20154,8 @@ constexpr T get_valid_capacity(T capacity) noexcept
     if constexpr (not uses_vector_load) { return cg_size; }
   }();
 
-  auto const c         = static_cast<T>(cuda::std::ceil(static_cast<double>(capacity)  / static_cast<double>(stride)));
+  auto const c =
+    static_cast<T>(cuda::std::ceil(static_cast<double>(capacity) / static_cast<double>(stride)));
   auto const min_prime = std::lower_bound(primes.begin(), primes.end(), c);
   return *min_prime * stride;
 }
diff --git a/include/cuco/detail/utility/cuda.hpp b/include/cuco/detail/utility/cuda.hpp
index 37f35cfdb..7cde2d572 100644
--- a/include/cuco/detail/utility/cuda.hpp
+++ b/include/cuco/detail/utility/cuda.hpp
@@ -16,6 +16,7 @@
 #pragma once
 
 #include <cuco/detail/error.hpp>
+
 #include <cuda/std/cmath>
 
 namespace cuco {
@@ -44,8 +45,7 @@ constexpr auto grid_size(index_type num,
                          int32_t block_size = default_block_size()) noexcept
 {
   return static_cast<int32_t>(
-    cuda::std::ceil(static_cast<double>(cg_size * num) / static_cast<double> (stride * block_size))
-  );
+    cuda::std::ceil(static_cast<double>(cg_size * num) / static_cast<double>(stride * block_size)));
 }
 
 /**