[INFRA] Import NVIDIA/CCCL upstream as optimization reference library

CCCL (CUDA C++ Core Libraries) provides:
- CUB: device/block/warp-level GPU primitives (reduce, scan, sort, topk)
- Thrust: high-level parallel algorithms (transform_reduce, sort, scan)
- libcudacxx: CUDA C++ standard library (atomics, barriers, memory)
- cudax: experimental features (memory resources, allocators)
- Tuning policies: per-SM hardware-specific algorithm parameters

Competition optimization vectors mapped to CCCL:
- Output TPS (83% weight): warp_reduce, block_reduce, device_topk
- Input TPS (14% weight): device_scan, block_load, prefetch
- Cache TPS (3% weight): prefix caching strategy patterns
- Memory (0.9 util): pooled/cached/buddy allocators

Source: https://github.com/NVIDIA/cccl (shallow clone, HEAD only)
License: Apache-2.0

cccl_upstream/thrust/testing/reverse_iterator.cu

@@ -0,0 +1,156 @@
+#include <thrust/iterator/reverse_iterator.h>
+#include <thrust/scan.h>
+#include <thrust/sequence.h>
+
+#include <cuda/std/iterator>
+#include <cuda/std/type_traits>
+
+#include <unittest/unittest.h>
+
+// ensure that we properly support thrust::reverse_iterator from cuda::std
+void TestReverseIteratorTraits()
+{
+  using base_it = thrust::host_vector<int>::iterator;
+
+  using it       = thrust::reverse_iterator<base_it>;
+  using traits   = cuda::std::iterator_traits<it>;
+  using category = ::cuda::std::random_access_iterator_tag;
+
+  static_assert(cuda::std::is_same_v<traits::difference_type, ptrdiff_t>);
+  static_assert(cuda::std::is_same_v<traits::value_type, int>);
+  static_assert(cuda::std::is_same_v<traits::pointer, void>);
+  static_assert(cuda::std::is_same_v<traits::reference, int&>);
+  static_assert(cuda::std::is_same_v<traits::iterator_category, category>);
+
+  static_assert(cuda::std::is_same_v<thrust::iterator_traversal_t<it>, thrust::random_access_traversal_tag>);
+
+  static_assert(cuda::std::__has_random_access_traversal<it>);
+
+  static_assert(cuda::std::output_iterator<it, int>);
+  static_assert(cuda::std::input_iterator<it>);
+  static_assert(cuda::std::forward_iterator<it>);
+  static_assert(cuda::std::bidirectional_iterator<it>);
+  static_assert(cuda::std::random_access_iterator<it>);
+  static_assert(!cuda::std::contiguous_iterator<it>);
+}
+DECLARE_UNITTEST(TestReverseIteratorTraits);
+
+void TestReverseIteratorCopyConstructor()
+{
+  thrust::host_vector<int> h_v(1, 13);
+
+  thrust::reverse_iterator<thrust::host_vector<int>::iterator> h_iter0(h_v.end());
+  thrust::reverse_iterator<thrust::host_vector<int>::iterator> h_iter1(h_iter0);
+
+  ASSERT_EQUAL_QUIET(h_iter0, h_iter1);
+  ASSERT_EQUAL(*h_iter0, *h_iter1);
+
+  thrust::device_vector<int> d_v(1, 13);
+
+  thrust::reverse_iterator<thrust::device_vector<int>::iterator> d_iter2(d_v.end());
+  thrust::reverse_iterator<thrust::device_vector<int>::iterator> d_iter3(d_iter2);
+
+  ASSERT_EQUAL_QUIET(d_iter2, d_iter3);
+  ASSERT_EQUAL(*d_iter2, *d_iter3);
+}
+DECLARE_UNITTEST(TestReverseIteratorCopyConstructor);
+static_assert(cuda::std::is_trivially_copy_constructible<thrust::reverse_iterator<int*>>::value);
+static_assert(cuda::std::is_trivially_copyable<thrust::reverse_iterator<int*>>::value);
+
+void TestReverseIteratorIncrement()
+{
+  thrust::host_vector<int> h_v(4);
+  thrust::sequence(h_v.begin(), h_v.end());
+
+  thrust::reverse_iterator<thrust::host_vector<int>::iterator> h_iter(h_v.end());
+
+  ASSERT_EQUAL(*h_iter, 3);
+
+  h_iter++;
+  ASSERT_EQUAL(*h_iter, 2);
+
+  h_iter++;
+  ASSERT_EQUAL(*h_iter, 1);
+
+  h_iter++;
+  ASSERT_EQUAL(*h_iter, 0);
+
+  thrust::device_vector<int> d_v(4);
+  thrust::sequence(d_v.begin(), d_v.end());
+
+  thrust::reverse_iterator<thrust::device_vector<int>::iterator> d_iter(d_v.end());
+
+  ASSERT_EQUAL(*d_iter, 3);
+
+  d_iter++;
+  ASSERT_EQUAL(*d_iter, 2);
+
+  d_iter++;
+  ASSERT_EQUAL(*d_iter, 1);
+
+  d_iter++;
+  ASSERT_EQUAL(*d_iter, 0);
+}
+DECLARE_UNITTEST(TestReverseIteratorIncrement);
+
+template <typename Vector>
+void TestReverseIteratorCopy()
+{
+  Vector source{10, 20, 30, 40};
+
+  Vector destination(8, 0); // arm gcc is complaining here
+
+  thrust::copy(
+    thrust::make_reverse_iterator(source.end()), thrust::make_reverse_iterator(source.begin()), destination.begin());
+
+  destination.resize(4);
+  Vector ref{40, 30, 20, 10};
+  ASSERT_EQUAL(destination, ref);
+}
+DECLARE_VECTOR_UNITTEST(TestReverseIteratorCopy);
+
+void TestReverseIteratorExclusiveScanSimple()
+{
+  using T        = int;
+  const size_t n = 10;
+
+  thrust::host_vector<T> h_data(n);
+  thrust::sequence(h_data.begin(), h_data.end());
+
+  thrust::device_vector<T> d_data = h_data;
+
+  thrust::host_vector<T> h_result(h_data.size());
+  thrust::device_vector<T> d_result(d_data.size());
+
+  thrust::exclusive_scan(
+    thrust::make_reverse_iterator(h_data.end()), thrust::make_reverse_iterator(h_data.begin()), h_result.begin());
+
+  thrust::exclusive_scan(
+    thrust::make_reverse_iterator(d_data.end()), thrust::make_reverse_iterator(d_data.begin()), d_result.begin());
+
+  ASSERT_EQUAL_QUIET(h_result, d_result);
+}
+DECLARE_UNITTEST(TestReverseIteratorExclusiveScanSimple);
+
+template <typename T>
+struct TestReverseIteratorExclusiveScan
+{
+  void operator()(const size_t n)
+  {
+    thrust::host_vector<T> h_data = unittest::random_samples<T>(n);
+
+    thrust::device_vector<T> d_data = h_data;
+
+    thrust::host_vector<T> h_result(n);
+    thrust::device_vector<T> d_result(n);
+
+    thrust::exclusive_scan(
+      thrust::make_reverse_iterator(h_data.end()), thrust::make_reverse_iterator(h_data.begin()), h_result.begin());
+
+    thrust::exclusive_scan(
+      thrust::make_reverse_iterator(d_data.end()), thrust::make_reverse_iterator(d_data.begin()), d_result.begin());
+
+    ASSERT_EQUAL_QUIET(h_result, d_result);
+  }
+};
+VariableUnitTest<TestReverseIteratorExclusiveScan, IntegralTypes> TestReverseIteratorExclusiveScanInstance;