Improve C++ docs

cpp/examples/basic/CMakeLists.txt

@@ -7,7 +7,7 @@ cmake_minimum_required(VERSION 3.30.4)
 
 include(../set_cuda_architecture.cmake)
 
-# initialize cuda architecture
+# initialize CUDA architectures
 rapids_cuda_init_architectures(basic_example)
 
 project(

cpp/include/doxygen_groups.h

@@ -48,7 +48,6 @@
  * @defgroup cuda_streams CUDA Streams
  * @defgroup data_containers Data Containers
  * @defgroup errors Errors
- * @defgroup logging Logging
  * @defgroup thrust_integrations Thrust Integrations
  * @defgroup utilities Utilities
  */

cpp/include/rmm/cuda_stream.hpp

@@ -56,7 +56,7 @@ class cuda_stream {
   cuda_stream& operator=(cuda_stream&)  = delete;
 
   /**
-   * @brief Construct a new cuda stream object
+   * @brief Construct a new CUDA stream object
    *
    * @param flags Stream creation flags.
    *

cpp/include/rmm/cuda_stream_pool.hpp

@@ -33,7 +33,7 @@ class cuda_stream_pool {
   static constexpr std::size_t default_size{16};  ///< Default stream pool size
 
   /**
-   * @brief Construct a new cuda stream pool object of the given non-zero size
+   * @brief Construct a new CUDA stream pool object of the given non-zero size
    *
    * @throws logic_error if `pool_size` is zero
    * @param pool_size The number of streams in the pool

cpp/include/rmm/detail/error.hpp

@@ -26,11 +26,11 @@
  * specified.
  *
  * Example usage:
- * ```
- * // throws rmm::logic_error
+ * ```cpp
+ * // Throws rmm::logic_error
  * RMM_EXPECTS(p != nullptr, "Unexpected null pointer");
  *
- * // throws std::runtime_error
+ * // Throws std::runtime_error
  * RMM_EXPECTS(p != nullptr, "Unexpected nullptr", std::runtime_error);
  * ```
  * @param ... This macro accepts either two or three arguments:
@@ -60,7 +60,7 @@
  * @brief Indicates that an erroneous code path has been taken.
  *
  * Example usage:
- * ```c++
+ * ```cpp
  * // Throws rmm::logic_error
  * RMM_FAIL("Unsupported code path");
  *
@@ -91,8 +91,7 @@
  * specified.
  *
  * Example:
- * ```c++
- *
+ * ```cpp
  * // Throws rmm::cuda_error if `cudaMalloc` fails
  * RMM_CUDA_TRY(cudaMalloc(&p, 100));
  *
@@ -176,14 +175,14 @@
  * equal to `cudaSuccess`.
  *
  *
- * Replaces usecases such as:
- * ```
+ * Replaces use cases such as:
+ * ```cpp
  * auto status = cudaRuntimeApi(...);
  * assert(status == cudaSuccess);
  * ```
  *
  * Example:
- * ```
+ * ```cpp
  * RMM_ASSERT_CUDA_SUCCESS(cudaRuntimeApi(...));
  * ```
  *

cpp/include/rmm/detail/nvtx/ranges.hpp

@@ -22,7 +22,7 @@ struct librmm_domain {
  * Customizes an NVTX range with the given input.
  *
  * Example:
- * ```
+ * ```cpp
  * void some_function(){
  *    rmm::scoped_range rng{"custom_name"}; // Customizes range name
  *    ...
@@ -41,7 +41,7 @@ using scoped_range = ::nvtx3::scoped_range_in<librmm_domain>;
  * name the range.
  *
  * Example:
- * ```
+ * ```cpp
  * void some_function(){
  *    RMM_FUNC_RANGE();
  *    ...

cpp/include/rmm/device_buffer.hpp

@@ -34,28 +34,28 @@ namespace RMM_NAMESPACE {
  * behavior to read the contents of `data()` before first initializing it.
  *
  * Examples:
- * ```
- * //Allocates at least 100 bytes of device memory using the default memory
- * //resource and default stream.
+ * ```cpp
+ * // Allocates at least 100 bytes of device memory using the default memory
+ * // resource and default stream.
  * device_buffer buff(100);
  *
- * // allocates at least 100 bytes using the custom memory resource and
+ * // Allocates at least 100 bytes using the custom memory resource and
  * // specified stream
  * custom_memory_resource mr;
  * cuda_stream_view stream = cuda_stream_view{};
  * device_buffer custom_buff(100, stream, &mr);
  *
- * // deep copies `buff` into a new device buffer using the specified stream
+ * // Deep copies `buff` into a new device buffer using the specified stream
  * device_buffer buff_copy(buff, stream);
  *
- * // moves the memory in `from_buff` to `to_buff`. Deallocates previously allocated
+ * // Moves the memory in `from_buff` to `to_buff`. Deallocates previously allocated
  * // to_buff memory on `to_buff.stream()`.
  * device_buffer to_buff(std::move(from_buff));
  *
- * // deep copies `buff` into a new device buffer using the specified stream
+ * // Deep copies `buff` into a new device buffer using the specified stream
  * device_buffer buff_copy(buff, stream);
  *
- * // shallow copies `buff` into a new device_buffer, `buff` is now empty
+ * // Shallow copies `buff` into a new device_buffer, `buff` is now empty
  * device_buffer buff_move(std::move(buff));
  *
  * // Default construction. Buffer is empty
@@ -65,7 +65,7 @@ namespace RMM_NAMESPACE {
  * // deep copies any previous contents. Otherwise, simply updates the value of `size()` to the
  * // newly requested size without any allocations or copies. Uses the specified stream.
  * buff_default.resize(100, stream);
- *```
+ * ```
  */
 class device_buffer {
  public:

cpp/include/rmm/device_scalar.hpp

@@ -168,7 +168,7 @@ class device_scalar {
    * referenced by `v` should not be destroyed or modified until `stream` has been
    * synchronized. Otherwise, behavior is undefined.
    *
-   * @note: This function incurs a host to device memcpy or device memset and should be used
+   * @note This function incurs a host to device memcpy or device memset and should be used
    * carefully.
    *
    * Example:
@@ -209,7 +209,7 @@ class device_scalar {
    *
    * This function does not synchronize `stream` before returning.
    *
-   * @note: This function incurs a device memset and should be used carefully.
+   * @note This function incurs a device memset and should be used carefully.
    *
    * @param stream CUDA stream on which to perform the copy
    */

cpp/include/rmm/device_uvector.hpp

@@ -70,14 +70,14 @@ class device_uvector {
                 "device_uvector only supports types that are trivially copyable.");
 
  public:
-  using value_type = T;            ///< T; stored value type
+  using value_type = T;            ///< Stored value type
   using size_type  = std::size_t;  ///< The type used for the size of the vector
-  using reference = value_type&;  ///< value_type&; reference type returned by operator[](size_type)
-  using const_reference = value_type const&;  ///< value_type const&; constant reference type
-                                              ///< returned by operator[](size_type) const
-  using pointer        = value_type*;         ///< The type of the pointer returned by data()
-  using const_pointer  = value_type const*;   ///< The type of the pointer returned by data() const
-  using iterator       = pointer;             ///< The type of the iterator returned by begin()
+  using reference  = value_type&;  ///< Reference type returned by operator[](size_type)
+  using const_reference =
+    value_type const&;  ///< Constant reference type returned by operator[](size_type) const
+  using pointer        = value_type*;        ///< The type of the pointer returned by data()
+  using const_pointer  = value_type const*;  ///< The type of the pointer returned by data() const
+  using iterator       = pointer;            ///< The type of the iterator returned by begin()
   using const_iterator = const_pointer;  ///< The type of the const iterator returned by cbegin()
   using reverse_iterator =
     thrust::reverse_iterator<iterator>;  ///< The type of the iterator returned by rbegin()

cpp/include/rmm/mr/device/callback_memory_resource.hpp

@@ -23,7 +23,7 @@ namespace mr {
  * @brief Callback function type used by callback memory resource for allocation.
  *
  * The signature of the callback function is:
- *   `void* allocate_callback_t(std::size_t bytes, cuda_stream_view stream, void* arg);
+ *   `void* allocate_callback_t(std::size_t bytes, cuda_stream_view stream, void* arg);`
  *
  * * Returns a pointer to an allocation of at least `bytes` usable immediately on
  *   `stream`. The stream-ordered behavior requirements are identical to
@@ -40,7 +40,7 @@ using allocate_callback_t = std::function<void*(std::size_t, cuda_stream_view, v
  * @brief Callback function type used by callback_memory_resource for deallocation.
  *
  * The signature of the callback function is:
- *   `void deallocate_callback_t(void* ptr, std::size_t bytes, cuda_stream_view stream, void* arg);
+ *   `void deallocate_callback_t(void* ptr, std::size_t bytes, cuda_stream_view stream, void* arg);`
  *
  * * Deallocates memory pointed to by `ptr`. `bytes` specifies the size of the allocation
  *   in bytes, and must equal the value of `bytes` that was passed to the allocate callback

cpp/include/rmm/mr/device/owning_wrapper.hpp

@@ -112,7 +112,7 @@ class owning_wrapper : public device_memory_resource {
    * };
    *
    * using cuda = rmm::mr::cuda_memory_resource;
-   * using example = example_resource<cuda,cuda>;
+   * using example = example_resource<cuda, cuda>;
    * using wrapped_example = rmm::mr::owning_wrapper<example, cuda, cuda>;
    * auto cuda_mr = std::make_shared<cuda>();
    *

cpp/include/rmm/mr/device/polymorphic_allocator.hpp

@@ -175,7 +175,7 @@ class stream_allocator_adaptor {
   /**
    * @brief Construct a `stream_allocator_adaptor` using `a` as the underlying allocator.
    *
-   * @note: The `stream` must not be destroyed before the `stream_allocator_adaptor`, otherwise
+   * @note The `stream` must not be destroyed before the `stream_allocator_adaptor`, otherwise
    * behavior is undefined.
    *
    * @param allocator The stream ordered allocator to use as the underlying allocator

cpp/include/rmm/mr/device/thrust_allocator_adaptor.hpp

@@ -31,7 +31,7 @@ namespace mr {
  * allocate objects of a specific type `T`, but can be freely rebound to other
  * types.
  *
- * The allocator records the current cuda device and may only be used with a backing
+ * The allocator records the current CUDA device and may only be used with a backing
  * `device_async_resource_ref` valid for the same device.
  *
  * @tparam T The type of the objects that will be allocated by this allocator

cpp/include/rmm/mr/device/tracking_resource_adaptor.hpp

@@ -38,8 +38,7 @@ namespace mr {
  * the project is linked with `-rdynamic`. This can be accomplished with
  * `add_link_options(-rdynamic)` in cmake.
  *
- * @tparam Upstream Type of the upstream resource used for
- * allocation/deallocation.
+ * @tparam Upstream Type of the upstream resource used for allocation/deallocation.
  */
 template <typename Upstream>
 class tracking_resource_adaptor final : public device_memory_resource {

cpp/include/rmm/mr/pinned_host_memory_resource.hpp

@@ -22,7 +22,7 @@ namespace RMM_NAMESPACE {
 namespace mr {
 
 /**
- * @addtogroup memory_resources
+ * @addtogroup device_memory_resources
  * @{
  * @file
  */

cpp/tests/CMakeLists.txt

@@ -190,7 +190,7 @@ ConfigureTest(PINNED_MR_TEST mr/host/pinned_mr_tests.cpp)
 # pinned pool mr tests
 ConfigureTest(PINNED_POOL_MR_TEST mr/host/pinned_pool_mr_tests.cpp)
 
-# cuda stream tests
+# CUDA stream tests
 ConfigureTest(CUDA_STREAM_TEST cuda_stream_tests.cpp cuda_stream_pool_tests.cpp)
 
 # device buffer tests

cpp/tests/container_multidevice_tests.cu

@@ -28,7 +28,7 @@ TYPED_TEST_SUITE(ContainerMultiDeviceTest, containers);
 
 TYPED_TEST(ContainerMultiDeviceTest, CreateDestroyDifferentActiveDevice)
 {
-  // Get the number of cuda devices
+  // Get the number of CUDA devices
   int num_devices = rmm::get_num_cuda_devices();
 
   // only run on multidevice systems
@@ -55,7 +55,7 @@ TYPED_TEST(ContainerMultiDeviceTest, CreateDestroyDifferentActiveDevice)
 
 TYPED_TEST(ContainerMultiDeviceTest, CreateMoveDestroyDifferentActiveDevice)
 {
-  // Get the number of cuda devices
+  // Get the number of CUDA devices
   int num_devices = rmm::get_num_cuda_devices();
 
   // only run on multidevice systems
@@ -95,7 +95,7 @@ TYPED_TEST(ContainerMultiDeviceTest, CreateMoveDestroyDifferentActiveDevice)
 
 TYPED_TEST(ContainerMultiDeviceTest, ResizeDifferentActiveDevice)
 {
-  // Get the number of cuda devices
+  // Get the number of CUDA devices
   int num_devices = rmm::get_num_cuda_devices();
 
   // only run on multidevice systems
@@ -118,7 +118,7 @@ TYPED_TEST(ContainerMultiDeviceTest, ResizeDifferentActiveDevice)
 
 TYPED_TEST(ContainerMultiDeviceTest, ShrinkDifferentActiveDevice)
 {
-  // Get the number of cuda devices
+  // Get the number of CUDA devices
   int num_devices = rmm::get_num_cuda_devices();
 
   // only run on multidevice systems

cpp/tests/logger_tests.cpp

@@ -84,7 +84,7 @@ class raii_temp_directory {
  *
  * Events in the log file are expected to occur in the same order as in `expected_events`.
  *
- * @note: This function accounts for the fact that `device_memory_resource` automatically pads
+ * @note This function accounts for the fact that `device_memory_resource` automatically pads
  * allocations to a multiple of 8 bytes by rounding up the expected allocation sizes to a multiple
  * of 8.
  *

cpp/tests/mr/device/pool_mr_tests.cpp

@@ -143,7 +143,7 @@ TEST(PoolTest, MultidevicePool)
 {
   using MemoryResource = rmm::mr::pool_memory_resource<rmm::mr::cuda_memory_resource>;
 
-  // Get the number of cuda devices
+  // Get the number of CUDA devices
   int num_devices = rmm::get_num_cuda_devices();
 
   // only run on multidevice systems