Template Class device_matrix

• Defined in File device_matrix.h

Inheritance Relationships

Base Type

• private noncopyable

Class Documentation

template<typename T>
class icrar::cuda::device_matrix : private noncopyable

A cuda device buffer object that represents a memory buffer on a cuda device. Matrix size is fixed at construction and can only be resized using move semantics.

Note

See https://www.quantstart.com/articles/Matrix-Matrix-Multiplication-on-the-GPU-with-Nvidia-CUDA/

Note

See https://forums.developer.nvidia.com/t/guide-cudamalloc3d-and-cudaarrays/23421

Template Parameters

• T: numeric type

Public Functions

device_matrix()

Default constructor.

device_matrix(device_matrix &&other) noexcept

Move Constructor.

Parameters

• other:

device_matrix &operator=(device_matrix &&other) noexcept

Move Assignment Operator.

Return

device_matrix&

Parameters

• other:

device_matrix(size_t rows, size_t cols, const T *data = nullptr)

Construct a new device matrix object of fixed size and initialized asyncronously if data is provided.

Parameters

• rows: number of rows

• cols: number of columns

• data: constigous column major data of size rows*cols*sizeof(T) to copy to device

device_matrix(const Eigen::Matrix<T, Eigen::Dynamic, Eigen::Dynamic> &data)

template<int Rows, int Cols>
device_matrix(const Eigen::Matrix<T, Rows, Cols> &data)

~device_matrix()

__host__ __device__ T *Get()

__host__ __device__ const T *Get() const

__host__ __device__ size_t GetRows() const

__host__ __device__ size_t GetCols() const

__host__ __device__ size_t GetCount() const

__host__ __device__ size_t GetSize() const

__host__ void SetZeroAsync()

__host__ void SetDataSync(const T *data)

Performs a synchronous copy of data into the device buffer.

Return

host

Parameters

• data:

__host__ void SetDataAsync(const T *data)

Set the Data Async object.

Return

host

Parameters

• data:

__host__ void ToHost(T *out) const

__host__ void ToHost(std::vector<T> &out) const

__host__ void ToHost(Eigen::Matrix<T, Eigen::Dynamic, Eigen::Dynamic> &out) const

template<int Rows, int Cols>
__host__ void ToHost(Eigen::Matrix<T, Rows, Cols> &out) const

__host__ Eigen::Matrix<T, Eigen::Dynamic, Eigen::Dynamic> ToHost() const

__host__ void ToHostAsync(T *out) const

__host__ void ToHostAsync(Eigen::Matrix<T, Eigen::Dynamic, Eigen::Dynamic> &out) const

__host__ Eigen::Matrix<T, Eigen::Dynamic, Eigen::Dynamic> ToHostAsync() const