Data structures for interacting with CUDA and MAGMA environments. More...

#include "tsgAcceleratedHandles.hpp"

Include dependency graph for tsgAcceleratedDataStructures.hpp:

This graph shows which files directly or indirectly include this file:

Classes
class	TasGrid::GpuVector< T >
	Template class that wraps around a single GPU array, providing functionality that mimics std::vector. More...

struct	TasGrid::GpuEngine
	Wrapper class around calls GPU accelerated linear algebra libraries. More...

class	TasGrid::AccelerationDomainTransform
	Implements the domain transform algorithms in case the user data is provided on the GPU. More...

struct	TasGrid::AccelerationContext
	Wrapper class around GPU device ID, acceleration type and GpuEngine. More...

Namespaces
	TasGrid
	Encapsulates the Tasmanian Sparse Grid module.

	TasGrid::TasGpu
	Wrappers around custom CUDA kernels to handle domain transforms and basis evaluations, the kernels are instantiated in tsgCudaKernels.cu.

	TasGrid::AccelerationMeta
	Common methods for manipulating acceleration options and reading CUDA environment properties.

Functions
template<typename T >
void	TasGrid::TasGpu::dtrans2can (AccelerationContext const acc, bool use01, int dims, int num_x, int pad_size, const double gpu_trans_a, const double gpu_trans_b, const T gpu_x_transformed, T *gpu_x_canonical)
	Uses custom kernel to convert transformed points to canonical points, all arrays live on the CUDA device. More...

template<typename T >
void	TasGrid::TasGpu::devalpwpoly (AccelerationContext const acc, int order, TypeOneDRule rule, int num_dimensions, int num_x, int num_basis, const T gpu_x, const T gpu_nodes, const T gpu_support, T *gpu_y)
	Evaluate the basis functions for a local polynomial grid using the DENSE algorithm. More...

template<typename T >
void	TasGrid::TasGpu::devalpwpoly_sparse (AccelerationContext const acc, int order, TypeOneDRule rule, int dims, int num_x, const T gpu_x, const GpuVector< T > &gpu_nodes, const GpuVector< T > &gpu_support, const GpuVector< int > &gpu_hpntr, const GpuVector< int > &gpu_hindx, const GpuVector< int > &gpu_hroots, GpuVector< int > &gpu_spntr, GpuVector< int > &gpu_sindx, GpuVector< T > &gpu_svals)
	Evaluate the basis functions for a local polynomial grid using the SPARSE algorithm. More...

template<typename T >
void	TasGrid::TasGpu::devalseq (AccelerationContext const acc, int dims, int num_x, const std::vector< int > &max_levels, const T gpu_x, const GpuVector< int > &num_nodes, const GpuVector< int > &points, const GpuVector< T > &nodes, const GpuVector< T > &coeffs, T *gpu_result)
	Evaluate the basis for a Sequence grid. More...

template<typename T >
void	TasGrid::TasGpu::devalfor (AccelerationContext const acc, int dims, int num_x, const std::vector< int > &max_levels, const T gpu_x, const GpuVector< int > &num_nodes, const GpuVector< int > &points, T gpu_wreal, typename GpuVector< T >::value_type gpu_wimag)
	Evaluate the basis for a Fourier grid. More...

template<typename T >
void	TasGrid::TasGpu::devalglo (AccelerationContext const acc, bool is_nested, bool is_clenshawcurtis0, int dims, int num_x, int num_p, int num_basis, T const gpu_x, GpuVector< T > const &nodes, GpuVector< T > const &coeff, GpuVector< T > const &tensor_weights, GpuVector< int > const &nodes_per_level, GpuVector< int > const &offset_per_level, GpuVector< int > const &map_dimension, GpuVector< int > const &map_level, GpuVector< int > const &active_tensors, GpuVector< int > const &active_num_points, GpuVector< int > const &dim_offsets, GpuVector< int > const &map_tensor, GpuVector< int > const &map_index, GpuVector< int > const &map_reference, T *gpu_result)
	Evaluate the basis for Global grid. More...

void	TasGrid::TasGpu::fillDataGPU (AccelerationContext const *acc, double value, long long N, long long stride, double data[])
	Fills the data with the provided real number at the given stride.

template<typename T >
void	TasGrid::TasGpu::load_n (AccelerationContext const acc, T const cpu_data, size_t num_entries, T *gpu_data)
	Similar to copy_n, copies the data from the CPU to the GPU.

template<typename T , typename U >
Utils::use_if<!std::is_same< U, T >::value >	TasGrid::TasGpu::load_n (AccelerationContext const acc, U const cpu_data, size_t num_entries, T *gpu_data)
	Similar to copy_n, copies the data from the CPU to the GPU.

TypeAcceleration	TasGrid::AccelerationMeta::getIOAccelerationString (const char *name)
	Convert the string (coming from C or Python) into an enumerated type. More...

std::map< std::string, TypeAcceleration >	TasGrid::AccelerationMeta::getStringToAccelerationMap ()
	Creates a map with std::string rule names (used by C/Python/CLI) mapped to TypeAcceleration enums.

const char *	TasGrid::AccelerationMeta::getIOAccelerationString (TypeAcceleration accel)
	Convert the enumerated type to a string, the inverse of getIOAccelerationString() More...

int	TasGrid::AccelerationMeta::getIOAccelerationInt (TypeAcceleration accel)
	Convert the integer (coming from Fortran) into an enumerated type. More...

TypeAcceleration	TasGrid::AccelerationMeta::getIOIntAcceleration (int accel)
	Convert the enumerated type to an integer, the inverse of getIOAccelerationInt() More...

bool	TasGrid::AccelerationMeta::isAccTypeGPU (TypeAcceleration accel)
	Returns true if accele is cuda, cublas or magma. More...

bool	TasGrid::AccelerationMeta::isAvailable (TypeAcceleration accel)
	Identifies whether the acceleration mode is available.

TypeAcceleration	TasGrid::AccelerationMeta::getAvailableFallback (TypeAcceleration accel)
	Implements fallback logic, if accel has been enabled through CMake then this returns accel, otherwise it returns the "next-best-thing". More...

int	TasGrid::AccelerationMeta::getNumGpuDevices ()
	Return the number of visible GPU devices. More...

void	TasGrid::AccelerationMeta::setDefaultGpuDevice (int deviceID)
	Selects the active device for this CPU thread, not supported for DPC++. More...

unsigned long long	TasGrid::AccelerationMeta::getTotalGPUMemory (int deviceID)
	Return the memory available in the device (in units of bytes). More...

std::string	TasGrid::AccelerationMeta::getGpuDeviceName (int deviceID)
	Returns the name of the selected GPU device, empty string if no device is available or the index is out of bounds. More...

template<typename T >
void	TasGrid::AccelerationMeta::recvGpuArray (AccelerationContext const , size_t num_entries, const T gpu_data, std::vector< T > &cpu_data)
	Copy a device array to the main memory, used for testing only, always favor using GpuVector (if possible). More...

template<typename T >
void	TasGrid::AccelerationMeta::delGpuArray (AccelerationContext const , T x)
	Deallocate device array, used primarily for testing, always favor using GpuVector (if possible). More...

void *	TasGrid::AccelerationMeta::createCublasHandle ()
	Creates a new cuBlas handle, used in unit-testing only.

void	TasGrid::AccelerationMeta::deleteCublasHandle (void *)
	Destroys the cuBlas handle, used in unit-testing only.

Detailed Description

Data structures for interacting with CUDA and MAGMA environments.

Author: Miroslav Stoyanov

Classes and namespaces that wrap around basic CUDA and MAGMA functionality, the classes allow RAII style of memory management for CUDA GPU arrays, as well as handy encapsulation of the cuBlas/cuSparse/MAGMA handles and streams.

Classes

Namespaces

Functions

Detailed Description