Reference documentation for deal.II version Git 953c9590e9 2020-10-28 17:25:14 -0400
\(\newcommand{\dealvcentcolon}{\mathrel{\mathop{:}}}\) \(\newcommand{\dealcoloneq}{\dealvcentcolon\mathrel{\mkern-1.2mu}=}\) \(\newcommand{\jump}[1]{\left[\!\left[ #1 \right]\!\right]}\) \(\newcommand{\average}[1]{\left\{\!\left\{ #1 \right\}\!\right\}}\)
Namespaces | Classes | Enumerations | Functions
CUDAWrappers::internal Namespace Reference

Namespaces

 internal
 

Classes

struct  EvaluatorTensorProduct
 
struct  EvaluatorTensorProduct< evaluate_general, dim, fe_degree, n_q_points_1d, Number >
 
class  HangingNodes
 

Enumerations

enum  EvaluatorVariant { evaluate_general, evaluate_symmetric, evaluate_evenodd }
 

Functions

template<int dim, int n_points_1d>
unsigned int compute_index ()
 
template<int dim, int fe_degree, bool transpose, typename Number >
void resolve_hanging_nodes (const unsigned int constraint_mask, Number *values)
 

Function Documentation

◆ compute_index()

template<int dim, int n_points_1d>
unsigned int CUDAWrappers::internal::compute_index ( )
inline

Compute the dof/quad index for a given thread id, dimension, and number of points in each space dimensions.

Definition at line 49 of file cuda_fe_evaluation.h.

◆ resolve_hanging_nodes()

template<int dim, int fe_degree, bool transpose, typename Number >
void CUDAWrappers::internal::resolve_hanging_nodes ( const unsigned int  constraint_mask,
Number *  values 
)

This function resolves the hanging nodes using tensor product.

The implementation of this class is explained in Matrix-Free Finite-Element Computations On Graphics Processors With Adaptively Refined Unstructured Meshes by Karl Ljungkvist, SpringSim-HPC, 2017 April 23-26.

Definition at line 975 of file cuda_hanging_nodes_internal.h.