OpenFPM_pdata  4.1.0
Project that contain the implementation of distributed structures
 
Loading...
Searching...
No Matches
cub::BlockStore< T, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::StoreInternal< BLOCK_STORE_VECTORIZE, DUMMY > Struct Template Reference

Detailed Description

template<typename T, int BLOCK_DIM_X, int ITEMS_PER_THREAD, BlockStoreAlgorithm ALGORITHM = BLOCK_STORE_DIRECT, int BLOCK_DIM_Y = 1, int BLOCK_DIM_Z = 1, int PTX_ARCH = CUB_PTX_ARCH>
template<int DUMMY>
struct cub::BlockStore< T, BLOCK_DIM_X, ITEMS_PER_THREAD, ALGORITHM, BLOCK_DIM_Y, BLOCK_DIM_Z, PTX_ARCH >::StoreInternal< BLOCK_STORE_VECTORIZE, DUMMY >

BLOCK_STORE_VECTORIZE specialization of store helper

Definition at line 584 of file block_store.cuh.

Public Types

typedef NullType TempStorage
 Shared memory storage layout type.
 

Public Member Functions

__device__ __forceinline__ StoreInternal (TempStorage &, int linear_tid)
 Constructor.
 
__device__ __forceinline__ void Store (T *block_ptr, T(&items)[ITEMS_PER_THREAD])
 Store items into a linear segment of memory, specialized for native pointer types (attempts vectorization)
 
template<typename OutputIteratorT >
__device__ __forceinline__ void Store (OutputIteratorT block_itr, T(&items)[ITEMS_PER_THREAD])
 Store items into a linear segment of memory, specialized for opaque input iterators (skips vectorization)
 
template<typename OutputIteratorT >
__device__ __forceinline__ void Store (OutputIteratorT block_itr, T(&items)[ITEMS_PER_THREAD], int valid_items)
 Store items into a linear segment of memory, guarded by range.
 

Data Fields

int linear_tid
 Linear thread-id.
 

The documentation for this struct was generated from the following file: