cub::DispatchSegmentedReduce< InputIteratorT, OutputIteratorT, OffsetIteratorT, OffsetT, ReductionOpT > Struct Template Reference

< Binary reduction functor type having member T operator()(const T &a, const T &b) More...

Detailed Description

template<typename InputIteratorT, typename OutputIteratorT, typename OffsetIteratorT, typename OffsetT, typename ReductionOpT>
struct cub::DispatchSegmentedReduce< InputIteratorT, OutputIteratorT, OffsetIteratorT, OffsetT, ReductionOpT >

< Binary reduction functor type having member T operator()(const T &a, const T &b)

Utility class for dispatching the appropriately-tuned kernels for device-wide reduction

Definition at line 681 of file dispatch_reduce.cuh.

Inheritance diagram for cub::DispatchSegmentedReduce< InputIteratorT, OutputIteratorT, OffsetIteratorT, OffsetT, ReductionOpT >:

Public Types
typedef If<(Equals< typename std::iterator_traits< OutputIteratorT >::value_type, void >::VALUE), typename std::iterator_traits< InputIteratorT >::value_type, typename std::iterator_traits< OutputIteratorT >::value_type >::Type	OutputT
	The output value type.
Public Types inherited from cub::DeviceReducePolicy< std::iterator_traits< InputIteratorT >::value_type, OffsetT, ReductionOpT >
typedef Policy600	MaxPolicy
	MaxPolicy.

Public Member Functions
CUB_RUNTIME_FUNCTION __forceinline__	DispatchSegmentedReduce (void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, OffsetT num_segments, OffsetIteratorT d_begin_offsets, OffsetIteratorT d_end_offsets, ReductionOpT reduction_op, OutputT init, cudaStream_t stream, bool debug_synchronous, int ptx_version)
	Constructor.
template<typename ActivePolicyT , typename DeviceSegmentedReduceKernelT >
CUB_RUNTIME_FUNCTION __forceinline__ cudaError_t	InvokePasses (DeviceSegmentedReduceKernelT segmented_reduce_kernel)
	Invocation. More...
template<typename ActivePolicyT >
CUB_RUNTIME_FUNCTION __forceinline__ cudaError_t	Invoke ()
	Invocation.

Static Public Member Functions
CUB_RUNTIME_FUNCTION static __forceinline__ cudaError_t	Dispatch (void *d_temp_storage, size_t &temp_storage_bytes, InputIteratorT d_in, OutputIteratorT d_out, int num_segments, OffsetIteratorT d_begin_offsets, OffsetIteratorT d_end_offsets, ReductionOpT reduction_op, OutputT init, cudaStream_t stream, bool debug_synchronous)

Data Fields
void *	d_temp_storage
	[in] Device-accessible allocation of temporary storage. When NULL, the required allocation size is written to temp_storage_bytes and no work is done.
size_t &	temp_storage_bytes
	[in,out] Reference to size in bytes of d_temp_storage allocation
InputIteratorT	d_in
	[in] Pointer to the input sequence of data items
OutputIteratorT	d_out
	[out] Pointer to the output aggregate
OffsetT	num_segments
	[in] The number of segments that comprise the sorting data
OffsetIteratorT	d_begin_offsets
	[in] Pointer to the sequence of beginning offsets of length num_segments, such that d_begin_offsets[i] is the first element of the ith data segment in d_keys_* and d_values_*
OffsetIteratorT	d_end_offsets
	[in] Pointer to the sequence of ending offsets of length num_segments, such that d_end_offsets[i]-1 is the last element of the ith data segment in d_keys_* and d_values_*. If d_end_offsets[i]-1 <= d_begin_offsets[i], the ith is considered empty.
ReductionOpT	reduction_op
	[in] Binary reduction functor
OutputT	init
	[in] The initial value of the reduction
cudaStream_t	stream
	[in] CUDA stream to launch kernels within. Default is stream0.
bool	debug_synchronous
	[in] Whether or not to synchronize the stream after every kernel launch to check for errors. Also causes launch configurations to be printed to the console. Default is false.
int	ptx_version
	[in] PTX version

Member Function Documentation

Dispatch()

template<typename InputIteratorT , typename OutputIteratorT , typename OffsetIteratorT , typename OffsetT , typename ReductionOpT >

CUB_RUNTIME_FUNCTION static __forceinline__ cudaError_t cub::DispatchSegmentedReduce< InputIteratorT, OutputIteratorT, OffsetIteratorT, OffsetT, ReductionOpT >::Dispatch ( void *  d_temp_storage, size_t &  temp_storage_bytes, InputIteratorT  d_in, OutputIteratorT  d_out, int  num_segments, OffsetIteratorT  d_begin_offsets, OffsetIteratorT  d_end_offsets, ReductionOpT  reduction_op, OutputT  init, cudaStream_t  stream, bool  debug_synchronous  )

inlinestatic

Internal dispatch routine for computing a device-wide reduction

Parameters

[in]	d_temp_storage	Device-accessible allocation of temporary storage. When NULL, the required allocation size is written to temp_storage_bytes and no work is done.
[in,out]	temp_storage_bytes	Reference to size in bytes of d_temp_storage allocation
[in]	d_in	Pointer to the input sequence of data items
[out]	d_out	Pointer to the output aggregate
[in]	num_segments	The number of segments that comprise the sorting data
[in]	d_begin_offsets	Pointer to the sequence of beginning offsets of length num_segments, such that d_begin_offsets[i] is the first element of the ith data segment in d_keys_* and d_values_*
[in]	d_end_offsets	Pointer to the sequence of ending offsets of length num_segments, such that d_end_offsets[i]-1 is the last element of the ith data segment in d_keys_* and d_values_*. If d_end_offsets[i]-1 <= d_begin_offsets[i], the ith is considered empty.
[in]	reduction_op	Binary reduction functor
[in]	init	The initial value of the reduction
[in]	stream	[optional] CUDA stream to launch kernels within. Default is stream0.
[in]	debug_synchronous	[optional] Whether or not to synchronize the stream after every kernel launch to check for errors. Also causes launch configurations to be printed to the console. Default is false.

Definition at line 835 of file dispatch_reduce.cuh.

InvokePasses()

template<typename InputIteratorT , typename OutputIteratorT , typename OffsetIteratorT , typename OffsetT , typename ReductionOpT >

template<typename ActivePolicyT , typename DeviceSegmentedReduceKernelT >

CUB_RUNTIME_FUNCTION __forceinline__ cudaError_t cub::DispatchSegmentedReduce< InputIteratorT, OutputIteratorT, OffsetIteratorT, OffsetT, ReductionOpT >::InvokePasses ( DeviceSegmentedReduceKernelT  segmented_reduce_kernel )

inline

Invocation.

< Function type of cub::DeviceSegmentedReduceKernel

Parameters

[in]

segmented_reduce_kernel

Kernel function pointer to parameterization of cub::DeviceSegmentedReduceKernel

Definition at line 759 of file dispatch_reduce.cuh.

---

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

• openfpm_data/src/util/cuda/cub_old/device/dispatch/dispatch_reduce.cuh