parallel_active_index.h File Reference

#include "util/atomic.h"

Go to the source code of this file.

Macros
#define	GPU_PARALLEL_ACTIVE_INDEX_DEFAULT_BLOCK_SIZE   512
#define	gpu_parallel_active_index_array(num_states, indices, num_indices, is_active_op)

Functions
template<typename IsActiveOp>
__device__ void	gpu_parallel_active_index_array_impl (const uint num_states, ccl_global int *indices, ccl_global int *num_indices, IsActiveOp is_active_op)

Macro Definition Documentation

gpu_parallel_active_index_array

#define gpu_parallel_active_index_array	(		num_states,
			indices,
			num_indices,
			is_active_op )

Value:

gpu_parallel_active_index_array_impl(num_states, indices, num_indices, is_active_op)

Definition at line 190 of file parallel_active_index.h.

GPU_PARALLEL_ACTIVE_INDEX_DEFAULT_BLOCK_SIZE

#define GPU_PARALLEL_ACTIVE_INDEX_DEFAULT_BLOCK_SIZE   512

Definition at line 19 of file parallel_active_index.h.

Function Documentation

gpu_parallel_active_index_array_impl()

template<typename IsActiveOp>

__device__ void gpu_parallel_active_index_array_impl	(	const uint	num_states,
		ccl_global int *	indices,
		ccl_global int *	num_indices,
		IsActiveOp	is_active_op )

Definition at line 74 of file parallel_active_index.h.

References __KERNEL_METAL__, atomic_fetch_and_add_uint32(), ccl_global, ccl_gpu_ballot, ccl_gpu_block_dim_x, ccl_gpu_block_idx_x, ccl_gpu_local_syncthreads, ccl_gpu_shared, ccl_gpu_syncthreads, ccl_gpu_thread_idx_x, ccl_gpu_thread_mask, ccl_gpu_warp_size, i, indices, num_states, and popcount().