CCCL (CUDA C++ Core Libraries) provides: - CUB: device/block/warp-level GPU primitives (reduce, scan, sort, topk) - Thrust: high-level parallel algorithms (transform_reduce, sort, scan) - libcudacxx: CUDA C++ standard library (atomics, barriers, memory) - cudax: experimental features (memory resources, allocators) - Tuning policies: per-SM hardware-specific algorithm parameters Competition optimization vectors mapped to CCCL: - Output TPS (83% weight): warp_reduce, block_reduce, device_topk - Input TPS (14% weight): device_scan, block_load, prefetch - Cache TPS (3% weight): prefix caching strategy patterns - Memory (0.9 util): pooled/cached/buddy allocators Source: https://github.com/NVIDIA/cccl (shallow clone, HEAD only) License: Apache-2.0
62 lines
2.2 KiB
C++
62 lines
2.2 KiB
C++
// SPDX-FileCopyrightText: Copyright (c) 2008-2013, NVIDIA Corporation. All rights reserved.
|
|
// SPDX-License-Identifier: Apache-2.0
|
|
|
|
#pragma once
|
|
|
|
#include <thrust/detail/config.h>
|
|
|
|
#if defined(_CCCL_IMPLICIT_SYSTEM_HEADER_GCC)
|
|
# pragma GCC system_header
|
|
#elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_CLANG)
|
|
# pragma clang system_header
|
|
#elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_MSVC)
|
|
# pragma system_header
|
|
#endif // no system header
|
|
#include <thrust/iterator/iterator_traits.h>
|
|
#include <thrust/system/detail/generic/select_system.h>
|
|
#include <thrust/tabulate.h>
|
|
|
|
// Include all active backend system implementations (generic, sequential, host and device)
|
|
#include <thrust/system/detail/generic/tabulate.h>
|
|
#include <thrust/system/detail/sequential/tabulate.h>
|
|
#include __THRUST_HOST_SYSTEM_ALGORITH_DETAIL_HEADER_INCLUDE(tabulate.h)
|
|
#include __THRUST_DEVICE_SYSTEM_ALGORITH_DETAIL_HEADER_INCLUDE(tabulate.h)
|
|
|
|
// Some build systems need a hint to know which files we could include
|
|
#if 0
|
|
# include <thrust/system/cpp/detail/tabulate.h>
|
|
# include <thrust/system/cuda/detail/tabulate.h>
|
|
# include <thrust/system/omp/detail/tabulate.h>
|
|
# include <thrust/system/tbb/detail/tabulate.h>
|
|
#endif
|
|
|
|
THRUST_NAMESPACE_BEGIN
|
|
|
|
_CCCL_EXEC_CHECK_DISABLE
|
|
template <typename DerivedPolicy, typename ForwardIterator, typename UnaryOperation>
|
|
_CCCL_HOST_DEVICE void
|
|
tabulate(const thrust::detail::execution_policy_base<DerivedPolicy>& exec,
|
|
ForwardIterator first,
|
|
ForwardIterator last,
|
|
UnaryOperation unary_op)
|
|
{
|
|
_CCCL_NVTX_RANGE_SCOPE("thrust::tabulate");
|
|
using thrust::system::detail::generic::tabulate;
|
|
return tabulate(thrust::detail::derived_cast(thrust::detail::strip_const(exec)), first, last, unary_op);
|
|
} // end tabulate()
|
|
|
|
template <typename ForwardIterator, typename UnaryOperation>
|
|
void tabulate(ForwardIterator first, ForwardIterator last, UnaryOperation unary_op)
|
|
{
|
|
_CCCL_NVTX_RANGE_SCOPE("thrust::tabulate");
|
|
using thrust::system::detail::generic::select_system;
|
|
|
|
using System = typename thrust::iterator_system<ForwardIterator>::type;
|
|
|
|
System system;
|
|
|
|
return thrust::tabulate(select_system(system), first, last, unary_op);
|
|
} // end tabulate()
|
|
|
|
THRUST_NAMESPACE_END
|