56fd68e7dd
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
35 lines
1.1 KiB
C++
35 lines
1.1 KiB
C++
// SPDX-FileCopyrightText: Copyright (c) 2020, 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/mr/allocator.h>
|
|
#include <thrust/mr/device_memory_resource.h>
|
|
#include <thrust/mr/disjoint_tls_pool.h>
|
|
#include <thrust/mr/new.h>
|
|
|
|
THRUST_NAMESPACE_BEGIN
|
|
|
|
namespace detail
|
|
{
|
|
inline thrust::mr::allocator<
|
|
char,
|
|
thrust::mr::disjoint_unsynchronized_pool_resource<thrust::device_memory_resource, thrust::mr::new_delete_resource>>
|
|
single_device_tls_caching_allocator()
|
|
{
|
|
return {&thrust::mr::tls_disjoint_pool(thrust::mr::get_global_resource<thrust::device_memory_resource>(),
|
|
thrust::mr::get_global_resource<thrust::mr::new_delete_resource>())};
|
|
}
|
|
} // namespace detail
|
|
|
|
THRUST_NAMESPACE_END
|