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
46 lines
1.2 KiB
C++
46 lines
1.2 KiB
C++
//===----------------------------------------------------------------------===//
|
|
//
|
|
// Part of libcu++, the C++ Standard Library for your entire system,
|
|
// under the Apache License v2.0 with LLVM Exceptions.
|
|
// See https://llvm.org/LICENSE.txt for license information.
|
|
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
|
|
// SPDX-FileCopyrightText: Copyright (c) 2024 NVIDIA CORPORATION & AFFILIATES.
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
#include <fstream>
|
|
#include <iostream>
|
|
#include <ostream>
|
|
|
|
#include "generators/compare_and_swap.h"
|
|
#include "generators/exchange.h"
|
|
#include "generators/fence.h"
|
|
#include "generators/fetch_ops.h"
|
|
#include "generators/header.h"
|
|
#include "generators/ld_st.h"
|
|
|
|
using namespace std::string_literals;
|
|
|
|
int main(int argc, char** argv)
|
|
{
|
|
std::fstream filestream;
|
|
|
|
if (argc == 2)
|
|
{
|
|
filestream.open(argv[1], filestream.out);
|
|
}
|
|
|
|
std::ostream& stream = filestream.is_open() ? filestream : std::cout;
|
|
|
|
FormatHeader(stream);
|
|
FormatFence(stream);
|
|
FormatLoad(stream);
|
|
FormatStore(stream);
|
|
FormatCompareAndSwap(stream);
|
|
FormatExchange(stream);
|
|
FormatFetchOps(stream);
|
|
FormatTail(stream);
|
|
|
|
return 0;
|
|
}
|