Spaces:
Build error
Build error
File size: 3,171 Bytes
28451f7 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 |
/*
* SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
* SPDX-License-Identifier: Apache-2.0
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
// This file was taken from the tev image viewer and is re-released here
// under the NVIDIA Source Code License with permission from the author.
#pragma once
#include <neural-graphics-primitives/common.h>
#include <atomic>
#include <deque>
#include <functional>
#include <future>
#include <thread>
#include <vector>
namespace ngp {
template <typename T>
void wait_all(T&& futures) {
for (auto& f : futures) {
f.get();
}
}
class ThreadPool {
public:
ThreadPool();
ThreadPool(size_t maxNum_threads, bool force = false);
virtual ~ThreadPool();
template <class F>
auto enqueue_task(F&& f, bool high_priority = false) -> std::future<std::result_of_t <F()>> {
using return_type = std::result_of_t<F()>;
auto task = std::make_shared<std::packaged_task<return_type()>>(std::forward<F>(f));
auto res = task->get_future();
{
std::lock_guard<std::mutex> lock{m_task_queue_mutex};
if (high_priority) {
m_task_queue.emplace_front([task]() { (*task)(); });
} else {
m_task_queue.emplace_back([task]() { (*task)(); });
}
}
m_worker_condition.notify_one();
return res;
}
void start_threads(size_t num);
void shutdown_threads(size_t num);
void set_n_threads(size_t num);
void wait_until_queue_completed();
void flush_queue();
template <typename Int, typename F>
void parallel_for_async(Int start, Int end, F body, std::vector<std::future<void>>& futures) {
Int local_num_threads = (Int)m_num_threads;
Int range = end - start;
Int chunk = (range / local_num_threads) + 1;
for (Int i = 0; i < local_num_threads; ++i) {
futures.emplace_back(enqueue_task([i, chunk, start, end, body] {
Int inner_start = start + i * chunk;
Int inner_end = std::min(end, start + (i + 1) * chunk);
for (Int j = inner_start; j < inner_end; ++j) {
body(j);
}
}));
}
}
template <typename Int, typename F>
std::vector<std::future<void>> parallel_for_async(Int start, Int end, F body) {
std::vector<std::future<void>> futures;
parallel_for_async(start, end, body, futures);
return futures;
}
template <typename Int, typename F>
void parallel_for(Int start, Int end, F body) {
wait_all(parallel_for_async(start, end, body));
}
private:
size_t m_num_threads = 0;
std::vector<std::thread> m_threads;
std::deque<std::function<void()>> m_task_queue;
std::mutex m_task_queue_mutex;
std::condition_variable m_worker_condition;
std::condition_variable m_task_queue_completed_condition;
};
}
|