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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323
|
// Copyright 2022 Huawei Cloud Computing Technology Co., Ltd.
//
// 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.
#include "src/buildtool/multithreading/task_system.hpp"
#include <atomic>
#include <chrono>
#include <compare>
#include <condition_variable>
#include <cstddef>
#include <cstdint>
#include <functional>
#include <iterator>
#include <mutex>
#include <numeric> // std::iota
#include <ratio>
#include <string>
#include <thread>
#include <unordered_set>
#include <utility>
#include <vector>
#include "catch2/catch_test_macros.hpp"
#include "catch2/generators/catch_generators_all.hpp"
#include "catch2/matchers/catch_matchers_all.hpp"
#include "src/utils/cpp/atomic.hpp"
#include "test/utils/container_matchers.hpp"
namespace {
enum class CallStatus : std::uint8_t { kNotExecuted, kExecuted };
} // namespace
TEST_CASE("Basic", "[task_system]") {
SECTION("Empty task system terminates") {
{ TaskSystem ts; }
CHECK(true);
}
SECTION("0-arguments constructor") {
TaskSystem ts;
CHECK(ts.NumberOfThreads() == std::thread::hardware_concurrency());
}
SECTION("1-argument constructor") {
std::size_t const desired_number_of_threads_in_ts =
GENERATE(1U, 2U, 5U, 10U, std::thread::hardware_concurrency());
TaskSystem ts(desired_number_of_threads_in_ts);
CHECK(ts.NumberOfThreads() == desired_number_of_threads_in_ts);
}
}
TEST_CASE("Side effects of tasks are reflected out of ts", "[task_system]") {
SECTION("Lambda function") {
auto status = CallStatus::kNotExecuted;
{ // Make sure that all tasks will be completed before the checks
TaskSystem ts;
ts.QueueTask([&status]() { status = CallStatus::kExecuted; });
}
CHECK(status == CallStatus::kExecuted);
}
SECTION("std::function") {
auto status = CallStatus::kNotExecuted;
{
TaskSystem ts;
std::function<void()> f{
[&status]() { status = CallStatus::kExecuted; }};
ts.QueueTask(f);
}
CHECK(status == CallStatus::kExecuted);
}
SECTION("Struct") {
auto s = CallStatus::kNotExecuted;
struct Callable {
explicit Callable(CallStatus* cs) : status{cs} {}
void operator()() const { *status = CallStatus::kExecuted; }
CallStatus* status;
};
Callable c{&s};
{
TaskSystem ts;
ts.QueueTask(c);
}
CHECK(&s == c.status);
CHECK(s == CallStatus::kExecuted);
}
SECTION("Lambda capturing `this` inside struct") {
std::string ext_name{};
struct Wrapper {
std::string name;
// ts must be second, otherwise name will get destroyed before the
// task system is finished.
TaskSystem ts;
explicit Wrapper(std::string n) : name{std::move(n)} {}
void QueueSetAndCheck(std::string* ext) {
ts.QueueTask([this, ext]() {
SetDefaultName();
CheckDefaultName(ext);
});
}
void SetDefaultName() { name = "Default"; }
void CheckDefaultName(std::string* ext) const {
*ext = name;
CHECK(name == "Default");
}
};
{
Wrapper w{"Non-default name"};
w.QueueSetAndCheck(&ext_name);
}
CHECK(ext_name == "Default");
}
}
TEST_CASE("All tasks are executed", "[task_system]") {
std::size_t const number_of_tasks = 1000;
std::vector<int> tasks_executed;
std::vector<int> queued_tasks(number_of_tasks);
std::iota(std::begin(queued_tasks), std::end(queued_tasks), 0);
std::mutex m;
{
TaskSystem ts;
for (auto task_num : queued_tasks) {
ts.QueueTask([&tasks_executed, &m, task_num]() {
std::unique_lock l{m};
tasks_executed.push_back(task_num);
});
}
}
CHECK_THAT(tasks_executed,
HasSameElementsAs<std::vector<int>>(queued_tasks));
}
TEST_CASE("Task is executed even if it needs to wait for a long while",
"[task_system]") {
auto status = CallStatus::kNotExecuted;
// Calculate what would take for the task system to be constructed, queue a
// non-sleeping task, execute it and be destructed
auto const start_no_sleep = std::chrono::high_resolution_clock::now();
{
TaskSystem ts;
ts.QueueTask([&status]() { status = CallStatus::kExecuted; });
}
auto const end_no_sleep = std::chrono::high_resolution_clock::now();
status = CallStatus::kNotExecuted;
std::chrono::nanoseconds const sleep_time =
10 * std::chrono::duration_cast<std::chrono::nanoseconds>(
end_no_sleep - start_no_sleep);
auto const start = std::chrono::high_resolution_clock::now();
{
TaskSystem ts;
ts.QueueTask([&status, sleep_time]() {
std::this_thread::sleep_for(sleep_time);
status = CallStatus::kExecuted;
});
}
auto const end = std::chrono::high_resolution_clock::now();
CHECK(end - start > sleep_time);
CHECK(status == CallStatus::kExecuted);
}
TEST_CASE("All threads run until work is done", "[task_system]") {
using namespace std::chrono_literals;
static auto const kNumThreads = std::thread::hardware_concurrency();
static auto const kFailTimeout = 10s;
std::mutex mutex{};
std::condition_variable cv{};
std::unordered_set<std::thread::id> tids{};
// Add thread id to set and wait for others to do the same.
auto store_id = [&tids, &mutex, &cv]() -> void {
std::unique_lock lock(mutex);
tids.emplace(std::this_thread::get_id());
cv.notify_all();
cv.wait_for(
lock, kFailTimeout, [&tids] { return tids.size() == kNumThreads; });
};
SECTION("single task produces multiple tasks") {
{
TaskSystem ts{kNumThreads};
// Wait some time for all threads to go to sleep.
std::this_thread::sleep_for(1s);
// All threads should stay alive until their corresponding queue is
// filled. One task per thread (assumes round-robin push to queues).
for (std::size_t i{}; i < ts.NumberOfThreads(); ++i) {
ts.QueueTask([&store_id] { store_id(); });
}
}
CHECK(tids.size() == kNumThreads);
}
SECTION("multiple tasks reduce to one, which produces multiple tasks") {
atomic<std::size_t> counter{};
// All threads wait for counter, last thread creates 'store_id' tasks.
auto barrier = [&counter, &store_id](TaskSystem* ts) {
auto value = ++counter;
if (value == kNumThreads) {
counter.notify_all();
// Wait some time for other threads to go to sleep.
std::this_thread::sleep_for(1s);
// One task per thread (assumes round-robin push to queues).
for (std::size_t i{}; i < ts->NumberOfThreads(); ++i) {
ts->QueueTask([&store_id] { store_id(); });
}
}
else {
while (value != kNumThreads) {
counter.wait(value);
value = counter;
}
}
};
{
TaskSystem ts{kNumThreads};
// Wait some time for all threads to go to sleep.
std::this_thread::sleep_for(1s);
// One task per thread (assumes round-robin push to queues).
for (std::size_t i{}; i < ts.NumberOfThreads(); ++i) {
ts.QueueTask([&barrier, &ts] { barrier(&ts); });
}
}
CHECK(tids.size() == kNumThreads);
}
}
TEST_CASE("Use finish as system-wide barrier", "[task_system]") {
using namespace std::chrono_literals;
static auto const kNumThreads = std::thread::hardware_concurrency();
std::vector<int> vec(kNumThreads, 0);
std::vector<int> exp0(kNumThreads, 0);
std::vector<int> exp1(kNumThreads, 1);
std::vector<int> exp2(kNumThreads, 2);
{
TaskSystem ts{kNumThreads};
// Wait for all threads to go to sleep.
ts.Finish();
CHECK(vec == exp0);
for (std::size_t i{}; i < ts.NumberOfThreads(); ++i) {
ts.QueueTask([&vec, i] {
std::this_thread::sleep_for(1s);
vec[i] = 1;
});
}
ts.Finish();
CHECK(vec == exp1);
for (std::size_t i{}; i < ts.NumberOfThreads(); ++i) {
ts.QueueTask([&vec, i] {
std::this_thread::sleep_for(1s);
vec[i] = 2;
});
}
}
CHECK(vec == exp2);
}
TEST_CASE("Shut down a running task system", "[task_system]") {
using namespace std::chrono_literals;
static auto const kNumThreads = std::thread::hardware_concurrency();
std::atomic<int> count{0};
std::atomic<bool> finished{false};
std::function<void()> sleeper{};
{
TaskSystem ts{kNumThreads};
// sleeper, recursively runs forever
sleeper = [&count, &ts, &sleeper]() {
++count;
std::this_thread::sleep_for(1s);
ts.QueueTask(sleeper);
};
// waiter, asynchronous task waiting for task system to finish
std::thread waiter{[&finished, &ts] {
ts.Finish();
finished = true;
}};
// run sleeper
ts.QueueTask(sleeper);
std::this_thread::sleep_for(1s);
// initiate shutdown and join with waiter
ts.Shutdown();
waiter.join();
}
CHECK(count > 0);
CHECK(finished);
}
|