mirror of
https://github.com/boostorg/atomic.git
synced 2026-02-02 20:32:09 +00:00
76e25f36a3
The generic implementation is based on the lock pool. A list of condition variables (or waiting futexes) is added per lock. Basically, the lock pool serves as a global hash table, where each lock represents a bucket and each wait state is an element. Every wait operation allocates a wait state keyed on the pointer to the atomic object. Notify operations look up the wait state by the atomic pointer and notify the condition variable/futex. The corresponding lock needs to be acquired to protect the wait state list during all wait/notify operations. Backends not involving the lock pool are going to be added later. The implementation of wait operation extends the C++20 definition in that it returns the newly loaded value instead of void. This allows the caller to avoid loading the value himself. The waiting/notifying operations are not address-free. Address-free variants will be added later. Added tests for the new operations and refactored existing tests for atomic operations. Added docs for the new operations.
78 lines
2.3 KiB
C++
78 lines
2.3 KiB
C++
// Copyright (c) 2020 Andrey Semashev
|
|
//
|
|
// Distributed under the Boost Software License, Version 1.0.
|
|
// See accompanying file LICENSE_1_0.txt or copy at
|
|
// http://www.boost.org/LICENSE_1_0.txt)
|
|
|
|
// This is a fuzzing test for waiting and notifying operations.
|
|
// The test creates a number of threads exceeding the number of hardware threads, each of which
|
|
// blocks on the atomic object. The main thread then notifies one or all threads repeatedly,
|
|
// while incrementing the atomic object. The test ends when the atomic counter reaches the predefined limit.
|
|
// The goal of the test is to verify that (a) it doesn't crash and (b) all threads get unblocked in the end.
|
|
|
|
#include <boost/atomic.hpp>
|
|
|
|
#include <iostream>
|
|
#include <boost/config.hpp>
|
|
#include <boost/bind/bind.hpp>
|
|
#include <boost/chrono/chrono.hpp>
|
|
#include <boost/thread/thread.hpp>
|
|
#include <boost/thread/barrier.hpp>
|
|
#include <boost/smart_ptr/scoped_array.hpp>
|
|
|
|
namespace chrono = boost::chrono;
|
|
|
|
boost::atomic< unsigned int > g_atomic(0u);
|
|
|
|
BOOST_CONSTEXPR_OR_CONST unsigned int loop_count = 4096u;
|
|
|
|
void thread_func(boost::barrier* barrier)
|
|
{
|
|
barrier->wait();
|
|
|
|
unsigned int old_count = 0u;
|
|
while (true)
|
|
{
|
|
unsigned int new_count = g_atomic.wait(old_count, boost::memory_order_relaxed);
|
|
if (new_count >= loop_count)
|
|
break;
|
|
|
|
old_count = new_count;
|
|
}
|
|
}
|
|
|
|
int main()
|
|
{
|
|
const unsigned int thread_count = boost::thread::hardware_concurrency() + 4u;
|
|
boost::barrier barrier(thread_count + 1u);
|
|
boost::scoped_array< boost::thread > threads(new boost::thread[thread_count]);
|
|
|
|
for (unsigned int i = 0u; i < thread_count; ++i)
|
|
boost::thread(boost::bind(&thread_func, &barrier)).swap(threads[i]);
|
|
|
|
barrier.wait();
|
|
|
|
// Let the threads block on the atomic counter
|
|
boost::this_thread::sleep_for(chrono::milliseconds(100));
|
|
|
|
while (true)
|
|
{
|
|
for (unsigned int i = 0u; i < thread_count; ++i)
|
|
{
|
|
g_atomic.opaque_add(1u, boost::memory_order_relaxed);
|
|
g_atomic.notify_one();
|
|
}
|
|
|
|
unsigned int old_count = g_atomic.fetch_add(1u, boost::memory_order_relaxed);
|
|
g_atomic.notify_all();
|
|
|
|
if ((old_count + 1u) >= loop_count)
|
|
break;
|
|
}
|
|
|
|
for (unsigned int i = 0u; i < thread_count; ++i)
|
|
threads[i].join();
|
|
|
|
return 0u;
|
|
}
|