/* ISC License Copyright (c) 2026, Antonio SJ Musumeci Permission to use, copy, modify, and/or distribute this software for any purpose with or without fee is hereby granted, provided that the above copyright notice and this permission notice appear in all copies. THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. */ #pragma once #include "bounded_queue.hpp" #include "invocable.h" #include "mutex.hpp" #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include struct ThreadPoolTraits : public moodycamel::ConcurrentQueueDefaultTraits { static const int MAX_SEMA_SPINS = 1; }; class ThreadPool { public: using PToken = moodycamel::ProducerToken; using CToken = moodycamel::ConsumerToken; private: using Func = ofats::any_invocable; using Queue = BoundedQueue; public: struct ScalingConfig { std::size_t min_threads = 2; std::size_t max_threads = 64; std::int64_t sample_interval_usecs = 100000; // 100ms hill-climb sample period std::int64_t cooldown_usecs = 200000; // 200ms min time between scaling events std::int64_t idle_threshold_usecs = 5000000; // 5s worker idle timeout before self-exit }; explicit ThreadPool(const unsigned thread_count_ = std::thread::hardware_concurrency(), const unsigned max_queue_depth_ = std::thread::hardware_concurrency() * 2, const std::string &name_ = {}, const ScalingConfig *scaling_config_ = nullptr); ~ThreadPool(); ThreadPool(const ThreadPool&) = delete; ThreadPool& operator=(const ThreadPool&) = delete; ThreadPool(ThreadPool&&) = delete; ThreadPool& operator=(ThreadPool&&) = delete; enum WorkerAction { CONTINUE, BREAK, EXIT }; private: static void *start_routine(void *arg_); static void *start_routine_autoscale(void *arg_); static void *monitor_routine(void *arg_); WorkerAction _process_func(Func &func_); int _add_thread_scaling_locked(const std::string &name_); int _remove_thread_scaling_locked(); void _remove_self(pthread_t t_); bool _try_remove_self_if_above(pthread_t t_, std::size_t min_count_); void _maybe_grow_on_pressure(); static std::uint64_t _now_usecs(); std::atomic _remove_count{0}; public: int add_thread(const std::string &name = {}); int remove_thread(); int set_threads(const std::size_t count); public: template void enqueue_work(ThreadPool::PToken &ptok_, FuncType &&func_); template void enqueue_work(FuncType &&func_); template bool try_enqueue_work(ThreadPool::PToken &ptok_, FuncType &&func_); template bool try_enqueue_work(FuncType &&func_); template bool try_enqueue_work_for(ThreadPool::PToken &ptok_, std::int64_t timeout_usecs_, FuncType &&func_); template bool try_enqueue_work_for(std::int64_t timeout_usecs_, FuncType &&func_); template [[nodiscard]] std::future> enqueue_task(FuncType&& func_); // Auto-scaling enqueue: grows pool on queue pressure template void enqueue_work_autoscale(ThreadPool::PToken &ptok_, FuncType &&func_); template void enqueue_work_autoscale(FuncType &&func_); public: std::vector threads() const; ThreadPool::PToken ptoken(); // Introspection std::size_t thread_count() const; std::size_t queue_depth() const; std::size_t queue_capacity() const; std::uint64_t tasks_enqueued() const; std::uint64_t tasks_completed() const; std::uint64_t tasks_pending() const; private: Queue _queue; private: std::string const _name; std::vector _threads; mutable Mutex _threads_mutex; Mutex _scaling_mutex; std::atomic _stop{false}; // Monitoring counters std::atomic _tasks_enqueued{0}; std::atomic _tasks_completed{0}; // Auto-scaling state bool _autoscale_enabled; ScalingConfig _scaling_config; pthread_t _monitor_thread; std::atomic _last_scale_time_usecs{0}; }; inline std::uint64_t ThreadPool::_now_usecs() { struct timespec ts; clock_gettime(CLOCK_MONOTONIC,&ts); return (std::uint64_t)ts.tv_sec * 1000000ULL + (std::uint64_t)ts.tv_nsec / 1000ULL; } inline ThreadPool::ThreadPool(const unsigned thread_count_, const unsigned max_queue_depth_, const std::string &name_, const ScalingConfig *scaling_config_) : _queue(max_queue_depth_), _name(name_), _autoscale_enabled(scaling_config_ != nullptr), _scaling_config(), _monitor_thread() { if(scaling_config_) _scaling_config = *scaling_config_; // Clamp initial thread count to ScalingConfig bounds when auto-scaling. std::size_t count = thread_count_; if(_autoscale_enabled) { if(count < _scaling_config.min_threads) count = _scaling_config.min_threads; if(count > _scaling_config.max_threads) count = _scaling_config.max_threads; } sigset_t oldset; sigset_t newset; sigfillset(&newset); pthread_sigmask(SIG_BLOCK,&newset,&oldset); _threads.reserve(count); for(std::size_t i = 0; i < count; ++i) { int rv; pthread_t t; auto routine = _autoscale_enabled ? ThreadPool::start_routine_autoscale : ThreadPool::start_routine; rv = pthread_create(&t,NULL,routine,this); if(rv != 0) { syslog(LOG_WARNING, "threadpool (%s): error spawning thread - %d (%s)", _name.c_str(), rv, strerror(rv)); continue; } if(not _name.empty()) { std::string name = _name.substr(0, 15); pthread_setname_np(t, name.c_str()); } _threads.push_back(t); } pthread_sigmask(SIG_SETMASK,&oldset,NULL); if(_threads.empty()) throw std::runtime_error("threadpool: failed to spawn any threads"); syslog(LOG_DEBUG, "threadpool (%s): spawned %zu threads w/ max queue depth %u", _name.c_str(), _threads.size(), max_queue_depth_); // Start monitor thread if auto-scaling enabled if(_autoscale_enabled) { sigfillset(&newset); pthread_sigmask(SIG_BLOCK,&newset,&oldset); int rv = pthread_create(&_monitor_thread,NULL, ThreadPool::monitor_routine,this); pthread_sigmask(SIG_SETMASK,&oldset,NULL); if(rv != 0) { syslog(LOG_WARNING, "threadpool (%s): failed to spawn monitor thread - %d (%s)", _name.c_str(), rv, strerror(rv)); _autoscale_enabled = false; } else { pthread_setname_np(_monitor_thread, "tp.monitor"); syslog(LOG_DEBUG, "threadpool (%s): auto-scaling enabled [%zu, %zu]", _name.c_str(), _scaling_config.min_threads, _scaling_config.max_threads); } } } inline ThreadPool::~ThreadPool() { _stop.store(true,std::memory_order_release); // Stop monitor thread first if(_autoscale_enabled) { pthread_join(_monitor_thread,NULL); } // Snapshot threads under lock to close the race with _remove_self. // After _stop is set, _remove_self becomes a no-op, so no thread // will detach or erase itself from this point forward. std::vector snapshot; { mutex_lockguard(_threads_mutex); snapshot = _threads; } syslog(LOG_DEBUG, "threadpool (%s): destroying %zu threads", _name.c_str(), snapshot.size()); // Enqueue one poison pill per thread for(std::size_t i = 0; i < snapshot.size(); ++i) _queue.enqueue_unbounded(Func{}); for(auto t : snapshot) pthread_join(t,NULL); } // Shared post-dequeue logic for both worker loops. // Returns BREAK for poison pill, EXIT for _remove_count claim, CONTINUE otherwise. inline ThreadPool::WorkerAction ThreadPool::_process_func(Func &func_) { // Empty func is a poison pill - exit cleanly. // If _stop is set, destructor owns the join - don't self-remove. // Otherwise this is a set_threads/remove_thread shrink - detach. if(!func_) { if(!_stop.load(std::memory_order_acquire)) _remove_self(pthread_self()); return BREAK; } try { func_(); } catch(const std::exception &e) { syslog(LOG_CRIT, "threadpool (%s): uncaught exception caught by worker - %s", _name.c_str(), e.what()); } catch(...) { syslog(LOG_CRIT, "threadpool (%s): uncaught non-standard exception caught by worker", _name.c_str()); } // Force destruction to release resources func_ = {}; _tasks_completed.fetch_add(1,std::memory_order_relaxed); // _remove_thread_scaling_locked increments _remove_count. // Workers race to claim a decrement after each task; the winner // detaches itself and exits. During shutdown (_stop set), // _remove_self is a no-op, so the destructor still joins us. { std::size_t count = _remove_count.load(std::memory_order_relaxed); while(count > 0) { if(_remove_count.compare_exchange_weak(count,count - 1, std::memory_order_relaxed)) { _remove_self(pthread_self()); return EXIT; } } } return CONTINUE; } // Standard worker loop: blocks indefinitely on dequeue. inline void* ThreadPool::start_routine(void *arg_) { ThreadPool *btp = static_cast(arg_); ThreadPool::Func func; ThreadPool::Queue &q = btp->_queue; ThreadPool::CToken ctok(btp->_queue.make_ctoken()); while(true) { q.wait_dequeue(ctok,func); WorkerAction action = btp->_process_func(func); if(action != CONTINUE) break; } return nullptr; } // Auto-scaling worker loop: uses timed dequeue so idle threads can self-exit. inline void* ThreadPool::start_routine_autoscale(void *arg_) { ThreadPool *btp = static_cast(arg_); ThreadPool::Func func; ThreadPool::Queue &q = btp->_queue; ThreadPool::CToken ctok(btp->_queue.make_ctoken()); std::string const name = btp->_name; while(true) { bool got = q.try_wait_dequeue_for(ctok,func, btp->_scaling_config.idle_threshold_usecs); if(!got) { // Timed out - check if we should shrink if(btp->_stop.load(std::memory_order_acquire)) break; // Copy state needed for the log message BEFORE detaching, // because _try_remove_self_if_above detaches the thread and // the destructor won't wait for us — btp may be destroyed // by the time we reach the syslog call. // // Safety: after _try_remove_self_if_above returns true the // thread is detached and removed from _threads. The // destructor's snapshot won't include it, so pthread_join // is never called. From this point we touch only locals // (name is a std::string copy made at function entry), so // there is no use-after-free even if ~ThreadPool runs // concurrently on another thread. if(btp->_try_remove_self_if_above(pthread_self(), btp->_scaling_config.min_threads)) { syslog(LOG_DEBUG, "threadpool (%s): idle self-exit", name.c_str()); return nullptr; } continue; } WorkerAction action = btp->_process_func(func); if(action != CONTINUE) break; } return nullptr; } // Hill-climbing monitor thread. // Periodically samples throughput, perturbs thread count by +/-1, // and moves in the direction that improves throughput. // Uses an EMA to smooth throughput noise and requires two consecutive // declining samples before reversing direction. inline void* ThreadPool::monitor_routine(void *arg_) { ThreadPool *btp = static_cast(arg_); const std::int64_t interval = btp->_scaling_config.sample_interval_usecs; int direction = 1; int decline_streak = 0; std::uint64_t prev_completed = btp->_tasks_completed.load(std::memory_order_relaxed); double ema_throughput = 0.0; constexpr double ema_alpha = 0.3; // weight for new samples constexpr int decline_threshold = 2; int warmup_samples = 0; const struct timespec sleep_ts = { static_cast(interval / 1000000), static_cast((interval % 1000000) * 1000) }; while(!btp->_stop.load(std::memory_order_acquire)) { { struct timespec remaining = sleep_ts; while(nanosleep(&remaining,&remaining) == -1 && errno == EINTR) { if(btp->_stop.load(std::memory_order_acquire)) break; } } if(btp->_stop.load(std::memory_order_acquire)) break; std::uint64_t completed = btp->_tasks_completed.load(std::memory_order_relaxed); std::uint64_t raw_throughput = completed - prev_completed; prev_completed = completed; // Warm up: seed the EMA with the first two samples before acting. if(warmup_samples < 2) { ema_throughput = (warmup_samples == 0) ? (double)raw_throughput : ema_alpha * (double)raw_throughput + (1.0 - ema_alpha) * ema_throughput; ++warmup_samples; continue; } double prev_ema = ema_throughput; ema_throughput = ema_alpha * (double)raw_throughput + (1.0 - ema_alpha) * ema_throughput; // Only adjust if there's meaningful work happening. // Reset direction to +1 (bias toward growth) so a stale // shrink direction from before the idle period doesn't // penalize the pool when work resumes. if(ema_throughput < 0.5 && prev_ema < 0.5) { decline_streak = 0; direction = 1; continue; } // Hill-climb: did smoothed throughput improve since last sample? if(ema_throughput >= prev_ema) { decline_streak = 0; } else { ++decline_streak; if(decline_streak >= decline_threshold) { direction = -direction; decline_streak = 0; } else { continue; // wait for confirmation before reversing } } // Respect the global cooldown so the monitor and fast-path // pressure grow don't step on each other. { std::uint64_t now = _now_usecs(); std::uint64_t last = btp->_last_scale_time_usecs.load(std::memory_order_relaxed); if((now - last) < (std::uint64_t)btp->_scaling_config.cooldown_usecs) continue; } // Read count, act, then re-read for an accurate log message. std::size_t count = btp->thread_count(); if(direction > 0 && count < btp->_scaling_config.max_threads) { btp->add_thread(); btp->_last_scale_time_usecs.store(_now_usecs(), std::memory_order_relaxed); syslog(LOG_DEBUG, "threadpool (%s): hill-climb grow to %zu threads " "(ema %.1f -> %.1f)", btp->_name.c_str(), btp->thread_count(), prev_ema, ema_throughput); } else if(direction < 0 && count > btp->_scaling_config.min_threads) { btp->remove_thread(); btp->_last_scale_time_usecs.store(_now_usecs(), std::memory_order_relaxed); syslog(LOG_DEBUG, "threadpool (%s): hill-climb shrink to %zu threads " "(ema %.1f -> %.1f)", btp->_name.c_str(), btp->thread_count(), prev_ema, ema_throughput); } } return nullptr; } inline void ThreadPool::_remove_self(pthread_t t_) { // During shutdown the destructor owns all joins; don't detach. if(_stop.load(std::memory_order_acquire)) return; mutex_lockguard(_threads_mutex); auto it = std::find(_threads.begin(),_threads.end(),t_); if(it != _threads.end()) { _threads.erase(it); pthread_detach(t_); } } // Atomically check thread count and remove self if above min_count_. // Returns true if removed, false if at or below threshold. inline bool ThreadPool::_try_remove_self_if_above(pthread_t t_, std::size_t min_count_) { if(_stop.load(std::memory_order_acquire)) return false; mutex_lockguard(_threads_mutex); if(_threads.size() <= min_count_) return false; auto it = std::find(_threads.begin(),_threads.end(),t_); if(it == _threads.end()) return false; _threads.erase(it); pthread_detach(t_); return true; } inline void ThreadPool::_maybe_grow_on_pressure() { if(!_autoscale_enabled) return; std::uint64_t now = _now_usecs(); std::uint64_t last = _last_scale_time_usecs.load(std::memory_order_relaxed); if((now - last) < (std::uint64_t)_scaling_config.cooldown_usecs) return; // Hold _scaling_mutex so the thread-count check and add are atomic // with respect to other scaling operations. mutex_lockguard(_scaling_mutex); std::size_t count; { mutex_lockguard(_threads_mutex); count = _threads.size(); } if(count >= _scaling_config.max_threads) return; // CAS to claim the cooldown window — only one caller wins. // Done after the max-threads check so we don't burn the cooldown // when the pool is already at capacity. if(!_last_scale_time_usecs.compare_exchange_strong(last,now, std::memory_order_relaxed)) return; _add_thread_scaling_locked({}); syslog(LOG_DEBUG, "threadpool (%s): fast-path grow to %zu threads (queue full)", _name.c_str(), count + 1); } inline int ThreadPool::_add_thread_scaling_locked(const std::string &name_) { int rv; pthread_t t; sigset_t oldset; sigset_t newset; std::string name; name = (name_.empty() ? _name : name_); auto routine = _autoscale_enabled ? ThreadPool::start_routine_autoscale : ThreadPool::start_routine; sigfillset(&newset); pthread_sigmask(SIG_BLOCK,&newset,&oldset); rv = pthread_create(&t,NULL,routine,this); pthread_sigmask(SIG_SETMASK,&oldset,NULL); if(rv != 0) { syslog(LOG_WARNING, "threadpool (%s): error spawning thread - %d (%s)", _name.c_str(), rv, strerror(rv)); return -rv; } if(not name.empty()) { std::string n = name.substr(0, 15); pthread_setname_np(t,n.c_str()); } { mutex_lockguard(_threads_mutex); _threads.push_back(t); } return 0; } inline int ThreadPool::add_thread(const std::string &name_) { mutex_lockguard(_scaling_mutex); return _add_thread_scaling_locked(name_); } inline int ThreadPool::_remove_thread_scaling_locked() { { mutex_lockguard(_threads_mutex); std::size_t min_allowed = 1; if(_autoscale_enabled && _scaling_config.min_threads > min_allowed) min_allowed = _scaling_config.min_threads; if(_threads.size() <= min_allowed) return -EINVAL; } // Increment the atomic counter. The next worker to finish a task // will claim the decrement in _process_func, detach itself via // _remove_self, and exit. No heap allocation, no blocking. _remove_count.fetch_add(1,std::memory_order_relaxed); return 0; } inline int ThreadPool::remove_thread() { mutex_lockguard(_scaling_mutex); return _remove_thread_scaling_locked(); } inline int ThreadPool::set_threads(const std::size_t count_) { mutex_lockguard(_scaling_mutex); if(count_ == 0) return -EINVAL; // Clamp to autoscale bounds so external callers can't push the // pool outside the range the monitor and idle-exit logic expect. std::size_t count = count_; if(_autoscale_enabled) { if(count < _scaling_config.min_threads) count = _scaling_config.min_threads; if(count > _scaling_config.max_threads) count = _scaling_config.max_threads; } std::size_t current; { mutex_lockguard(_threads_mutex); current = _threads.size(); } int rv = 0; for(std::size_t i = current; i < count; ++i) { rv = _add_thread_scaling_locked({}); if(rv != 0) return rv; } for(std::size_t i = count; i < current; ++i) { rv = _remove_thread_scaling_locked(); if(rv != 0) return rv; } return 0; } template inline void ThreadPool::enqueue_work(ThreadPool::PToken &ptok_, FuncType &&func_) { _queue.enqueue(ptok_, std::forward(func_)); _tasks_enqueued.fetch_add(1,std::memory_order_relaxed); } template inline void ThreadPool::enqueue_work(FuncType &&func_) { _queue.enqueue(std::forward(func_)); _tasks_enqueued.fetch_add(1,std::memory_order_relaxed); } template inline bool ThreadPool::try_enqueue_work(ThreadPool::PToken &ptok_, FuncType &&func_) { if(!_queue.try_enqueue(ptok_,std::forward(func_))) return false; _tasks_enqueued.fetch_add(1,std::memory_order_relaxed); return true; } template inline bool ThreadPool::try_enqueue_work(FuncType &&func_) { if(!_queue.try_enqueue(std::forward(func_))) return false; _tasks_enqueued.fetch_add(1,std::memory_order_relaxed); return true; } template inline bool ThreadPool::try_enqueue_work_for(ThreadPool::PToken &ptok_, std::int64_t timeout_usecs_, FuncType &&func_) { if(!_queue.try_enqueue_for(ptok_,timeout_usecs_,std::forward(func_))) return false; _tasks_enqueued.fetch_add(1,std::memory_order_relaxed); return true; } template inline bool ThreadPool::try_enqueue_work_for(std::int64_t timeout_usecs_, FuncType &&func_) { if(!_queue.try_enqueue_for(timeout_usecs_,std::forward(func_))) return false; _tasks_enqueued.fetch_add(1,std::memory_order_relaxed); return true; } template inline void ThreadPool::enqueue_work_autoscale(ThreadPool::PToken &ptok_, FuncType &&func_) { // Materialize into a Func up front so we own the value across // the try/block-enqueue boundary without a double std::forward. // // Safe: BoundedQueue::try_enqueue checks the semaphore before // forwarding, so f is not moved-from on failure. Func f(std::forward(func_)); if(_queue.try_enqueue(std::move(f))) { _tasks_enqueued.fetch_add(1,std::memory_order_relaxed); return; } _maybe_grow_on_pressure(); // Block-enqueue (will succeed once a worker frees a slot) _queue.enqueue(ptok_,std::move(f)); _tasks_enqueued.fetch_add(1,std::memory_order_relaxed); } template inline void ThreadPool::enqueue_work_autoscale(FuncType &&func_) { // See ptok_ overload for safety note on try_enqueue + move. Func f(std::forward(func_)); if(_queue.try_enqueue(std::move(f))) { _tasks_enqueued.fetch_add(1,std::memory_order_relaxed); return; } _maybe_grow_on_pressure(); // Block-enqueue (will succeed once a worker frees a slot) _queue.enqueue(std::move(f)); _tasks_enqueued.fetch_add(1,std::memory_order_relaxed); } template [[nodiscard]] inline std::future> ThreadPool::enqueue_task(FuncType&& func_) { using TaskReturnType = std::invoke_result_t; using Promise = std::promise; Promise promise; auto future = promise.get_future(); auto work = [promise_ = std::move(promise), func_ = std::forward(func_)]() mutable { try { if constexpr (std::is_void_v) { func_(); promise_.set_value(); } else { promise_.set_value(func_()); } } catch(...) { promise_.set_exception(std::current_exception()); } }; _queue.enqueue(std::move(work)); _tasks_enqueued.fetch_add(1,std::memory_order_relaxed); return future; } inline std::vector ThreadPool::threads() const { mutex_lockguard(_threads_mutex); return _threads; } inline ThreadPool::PToken ThreadPool::ptoken() { return _queue.make_ptoken(); } inline std::size_t ThreadPool::thread_count() const { mutex_lockguard(_threads_mutex); return _threads.size(); } inline std::size_t ThreadPool::queue_depth() const { return _queue.size_approx(); } inline std::size_t ThreadPool::queue_capacity() const { return _queue.max_depth(); } inline std::uint64_t ThreadPool::tasks_enqueued() const { return _tasks_enqueued.load(std::memory_order_relaxed); } inline std::uint64_t ThreadPool::tasks_completed() const { return _tasks_completed.load(std::memory_order_relaxed); } // Approximate: the two counters are loaded non-atomically with // relaxed ordering, so the result may momentarily read zero (or // a lower value) while tasks complete between the two loads. // Suitable for monitoring and heuristics, not for synchronization. inline std::uint64_t ThreadPool::tasks_pending() const { auto e = _tasks_enqueued.load(std::memory_order_relaxed); auto c = _tasks_completed.load(std::memory_order_relaxed); return (e > c) ? (e - c) : 0; }