Make it possible to override the synchonization primitives used by the threadpool using macros.

Eigen/ThreadPool

@@ -51,6 +51,16 @@
 #include "src/Core/util/Meta.h"
 #include "src/Core/util/MaxSizeVector.h"
 
+#ifndef EIGEN_MUTEX
+#define EIGEN_MUTEX std::mutex
+#endif
+#ifndef EIGEN_MUTEX_LOCK
+#define EIGEN_MUTEX_LOCK std::unique_lock<std::mutex>
+#endif
+#ifndef EIGEN_CONDVAR
+#define EIGEN_CONDVAR std::condition_variable
+#endif
+
 // IWYU pragma: begin_exports
 #include "src/ThreadPool/ThreadLocal.h"
 #include "src/ThreadPool/ThreadYield.h"

Eigen/src/ThreadPool/Barrier.h

@@ -33,7 +33,7 @@ class Barrier {
       eigen_plain_assert(((v + 2) & ~1) != 0);
       return;  // either count has not dropped to 0, or waiter is not waiting
     }
-    std::unique_lock<std::mutex> l(mu_);
+    EIGEN_MUTEX_LOCK l(mu_);
     eigen_plain_assert(!notified_);
     notified_ = true;
     cv_.notify_all();
@@ -42,15 +42,15 @@ class Barrier {
   void Wait() {
     unsigned int v = state_.fetch_or(1, std::memory_order_acq_rel);
     if ((v >> 1) == 0) return;
-    std::unique_lock<std::mutex> l(mu_);
+    EIGEN_MUTEX_LOCK l(mu_);
     while (!notified_) {
       cv_.wait(l);
     }
   }
 
  private:
-  std::mutex mu_;
+  EIGEN_MUTEX mu_;
-  std::condition_variable cv_;
+  EIGEN_CONDVAR cv_;
   std::atomic<unsigned int> state_;  // low bit is waiter flag
   bool notified_;
 };

Eigen/src/ThreadPool/EventCount.h

@@ -171,8 +171,8 @@ class EventCount {
     // Align to 128 byte boundary to prevent false sharing with other Waiter
     // objects in the same vector.
     EIGEN_ALIGN_TO_BOUNDARY(128) std::atomic<uint64_t> next;
-    std::mutex mu;
+    EIGEN_MUTEX mu;
-    std::condition_variable cv;
+    EIGEN_CONDVAR cv;
     uint64_t epoch = 0;
     unsigned state = kNotSignaled;
     enum {
@@ -220,7 +220,7 @@ class EventCount {
   }
 
   void Park(Waiter* w) {
-    std::unique_lock<std::mutex> lock(w->mu);
+    EIGEN_MUTEX_LOCK lock(w->mu);
     while (w->state != Waiter::kSignaled) {
       w->state = Waiter::kWaiting;
       w->cv.wait(lock);
@@ -233,7 +233,7 @@ class EventCount {
       next = wnext == kStackMask ? nullptr : &waiters_[wnext];
       unsigned state;
       {
-        std::unique_lock<std::mutex> lock(w->mu);
+        EIGEN_MUTEX_LOCK lock(w->mu);
         state = w->state;
         w->state = Waiter::kSignaled;
       }

Eigen/src/ThreadPool/NonBlockingThreadPool.h

@@ -248,7 +248,7 @@ class ThreadPoolTempl : public Eigen::ThreadPoolInterface {
   EventCount ec_;
 #ifndef EIGEN_THREAD_LOCAL
   std::unique_ptr<Barrier> init_barrier_;
-  std::mutex per_thread_map_mutex_;  // Protects per_thread_map_.
+  EIGEN_MUTEX per_thread_map_mutex_;  // Protects per_thread_map_.
   std::unordered_map<uint64_t, std::unique_ptr<PerThread>> per_thread_map_;
 #endif
 

Eigen/src/ThreadPool/RunQueue.h

@@ -84,7 +84,7 @@ class RunQueue {
   // PushBack adds w at the end of the queue.
   // If queue is full returns w, otherwise returns default-constructed Work.
   Work PushBack(Work w) {
-    std::unique_lock<std::mutex> lock(mutex_);
+    EIGEN_MUTEX_LOCK lock(mutex_);
     unsigned back = back_.load(std::memory_order_relaxed);
     Elem* e = &array_[(back - 1) & kMask];
     uint8_t s = e->state.load(std::memory_order_relaxed);
@@ -101,7 +101,7 @@ class RunQueue {
   // PopBack removes and returns the last elements in the queue.
   Work PopBack() {
     if (Empty()) return Work();
-    std::unique_lock<std::mutex> lock(mutex_);
+    EIGEN_MUTEX_LOCK lock(mutex_);
     unsigned back = back_.load(std::memory_order_relaxed);
     Elem* e = &array_[back & kMask];
     uint8_t s = e->state.load(std::memory_order_relaxed);
@@ -118,7 +118,7 @@ class RunQueue {
   // Returns number of elements removed.
   unsigned PopBackHalf(std::vector<Work>* result) {
     if (Empty()) return 0;
-    std::unique_lock<std::mutex> lock(mutex_);
+    EIGEN_MUTEX_LOCK lock(mutex_);
     unsigned back = back_.load(std::memory_order_relaxed);
     unsigned size = Size();
     unsigned mid = back;
@@ -174,7 +174,7 @@ class RunQueue {
     kBusy,
     kReady,
   };
-  std::mutex mutex_;
+  EIGEN_MUTEX mutex_;
   // Low log(kSize) + 1 bits in front_ and back_ contain rolling index of
   // front/back, respectively. The remaining bits contain modification counters
   // that are incremented on Push operations. This allows us to (1) distinguish

Eigen/src/ThreadPool/ThreadLocal.h

@@ -225,7 +225,7 @@ class ThreadLocal {
     if (filled_records_.load(std::memory_order_relaxed) < capacity_) return;
 
     // Adds a happens before edge from the last call to SpilledLocal().
-    std::unique_lock<std::mutex> lock(mu_);
+    EIGEN_MUTEX_LOCK lock(mu_);
     for (auto& kv : per_thread_map_) {
       f(kv.first, kv.second);
     }
@@ -245,7 +245,7 @@ class ThreadLocal {
     if (filled_records_.load(std::memory_order_relaxed) < capacity_) return;
 
     // Adds a happens before edge from the last call to SpilledLocal().
-    std::unique_lock<std::mutex> lock(mu_);
+    EIGEN_MUTEX_LOCK lock(mu_);
     for (auto& kv : per_thread_map_) {
       release_(kv.second);
     }
@@ -259,7 +259,7 @@ class ThreadLocal {
 
   // Use unordered map guarded by a mutex when lock free storage is full.
   T& SpilledLocal(std::thread::id this_thread) {
-    std::unique_lock<std::mutex> lock(mu_);
+    EIGEN_MUTEX_LOCK lock(mu_);
 
     auto it = per_thread_map_.find(this_thread);
     if (it == per_thread_map_.end()) {
@@ -290,7 +290,7 @@ class ThreadLocal {
   // We fallback on per thread map if lock-free storage is full. In practice
   // this should never happen, if `capacity_` is a reasonable estimate of the
   // number of threads running in a system.
-  std::mutex mu_;  // Protects per_thread_map_.
+  EIGEN_MUTEX mu_;  // Protects per_thread_map_.
   std::unordered_map<std::thread::id, T> per_thread_map_;
 };