Index: include/future
===================================================================
--- include/future
+++ include/future
@@ -367,7 +367,8 @@
 #include <memory>
 #include <chrono>
 #include <exception>
-#include <mutex>
+#include <atomic>
+#include <condition_variable>
 #include <thread>
 
 #if !defined(_LIBCPP_HAS_NO_PRAGMA_SYSTEM_HEADER)
@@ -533,12 +534,24 @@
 {
 protected:
     exception_ptr __exception_;
-    mutable mutex __mut_;
-    mutable condition_variable __cv_;
+
+    mutable class __spin_lock
+    {
+        atomic_flag __locked_ = ATOMIC_FLAG_INIT ;
+    public:
+        void lock() {
+            while (__locked_.test_and_set(memory_order_acquire)) { ; }
+        }
+        void unlock() {
+            __locked_.clear(memory_order_release);
+        }
+    } __mut_;
+
+    mutable condition_variable_any __cv_;
     unsigned __state_;
 
     virtual void __on_zero_shared() _NOEXCEPT;
-    void __sub_wait(unique_lock<mutex>& __lk);
+    void __sub_wait(unique_lock<__spin_lock>& __lk);
 public:
     enum
     {
@@ -558,7 +571,7 @@
     _LIBCPP_INLINE_VISIBILITY
     void __set_future_attached()
     {
-        lock_guard<mutex> __lk(__mut_);
+        lock_guard<__spin_lock> __lk(__mut_);
         __state_ |= __future_attached;
     }
     _LIBCPP_INLINE_VISIBILITY
@@ -596,7 +609,7 @@
 future_status
 __assoc_sub_state::wait_until(const chrono::time_point<_Clock, _Duration>& __abs_time) const
 {
-    unique_lock<mutex> __lk(__mut_);
+    unique_lock<__spin_lock> __lk(__mut_);
     if (__state_ & deferred)
         return future_status::deferred;
     while (!(__state_ & ready) && _Clock::now() < __abs_time)
@@ -663,7 +676,7 @@
 __assoc_state<_Rp>::set_value(_Arg& __arg)
 #endif
 {
-    unique_lock<mutex> __lk(this->__mut_);
+    unique_lock<__spin_lock> __lk(this->__mut_);
     if (this->__has_value())
         __throw_future_error(future_errc::promise_already_satisfied);
     ::new(&__value_) _Rp(_VSTD::forward<_Arg>(__arg));
@@ -680,7 +693,7 @@
 __assoc_state<_Rp>::set_value_at_thread_exit(_Arg& __arg)
 #endif
 {
-    unique_lock<mutex> __lk(this->__mut_);
+    unique_lock<__spin_lock> __lk(this->__mut_);
     if (this->__has_value())
         __throw_future_error(future_errc::promise_already_satisfied);
     ::new(&__value_) _Rp(_VSTD::forward<_Arg>(__arg));
@@ -692,7 +705,7 @@
 _Rp
 __assoc_state<_Rp>::move()
 {
-    unique_lock<mutex> __lk(this->__mut_);
+    unique_lock<__spin_lock> __lk(this->__mut_);
     this->__sub_wait(__lk);
     if (this->__exception_ != nullptr)
         rethrow_exception(this->__exception_);
@@ -703,7 +716,7 @@
 typename add_lvalue_reference<_Rp>::type
 __assoc_state<_Rp>::copy()
 {
-    unique_lock<mutex> __lk(this->__mut_);
+    unique_lock<__spin_lock> __lk(this->__mut_);
     this->__sub_wait(__lk);
     if (this->__exception_ != nullptr)
         rethrow_exception(this->__exception_);
@@ -739,7 +752,7 @@
 void
 __assoc_state<_Rp&>::set_value(_Rp& __arg)
 {
-    unique_lock<mutex> __lk(this->__mut_);
+    unique_lock<__spin_lock> __lk(this->__mut_);
     if (this->__has_value())
         __throw_future_error(future_errc::promise_already_satisfied);
     __value_ = _VSTD::addressof(__arg);
@@ -751,7 +764,7 @@
 void
 __assoc_state<_Rp&>::set_value_at_thread_exit(_Rp& __arg)
 {
-    unique_lock<mutex> __lk(this->__mut_);
+    unique_lock<__spin_lock> __lk(this->__mut_);
     if (this->__has_value())
         __throw_future_error(future_errc::promise_already_satisfied);
     __value_ = _VSTD::addressof(__arg);
@@ -763,7 +776,7 @@
 _Rp&
 __assoc_state<_Rp&>::copy()
 {
-    unique_lock<mutex> __lk(this->__mut_);
+    unique_lock<__spin_lock> __lk(this->__mut_);
     this->__sub_wait(__lk);
     if (this->__exception_ != nullptr)
         rethrow_exception(this->__exception_);
Index: src/future.cpp
===================================================================
--- src/future.cpp
+++ src/future.cpp
@@ -91,7 +91,7 @@
 void
 __assoc_sub_state::set_value()
 {
-    unique_lock<mutex> __lk(__mut_);
+    unique_lock<__spin_lock> __lk(__mut_);
 #ifndef _LIBCPP_NO_EXCEPTIONS
     if (__has_value())
         throw future_error(make_error_code(future_errc::promise_already_satisfied));
@@ -103,7 +103,7 @@
 void
 __assoc_sub_state::set_value_at_thread_exit()
 {
-    unique_lock<mutex> __lk(__mut_);
+    unique_lock<__spin_lock> __lk(__mut_);
 #ifndef _LIBCPP_NO_EXCEPTIONS
     if (__has_value())
         throw future_error(make_error_code(future_errc::promise_already_satisfied));
@@ -115,7 +115,7 @@
 void
 __assoc_sub_state::set_exception(exception_ptr __p)
 {
-    unique_lock<mutex> __lk(__mut_);
+    unique_lock<__spin_lock> __lk(__mut_);
 #ifndef _LIBCPP_NO_EXCEPTIONS
     if (__has_value())
         throw future_error(make_error_code(future_errc::promise_already_satisfied));
@@ -128,7 +128,7 @@
 void
 __assoc_sub_state::set_exception_at_thread_exit(exception_ptr __p)
 {
-    unique_lock<mutex> __lk(__mut_);
+    unique_lock<__spin_lock> __lk(__mut_);
 #ifndef _LIBCPP_NO_EXCEPTIONS
     if (__has_value())
         throw future_error(make_error_code(future_errc::promise_already_satisfied));
@@ -140,7 +140,7 @@
 void
 __assoc_sub_state::__make_ready()
 {
-    unique_lock<mutex> __lk(__mut_);
+    unique_lock<__spin_lock> __lk(__mut_);
     __state_ |= ready;
     __cv_.notify_all();
 }
@@ -148,7 +148,7 @@
 void
 __assoc_sub_state::copy()
 {
-    unique_lock<mutex> __lk(__mut_);
+    unique_lock<__spin_lock> __lk(__mut_);
     __sub_wait(__lk);
     if (__exception_ != nullptr)
         rethrow_exception(__exception_);
@@ -157,12 +157,12 @@
 void
 __assoc_sub_state::wait()
 {
-    unique_lock<mutex> __lk(__mut_);
+    unique_lock<__spin_lock> __lk(__mut_);
     __sub_wait(__lk);
 }
 
 void
-__assoc_sub_state::__sub_wait(unique_lock<mutex>& __lk)
+__assoc_sub_state::__sub_wait(unique_lock<__spin_lock>& __lk)
 {
     if (!__is_ready())
     {