[K/N][Runtime] Use std::atomic_flag in SpinLock

kotlin-native/runtime/src/main/cpp/Mutex.hpp

@@ -17,7 +17,7 @@
 #ifndef RUNTIME_MUTEX_H
 #define RUNTIME_MUTEX_H
 
-#include <cstdint>
+#include <atomic>
 #include <thread>
 
 #include "KAssert.h"
@@ -37,20 +37,17 @@ template <>
 class SpinLock<MutexThreadStateHandling::kIgnore> : private Pinned {
 public:
     void lock() noexcept {
-        while (!__sync_bool_compare_and_swap(&atomicInt, 0, 1)) {
+        while(flag_.test_and_set(std::memory_order_acquire)) {
             std::this_thread::yield();
         }
     }
 
     void unlock() noexcept {
-        if (!__sync_bool_compare_and_swap(&atomicInt, 1, 0)) {
-            RuntimeAssert(false, "Unable to unlock");
-        }
+        flag_.clear(std::memory_order_release);
     }
 
 private:
-    // TODO: Consider using `std::atomic_flag`.
-    int32_t atomicInt = 0;
+    std::atomic_flag flag_ = ATOMIC_FLAG_INIT;
 };
 
 template <>
@@ -58,25 +55,22 @@ class SpinLock<MutexThreadStateHandling::kSwitchIfRegistered> : private Pinned {
 public:
     void lock() noexcept {
         // Fast path without thread state switching.
-        if (__sync_bool_compare_and_swap(&atomicInt, 0, 1)) {
+        if (!flag_.test_and_set(std::memory_order_acquire)) {
             return;
         }
 
         kotlin::NativeOrUnregisteredThreadGuard guard(/* reentrant = */ true);
-        while (!__sync_bool_compare_and_swap(&atomicInt, 0, 1)) {
+        while (flag_.test_and_set(std::memory_order_acquire)) {
             std::this_thread::yield();
         }
     }
 
     void unlock() noexcept {
-        if (!__sync_bool_compare_and_swap(&atomicInt, 1, 0)) {
-            RuntimeAssert(false, "Unable to unlock");
-        }
+        flag_.clear(std::memory_order_release);
     }
 
 private:
-    // TODO: Consider using `std::atomic_flag`.
-    int32_t atomicInt = 0;
+    std::atomic_flag flag_ = ATOMIC_FLAG_INIT;
 };
 
 } // namespace kotlin