Merge pull request #9215 from liamwhite/swordfight
Ensure correctness of atomic store ordering
This commit is contained in:
		| @@ -384,7 +384,8 @@ void KScheduler::SwitchThread(KThread* next_thread) { | |||||||
|  |  | ||||||
| void KScheduler::ScheduleImpl() { | void KScheduler::ScheduleImpl() { | ||||||
|     // First, clear the needs scheduling bool. |     // First, clear the needs scheduling bool. | ||||||
|     m_state.needs_scheduling.store(false, std::memory_order_seq_cst); |     m_state.needs_scheduling.store(false, std::memory_order_relaxed); | ||||||
|  |     std::atomic_thread_fence(std::memory_order_seq_cst); | ||||||
|  |  | ||||||
|     // Load the appropriate thread pointers for scheduling. |     // Load the appropriate thread pointers for scheduling. | ||||||
|     KThread* const cur_thread{GetCurrentThreadPointer(kernel)}; |     KThread* const cur_thread{GetCurrentThreadPointer(kernel)}; | ||||||
| @@ -400,7 +401,8 @@ void KScheduler::ScheduleImpl() { | |||||||
|     // If there aren't, we want to check if the highest priority thread is the same as the current |     // If there aren't, we want to check if the highest priority thread is the same as the current | ||||||
|     // thread. |     // thread. | ||||||
|     if (highest_priority_thread == cur_thread) { |     if (highest_priority_thread == cur_thread) { | ||||||
|         // If they're the same, then we can just return. |         // If they're the same, then we can just issue a memory barrier and return. | ||||||
|  |         std::atomic_thread_fence(std::memory_order_seq_cst); | ||||||
|         return; |         return; | ||||||
|     } |     } | ||||||
|  |  | ||||||
| @@ -476,7 +478,8 @@ void KScheduler::ScheduleImplFiber() { | |||||||
|  |  | ||||||
|         // We failed to successfully do the context switch, and need to retry. |         // We failed to successfully do the context switch, and need to retry. | ||||||
|         // Clear needs_scheduling. |         // Clear needs_scheduling. | ||||||
|         m_state.needs_scheduling.store(false, std::memory_order_seq_cst); |         m_state.needs_scheduling.store(false, std::memory_order_relaxed); | ||||||
|  |         std::atomic_thread_fence(std::memory_order_seq_cst); | ||||||
|  |  | ||||||
|         // Refresh the highest priority thread. |         // Refresh the highest priority thread. | ||||||
|         highest_priority_thread = m_state.highest_priority_thread; |         highest_priority_thread = m_state.highest_priority_thread; | ||||||
|   | |||||||
| @@ -60,6 +60,9 @@ public: | |||||||
|  |  | ||||||
|         // Release an instance of the lock. |         // Release an instance of the lock. | ||||||
|         if ((--lock_count) == 0) { |         if ((--lock_count) == 0) { | ||||||
|  |             // Perform a memory barrier here. | ||||||
|  |             std::atomic_thread_fence(std::memory_order_seq_cst); | ||||||
|  |  | ||||||
|             // We're no longer going to hold the lock. Take note of what cores need scheduling. |             // We're no longer going to hold the lock. Take note of what cores need scheduling. | ||||||
|             const u64 cores_needing_scheduling = |             const u64 cores_needing_scheduling = | ||||||
|                 SchedulerType::UpdateHighestPriorityThreads(kernel); |                 SchedulerType::UpdateHighestPriorityThreads(kernel); | ||||||
|   | |||||||
		Reference in New Issue
	
	Block a user