diff --git a/README.md b/README.md
index 313510caa..475a2f7cb 100755
--- a/README.md
+++ b/README.md
@@ -1,7 +1,7 @@
 yuzu emulator early access
 =============
 
-This is the source code for early-access 1269.
+This is the source code for early-access 1272.
 
 ## Legal Notice
 
diff --git a/src/common/CMakeLists.txt b/src/common/CMakeLists.txt
index 5c8003eb1..64b8cf883 100755
--- a/src/common/CMakeLists.txt
+++ b/src/common/CMakeLists.txt
@@ -123,6 +123,7 @@ add_library(common STATIC
     hash.h
     hex_util.cpp
     hex_util.h
+    intrusive_red_black_tree.h
     logging/backend.cpp
     logging/backend.h
     logging/filter.cpp
@@ -145,6 +146,7 @@ add_library(common STATIC
     page_table.h
     param_package.cpp
     param_package.h
+    parent_of_member.h
     quaternion.h
     ring_buffer.h
     scm_rev.cpp
@@ -169,6 +171,7 @@ add_library(common STATIC
     time_zone.h
     timer.cpp
     timer.h
+    tree.h
     uint128.cpp
     uint128.h
     uuid.cpp
diff --git a/src/common/common_funcs.h b/src/common/common_funcs.h
index 367b6bf6e..c90978f9c 100755
--- a/src/common/common_funcs.h
+++ b/src/common/common_funcs.h
@@ -93,6 +93,14 @@ __declspec(dllimport) void __stdcall DebugBreak(void);
         return static_cast<T>(key) == 0;                                                           \
     }
 
+/// Evaluates a boolean expression, and returns a result unless that expression is true.
+#define R_UNLESS(expr, res)                                                                        \
+    {                                                                                              \
+        if (!(expr)) {                                                                             \
+            return res;                                                                            \
+        }                                                                                          \
+    }
+
 namespace Common {
 
 [[nodiscard]] constexpr u32 MakeMagic(char a, char b, char c, char d) {
diff --git a/src/core/CMakeLists.txt b/src/core/CMakeLists.txt
index 4b08adb80..fbca16015 100755
--- a/src/core/CMakeLists.txt
+++ b/src/core/CMakeLists.txt
@@ -141,8 +141,6 @@ add_library(core STATIC
     hardware_interrupt_manager.h
     hle/ipc.h
     hle/ipc_helpers.h
-    hle/kernel/address_arbiter.cpp
-    hle/kernel/address_arbiter.h
     hle/kernel/client_port.cpp
     hle/kernel/client_port.h
     hle/kernel/client_session.cpp
@@ -156,13 +154,19 @@ add_library(core STATIC
     hle/kernel/handle_table.h
     hle/kernel/hle_ipc.cpp
     hle/kernel/hle_ipc.h
+    hle/kernel/k_address_arbiter.cpp
+    hle/kernel/k_address_arbiter.h
     hle/kernel/k_affinity_mask.h
+    hle/kernel/k_condition_variable.cpp
+    hle/kernel/k_condition_variable.h
     hle/kernel/k_priority_queue.h
     hle/kernel/k_scheduler.cpp
     hle/kernel/k_scheduler.h
     hle/kernel/k_scheduler_lock.h
     hle/kernel/k_scoped_lock.h
     hle/kernel/k_scoped_scheduler_lock_and_sleep.h
+    hle/kernel/k_synchronization_object.cpp
+    hle/kernel/k_synchronization_object.h
     hle/kernel/kernel.cpp
     hle/kernel/kernel.h
     hle/kernel/memory/address_space_info.cpp
@@ -182,8 +186,6 @@ add_library(core STATIC
     hle/kernel/memory/slab_heap.h
     hle/kernel/memory/system_control.cpp
     hle/kernel/memory/system_control.h
-    hle/kernel/mutex.cpp
-    hle/kernel/mutex.h
     hle/kernel/object.cpp
     hle/kernel/object.h
     hle/kernel/physical_core.cpp
@@ -209,12 +211,10 @@ add_library(core STATIC
     hle/kernel/shared_memory.h
     hle/kernel/svc.cpp
     hle/kernel/svc.h
+    hle/kernel/svc_common.h
+    hle/kernel/svc_results.h
     hle/kernel/svc_types.h
     hle/kernel/svc_wrap.h
-    hle/kernel/synchronization_object.cpp
-    hle/kernel/synchronization_object.h
-    hle/kernel/synchronization.cpp
-    hle/kernel/synchronization.h
     hle/kernel/thread.cpp
     hle/kernel/thread.h
     hle/kernel/time_manager.cpp
diff --git a/src/core/core_timing.cpp b/src/core/core_timing.cpp
index e6c8461a5..874b5673a 100755
--- a/src/core/core_timing.cpp
+++ b/src/core/core_timing.cpp
@@ -49,6 +49,7 @@ void CoreTiming::ThreadEntry(CoreTiming& instance) {
     Common::SetCurrentThreadPriority(Common::ThreadPriority::VeryHigh);
     instance.on_thread_init();
     instance.ThreadLoop();
+    MicroProfileOnThreadExit();
 }
 
 void CoreTiming::Initialize(std::function<void()>&& on_thread_init_) {
diff --git a/src/core/hle/kernel/client_port.cpp b/src/core/hle/kernel/client_port.cpp
index 8aff2227a..f8f005f15 100755
--- a/src/core/hle/kernel/client_port.cpp
+++ b/src/core/hle/kernel/client_port.cpp
@@ -33,9 +33,6 @@ ResultVal<std::shared_ptr<ClientSession>> ClientPort::Connect() {
         server_port->AppendPendingSession(std::move(server));
     }
 
-    // Wake the threads waiting on the ServerPort
-    server_port->Signal();
-
     return MakeResult(std::move(client));
 }
 
diff --git a/src/core/hle/kernel/client_session.cpp b/src/core/hle/kernel/client_session.cpp
index be9eba519..e8e52900d 100755
--- a/src/core/hle/kernel/client_session.cpp
+++ b/src/core/hle/kernel/client_session.cpp
@@ -12,7 +12,7 @@
 
 namespace Kernel {
 
-ClientSession::ClientSession(KernelCore& kernel) : SynchronizationObject{kernel} {}
+ClientSession::ClientSession(KernelCore& kernel) : KSynchronizationObject{kernel} {}
 
 ClientSession::~ClientSession() {
     // This destructor will be called automatically when the last ClientSession handle is closed by
@@ -22,15 +22,6 @@ ClientSession::~ClientSession() {
     }
 }
 
-bool ClientSession::ShouldWait(const Thread* thread) const {
-    UNIMPLEMENTED();
-    return {};
-}
-
-void ClientSession::Acquire(Thread* thread) {
-    UNIMPLEMENTED();
-}
-
 bool ClientSession::IsSignaled() const {
     UNIMPLEMENTED();
     return true;
diff --git a/src/core/hle/kernel/client_session.h b/src/core/hle/kernel/client_session.h
index e5e0690c2..d5c9ebee8 100755
--- a/src/core/hle/kernel/client_session.h
+++ b/src/core/hle/kernel/client_session.h
@@ -7,7 +7,7 @@
 #include <memory>
 #include <string>
 
-#include "core/hle/kernel/synchronization_object.h"
+#include "core/hle/kernel/k_synchronization_object.h"
 #include "core/hle/result.h"
 
 union ResultCode;
@@ -26,7 +26,7 @@ class KernelCore;
 class Session;
 class Thread;
 
-class ClientSession final : public SynchronizationObject {
+class ClientSession final : public KSynchronizationObject {
 public:
     explicit ClientSession(KernelCore& kernel);
     ~ClientSession() override;
@@ -49,10 +49,6 @@ public:
     ResultCode SendSyncRequest(std::shared_ptr<Thread> thread, Core::Memory::Memory& memory,
                                Core::Timing::CoreTiming& core_timing);
 
-    bool ShouldWait(const Thread* thread) const override;
-
-    void Acquire(Thread* thread) override;
-
     bool IsSignaled() const override;
 
 private:
diff --git a/src/core/hle/kernel/errors.h b/src/core/hle/kernel/errors.h
index d4e5d88cf..7d32a39f0 100755
--- a/src/core/hle/kernel/errors.h
+++ b/src/core/hle/kernel/errors.h
@@ -13,12 +13,14 @@ namespace Kernel {
 constexpr ResultCode ERR_MAX_CONNECTIONS_REACHED{ErrorModule::Kernel, 7};
 constexpr ResultCode ERR_INVALID_CAPABILITY_DESCRIPTOR{ErrorModule::Kernel, 14};
 constexpr ResultCode ERR_THREAD_TERMINATING{ErrorModule::Kernel, 59};
+constexpr ResultCode ERR_TERMINATION_REQUESTED{ErrorModule::Kernel, 59};
 constexpr ResultCode ERR_INVALID_SIZE{ErrorModule::Kernel, 101};
 constexpr ResultCode ERR_INVALID_ADDRESS{ErrorModule::Kernel, 102};
 constexpr ResultCode ERR_OUT_OF_RESOURCES{ErrorModule::Kernel, 103};
 constexpr ResultCode ERR_OUT_OF_MEMORY{ErrorModule::Kernel, 104};
 constexpr ResultCode ERR_HANDLE_TABLE_FULL{ErrorModule::Kernel, 105};
 constexpr ResultCode ERR_INVALID_ADDRESS_STATE{ErrorModule::Kernel, 106};
+constexpr ResultCode ERR_INVALID_CURRENT_MEMORY{ErrorModule::Kernel, 106};
 constexpr ResultCode ERR_INVALID_MEMORY_PERMISSIONS{ErrorModule::Kernel, 108};
 constexpr ResultCode ERR_INVALID_MEMORY_RANGE{ErrorModule::Kernel, 110};
 constexpr ResultCode ERR_INVALID_PROCESSOR_ID{ErrorModule::Kernel, 113};
@@ -28,6 +30,7 @@ constexpr ResultCode ERR_INVALID_POINTER{ErrorModule::Kernel, 115};
 constexpr ResultCode ERR_INVALID_COMBINATION{ErrorModule::Kernel, 116};
 constexpr ResultCode RESULT_TIMEOUT{ErrorModule::Kernel, 117};
 constexpr ResultCode ERR_SYNCHRONIZATION_CANCELED{ErrorModule::Kernel, 118};
+constexpr ResultCode ERR_CANCELLED{ErrorModule::Kernel, 118};
 constexpr ResultCode ERR_OUT_OF_RANGE{ErrorModule::Kernel, 119};
 constexpr ResultCode ERR_INVALID_ENUM_VALUE{ErrorModule::Kernel, 120};
 constexpr ResultCode ERR_NOT_FOUND{ErrorModule::Kernel, 121};
diff --git a/src/core/hle/kernel/k_scheduler.cpp b/src/core/hle/kernel/k_scheduler.cpp
index c5fd82a6b..42f0ea483 100755
--- a/src/core/hle/kernel/k_scheduler.cpp
+++ b/src/core/hle/kernel/k_scheduler.cpp
@@ -180,22 +180,22 @@ u64 KScheduler::UpdateHighestPriorityThreadsImpl(KernelCore& kernel) {
     return cores_needing_scheduling;
 }
 
-void KScheduler::OnThreadStateChanged(KernelCore& kernel, Thread* thread, u32 old_state) {
+void KScheduler::OnThreadStateChanged(KernelCore& kernel, Thread* thread, ThreadState old_state) {
     ASSERT(kernel.GlobalSchedulerContext().IsLocked());
 
     // Check if the state has changed, because if it hasn't there's nothing to do.
-    const auto cur_state = thread->scheduling_state;
+    const auto cur_state = thread->GetRawState();
     if (cur_state == old_state) {
         return;
     }
 
     // Update the priority queues.
-    if (old_state == static_cast<u32>(ThreadSchedStatus::Runnable)) {
+    if (old_state == ThreadState::Runnable) {
         // If we were previously runnable, then we're not runnable now, and we should remove.
         GetPriorityQueue(kernel).Remove(thread);
         IncrementScheduledCount(thread);
         SetSchedulerUpdateNeeded(kernel);
-    } else if (cur_state == static_cast<u32>(ThreadSchedStatus::Runnable)) {
+    } else if (cur_state == ThreadState::Runnable) {
         // If we're now runnable, then we weren't previously, and we should add.
         GetPriorityQueue(kernel).PushBack(thread);
         IncrementScheduledCount(thread);
@@ -203,13 +203,11 @@ void KScheduler::OnThreadStateChanged(KernelCore& kernel, Thread* thread, u32 ol
     }
 }
 
-void KScheduler::OnThreadPriorityChanged(KernelCore& kernel, Thread* thread, Thread* current_thread,
-                                         u32 old_priority) {
-
+void KScheduler::OnThreadPriorityChanged(KernelCore& kernel, Thread* thread, s32 old_priority) {
     ASSERT(kernel.GlobalSchedulerContext().IsLocked());
 
     // If the thread is runnable, we want to change its priority in the queue.
-    if (thread->scheduling_state == static_cast<u32>(ThreadSchedStatus::Runnable)) {
+    if (thread->GetRawState() == ThreadState::Runnable) {
         GetPriorityQueue(kernel).ChangePriority(
             old_priority, thread == kernel.CurrentScheduler()->GetCurrentThread(), thread);
         IncrementScheduledCount(thread);
@@ -222,7 +220,7 @@ void KScheduler::OnThreadAffinityMaskChanged(KernelCore& kernel, Thread* thread,
     ASSERT(kernel.GlobalSchedulerContext().IsLocked());
 
     // If the thread is runnable, we want to change its affinity in the queue.
-    if (thread->scheduling_state == static_cast<u32>(ThreadSchedStatus::Runnable)) {
+    if (thread->GetRawState() == ThreadState::Runnable) {
         GetPriorityQueue(kernel).ChangeAffinityMask(old_core, old_affinity, thread);
         IncrementScheduledCount(thread);
         SetSchedulerUpdateNeeded(kernel);
@@ -292,7 +290,7 @@ void KScheduler::RotateScheduledQueue(s32 core_id, s32 priority) {
 
         // If the best thread we can choose has a priority the same or worse than ours, try to
         // migrate a higher priority thread.
-        if (best_thread != nullptr && best_thread->GetPriority() >= static_cast<u32>(priority)) {
+        if (best_thread != nullptr && best_thread->GetPriority() >= priority) {
             Thread* suggested = priority_queue.GetSuggestedFront(core_id);
             while (suggested != nullptr) {
                 // If the suggestion's priority is the same as ours, don't bother.
@@ -395,8 +393,8 @@ void KScheduler::YieldWithoutCoreMigration() {
     {
         KScopedSchedulerLock lock(kernel);
 
-        const auto cur_state = cur_thread.scheduling_state;
-        if (cur_state == static_cast<u32>(ThreadSchedStatus::Runnable)) {
+        const auto cur_state = cur_thread.GetRawState();
+        if (cur_state == ThreadState::Runnable) {
             // Put the current thread at the back of the queue.
             Thread* next_thread = priority_queue.MoveToScheduledBack(std::addressof(cur_thread));
             IncrementScheduledCount(std::addressof(cur_thread));
@@ -436,8 +434,8 @@ void KScheduler::YieldWithCoreMigration() {
     {
         KScopedSchedulerLock lock(kernel);
 
-        const auto cur_state = cur_thread.scheduling_state;
-        if (cur_state == static_cast<u32>(ThreadSchedStatus::Runnable)) {
+        const auto cur_state = cur_thread.GetRawState();
+        if (cur_state == ThreadState::Runnable) {
             // Get the current active core.
             const s32 core_id = cur_thread.GetActiveCore();
 
@@ -526,8 +524,8 @@ void KScheduler::YieldToAnyThread() {
     {
         KScopedSchedulerLock lock(kernel);
 
-        const auto cur_state = cur_thread.scheduling_state;
-        if (cur_state == static_cast<u32>(ThreadSchedStatus::Runnable)) {
+        const auto cur_state = cur_thread.GetRawState();
+        if (cur_state == ThreadState::Runnable) {
             // Get the current active core.
             const s32 core_id = cur_thread.GetActiveCore();
 
@@ -645,8 +643,7 @@ void KScheduler::Unload(Thread* thread) {
 
 void KScheduler::Reload(Thread* thread) {
     if (thread) {
-        ASSERT_MSG(thread->GetSchedulingStatus() == ThreadSchedStatus::Runnable,
-                   "Thread must be runnable.");
+        ASSERT_MSG(thread->GetState() == ThreadState::Runnable, "Thread must be runnable.");
 
         // Cancel any outstanding wakeup events for this thread
         thread->SetIsRunning(true);
@@ -725,7 +722,7 @@ void KScheduler::SwitchToCurrent() {
         do {
             if (current_thread != nullptr && !current_thread->IsHLEThread()) {
                 current_thread->context_guard.lock();
-                if (!current_thread->IsRunnable()) {
+                if (current_thread->GetRawState() != ThreadState::Runnable) {
                     current_thread->context_guard.unlock();
                     break;
                 }
@@ -772,7 +769,7 @@ void KScheduler::Initialize() {
 
     {
         KScopedSchedulerLock lock{system.Kernel()};
-        idle_thread->SetStatus(ThreadStatus::Ready);
+        idle_thread->SetState(ThreadState::Runnable);
     }
 }
 
diff --git a/src/core/hle/kernel/k_scheduler.h b/src/core/hle/kernel/k_scheduler.h
index e84abc84c..783665123 100755
--- a/src/core/hle/kernel/k_scheduler.h
+++ b/src/core/hle/kernel/k_scheduler.h
@@ -100,11 +100,10 @@ public:
     void YieldToAnyThread();
 
     /// Notify the scheduler a thread's status has changed.
-    static void OnThreadStateChanged(KernelCore& kernel, Thread* thread, u32 old_state);
+    static void OnThreadStateChanged(KernelCore& kernel, Thread* thread, ThreadState old_state);
 
     /// Notify the scheduler a thread's priority has changed.
-    static void OnThreadPriorityChanged(KernelCore& kernel, Thread* thread, Thread* current_thread,
-                                        u32 old_priority);
+    static void OnThreadPriorityChanged(KernelCore& kernel, Thread* thread, s32 old_priority);
 
     /// Notify the scheduler a thread's core and/or affinity mask has changed.
     static void OnThreadAffinityMaskChanged(KernelCore& kernel, Thread* thread,
diff --git a/src/core/hle/kernel/kernel.cpp b/src/core/hle/kernel/kernel.cpp
index e8ece8164..8d03f16fb 100755
--- a/src/core/hle/kernel/kernel.cpp
+++ b/src/core/hle/kernel/kernel.cpp
@@ -38,7 +38,6 @@
 #include "core/hle/kernel/resource_limit.h"
 #include "core/hle/kernel/service_thread.h"
 #include "core/hle/kernel/shared_memory.h"
-#include "core/hle/kernel/synchronization.h"
 #include "core/hle/kernel/thread.h"
 #include "core/hle/kernel/time_manager.h"
 #include "core/hle/lock.h"
@@ -51,8 +50,7 @@ namespace Kernel {
 
 struct KernelCore::Impl {
     explicit Impl(Core::System& system, KernelCore& kernel)
-        : synchronization{system}, time_manager{system}, global_handle_table{kernel}, system{
-                                                                                          system} {}
+        : time_manager{system}, global_handle_table{kernel}, system{system} {}
 
     void SetMulticore(bool is_multicore) {
         this->is_multicore = is_multicore;
@@ -307,7 +305,6 @@ struct KernelCore::Impl {
     std::vector<std::shared_ptr<Process>> process_list;
     Process* current_process = nullptr;
     std::unique_ptr<Kernel::GlobalSchedulerContext> global_scheduler_context;
-    Kernel::Synchronization synchronization;
     Kernel::TimeManager time_manager;
 
     std::shared_ptr<ResourceLimit> system_resource_limit;
@@ -461,14 +458,6 @@ const std::array<Core::CPUInterruptHandler, Core::Hardware::NUM_CPU_CORES>& Kern
     return impl->interrupts;
 }
 
-Kernel::Synchronization& KernelCore::Synchronization() {
-    return impl->synchronization;
-}
-
-const Kernel::Synchronization& KernelCore::Synchronization() const {
-    return impl->synchronization;
-}
-
 Kernel::TimeManager& KernelCore::TimeManager() {
     return impl->time_manager;
 }
@@ -613,9 +602,9 @@ void KernelCore::Suspend(bool in_suspention) {
     const bool should_suspend = exception_exited || in_suspention;
     {
         KScopedSchedulerLock lock(*this);
-        ThreadStatus status = should_suspend ? ThreadStatus::Ready : ThreadStatus::WaitSleep;
+        const auto state = should_suspend ? ThreadState::Runnable : ThreadState::Waiting;
         for (std::size_t i = 0; i < Core::Hardware::NUM_CPU_CORES; i++) {
-            impl->suspend_threads[i]->SetStatus(status);
+            impl->suspend_threads[i]->SetState(state);
         }
     }
 }
diff --git a/src/core/hle/kernel/kernel.h b/src/core/hle/kernel/kernel.h
index e3169f5a7..933d9a7d6 100755
--- a/src/core/hle/kernel/kernel.h
+++ b/src/core/hle/kernel/kernel.h
@@ -33,7 +33,6 @@ template <typename T>
 class SlabHeap;
 } // namespace Memory
 
-class AddressArbiter;
 class ClientPort;
 class GlobalSchedulerContext;
 class HandleTable;
@@ -129,12 +128,6 @@ public:
     /// Gets the an instance of the current physical CPU core.
     const Kernel::PhysicalCore& CurrentPhysicalCore() const;
 
-    /// Gets the an instance of the Synchronization Interface.
-    Kernel::Synchronization& Synchronization();
-
-    /// Gets the an instance of the Synchronization Interface.
-    const Kernel::Synchronization& Synchronization() const;
-
     /// Gets the an instance of the TimeManager Interface.
     Kernel::TimeManager& TimeManager();
 
diff --git a/src/core/hle/kernel/memory/memory_layout.h b/src/core/hle/kernel/memory/memory_layout.h
index 9b3d6267a..c7c0b2f49 100755
--- a/src/core/hle/kernel/memory/memory_layout.h
+++ b/src/core/hle/kernel/memory/memory_layout.h
@@ -5,9 +5,28 @@
 #pragma once
 
 #include "common/common_types.h"
+#include "core/device_memory.h"
 
 namespace Kernel::Memory {
 
+constexpr std::size_t KernelAslrAlignment = 2 * 1024 * 1024;
+constexpr std::size_t KernelVirtualAddressSpaceWidth = 1ULL << 39;
+constexpr std::size_t KernelPhysicalAddressSpaceWidth = 1ULL << 48;
+constexpr std::size_t KernelVirtualAddressSpaceBase = 0ULL - KernelVirtualAddressSpaceWidth;
+constexpr std::size_t KernelVirtualAddressSpaceEnd =
+    KernelVirtualAddressSpaceBase + (KernelVirtualAddressSpaceWidth - KernelAslrAlignment);
+constexpr std::size_t KernelVirtualAddressSpaceLast = KernelVirtualAddressSpaceEnd - 1;
+constexpr std::size_t KernelVirtualAddressSpaceSize =
+    KernelVirtualAddressSpaceEnd - KernelVirtualAddressSpaceBase;
+
+constexpr bool IsKernelAddressKey(VAddr key) {
+    return KernelVirtualAddressSpaceBase <= key && key <= KernelVirtualAddressSpaceLast;
+}
+
+constexpr bool IsKernelAddress(VAddr address) {
+    return KernelVirtualAddressSpaceBase <= address && address < KernelVirtualAddressSpaceEnd;
+}
+
 class MemoryRegion final {
     friend class MemoryLayout;
 
diff --git a/src/core/hle/kernel/object.h b/src/core/hle/kernel/object.h
index e3391e2af..27124ef67 100755
--- a/src/core/hle/kernel/object.h
+++ b/src/core/hle/kernel/object.h
@@ -50,6 +50,11 @@ public:
     }
     virtual HandleType GetHandleType() const = 0;
 
+    void Close() {
+        // TODO(bunnei): This is a placeholder to decrement the reference count, which we will use
+        // when we implement KAutoObject instead of using shared_ptr.
+    }
+
     /**
      * Check if a thread can wait on the object
      * @return True if a thread can wait on the object, otherwise false
diff --git a/src/core/hle/kernel/process.cpp b/src/core/hle/kernel/process.cpp
index b905b486a..37b77fa6e 100755
--- a/src/core/hle/kernel/process.cpp
+++ b/src/core/hle/kernel/process.cpp
@@ -55,7 +55,7 @@ void SetupMainThread(Core::System& system, Process& owner_process, u32 priority,
     // Threads by default are dormant, wake up the main thread so it runs when the scheduler fires
     {
         KScopedSchedulerLock lock{kernel};
-        thread->SetStatus(ThreadStatus::Ready);
+        thread->SetState(ThreadState::Runnable);
     }
 }
 } // Anonymous namespace
@@ -162,48 +162,6 @@ u64 Process::GetTotalPhysicalMemoryUsedWithoutSystemResource() const {
     return GetTotalPhysicalMemoryUsed() - GetSystemResourceUsage();
 }
 
-void Process::InsertConditionVariableThread(std::shared_ptr<Thread> thread) {
-    VAddr cond_var_addr = thread->GetCondVarWaitAddress();
-    std::list<std::shared_ptr<Thread>>& thread_list = cond_var_threads[cond_var_addr];
-    auto it = thread_list.begin();
-    while (it != thread_list.end()) {
-        const std::shared_ptr<Thread> current_thread = *it;
-        if (current_thread->GetPriority() > thread->GetPriority()) {
-            thread_list.insert(it, thread);
-            return;
-        }
-        ++it;
-    }
-    thread_list.push_back(thread);
-}
-
-void Process::RemoveConditionVariableThread(std::shared_ptr<Thread> thread) {
-    VAddr cond_var_addr = thread->GetCondVarWaitAddress();
-    std::list<std::shared_ptr<Thread>>& thread_list = cond_var_threads[cond_var_addr];
-    auto it = thread_list.begin();
-    while (it != thread_list.end()) {
-        const std::shared_ptr<Thread> current_thread = *it;
-        if (current_thread.get() == thread.get()) {
-            thread_list.erase(it);
-            return;
-        }
-        ++it;
-    }
-}
-
-std::vector<std::shared_ptr<Thread>> Process::GetConditionVariableThreads(
-    const VAddr cond_var_addr) {
-    std::vector<std::shared_ptr<Thread>> result{};
-    std::list<std::shared_ptr<Thread>>& thread_list = cond_var_threads[cond_var_addr];
-    auto it = thread_list.begin();
-    while (it != thread_list.end()) {
-        std::shared_ptr<Thread> current_thread = *it;
-        result.push_back(current_thread);
-        ++it;
-    }
-    return result;
-}
-
 void Process::RegisterThread(const Thread* thread) {
     thread_list.push_back(thread);
 }
@@ -318,7 +276,7 @@ void Process::PrepareForTermination() {
                 continue;
 
             // TODO(Subv): When are the other running/ready threads terminated?
-            ASSERT_MSG(thread->GetStatus() == ThreadStatus::WaitSynch,
+            ASSERT_MSG(thread->GetState() == ThreadState::Waiting,
                        "Exiting processes with non-waiting threads is currently unimplemented");
 
             thread->Stop();
@@ -406,21 +364,18 @@ void Process::LoadModule(CodeSet code_set, VAddr base_addr) {
     ReprotectSegment(code_set.DataSegment(), Memory::MemoryPermission::ReadAndWrite);
 }
 
+bool Process::IsSignaled() const {
+    ASSERT(kernel.GlobalSchedulerContext().IsLocked());
+    return is_signaled;
+}
+
 Process::Process(Core::System& system)
-    : SynchronizationObject{system.Kernel()}, page_table{std::make_unique<Memory::PageTable>(
-                                                  system)},
-      handle_table{system.Kernel()}, address_arbiter{system}, mutex{system}, system{system} {}
+    : KSynchronizationObject{system.Kernel()},
+      page_table{std::make_unique<Memory::PageTable>(system)}, handle_table{system.Kernel()},
+      address_arbiter{system}, condition_var{system}, system{system} {}
 
 Process::~Process() = default;
 
-void Process::Acquire(Thread* thread) {
-    ASSERT_MSG(!ShouldWait(thread), "Object unavailable!");
-}
-
-bool Process::ShouldWait(const Thread* thread) const {
-    return !is_signaled;
-}
-
 void Process::ChangeStatus(ProcessStatus new_status) {
     if (status == new_status) {
         return;
@@ -428,7 +383,7 @@ void Process::ChangeStatus(ProcessStatus new_status) {
 
     status = new_status;
     is_signaled = true;
-    Signal();
+    NotifyAvailable();
 }
 
 ResultCode Process::AllocateMainThreadStack(std::size_t stack_size) {
diff --git a/src/core/hle/kernel/process.h b/src/core/hle/kernel/process.h
index e412e58aa..21b6f9bd9 100755
--- a/src/core/hle/kernel/process.h
+++ b/src/core/hle/kernel/process.h
@@ -11,11 +11,11 @@
 #include <unordered_map>
 #include <vector>
 #include "common/common_types.h"
-#include "core/hle/kernel/address_arbiter.h"
 #include "core/hle/kernel/handle_table.h"
-#include "core/hle/kernel/mutex.h"
+#include "core/hle/kernel/k_address_arbiter.h"
+#include "core/hle/kernel/k_condition_variable.h"
+#include "core/hle/kernel/k_synchronization_object.h"
 #include "core/hle/kernel/process_capability.h"
-#include "core/hle/kernel/synchronization_object.h"
 #include "core/hle/result.h"
 
 namespace Core {
@@ -63,7 +63,7 @@ enum class ProcessStatus {
     DebugBreak,
 };
 
-class Process final : public SynchronizationObject {
+class Process final : public KSynchronizationObject {
 public:
     explicit Process(Core::System& system);
     ~Process() override;
@@ -123,24 +123,30 @@ public:
         return handle_table;
     }
 
-    /// Gets a reference to the process' address arbiter.
-    AddressArbiter& GetAddressArbiter() {
-        return address_arbiter;
+    ResultCode SignalToAddress(VAddr address) {
+        return condition_var.SignalToAddress(address);
     }
 
-    /// Gets a const reference to the process' address arbiter.
-    const AddressArbiter& GetAddressArbiter() const {
-        return address_arbiter;
+    ResultCode WaitForAddress(Handle handle, VAddr address, u32 tag) {
+        return condition_var.WaitForAddress(handle, address, tag);
     }
 
-    /// Gets a reference to the process' mutex lock.
-    Mutex& GetMutex() {
-        return mutex;
+    void SignalConditionVariable(u64 cv_key, int32_t count) {
+        return condition_var.Signal(cv_key, count);
     }
 
-    /// Gets a const reference to the process' mutex lock
-    const Mutex& GetMutex() const {
-        return mutex;
+    ResultCode WaitConditionVariable(VAddr address, u64 cv_key, u32 tag, s64 ns) {
+        return condition_var.Wait(address, cv_key, tag, ns);
+    }
+
+    ResultCode SignalAddressArbiter(VAddr address, Svc::SignalType signal_type, s32 value,
+                                    s32 count) {
+        return address_arbiter.SignalToAddress(address, signal_type, value, count);
+    }
+
+    ResultCode WaitAddressArbiter(VAddr address, Svc::ArbitrationType arb_type, s32 value,
+                                  s64 timeout) {
+        return address_arbiter.WaitForAddress(address, arb_type, value, timeout);
     }
 
     /// Gets the address to the process' dedicated TLS region.
@@ -250,15 +256,6 @@ public:
         return thread_list;
     }
 
-    /// Insert a thread into the condition variable wait container
-    void InsertConditionVariableThread(std::shared_ptr<Thread> thread);
-
-    /// Remove a thread from the condition variable wait container
-    void RemoveConditionVariableThread(std::shared_ptr<Thread> thread);
-
-    /// Obtain all condition variable threads waiting for some address
-    std::vector<std::shared_ptr<Thread>> GetConditionVariableThreads(VAddr cond_var_addr);
-
     /// Registers a thread as being created under this process,
     /// adding it to this process' thread list.
     void RegisterThread(const Thread* thread);
@@ -304,6 +301,8 @@ public:
 
     void LoadModule(CodeSet code_set, VAddr base_addr);
 
+    virtual bool IsSignaled() const override;
+
     ///////////////////////////////////////////////////////////////////////////////////////////////
     // Thread-local storage management
 
@@ -314,12 +313,6 @@ public:
     void FreeTLSRegion(VAddr tls_address);
 
 private:
-    /// Checks if the specified thread should wait until this process is available.
-    bool ShouldWait(const Thread* thread) const override;
-
-    /// Acquires/locks this process for the specified thread if it's available.
-    void Acquire(Thread* thread) override;
-
     /// Changes the process status. If the status is different
     /// from the current process status, then this will trigger
     /// a process signal.
@@ -373,12 +366,12 @@ private:
     HandleTable handle_table;
 
     /// Per-process address arbiter.
-    AddressArbiter address_arbiter;
+    KAddressArbiter address_arbiter;
 
     /// The per-process mutex lock instance used for handling various
     /// forms of services, such as lock arbitration, and condition
     /// variable related facilities.
-    Mutex mutex;
+    KConditionVariable condition_var;
 
     /// Address indicating the location of the process' dedicated TLS region.
     VAddr tls_region_address = 0;
@@ -389,9 +382,6 @@ private:
     /// List of threads that are running with this process as their owner.
     std::list<const Thread*> thread_list;
 
-    /// List of threads waiting for a condition variable
-    std::unordered_map<VAddr, std::list<std::shared_ptr<Thread>>> cond_var_threads;
-
     /// Address of the top of the main thread's stack
     VAddr main_thread_stack_top{};
 
@@ -410,6 +400,8 @@ private:
     /// Schedule count of this process
     s64 schedule_count{};
 
+    bool is_signaled{};
+
     /// System context
     Core::System& system;
 };
diff --git a/src/core/hle/kernel/readable_event.cpp b/src/core/hle/kernel/readable_event.cpp
index cea262ce0..99ed0857e 100755
--- a/src/core/hle/kernel/readable_event.cpp
+++ b/src/core/hle/kernel/readable_event.cpp
@@ -14,24 +14,22 @@
 
 namespace Kernel {
 
-ReadableEvent::ReadableEvent(KernelCore& kernel) : SynchronizationObject{kernel} {}
+ReadableEvent::ReadableEvent(KernelCore& kernel) : KSynchronizationObject{kernel} {}
 ReadableEvent::~ReadableEvent() = default;
 
-bool ReadableEvent::ShouldWait(const Thread* thread) const {
-    return !is_signaled;
-}
-
-void ReadableEvent::Acquire(Thread* thread) {
-    ASSERT_MSG(IsSignaled(), "object unavailable!");
-}
-
 void ReadableEvent::Signal() {
     if (is_signaled) {
         return;
     }
 
     is_signaled = true;
-    SynchronizationObject::Signal();
+    NotifyAvailable();
+}
+
+bool ReadableEvent::IsSignaled() const {
+    ASSERT(kernel.GlobalSchedulerContext().IsLocked());
+
+    return is_signaled;
 }
 
 void ReadableEvent::Clear() {
diff --git a/src/core/hle/kernel/readable_event.h b/src/core/hle/kernel/readable_event.h
index 3264dd066..95894db8d 100755
--- a/src/core/hle/kernel/readable_event.h
+++ b/src/core/hle/kernel/readable_event.h
@@ -4,8 +4,8 @@
 
 #pragma once
 
+#include "core/hle/kernel/k_synchronization_object.h"
 #include "core/hle/kernel/object.h"
-#include "core/hle/kernel/synchronization_object.h"
 
 union ResultCode;
 
@@ -14,7 +14,7 @@ namespace Kernel {
 class KernelCore;
 class WritableEvent;
 
-class ReadableEvent final : public SynchronizationObject {
+class ReadableEvent final : public KSynchronizationObject {
     friend class WritableEvent;
 
 public:
@@ -32,9 +32,6 @@ public:
         return HANDLE_TYPE;
     }
 
-    bool ShouldWait(const Thread* thread) const override;
-    void Acquire(Thread* thread) override;
-
     /// Unconditionally clears the readable event's state.
     void Clear();
 
@@ -46,11 +43,14 @@ public:
     ///      then ERR_INVALID_STATE will be returned.
     ResultCode Reset();
 
-    void Signal() override;
+    void Signal();
+
+    virtual bool IsSignaled() const override;
 
 private:
     explicit ReadableEvent(KernelCore& kernel);
 
+    bool is_signaled{};
     std::string name; ///< Name of event (optional)
 };
 
diff --git a/src/core/hle/kernel/server_port.cpp b/src/core/hle/kernel/server_port.cpp
index a549ae9d7..82857f93b 100755
--- a/src/core/hle/kernel/server_port.cpp
+++ b/src/core/hle/kernel/server_port.cpp
@@ -13,7 +13,7 @@
 
 namespace Kernel {
 
-ServerPort::ServerPort(KernelCore& kernel) : SynchronizationObject{kernel} {}
+ServerPort::ServerPort(KernelCore& kernel) : KSynchronizationObject{kernel} {}
 ServerPort::~ServerPort() = default;
 
 ResultVal<std::shared_ptr<ServerSession>> ServerPort::Accept() {
@@ -28,15 +28,9 @@ ResultVal<std::shared_ptr<ServerSession>> ServerPort::Accept() {
 
 void ServerPort::AppendPendingSession(std::shared_ptr<ServerSession> pending_session) {
     pending_sessions.push_back(std::move(pending_session));
-}
-
-bool ServerPort::ShouldWait(const Thread* thread) const {
-    // If there are no pending sessions, we wait until a new one is added.
-    return pending_sessions.empty();
-}
-
-void ServerPort::Acquire(Thread* thread) {
-    ASSERT_MSG(!ShouldWait(thread), "object unavailable!");
+    if (pending_sessions.size() == 1) {
+        NotifyAvailable();
+    }
 }
 
 bool ServerPort::IsSignaled() const {
diff --git a/src/core/hle/kernel/server_port.h b/src/core/hle/kernel/server_port.h
index 41b191b86..587d01e0a 100755
--- a/src/core/hle/kernel/server_port.h
+++ b/src/core/hle/kernel/server_port.h
@@ -9,8 +9,8 @@
 #include <utility>
 #include <vector>
 #include "common/common_types.h"
+#include "core/hle/kernel/k_synchronization_object.h"
 #include "core/hle/kernel/object.h"
-#include "core/hle/kernel/synchronization_object.h"
 #include "core/hle/result.h"
 
 namespace Kernel {
@@ -20,7 +20,7 @@ class KernelCore;
 class ServerSession;
 class SessionRequestHandler;
 
-class ServerPort final : public SynchronizationObject {
+class ServerPort final : public KSynchronizationObject {
 public:
     explicit ServerPort(KernelCore& kernel);
     ~ServerPort() override;
@@ -79,10 +79,7 @@ public:
     /// waiting to be accepted by this port.
     void AppendPendingSession(std::shared_ptr<ServerSession> pending_session);
 
-    bool ShouldWait(const Thread* thread) const override;
-    void Acquire(Thread* thread) override;
-
-    bool IsSignaled() const override;
+    virtual bool IsSignaled() const override;
 
 private:
     /// ServerSessions waiting to be accepted by the port
diff --git a/src/core/hle/kernel/server_session.cpp b/src/core/hle/kernel/server_session.cpp
index b40fe3916..4f2bb7822 100755
--- a/src/core/hle/kernel/server_session.cpp
+++ b/src/core/hle/kernel/server_session.cpp
@@ -24,7 +24,7 @@
 
 namespace Kernel {
 
-ServerSession::ServerSession(KernelCore& kernel) : SynchronizationObject{kernel} {}
+ServerSession::ServerSession(KernelCore& kernel) : KSynchronizationObject{kernel} {}
 
 ServerSession::~ServerSession() {
     kernel.ReleaseServiceThread(service_thread);
@@ -42,16 +42,6 @@ ResultVal<std::shared_ptr<ServerSession>> ServerSession::Create(KernelCore& kern
     return MakeResult(std::move(session));
 }
 
-bool ServerSession::ShouldWait(const Thread* thread) const {
-    // Closed sessions should never wait, an error will be returned from svcReplyAndReceive.
-    if (!parent->Client()) {
-        return false;
-    }
-
-    // Wait if we have no pending requests, or if we're currently handling a request.
-    return pending_requesting_threads.empty() || currently_handling != nullptr;
-}
-
 bool ServerSession::IsSignaled() const {
     // Closed sessions should never wait, an error will be returned from svcReplyAndReceive.
     if (!parent->Client()) {
@@ -62,15 +52,6 @@ bool ServerSession::IsSignaled() const {
     return !pending_requesting_threads.empty() && currently_handling == nullptr;
 }
 
-void ServerSession::Acquire(Thread* thread) {
-    ASSERT_MSG(!ShouldWait(thread), "object unavailable!");
-    // We are now handling a request, pop it from the stack.
-    // TODO(Subv): What happens if the client endpoint is closed before any requests are made?
-    ASSERT(!pending_requesting_threads.empty());
-    currently_handling = pending_requesting_threads.back();
-    pending_requesting_threads.pop_back();
-}
-
 void ServerSession::ClientDisconnected() {
     // We keep a shared pointer to the hle handler to keep it alive throughout
     // the call to ClientDisconnected, as ClientDisconnected invalidates the
@@ -172,7 +153,7 @@ ResultCode ServerSession::CompleteSyncRequest(HLERequestContext& context) {
     {
         KScopedSchedulerLock lock(kernel);
         if (!context.IsThreadWaiting()) {
-            context.GetThread().ResumeFromWait();
+            context.GetThread().Wakeup();
             context.GetThread().SetSynchronizationResults(nullptr, result);
         }
     }
diff --git a/src/core/hle/kernel/server_session.h b/src/core/hle/kernel/server_session.h
index e8d1d99ea..bfeb05c8d 100755
--- a/src/core/hle/kernel/server_session.h
+++ b/src/core/hle/kernel/server_session.h
@@ -10,8 +10,8 @@
 #include <vector>
 
 #include "common/threadsafe_queue.h"
+#include "core/hle/kernel/k_synchronization_object.h"
 #include "core/hle/kernel/service_thread.h"
-#include "core/hle/kernel/synchronization_object.h"
 #include "core/hle/result.h"
 
 namespace Core::Memory {
@@ -43,7 +43,7 @@ class Thread;
  * After the server replies to the request, the response is marshalled back to the caller's
  * TLS buffer and control is transferred back to it.
  */
-class ServerSession final : public SynchronizationObject {
+class ServerSession final : public KSynchronizationObject {
     friend class ServiceThread;
 
 public:
@@ -77,8 +77,6 @@ public:
         return parent.get();
     }
 
-    bool IsSignaled() const override;
-
     /**
      * Sets the HLE handler for the session. This handler will be called to service IPC requests
      * instead of the regular IPC machinery. (The regular IPC machinery is currently not
@@ -100,10 +98,6 @@ public:
     ResultCode HandleSyncRequest(std::shared_ptr<Thread> thread, Core::Memory::Memory& memory,
                                  Core::Timing::CoreTiming& core_timing);
 
-    bool ShouldWait(const Thread* thread) const override;
-
-    void Acquire(Thread* thread) override;
-
     /// Called when a client disconnection occurs.
     void ClientDisconnected();
 
@@ -130,6 +124,8 @@ public:
         convert_to_domain = true;
     }
 
+    virtual bool IsSignaled() const override;
+
 private:
     /// Queues a sync request from the emulated application.
     ResultCode QueueSyncRequest(std::shared_ptr<Thread> thread, Core::Memory::Memory& memory);
diff --git a/src/core/hle/kernel/session.cpp b/src/core/hle/kernel/session.cpp
index e4dd53e24..75304b961 100755
--- a/src/core/hle/kernel/session.cpp
+++ b/src/core/hle/kernel/session.cpp
@@ -9,7 +9,7 @@
 
 namespace Kernel {
 
-Session::Session(KernelCore& kernel) : SynchronizationObject{kernel} {}
+Session::Session(KernelCore& kernel) : KSynchronizationObject{kernel} {}
 Session::~Session() = default;
 
 Session::SessionPair Session::Create(KernelCore& kernel, std::string name) {
@@ -24,18 +24,9 @@ Session::SessionPair Session::Create(KernelCore& kernel, std::string name) {
     return std::make_pair(std::move(client_session), std::move(server_session));
 }
 
-bool Session::ShouldWait(const Thread* thread) const {
-    UNIMPLEMENTED();
-    return {};
-}
-
 bool Session::IsSignaled() const {
     UNIMPLEMENTED();
     return true;
 }
 
-void Session::Acquire(Thread* thread) {
-    UNIMPLEMENTED();
-}
-
 } // namespace Kernel
diff --git a/src/core/hle/kernel/session.h b/src/core/hle/kernel/session.h
index 7cd9c0d77..f6dd2c1d2 100755
--- a/src/core/hle/kernel/session.h
+++ b/src/core/hle/kernel/session.h
@@ -8,7 +8,7 @@
 #include <string>
 #include <utility>
 
-#include "core/hle/kernel/synchronization_object.h"
+#include "core/hle/kernel/k_synchronization_object.h"
 
 namespace Kernel {
 
@@ -19,7 +19,7 @@ class ServerSession;
  * Parent structure to link the client and server endpoints of a session with their associated
  * client port.
  */
-class Session final : public SynchronizationObject {
+class Session final : public KSynchronizationObject {
 public:
     explicit Session(KernelCore& kernel);
     ~Session() override;
@@ -37,12 +37,8 @@ public:
         return HANDLE_TYPE;
     }
 
-    bool ShouldWait(const Thread* thread) const override;
-
     bool IsSignaled() const override;
 
-    void Acquire(Thread* thread) override;
-
     std::shared_ptr<ClientSession> Client() {
         if (auto result{client.lock()}) {
             return result;
diff --git a/src/core/hle/kernel/svc.cpp b/src/core/hle/kernel/svc.cpp
index de3ed25da..99bb4ea20 100755
--- a/src/core/hle/kernel/svc.cpp
+++ b/src/core/hle/kernel/svc.cpp
@@ -10,6 +10,7 @@
 
 #include "common/alignment.h"
 #include "common/assert.h"
+#include "common/common_funcs.h"
 #include "common/fiber.h"
 #include "common/logging/log.h"
 #include "common/microprofile.h"
@@ -19,26 +20,28 @@
 #include "core/core_timing.h"
 #include "core/core_timing_util.h"
 #include "core/cpu_manager.h"
-#include "core/hle/kernel/address_arbiter.h"
 #include "core/hle/kernel/client_port.h"
 #include "core/hle/kernel/client_session.h"
 #include "core/hle/kernel/errors.h"
 #include "core/hle/kernel/handle_table.h"
+#include "core/hle/kernel/k_address_arbiter.h"
+#include "core/hle/kernel/k_condition_variable.h"
 #include "core/hle/kernel/k_scheduler.h"
 #include "core/hle/kernel/k_scoped_scheduler_lock_and_sleep.h"
+#include "core/hle/kernel/k_synchronization_object.h"
 #include "core/hle/kernel/kernel.h"
 #include "core/hle/kernel/memory/memory_block.h"
+#include "core/hle/kernel/memory/memory_layout.h"
 #include "core/hle/kernel/memory/page_table.h"
-#include "core/hle/kernel/mutex.h"
 #include "core/hle/kernel/physical_core.h"
 #include "core/hle/kernel/process.h"
 #include "core/hle/kernel/readable_event.h"
 #include "core/hle/kernel/resource_limit.h"
 #include "core/hle/kernel/shared_memory.h"
 #include "core/hle/kernel/svc.h"
+#include "core/hle/kernel/svc_results.h"
 #include "core/hle/kernel/svc_types.h"
 #include "core/hle/kernel/svc_wrap.h"
-#include "core/hle/kernel/synchronization.h"
 #include "core/hle/kernel/thread.h"
 #include "core/hle/kernel/time_manager.h"
 #include "core/hle/kernel/transfer_memory.h"
@@ -343,27 +346,10 @@ static ResultCode SendSyncRequest(Core::System& system, Handle handle) {
     auto thread = kernel.CurrentScheduler()->GetCurrentThread();
     {
         KScopedSchedulerLock lock(kernel);
-        thread->InvalidateHLECallback();
-        thread->SetStatus(ThreadStatus::WaitIPC);
+        thread->SetState(ThreadState::Waiting);
         session->SendSyncRequest(SharedFrom(thread), system.Memory(), system.CoreTiming());
     }
 
-    if (thread->HasHLECallback()) {
-        Handle event_handle = thread->GetHLETimeEvent();
-        if (event_handle != InvalidHandle) {
-            auto& time_manager = kernel.TimeManager();
-            time_manager.UnscheduleTimeEvent(event_handle);
-        }
-
-        {
-            KScopedSchedulerLock lock(kernel);
-            auto* sync_object = thread->GetHLESyncObject();
-            sync_object->RemoveWaitingThread(SharedFrom(thread));
-        }
-
-        thread->InvokeHLECallback(SharedFrom(thread));
-    }
-
     return thread->GetSignalingResult();
 }
 
@@ -436,7 +422,7 @@ static ResultCode GetProcessId32(Core::System& system, u32* process_id_low, u32*
 }
 
 /// Wait for the given handles to synchronize, timeout after the specified nanoseconds
-static ResultCode WaitSynchronization(Core::System& system, Handle* index, VAddr handles_address,
+static ResultCode WaitSynchronization(Core::System& system, s32* index, VAddr handles_address,
                                       u64 handle_count, s64 nano_seconds) {
     LOG_TRACE(Kernel_SVC, "called handles_address=0x{:X}, handle_count={}, nano_seconds={}",
               handles_address, handle_count, nano_seconds);
@@ -458,28 +444,26 @@ static ResultCode WaitSynchronization(Core::System& system, Handle* index, VAddr
     }
 
     auto& kernel = system.Kernel();
-    Thread::ThreadSynchronizationObjects objects(handle_count);
+    std::vector<KSynchronizationObject*> objects(handle_count);
     const auto& handle_table = kernel.CurrentProcess()->GetHandleTable();
 
     for (u64 i = 0; i < handle_count; ++i) {
         const Handle handle = memory.Read32(handles_address + i * sizeof(Handle));
-        const auto object = handle_table.Get<SynchronizationObject>(handle);
+        const auto object = handle_table.Get<KSynchronizationObject>(handle);
 
         if (object == nullptr) {
             LOG_ERROR(Kernel_SVC, "Object is a nullptr");
             return ERR_INVALID_HANDLE;
         }
 
-        objects[i] = object;
+        objects[i] = object.get();
     }
-    auto& synchronization = kernel.Synchronization();
-    const auto [result, handle_result] = synchronization.WaitFor(objects, nano_seconds);
-    *index = handle_result;
-    return result;
+    return KSynchronizationObject::Wait(kernel, index, objects.data(),
+                                        static_cast<s32>(objects.size()), nano_seconds);
 }
 
 static ResultCode WaitSynchronization32(Core::System& system, u32 timeout_low, u32 handles_address,
-                                        s32 handle_count, u32 timeout_high, Handle* index) {
+                                        s32 handle_count, u32 timeout_high, s32* index) {
     const s64 nano_seconds{(static_cast<s64>(timeout_high) << 32) | static_cast<s64>(timeout_low)};
     return WaitSynchronization(system, index, handles_address, handle_count, nano_seconds);
 }
@@ -504,56 +488,37 @@ static ResultCode CancelSynchronization32(Core::System& system, Handle thread_ha
     return CancelSynchronization(system, thread_handle);
 }
 
-/// Attempts to locks a mutex, creating it if it does not already exist
-static ResultCode ArbitrateLock(Core::System& system, Handle holding_thread_handle,
-                                VAddr mutex_addr, Handle requesting_thread_handle) {
-    LOG_TRACE(Kernel_SVC,
-              "called holding_thread_handle=0x{:08X}, mutex_addr=0x{:X}, "
-              "requesting_current_thread_handle=0x{:08X}",
-              holding_thread_handle, mutex_addr, requesting_thread_handle);
+/// Attempts to locks a mutex
+static ResultCode ArbitrateLock(Core::System& system, Handle thread_handle, VAddr address,
+                                u32 tag) {
+    LOG_TRACE(Kernel_SVC, "called thread_handle=0x{:08X}, address=0x{:X}, tag=0x{:08X}",
+              thread_handle, address, tag);
 
-    if (Core::Memory::IsKernelVirtualAddress(mutex_addr)) {
-        LOG_ERROR(Kernel_SVC, "Mutex Address is a kernel virtual address, mutex_addr={:016X}",
-                  mutex_addr);
-        return ERR_INVALID_ADDRESS_STATE;
-    }
+    // Validate the input address.
+    R_UNLESS(!Memory::IsKernelAddress(address), Svc::ResultInvalidCurrentMemory);
+    R_UNLESS(Common::IsAligned(address, sizeof(u32)), Svc::ResultInvalidAddress);
 
-    if (!Common::IsWordAligned(mutex_addr)) {
-        LOG_ERROR(Kernel_SVC, "Mutex Address is not word aligned, mutex_addr={:016X}", mutex_addr);
-        return ERR_INVALID_ADDRESS;
-    }
-
-    auto* const current_process = system.Kernel().CurrentProcess();
-    return current_process->GetMutex().TryAcquire(mutex_addr, holding_thread_handle,
-                                                  requesting_thread_handle);
+    return system.Kernel().CurrentProcess()->WaitForAddress(thread_handle, address, tag);
 }
 
-static ResultCode ArbitrateLock32(Core::System& system, Handle holding_thread_handle,
-                                  u32 mutex_addr, Handle requesting_thread_handle) {
-    return ArbitrateLock(system, holding_thread_handle, mutex_addr, requesting_thread_handle);
+static ResultCode ArbitrateLock32(Core::System& system, Handle thread_handle, u32 address,
+                                  u32 tag) {
+    return ArbitrateLock(system, thread_handle, address, tag);
 }
 
 /// Unlock a mutex
-static ResultCode ArbitrateUnlock(Core::System& system, VAddr mutex_addr) {
-    LOG_TRACE(Kernel_SVC, "called mutex_addr=0x{:X}", mutex_addr);
+static ResultCode ArbitrateUnlock(Core::System& system, VAddr address) {
+    LOG_TRACE(Kernel_SVC, "called address=0x{:X}", address);
 
-    if (Core::Memory::IsKernelVirtualAddress(mutex_addr)) {
-        LOG_ERROR(Kernel_SVC, "Mutex Address is a kernel virtual address, mutex_addr={:016X}",
-                  mutex_addr);
-        return ERR_INVALID_ADDRESS_STATE;
-    }
+    // Validate the input address.
+    R_UNLESS(!Memory::IsKernelAddress(address), Svc::ResultInvalidCurrentMemory);
+    R_UNLESS(Common::IsAligned(address, sizeof(u32)), Svc::ResultInvalidAddress);
 
-    if (!Common::IsWordAligned(mutex_addr)) {
-        LOG_ERROR(Kernel_SVC, "Mutex Address is not word aligned, mutex_addr={:016X}", mutex_addr);
-        return ERR_INVALID_ADDRESS;
-    }
-
-    auto* const current_process = system.Kernel().CurrentProcess();
-    return current_process->GetMutex().Release(mutex_addr);
+    return system.Kernel().CurrentProcess()->SignalToAddress(address);
 }
 
-static ResultCode ArbitrateUnlock32(Core::System& system, u32 mutex_addr) {
-    return ArbitrateUnlock(system, mutex_addr);
+static ResultCode ArbitrateUnlock32(Core::System& system, u32 address) {
+    return ArbitrateUnlock(system, address);
 }
 
 enum class BreakType : u32 {
@@ -1180,7 +1145,7 @@ static ResultCode SetThreadPriority(Core::System& system, Handle handle, u32 pri
         return ERR_INVALID_HANDLE;
     }
 
-    thread->SetPriority(priority);
+    thread->SetBasePriority(priority);
 
     return RESULT_SUCCESS;
 }
@@ -1559,7 +1524,7 @@ static ResultCode StartThread(Core::System& system, Handle thread_handle) {
         return ERR_INVALID_HANDLE;
     }
 
-    ASSERT(thread->GetStatus() == ThreadStatus::Dormant);
+    ASSERT(thread->GetState() == ThreadState::Initialized);
 
     return thread->Start();
 }
@@ -1620,224 +1585,135 @@ static void SleepThread32(Core::System& system, u32 nanoseconds_low, u32 nanosec
 }
 
 /// Wait process wide key atomic
-static ResultCode WaitProcessWideKeyAtomic(Core::System& system, VAddr mutex_addr,
-                                           VAddr condition_variable_addr, Handle thread_handle,
-                                           s64 nano_seconds) {
-    LOG_TRACE(
-        Kernel_SVC,
-        "called mutex_addr={:X}, condition_variable_addr={:X}, thread_handle=0x{:08X}, timeout={}",
-        mutex_addr, condition_variable_addr, thread_handle, nano_seconds);
+static ResultCode WaitProcessWideKeyAtomic(Core::System& system, VAddr address, VAddr cv_key,
+                                           u32 tag, s64 timeout_ns) {
+    LOG_TRACE(Kernel_SVC, "called address={:X}, cv_key={:X}, tag=0x{:08X}, timeout_ns={}", address,
+              cv_key, tag, timeout_ns);
 
-    if (Core::Memory::IsKernelVirtualAddress(mutex_addr)) {
-        LOG_ERROR(
-            Kernel_SVC,
-            "Given mutex address must not be within the kernel address space. address=0x{:016X}",
-            mutex_addr);
-        return ERR_INVALID_ADDRESS_STATE;
-    }
+    // Validate input.
+    R_UNLESS(!Memory::IsKernelAddress(address), Svc::ResultInvalidCurrentMemory);
+    R_UNLESS(Common::IsAligned(address, sizeof(int32_t)), Svc::ResultInvalidAddress);
 
-    if (!Common::IsWordAligned(mutex_addr)) {
-        LOG_ERROR(Kernel_SVC, "Given mutex address must be word-aligned. address=0x{:016X}",
-                  mutex_addr);
-        return ERR_INVALID_ADDRESS;
-    }
-
-    ASSERT(condition_variable_addr == Common::AlignDown(condition_variable_addr, 4));
-    auto& kernel = system.Kernel();
-    Handle event_handle;
-    Thread* current_thread = kernel.CurrentScheduler()->GetCurrentThread();
-    auto* const current_process = kernel.CurrentProcess();
-    {
-        KScopedSchedulerLockAndSleep lock(kernel, event_handle, current_thread, nano_seconds);
-        const auto& handle_table = current_process->GetHandleTable();
-        std::shared_ptr<Thread> thread = handle_table.Get<Thread>(thread_handle);
-        ASSERT(thread);
-
-        current_thread->SetSynchronizationResults(nullptr, RESULT_TIMEOUT);
-
-        if (thread->IsPendingTermination()) {
-            lock.CancelSleep();
-            return ERR_THREAD_TERMINATING;
+    // Convert timeout from nanoseconds to ticks.
+    s64 timeout{};
+    if (timeout_ns > 0) {
+        const s64 offset_tick(timeout_ns);
+        if (offset_tick > 0) {
+            timeout = offset_tick + 2;
+            if (timeout <= 0) {
+                timeout = std::numeric_limits<s64>::max();
+            }
+        } else {
+            timeout = std::numeric_limits<s64>::max();
         }
-
-        const auto release_result = current_process->GetMutex().Release(mutex_addr);
-        if (release_result.IsError()) {
-            lock.CancelSleep();
-            return release_result;
-        }
-
-        if (nano_seconds == 0) {
-            lock.CancelSleep();
-            return RESULT_TIMEOUT;
-        }
-
-        current_thread->SetCondVarWaitAddress(condition_variable_addr);
-        current_thread->SetMutexWaitAddress(mutex_addr);
-        current_thread->SetWaitHandle(thread_handle);
-        current_thread->SetStatus(ThreadStatus::WaitCondVar);
-        current_process->InsertConditionVariableThread(SharedFrom(current_thread));
+    } else {
+        timeout = timeout_ns;
     }
 
-    if (event_handle != InvalidHandle) {
-        auto& time_manager = kernel.TimeManager();
-        time_manager.UnscheduleTimeEvent(event_handle);
-    }
-
-    {
-        KScopedSchedulerLock lock(kernel);
-
-        auto* owner = current_thread->GetLockOwner();
-        if (owner != nullptr) {
-            owner->RemoveMutexWaiter(SharedFrom(current_thread));
-        }
-
-        current_process->RemoveConditionVariableThread(SharedFrom(current_thread));
-    }
-    // Note: Deliberately don't attempt to inherit the lock owner's priority.
-
-    return current_thread->GetSignalingResult();
+    // Wait on the condition variable.
+    return system.Kernel().CurrentProcess()->WaitConditionVariable(
+        address, Common::AlignDown(cv_key, sizeof(u32)), tag, timeout);
 }
 
-static ResultCode WaitProcessWideKeyAtomic32(Core::System& system, u32 mutex_addr,
-                                             u32 condition_variable_addr, Handle thread_handle,
-                                             u32 nanoseconds_low, u32 nanoseconds_high) {
-    const auto nanoseconds = static_cast<s64>(nanoseconds_low | (u64{nanoseconds_high} << 32));
-    return WaitProcessWideKeyAtomic(system, mutex_addr, condition_variable_addr, thread_handle,
-                                    nanoseconds);
+static ResultCode WaitProcessWideKeyAtomic32(Core::System& system, u32 address, u32 cv_key, u32 tag,
+                                             u32 timeout_ns_low, u32 timeout_ns_high) {
+    const auto timeout_ns = static_cast<s64>(timeout_ns_low | (u64{timeout_ns_high} << 32));
+    return WaitProcessWideKeyAtomic(system, address, cv_key, tag, timeout_ns);
 }
 
 /// Signal process wide key
-static void SignalProcessWideKey(Core::System& system, VAddr condition_variable_addr, s32 target) {
-    LOG_TRACE(Kernel_SVC, "called, condition_variable_addr=0x{:X}, target=0x{:08X}",
-              condition_variable_addr, target);
+static void SignalProcessWideKey(Core::System& system, VAddr cv_key, s32 count) {
+    LOG_TRACE(Kernel_SVC, "called, cv_key=0x{:X}, count=0x{:08X}", cv_key, count);
 
-    ASSERT(condition_variable_addr == Common::AlignDown(condition_variable_addr, 4));
+    // Signal the condition variable.
+    return system.Kernel().CurrentProcess()->SignalConditionVariable(
+        Common::AlignDown(cv_key, sizeof(u32)), count);
+}
 
-    // Retrieve a list of all threads that are waiting for this condition variable.
-    auto& kernel = system.Kernel();
-    KScopedSchedulerLock lock(kernel);
-    auto* const current_process = kernel.CurrentProcess();
-    std::vector<std::shared_ptr<Thread>> waiting_threads =
-        current_process->GetConditionVariableThreads(condition_variable_addr);
+static void SignalProcessWideKey32(Core::System& system, u32 cv_key, s32 count) {
+    SignalProcessWideKey(system, cv_key, count);
+}
 
-    // Only process up to 'target' threads, unless 'target' is less equal 0, in which case process
-    // them all.
-    std::size_t last = waiting_threads.size();
-    if (target > 0) {
-        last = std::min(waiting_threads.size(), static_cast<std::size_t>(target));
-    }
-    for (std::size_t index = 0; index < last; ++index) {
-        auto& thread = waiting_threads[index];
+namespace {
 
-        ASSERT(thread->GetCondVarWaitAddress() == condition_variable_addr);
-
-        // liberate Cond Var Thread.
-        current_process->RemoveConditionVariableThread(thread);
-
-        const std::size_t current_core = system.CurrentCoreIndex();
-        auto& monitor = system.Monitor();
-
-        // Atomically read the value of the mutex.
-        u32 mutex_val = 0;
-        u32 update_val = 0;
-        const VAddr mutex_address = thread->GetMutexWaitAddress();
-        do {
-            // If the mutex is not yet acquired, acquire it.
-            mutex_val = monitor.ExclusiveRead32(current_core, mutex_address);
-
-            if (mutex_val != 0) {
-                update_val = mutex_val | Mutex::MutexHasWaitersFlag;
-            } else {
-                update_val = thread->GetWaitHandle();
-            }
-        } while (!monitor.ExclusiveWrite32(current_core, mutex_address, update_val));
-        monitor.ClearExclusive();
-        if (mutex_val == 0) {
-            // We were able to acquire the mutex, resume this thread.
-            auto* const lock_owner = thread->GetLockOwner();
-            if (lock_owner != nullptr) {
-                lock_owner->RemoveMutexWaiter(thread);
-            }
-
-            thread->SetLockOwner(nullptr);
-            thread->SetSynchronizationResults(nullptr, RESULT_SUCCESS);
-            thread->ResumeFromWait();
-        } else {
-            // The mutex is already owned by some other thread, make this thread wait on it.
-            const Handle owner_handle = static_cast<Handle>(mutex_val & Mutex::MutexOwnerMask);
-            const auto& handle_table = system.Kernel().CurrentProcess()->GetHandleTable();
-            auto owner = handle_table.Get<Thread>(owner_handle);
-            ASSERT(owner);
-            if (thread->GetStatus() == ThreadStatus::WaitCondVar) {
-                thread->SetStatus(ThreadStatus::WaitMutex);
-            }
-
-            owner->AddMutexWaiter(thread);
-        }
+constexpr bool IsValidSignalType(Svc::SignalType type) {
+    switch (type) {
+    case Svc::SignalType::Signal:
+    case Svc::SignalType::SignalAndIncrementIfEqual:
+    case Svc::SignalType::SignalAndModifyByWaitingCountIfEqual:
+        return true;
+    default:
+        return false;
     }
 }
 
-static void SignalProcessWideKey32(Core::System& system, u32 condition_variable_addr, s32 target) {
-    SignalProcessWideKey(system, condition_variable_addr, target);
+constexpr bool IsValidArbitrationType(Svc::ArbitrationType type) {
+    switch (type) {
+    case Svc::ArbitrationType::WaitIfLessThan:
+    case Svc::ArbitrationType::DecrementAndWaitIfLessThan:
+    case Svc::ArbitrationType::WaitIfEqual:
+        return true;
+    default:
+        return false;
+    }
 }
 
+} // namespace
+
 // Wait for an address (via Address Arbiter)
-static ResultCode WaitForAddress(Core::System& system, VAddr address, u32 type, s32 value,
-                                 s64 timeout) {
-    LOG_TRACE(Kernel_SVC, "called, address=0x{:X}, type=0x{:X}, value=0x{:X}, timeout={}", address,
-              type, value, timeout);
+static ResultCode WaitForAddress(Core::System& system, VAddr address, Svc::ArbitrationType arb_type,
+                                 s32 value, s64 timeout_ns) {
+    LOG_TRACE(Kernel_SVC, "called, address=0x{:X}, arb_type=0x{:X}, value=0x{:X}, timeout_ns={}",
+              address, arb_type, value, timeout_ns);
 
-    // If the passed address is a kernel virtual address, return invalid memory state.
-    if (Core::Memory::IsKernelVirtualAddress(address)) {
-        LOG_ERROR(Kernel_SVC, "Address is a kernel virtual address, address={:016X}", address);
-        return ERR_INVALID_ADDRESS_STATE;
+    // Validate input.
+    R_UNLESS(!Memory::IsKernelAddress(address), Svc::ResultInvalidCurrentMemory);
+    R_UNLESS(Common::IsAligned(address, sizeof(int32_t)), Svc::ResultInvalidAddress);
+    R_UNLESS(IsValidArbitrationType(arb_type), Svc::ResultInvalidEnumValue);
+
+    // Convert timeout from nanoseconds to ticks.
+    s64 timeout{};
+    if (timeout_ns > 0) {
+        const s64 offset_tick(timeout_ns);
+        if (offset_tick > 0) {
+            timeout = offset_tick + 2;
+            if (timeout <= 0) {
+                timeout = std::numeric_limits<s64>::max();
+            }
+        } else {
+            timeout = std::numeric_limits<s64>::max();
+        }
+    } else {
+        timeout = timeout_ns;
     }
 
-    // If the address is not properly aligned to 4 bytes, return invalid address.
-    if (!Common::IsWordAligned(address)) {
-        LOG_ERROR(Kernel_SVC, "Address is not word aligned, address={:016X}", address);
-        return ERR_INVALID_ADDRESS;
-    }
-
-    const auto arbitration_type = static_cast<AddressArbiter::ArbitrationType>(type);
-    auto& address_arbiter = system.Kernel().CurrentProcess()->GetAddressArbiter();
-    const ResultCode result =
-        address_arbiter.WaitForAddress(address, arbitration_type, value, timeout);
-    return result;
+    return system.Kernel().CurrentProcess()->WaitAddressArbiter(address, arb_type, value, timeout);
 }
 
-static ResultCode WaitForAddress32(Core::System& system, u32 address, u32 type, s32 value,
-                                   u32 timeout_low, u32 timeout_high) {
-    const auto timeout = static_cast<s64>(timeout_low | (u64{timeout_high} << 32));
-    return WaitForAddress(system, address, type, value, timeout);
+static ResultCode WaitForAddress32(Core::System& system, u32 address, Svc::ArbitrationType arb_type,
+                                   s32 value, u32 timeout_ns_low, u32 timeout_ns_high) {
+    const auto timeout = static_cast<s64>(timeout_ns_low | (u64{timeout_ns_high} << 32));
+    return WaitForAddress(system, address, arb_type, value, timeout);
 }
 
 // Signals to an address (via Address Arbiter)
-static ResultCode SignalToAddress(Core::System& system, VAddr address, u32 type, s32 value,
-                                  s32 num_to_wake) {
-    LOG_TRACE(Kernel_SVC, "called, address=0x{:X}, type=0x{:X}, value=0x{:X}, num_to_wake=0x{:X}",
-              address, type, value, num_to_wake);
+static ResultCode SignalToAddress(Core::System& system, VAddr address, Svc::SignalType signal_type,
+                                  s32 value, s32 count) {
+    LOG_TRACE(Kernel_SVC, "called, address=0x{:X}, signal_type=0x{:X}, value=0x{:X}, count=0x{:X}",
+              address, signal_type, value, count);
 
-    // If the passed address is a kernel virtual address, return invalid memory state.
-    if (Core::Memory::IsKernelVirtualAddress(address)) {
-        LOG_ERROR(Kernel_SVC, "Address is a kernel virtual address, address={:016X}", address);
-        return ERR_INVALID_ADDRESS_STATE;
-    }
+    // Validate input.
+    R_UNLESS(!Memory::IsKernelAddress(address), Svc::ResultInvalidCurrentMemory);
+    R_UNLESS(Common::IsAligned(address, sizeof(s32)), Svc::ResultInvalidAddress);
+    R_UNLESS(IsValidSignalType(signal_type), Svc::ResultInvalidEnumValue);
 
-    // If the address is not properly aligned to 4 bytes, return invalid address.
-    if (!Common::IsWordAligned(address)) {
-        LOG_ERROR(Kernel_SVC, "Address is not word aligned, address={:016X}", address);
-        return ERR_INVALID_ADDRESS;
-    }
-
-    const auto signal_type = static_cast<AddressArbiter::SignalType>(type);
-    auto& address_arbiter = system.Kernel().CurrentProcess()->GetAddressArbiter();
-    return address_arbiter.SignalToAddress(address, signal_type, value, num_to_wake);
+    return system.Kernel().CurrentProcess()->SignalAddressArbiter(address, signal_type, value,
+                                                                  count);
 }
 
-static ResultCode SignalToAddress32(Core::System& system, u32 address, u32 type, s32 value,
-                                    s32 num_to_wake) {
-    return SignalToAddress(system, address, type, value, num_to_wake);
+static ResultCode SignalToAddress32(Core::System& system, u32 address, Svc::SignalType signal_type,
+                                    s32 value, s32 count) {
+    return SignalToAddress(system, address, signal_type, value, count);
 }
 
 static void KernelDebug([[maybe_unused]] Core::System& system,
diff --git a/src/core/hle/kernel/svc_types.h b/src/core/hle/kernel/svc_types.h
index 986724beb..356947741 100755
--- a/src/core/hle/kernel/svc_types.h
+++ b/src/core/hle/kernel/svc_types.h
@@ -65,4 +65,16 @@ struct MemoryInfo {
     u32 padding{};
 };
 
+enum class SignalType : u32 {
+    Signal = 0,
+    SignalAndIncrementIfEqual = 1,
+    SignalAndModifyByWaitingCountIfEqual = 2,
+};
+
+enum class ArbitrationType : u32 {
+    WaitIfLessThan = 0,
+    DecrementAndWaitIfLessThan = 1,
+    WaitIfEqual = 2,
+};
+
 } // namespace Kernel::Svc
diff --git a/src/core/hle/kernel/svc_wrap.h b/src/core/hle/kernel/svc_wrap.h
index 0b6dd9df0..a32750ed7 100755
--- a/src/core/hle/kernel/svc_wrap.h
+++ b/src/core/hle/kernel/svc_wrap.h
@@ -7,6 +7,7 @@
 #include "common/common_types.h"
 #include "core/arm/arm_interface.h"
 #include "core/core.h"
+#include "core/hle/kernel/svc_types.h"
 #include "core/hle/result.h"
 
 namespace Kernel {
@@ -215,9 +216,10 @@ void SvcWrap64(Core::System& system) {
         func(system, static_cast<u32>(Param(system, 0)), Param(system, 1), Param(system, 2)).raw);
 }
 
-template <ResultCode func(Core::System&, u32*, u64, u64, s64)>
+// Used by WaitSynchronization
+template <ResultCode func(Core::System&, s32*, u64, u64, s64)>
 void SvcWrap64(Core::System& system) {
-    u32 param_1 = 0;
+    s32 param_1 = 0;
     const u32 retval = func(system, &param_1, Param(system, 1), static_cast<u32>(Param(system, 2)),
                             static_cast<s64>(Param(system, 3)))
                            .raw;
@@ -276,18 +278,22 @@ void SvcWrap64(Core::System& system) {
     FuncReturn(system, retval);
 }
 
-template <ResultCode func(Core::System&, u64, u32, s32, s64)>
+// Used by WaitForAddress
+template <ResultCode func(Core::System&, u64, Svc::ArbitrationType, s32, s64)>
 void SvcWrap64(Core::System& system) {
-    FuncReturn(system, func(system, Param(system, 0), static_cast<u32>(Param(system, 1)),
-                            static_cast<s32>(Param(system, 2)), static_cast<s64>(Param(system, 3)))
-                           .raw);
+    FuncReturn(system,
+               func(system, Param(system, 0), static_cast<Svc::ArbitrationType>(Param(system, 1)),
+                    static_cast<s32>(Param(system, 2)), static_cast<s64>(Param(system, 3)))
+                   .raw);
 }
 
-template <ResultCode func(Core::System&, u64, u32, s32, s32)>
+// Used by SignalToAddress
+template <ResultCode func(Core::System&, u64, Svc::SignalType, s32, s32)>
 void SvcWrap64(Core::System& system) {
-    FuncReturn(system, func(system, Param(system, 0), static_cast<u32>(Param(system, 1)),
-                            static_cast<s32>(Param(system, 2)), static_cast<s32>(Param(system, 3)))
-                           .raw);
+    FuncReturn(system,
+               func(system, Param(system, 0), static_cast<Svc::SignalType>(Param(system, 1)),
+                    static_cast<s32>(Param(system, 2)), static_cast<s32>(Param(system, 3)))
+                   .raw);
 }
 
 ////////////////////////////////////////////////////////////////////////////////////////////////////
@@ -503,22 +509,23 @@ void SvcWrap32(Core::System& system) {
 }
 
 // Used by WaitForAddress32
-template <ResultCode func(Core::System&, u32, u32, s32, u32, u32)>
+template <ResultCode func(Core::System&, u32, Svc::ArbitrationType, s32, u32, u32)>
 void SvcWrap32(Core::System& system) {
     const u32 retval = func(system, static_cast<u32>(Param(system, 0)),
-                            static_cast<u32>(Param(system, 1)), static_cast<s32>(Param(system, 2)),
-                            static_cast<u32>(Param(system, 3)), static_cast<u32>(Param(system, 4)))
+                            static_cast<Svc::ArbitrationType>(Param(system, 1)),
+                            static_cast<s32>(Param(system, 2)), static_cast<u32>(Param(system, 3)),
+                            static_cast<u32>(Param(system, 4)))
                            .raw;
     FuncReturn(system, retval);
 }
 
 // Used by SignalToAddress32
-template <ResultCode func(Core::System&, u32, u32, s32, s32)>
+template <ResultCode func(Core::System&, u32, Svc::SignalType, s32, s32)>
 void SvcWrap32(Core::System& system) {
-    const u32 retval =
-        func(system, static_cast<u32>(Param(system, 0)), static_cast<u32>(Param(system, 1)),
-             static_cast<s32>(Param(system, 2)), static_cast<s32>(Param(system, 3)))
-            .raw;
+    const u32 retval = func(system, static_cast<u32>(Param(system, 0)),
+                            static_cast<Svc::SignalType>(Param(system, 1)),
+                            static_cast<s32>(Param(system, 2)), static_cast<s32>(Param(system, 3)))
+                           .raw;
     FuncReturn(system, retval);
 }
 
@@ -539,9 +546,9 @@ void SvcWrap32(Core::System& system) {
 }
 
 // Used by WaitSynchronization32
-template <ResultCode func(Core::System&, u32, u32, s32, u32, Handle*)>
+template <ResultCode func(Core::System&, u32, u32, s32, u32, s32*)>
 void SvcWrap32(Core::System& system) {
-    u32 param_1 = 0;
+    s32 param_1 = 0;
     const u32 retval = func(system, Param32(system, 0), Param32(system, 1), Param32(system, 2),
                             Param32(system, 3), &param_1)
                            .raw;
diff --git a/src/core/hle/kernel/thread.cpp b/src/core/hle/kernel/thread.cpp
index a4f9e0d97..eda56c31c 100755
--- a/src/core/hle/kernel/thread.cpp
+++ b/src/core/hle/kernel/thread.cpp
@@ -17,9 +17,11 @@
 #include "core/hardware_properties.h"
 #include "core/hle/kernel/errors.h"
 #include "core/hle/kernel/handle_table.h"
+#include "core/hle/kernel/k_condition_variable.h"
 #include "core/hle/kernel/k_scheduler.h"
 #include "core/hle/kernel/k_scoped_scheduler_lock_and_sleep.h"
 #include "core/hle/kernel/kernel.h"
+#include "core/hle/kernel/memory/memory_layout.h"
 #include "core/hle/kernel/object.h"
 #include "core/hle/kernel/process.h"
 #include "core/hle/kernel/thread.h"
@@ -34,26 +36,19 @@
 
 namespace Kernel {
 
-bool Thread::ShouldWait(const Thread* thread) const {
-    return status != ThreadStatus::Dead;
-}
-
 bool Thread::IsSignaled() const {
-    return status == ThreadStatus::Dead;
+    return signaled;
 }
 
-void Thread::Acquire(Thread* thread) {
-    ASSERT_MSG(!ShouldWait(thread), "object unavailable!");
-}
-
-Thread::Thread(KernelCore& kernel) : SynchronizationObject{kernel} {}
+Thread::Thread(KernelCore& kernel) : KSynchronizationObject{kernel} {}
 Thread::~Thread() = default;
 
 void Thread::Stop() {
     {
         KScopedSchedulerLock lock(kernel);
-        SetStatus(ThreadStatus::Dead);
-        Signal();
+        SetState(ThreadState::Terminated);
+        signaled = true;
+        NotifyAvailable();
         kernel.GlobalHandleTable().Close(global_handle);
 
         if (owner_process) {
@@ -67,59 +62,27 @@ void Thread::Stop() {
     global_handle = 0;
 }
 
-void Thread::ResumeFromWait() {
+void Thread::Wakeup() {
     KScopedSchedulerLock lock(kernel);
-    switch (status) {
-    case ThreadStatus::Paused:
-    case ThreadStatus::WaitSynch:
-    case ThreadStatus::WaitHLEEvent:
-    case ThreadStatus::WaitSleep:
-    case ThreadStatus::WaitIPC:
-    case ThreadStatus::WaitMutex:
-    case ThreadStatus::WaitCondVar:
-    case ThreadStatus::WaitArb:
-    case ThreadStatus::Dormant:
-        break;
-
-    case ThreadStatus::Ready:
-        // The thread's wakeup callback must have already been cleared when the thread was first
-        // awoken.
-        ASSERT(hle_callback == nullptr);
-        // If the thread is waiting on multiple wait objects, it might be awoken more than once
-        // before actually resuming. We can ignore subsequent wakeups if the thread status has
-        // already been set to ThreadStatus::Ready.
-        return;
-    case ThreadStatus::Dead:
-        // This should never happen, as threads must complete before being stopped.
-        DEBUG_ASSERT_MSG(false, "Thread with object id {} cannot be resumed because it's DEAD.",
-                         GetObjectId());
-        return;
-    }
-
-    SetStatus(ThreadStatus::Ready);
-}
-
-void Thread::OnWakeUp() {
-    KScopedSchedulerLock lock(kernel);
-    SetStatus(ThreadStatus::Ready);
+    SetState(ThreadState::Runnable);
 }
 
 ResultCode Thread::Start() {
     KScopedSchedulerLock lock(kernel);
-    SetStatus(ThreadStatus::Ready);
+    SetState(ThreadState::Runnable);
     return RESULT_SUCCESS;
 }
 
 void Thread::CancelWait() {
     KScopedSchedulerLock lock(kernel);
-    if (GetSchedulingStatus() != ThreadSchedStatus::Paused || !is_waiting_on_sync) {
+    if (GetState() != ThreadState::Waiting || !is_cancellable) {
         is_sync_cancelled = true;
         return;
     }
     // TODO(Blinkhawk): Implement cancel of server session
     is_sync_cancelled = false;
     SetSynchronizationResults(nullptr, ERR_SYNCHRONIZATION_CANCELED);
-    SetStatus(ThreadStatus::Ready);
+    SetState(ThreadState::Runnable);
 }
 
 static void ResetThreadContext32(Core::ARM_Interface::ThreadContext32& context, u32 stack_top,
@@ -183,25 +146,24 @@ ResultVal<std::shared_ptr<Thread>> Thread::Create(Core::System& system, ThreadTy
     std::shared_ptr<Thread> thread = std::make_shared<Thread>(kernel);
 
     thread->thread_id = kernel.CreateNewThreadID();
-    thread->status = ThreadStatus::Dormant;
+    thread->thread_state = ThreadState::Initialized;
     thread->entry_point = entry_point;
     thread->stack_top = stack_top;
     thread->disable_count = 1;
     thread->tpidr_el0 = 0;
-    thread->nominal_priority = thread->current_priority = priority;
+    thread->current_priority = priority;
+    thread->base_priority = priority;
+    thread->lock_owner = nullptr;
     thread->schedule_count = -1;
     thread->last_scheduled_tick = 0;
     thread->processor_id = processor_id;
     thread->ideal_core = processor_id;
     thread->affinity_mask.SetAffinity(processor_id, true);
-    thread->wait_objects = nullptr;
-    thread->mutex_wait_address = 0;
-    thread->condvar_wait_address = 0;
-    thread->wait_handle = 0;
     thread->name = std::move(name);
     thread->global_handle = kernel.GlobalHandleTable().Create(thread).Unwrap();
     thread->owner_process = owner_process;
     thread->type = type_flags;
+    thread->signaled = false;
     if ((type_flags & THREADTYPE_IDLE) == 0) {
         auto& scheduler = kernel.GlobalSchedulerContext();
         scheduler.AddThread(thread);
@@ -226,153 +188,182 @@ ResultVal<std::shared_ptr<Thread>> Thread::Create(Core::System& system, ThreadTy
     return MakeResult<std::shared_ptr<Thread>>(std::move(thread));
 }
 
-void Thread::SetPriority(u32 priority) {
-    KScopedSchedulerLock lock(kernel);
+void Thread::SetBasePriority(u32 priority) {
     ASSERT_MSG(priority <= THREADPRIO_LOWEST && priority >= THREADPRIO_HIGHEST,
                "Invalid priority value.");
-    nominal_priority = priority;
-    UpdatePriority();
+
+    KScopedSchedulerLock lock(kernel);
+
+    // Change our base priority.
+    base_priority = priority;
+
+    // Perform a priority restoration.
+    RestorePriority(kernel, this);
 }
 
-void Thread::SetSynchronizationResults(SynchronizationObject* object, ResultCode result) {
+void Thread::SetSynchronizationResults(KSynchronizationObject* object, ResultCode result) {
     signaling_object = object;
     signaling_result = result;
 }
 
-s32 Thread::GetSynchronizationObjectIndex(std::shared_ptr<SynchronizationObject> object) const {
-    ASSERT_MSG(!wait_objects->empty(), "Thread is not waiting for anything");
-    const auto match = std::find(wait_objects->rbegin(), wait_objects->rend(), object);
-    return static_cast<s32>(std::distance(match, wait_objects->rend()) - 1);
-}
-
 VAddr Thread::GetCommandBufferAddress() const {
     // Offset from the start of TLS at which the IPC command buffer begins.
     constexpr u64 command_header_offset = 0x80;
     return GetTLSAddress() + command_header_offset;
 }
 
-void Thread::SetStatus(ThreadStatus new_status) {
-    if (new_status == status) {
-        return;
-    }
+void Thread::SetState(ThreadState state) {
+    KScopedSchedulerLock sl(kernel);
 
-    switch (new_status) {
-    case ThreadStatus::Ready:
-        SetSchedulingStatus(ThreadSchedStatus::Runnable);
-        break;
-    case ThreadStatus::Dormant:
-        SetSchedulingStatus(ThreadSchedStatus::None);
-        break;
-    case ThreadStatus::Dead:
-        SetSchedulingStatus(ThreadSchedStatus::Exited);
-        break;
-    default:
-        SetSchedulingStatus(ThreadSchedStatus::Paused);
-        break;
+    SetMutexWaitAddressForDebugging(0);
+    const ThreadState old_state = thread_state;
+    thread_state =
+        static_cast<ThreadState>((old_state & ~ThreadState::Mask) | (state & ThreadState::Mask));
+    if (thread_state != old_state) {
+        KScheduler::OnThreadStateChanged(kernel, this, old_state);
     }
-
-    status = new_status;
 }
 
-void Thread::AddMutexWaiter(std::shared_ptr<Thread> thread) {
-    if (thread->lock_owner.get() == this) {
-        // If the thread is already waiting for this thread to release the mutex, ensure that the
-        // waiters list is consistent and return without doing anything.
-        const auto iter = std::find(wait_mutex_threads.begin(), wait_mutex_threads.end(), thread);
-        ASSERT(iter != wait_mutex_threads.end());
-        return;
+void Thread::AddWaiterImpl(Thread* thread) {
+    ASSERT(kernel.GlobalSchedulerContext().IsLocked());
+
+    // Find the right spot to insert the waiter.
+    auto it = waiter_list.begin();
+    while (it != waiter_list.end()) {
+        if (it->GetPriority() > thread->GetPriority()) {
+            break;
+        }
+        it++;
     }
 
-    // A thread can't wait on two different mutexes at the same time.
-    ASSERT(thread->lock_owner == nullptr);
+    // Keep track of how many kernel waiters we have.
+    if (Memory::IsKernelAddressKey(thread->GetAddressKey())) {
+        ASSERT((num_kernel_waiters++) >= 0);
+    }
 
-    // Ensure that the thread is not already in the list of mutex waiters
-    const auto iter = std::find(wait_mutex_threads.begin(), wait_mutex_threads.end(), thread);
-    ASSERT(iter == wait_mutex_threads.end());
-
-    // Keep the list in an ordered fashion
-    const auto insertion_point = std::find_if(
-        wait_mutex_threads.begin(), wait_mutex_threads.end(),
-        [&thread](const auto& entry) { return entry->GetPriority() > thread->GetPriority(); });
-    wait_mutex_threads.insert(insertion_point, thread);
-    thread->lock_owner = SharedFrom(this);
-
-    UpdatePriority();
+    // Insert the waiter.
+    waiter_list.insert(it, *thread);
+    thread->SetLockOwner(this);
 }
 
-void Thread::RemoveMutexWaiter(std::shared_ptr<Thread> thread) {
-    ASSERT(thread->lock_owner.get() == this);
+void Thread::RemoveWaiterImpl(Thread* thread) {
+    ASSERT(kernel.GlobalSchedulerContext().IsLocked());
 
-    // Ensure that the thread is in the list of mutex waiters
-    const auto iter = std::find(wait_mutex_threads.begin(), wait_mutex_threads.end(), thread);
-    ASSERT(iter != wait_mutex_threads.end());
+    // Keep track of how many kernel waiters we have.
+    if (Memory::IsKernelAddressKey(thread->GetAddressKey())) {
+        ASSERT((num_kernel_waiters--) > 0);
+    }
 
-    wait_mutex_threads.erase(iter);
-
-    thread->lock_owner = nullptr;
-    UpdatePriority();
+    // Remove the waiter.
+    waiter_list.erase(waiter_list.iterator_to(*thread));
+    thread->SetLockOwner(nullptr);
 }
 
-void Thread::UpdatePriority() {
-    // If any of the threads waiting on the mutex have a higher priority
-    // (taking into account priority inheritance), then this thread inherits
-    // that thread's priority.
-    u32 new_priority = nominal_priority;
-    if (!wait_mutex_threads.empty()) {
-        if (wait_mutex_threads.front()->current_priority < new_priority) {
-            new_priority = wait_mutex_threads.front()->current_priority;
+void Thread::RestorePriority(KernelCore& kernel, Thread* thread) {
+    ASSERT(kernel.GlobalSchedulerContext().IsLocked());
+
+    while (true) {
+        // We want to inherit priority where possible.
+        s32 new_priority = thread->GetBasePriority();
+        if (thread->HasWaiters()) {
+            new_priority = std::min(new_priority, thread->waiter_list.front().GetPriority());
+        }
+
+        // If the priority we would inherit is not different from ours, don't do anything.
+        if (new_priority == thread->GetPriority()) {
+            return;
+        }
+
+        // Ensure we don't violate condition variable red black tree invariants.
+        if (auto* cv_tree = thread->GetConditionVariableTree(); cv_tree != nullptr) {
+            BeforeUpdatePriority(kernel, cv_tree, thread);
+        }
+
+        // Change the priority.
+        const s32 old_priority = thread->GetPriority();
+        thread->SetPriority(new_priority);
+
+        // Restore the condition variable, if relevant.
+        if (auto* cv_tree = thread->GetConditionVariableTree(); cv_tree != nullptr) {
+            AfterUpdatePriority(kernel, cv_tree, thread);
+        }
+
+        // Update the scheduler.
+        KScheduler::OnThreadPriorityChanged(kernel, thread, old_priority);
+
+        // Keep the lock owner up to date.
+        Thread* lock_owner = thread->GetLockOwner();
+        if (lock_owner == nullptr) {
+            return;
+        }
+
+        // Update the thread in the lock owner's sorted list, and continue inheriting.
+        lock_owner->RemoveWaiterImpl(thread);
+        lock_owner->AddWaiterImpl(thread);
+        thread = lock_owner;
+    }
+}
+
+void Thread::AddWaiter(Thread* thread) {
+    AddWaiterImpl(thread);
+    RestorePriority(kernel, this);
+}
+
+void Thread::RemoveWaiter(Thread* thread) {
+    RemoveWaiterImpl(thread);
+    RestorePriority(kernel, this);
+}
+
+Thread* Thread::RemoveWaiterByKey(s32* out_num_waiters, VAddr key) {
+    ASSERT(kernel.GlobalSchedulerContext().IsLocked());
+
+    s32 num_waiters{};
+    Thread* next_lock_owner{};
+    auto it = waiter_list.begin();
+    while (it != waiter_list.end()) {
+        if (it->GetAddressKey() == key) {
+            Thread* thread = std::addressof(*it);
+
+            // Keep track of how many kernel waiters we have.
+            if (Memory::IsKernelAddressKey(thread->GetAddressKey())) {
+                ASSERT((num_kernel_waiters--) > 0);
+            }
+            it = waiter_list.erase(it);
+
+            // Update the next lock owner.
+            if (next_lock_owner == nullptr) {
+                next_lock_owner = thread;
+                next_lock_owner->SetLockOwner(nullptr);
+            } else {
+                next_lock_owner->AddWaiterImpl(thread);
+            }
+            num_waiters++;
+        } else {
+            it++;
         }
     }
 
-    if (new_priority == current_priority) {
-        return;
+    // Do priority updates, if we have a next owner.
+    if (next_lock_owner) {
+        RestorePriority(kernel, this);
+        RestorePriority(kernel, next_lock_owner);
     }
 
-    if (GetStatus() == ThreadStatus::WaitCondVar) {
-        owner_process->RemoveConditionVariableThread(SharedFrom(this));
-    }
-
-    SetCurrentPriority(new_priority);
-
-    if (GetStatus() == ThreadStatus::WaitCondVar) {
-        owner_process->InsertConditionVariableThread(SharedFrom(this));
-    }
-
-    if (!lock_owner) {
-        return;
-    }
-
-    // Ensure that the thread is within the correct location in the waiting list.
-    auto old_owner = lock_owner;
-    lock_owner->RemoveMutexWaiter(SharedFrom(this));
-    old_owner->AddMutexWaiter(SharedFrom(this));
-
-    // Recursively update the priority of the thread that depends on the priority of this one.
-    lock_owner->UpdatePriority();
-}
-
-bool Thread::AllSynchronizationObjectsReady() const {
-    return std::none_of(wait_objects->begin(), wait_objects->end(),
-                        [this](const std::shared_ptr<SynchronizationObject>& object) {
-                            return object->ShouldWait(this);
-                        });
-}
-
-bool Thread::InvokeHLECallback(std::shared_ptr<Thread> thread) {
-    ASSERT(hle_callback);
-    return hle_callback(std::move(thread));
+    // Return output.
+    *out_num_waiters = num_waiters;
+    return next_lock_owner;
 }
 
 ResultCode Thread::SetActivity(ThreadActivity value) {
     KScopedSchedulerLock lock(kernel);
 
-    auto sched_status = GetSchedulingStatus();
+    auto sched_status = GetState();
 
-    if (sched_status != ThreadSchedStatus::Runnable && sched_status != ThreadSchedStatus::Paused) {
+    if (sched_status != ThreadState::Runnable && sched_status != ThreadState::Waiting) {
         return ERR_INVALID_STATE;
     }
 
-    if (IsPendingTermination()) {
+    if (IsTerminationRequested()) {
         return RESULT_SUCCESS;
     }
 
@@ -394,7 +385,7 @@ ResultCode Thread::Sleep(s64 nanoseconds) {
     Handle event_handle{};
     {
         KScopedSchedulerLockAndSleep lock(kernel, event_handle, this, nanoseconds);
-        SetStatus(ThreadStatus::WaitSleep);
+        SetState(ThreadState::Waiting);
     }
 
     if (event_handle != InvalidHandle) {
@@ -405,34 +396,21 @@ ResultCode Thread::Sleep(s64 nanoseconds) {
 }
 
 void Thread::AddSchedulingFlag(ThreadSchedFlags flag) {
-    const u32 old_state = scheduling_state;
+    const auto old_state = GetRawState();
     pausing_state |= static_cast<u32>(flag);
-    const u32 base_scheduling = static_cast<u32>(GetSchedulingStatus());
-    scheduling_state = base_scheduling | pausing_state;
+    const auto base_scheduling = GetState();
+    thread_state = base_scheduling | static_cast<ThreadState>(pausing_state);
     KScheduler::OnThreadStateChanged(kernel, this, old_state);
 }
 
 void Thread::RemoveSchedulingFlag(ThreadSchedFlags flag) {
-    const u32 old_state = scheduling_state;
+    const auto old_state = GetRawState();
     pausing_state &= ~static_cast<u32>(flag);
-    const u32 base_scheduling = static_cast<u32>(GetSchedulingStatus());
-    scheduling_state = base_scheduling | pausing_state;
+    const auto base_scheduling = GetState();
+    thread_state = base_scheduling | static_cast<ThreadState>(pausing_state);
     KScheduler::OnThreadStateChanged(kernel, this, old_state);
 }
 
-void Thread::SetSchedulingStatus(ThreadSchedStatus new_status) {
-    const u32 old_state = scheduling_state;
-    scheduling_state = (scheduling_state & static_cast<u32>(ThreadSchedMasks::HighMask)) |
-                       static_cast<u32>(new_status);
-    KScheduler::OnThreadStateChanged(kernel, this, old_state);
-}
-
-void Thread::SetCurrentPriority(u32 new_priority) {
-    const u32 old_priority = std::exchange(current_priority, new_priority);
-    KScheduler::OnThreadPriorityChanged(kernel, this, kernel.CurrentScheduler()->GetCurrentThread(),
-                                        old_priority);
-}
-
 ResultCode Thread::SetCoreAndAffinityMask(s32 new_core, u64 new_affinity_mask) {
     KScopedSchedulerLock lock(kernel);
     const auto HighestSetCore = [](u64 mask, u32 max_cores) {
diff --git a/src/core/hle/kernel/thread.h b/src/core/hle/kernel/thread.h
index 11ef29888..bf6bca2e4 100755
--- a/src/core/hle/kernel/thread.h
+++ b/src/core/hle/kernel/thread.h
@@ -10,12 +10,16 @@
 #include <utility>
 #include <vector>
 
+#include <boost/intrusive/list.hpp>
+
 #include "common/common_types.h"
+#include "common/intrusive_red_black_tree.h"
 #include "common/spin_lock.h"
 #include "core/arm/arm_interface.h"
 #include "core/hle/kernel/k_affinity_mask.h"
+#include "core/hle/kernel/k_synchronization_object.h"
 #include "core/hle/kernel/object.h"
-#include "core/hle/kernel/synchronization_object.h"
+#include "core/hle/kernel/svc_common.h"
 #include "core/hle/result.h"
 
 namespace Common {
@@ -73,19 +77,24 @@ enum ThreadProcessorId : s32 {
                                      (1 << THREADPROCESSORID_2) | (1 << THREADPROCESSORID_3)
 };
 
-enum class ThreadStatus {
-    Ready,        ///< Ready to run
-    Paused,       ///< Paused by SetThreadActivity or debug
-    WaitHLEEvent, ///< Waiting for hle event to finish
-    WaitSleep,    ///< Waiting due to a SleepThread SVC
-    WaitIPC,      ///< Waiting for the reply from an IPC request
-    WaitSynch,    ///< Waiting due to WaitSynchronization
-    WaitMutex,    ///< Waiting due to an ArbitrateLock svc
-    WaitCondVar,  ///< Waiting due to an WaitProcessWideKey svc
-    WaitArb,      ///< Waiting due to a SignalToAddress/WaitForAddress svc
-    Dormant,      ///< Created but not yet made ready
-    Dead          ///< Run to completion, or forcefully terminated
+enum class ThreadState : u16 {
+    Initialized = 0,
+    Waiting = 1,
+    Runnable = 2,
+    Terminated = 3,
+
+    SuspendShift = 4,
+    Mask = (1 << SuspendShift) - 1,
+
+    ProcessSuspended = (1 << (0 + SuspendShift)),
+    ThreadSuspended = (1 << (1 + SuspendShift)),
+    DebugSuspended = (1 << (2 + SuspendShift)),
+    BacktraceSuspended = (1 << (3 + SuspendShift)),
+    InitSuspended = (1 << (4 + SuspendShift)),
+
+    SuspendFlagMask = ((1 << 5) - 1) << SuspendShift,
 };
+DECLARE_ENUM_FLAG_OPERATORS(ThreadState);
 
 enum class ThreadWakeupReason {
     Signal, // The thread was woken up by WakeupAllWaitingThreads due to an object signal.
@@ -97,13 +106,6 @@ enum class ThreadActivity : u32 {
     Paused = 1,
 };
 
-enum class ThreadSchedStatus : u32 {
-    None = 0,
-    Paused = 1,
-    Runnable = 2,
-    Exited = 3,
-};
-
 enum class ThreadSchedFlags : u32 {
     ProcessPauseFlag = 1 << 4,
     ThreadPauseFlag = 1 << 5,
@@ -111,13 +113,10 @@ enum class ThreadSchedFlags : u32 {
     KernelInitPauseFlag = 1 << 8,
 };
 
-enum class ThreadSchedMasks : u32 {
-    LowMask = 0x000f,
-    HighMask = 0xfff0,
-    ForcePauseMask = 0x0070,
-};
+class Thread final : public KSynchronizationObject, public boost::intrusive::list_base_hook<> {
+    friend class KScheduler;
+    friend class Process;
 
-class Thread final : public SynchronizationObject {
 public:
     explicit Thread(KernelCore& kernel);
     ~Thread() override;
@@ -127,10 +126,6 @@ public:
     using ThreadContext32 = Core::ARM_Interface::ThreadContext32;
     using ThreadContext64 = Core::ARM_Interface::ThreadContext64;
 
-    using ThreadSynchronizationObjects = std::vector<std::shared_ptr<SynchronizationObject>>;
-
-    using HLECallback = std::function<bool(std::shared_ptr<Thread> thread)>;
-
     /**
      * Creates and returns a new thread. The new thread is immediately scheduled
      * @param system The instance of the whole system
@@ -186,59 +181,54 @@ public:
         return HANDLE_TYPE;
     }
 
-    bool ShouldWait(const Thread* thread) const override;
-    void Acquire(Thread* thread) override;
-    bool IsSignaled() const override;
-
     /**
      * Gets the thread's current priority
      * @return The current thread's priority
      */
-    u32 GetPriority() const {
+    [[nodiscard]] s32 GetPriority() const {
         return current_priority;
     }
 
+    /**
+     * Sets the thread's current priority.
+     * @param priority The new priority.
+     */
+    void SetPriority(s32 priority) {
+        current_priority = priority;
+    }
+
     /**
      * Gets the thread's nominal priority.
      * @return The current thread's nominal priority.
      */
-    u32 GetNominalPriority() const {
-        return nominal_priority;
+    [[nodiscard]] s32 GetBasePriority() const {
+        return base_priority;
     }
 
     /**
-     * Sets the thread's current priority
-     * @param priority The new priority
+     * Sets the thread's nominal priority.
+     * @param priority The new priority.
      */
-    void SetPriority(u32 priority);
-
-    /// Adds a thread to the list of threads that are waiting for a lock held by this thread.
-    void AddMutexWaiter(std::shared_ptr<Thread> thread);
-
-    /// Removes a thread from the list of threads that are waiting for a lock held by this thread.
-    void RemoveMutexWaiter(std::shared_ptr<Thread> thread);
-
-    /// Recalculates the current priority taking into account priority inheritance.
-    void UpdatePriority();
+    void SetBasePriority(u32 priority);
 
     /// Changes the core that the thread is running or scheduled to run on.
-    ResultCode SetCoreAndAffinityMask(s32 new_core, u64 new_affinity_mask);
+    [[nodiscard]] ResultCode SetCoreAndAffinityMask(s32 new_core, u64 new_affinity_mask);
 
     /**
      * Gets the thread's thread ID
      * @return The thread's ID
      */
-    u64 GetThreadID() const {
+    [[nodiscard]] u64 GetThreadID() const {
         return thread_id;
     }
 
     /// Resumes a thread from waiting
-    void ResumeFromWait();
-
-    void OnWakeUp();
+    void Wakeup();
 
     ResultCode Start();
 
+    virtual bool IsSignaled() const override;
+
     /// Cancels a waiting operation that this thread may or may not be within.
     ///
     /// When the thread is within a waiting state, this will set the thread's
@@ -247,30 +237,21 @@ public:
     ///
     void CancelWait();
 
-    void SetSynchronizationResults(SynchronizationObject* object, ResultCode result);
+    void SetSynchronizationResults(KSynchronizationObject* object, ResultCode result);
 
-    SynchronizationObject* GetSignalingObject() const {
-        return signaling_object;
+    void SetSyncedObject(KSynchronizationObject* object, ResultCode result) {
+        SetSynchronizationResults(object, result);
+    }
+
+    ResultCode GetWaitResult(KSynchronizationObject** out) const {
+        *out = signaling_object;
+        return signaling_result;
     }
 
     ResultCode GetSignalingResult() const {
         return signaling_result;
     }
 
-    /**
-     * Retrieves the index that this particular object occupies in the list of objects
-     * that the thread passed to WaitSynchronization, starting the search from the last element.
-     *
-     * It is used to set the output index of WaitSynchronization when the thread is awakened.
-     *
-     * When a thread wakes up due to an object signal, the kernel will use the index of the last
-     * matching object in the wait objects list in case of having multiple instances of the same
-     * object in the list.
-     *
-     * @param object Object to query the index of.
-     */
-    s32 GetSynchronizationObjectIndex(std::shared_ptr<SynchronizationObject> object) const;
-
     /**
      * Stops a thread, invalidating it from further use
      */
@@ -341,18 +322,22 @@ public:
 
     std::shared_ptr<Common::Fiber>& GetHostContext();
 
-    ThreadStatus GetStatus() const {
-        return status;
+    ThreadState GetState() const {
+        return thread_state & ThreadState::Mask;
     }
 
-    void SetStatus(ThreadStatus new_status);
+    ThreadState GetRawState() const {
+        return thread_state;
+    }
+
+    void SetState(ThreadState state);
 
     s64 GetLastScheduledTick() const {
-        return this->last_scheduled_tick;
+        return last_scheduled_tick;
     }
 
     void SetLastScheduledTick(s64 tick) {
-        this->last_scheduled_tick = tick;
+        last_scheduled_tick = tick;
     }
 
     u64 GetTotalCPUTimeTicks() const {
@@ -387,98 +372,18 @@ public:
         return owner_process;
     }
 
-    const ThreadSynchronizationObjects& GetSynchronizationObjects() const {
-        return *wait_objects;
-    }
-
-    void SetSynchronizationObjects(ThreadSynchronizationObjects* objects) {
-        wait_objects = objects;
-    }
-
-    void ClearSynchronizationObjects() {
-        for (const auto& waiting_object : *wait_objects) {
-            waiting_object->RemoveWaitingThread(SharedFrom(this));
-        }
-        wait_objects->clear();
-    }
-
-    /// Determines whether all the objects this thread is waiting on are ready.
-    bool AllSynchronizationObjectsReady() const;
-
     const MutexWaitingThreads& GetMutexWaitingThreads() const {
         return wait_mutex_threads;
     }
 
     Thread* GetLockOwner() const {
-        return lock_owner.get();
+        return lock_owner;
     }
 
-    void SetLockOwner(std::shared_ptr<Thread> owner) {
-        lock_owner = std::move(owner);
+    void SetLockOwner(Thread* owner) {
+        lock_owner = owner;
     }
 
-    VAddr GetCondVarWaitAddress() const {
-        return condvar_wait_address;
-    }
-
-    void SetCondVarWaitAddress(VAddr address) {
-        condvar_wait_address = address;
-    }
-
-    VAddr GetMutexWaitAddress() const {
-        return mutex_wait_address;
-    }
-
-    void SetMutexWaitAddress(VAddr address) {
-        mutex_wait_address = address;
-    }
-
-    Handle GetWaitHandle() const {
-        return wait_handle;
-    }
-
-    void SetWaitHandle(Handle handle) {
-        wait_handle = handle;
-    }
-
-    VAddr GetArbiterWaitAddress() const {
-        return arb_wait_address;
-    }
-
-    void SetArbiterWaitAddress(VAddr address) {
-        arb_wait_address = address;
-    }
-
-    bool HasHLECallback() const {
-        return hle_callback != nullptr;
-    }
-
-    void SetHLECallback(HLECallback callback) {
-        hle_callback = std::move(callback);
-    }
-
-    void SetHLETimeEvent(Handle time_event) {
-        hle_time_event = time_event;
-    }
-
-    void SetHLESyncObject(SynchronizationObject* object) {
-        hle_object = object;
-    }
-
-    Handle GetHLETimeEvent() const {
-        return hle_time_event;
-    }
-
-    SynchronizationObject* GetHLESyncObject() const {
-        return hle_object;
-    }
-
-    void InvalidateHLECallback() {
-        SetHLECallback(nullptr);
-    }
-
-    bool InvokeHLECallback(std::shared_ptr<Thread> thread);
-
     u32 GetIdealCore() const {
         return ideal_core;
     }
@@ -493,20 +398,11 @@ public:
     ResultCode Sleep(s64 nanoseconds);
 
     s64 GetYieldScheduleCount() const {
-        return this->schedule_count;
+        return schedule_count;
     }
 
     void SetYieldScheduleCount(s64 count) {
-        this->schedule_count = count;
-    }
-
-    ThreadSchedStatus GetSchedulingStatus() const {
-        return static_cast<ThreadSchedStatus>(scheduling_state &
-                                              static_cast<u32>(ThreadSchedMasks::LowMask));
-    }
-
-    bool IsRunnable() const {
-        return scheduling_state == static_cast<u32>(ThreadSchedStatus::Runnable);
+        schedule_count = count;
     }
 
     bool IsRunning() const {
@@ -517,36 +413,32 @@ public:
         is_running = value;
     }
 
-    bool IsSyncCancelled() const {
+    bool IsWaitCancelled() const {
         return is_sync_cancelled;
     }
 
-    void SetSyncCancelled(bool value) {
-        is_sync_cancelled = value;
+    void ClearWaitCancelled() {
+        is_sync_cancelled = false;
     }
 
     Handle GetGlobalHandle() const {
         return global_handle;
     }
 
-    bool IsWaitingForArbitration() const {
-        return waiting_for_arbitration;
+    bool IsCancellable() const {
+        return is_cancellable;
     }
 
-    void WaitForArbitration(bool set) {
-        waiting_for_arbitration = set;
+    void SetCancellable() {
+        is_cancellable = true;
     }
 
-    bool IsWaitingSync() const {
-        return is_waiting_on_sync;
+    void ClearCancellable() {
+        is_cancellable = false;
     }
 
-    void SetWaitingSync(bool is_waiting) {
-        is_waiting_on_sync = is_waiting;
-    }
-
-    bool IsPendingTermination() const {
-        return will_be_terminated || GetSchedulingStatus() == ThreadSchedStatus::Exited;
+    bool IsTerminationRequested() const {
+        return will_be_terminated || GetRawState() == ThreadState::Terminated;
     }
 
     bool IsPaused() const {
@@ -578,21 +470,21 @@ public:
         constexpr QueueEntry() = default;
 
         constexpr void Initialize() {
-            this->prev = nullptr;
-            this->next = nullptr;
+            prev = nullptr;
+            next = nullptr;
         }
 
         constexpr Thread* GetPrev() const {
-            return this->prev;
+            return prev;
         }
         constexpr Thread* GetNext() const {
-            return this->next;
+            return next;
         }
         constexpr void SetPrev(Thread* thread) {
-            this->prev = thread;
+            prev = thread;
         }
         constexpr void SetNext(Thread* thread) {
-            this->next = thread;
+            next = thread;
         }
 
     private:
@@ -601,11 +493,11 @@ public:
     };
 
     QueueEntry& GetPriorityQueueEntry(s32 core) {
-        return this->per_core_priority_queue_entry[core];
+        return per_core_priority_queue_entry[core];
     }
 
     const QueueEntry& GetPriorityQueueEntry(s32 core) const {
-        return this->per_core_priority_queue_entry[core];
+        return per_core_priority_queue_entry[core];
     }
 
     s32 GetDisableDispatchCount() const {
@@ -622,24 +514,162 @@ public:
         disable_count--;
     }
 
-private:
-    friend class GlobalSchedulerContext;
-    friend class KScheduler;
-    friend class Process;
+    void SetWaitObjectsForDebugging(KSynchronizationObject** objects, s32 num_objects) {
+        wait_objects_for_debugging.clear();
+        wait_objects_for_debugging.reserve(num_objects);
+        for (auto i = 0; i < num_objects; ++i) {
+            wait_objects_for_debugging.emplace_back(objects[i]);
+        }
+    }
 
-    void SetSchedulingStatus(ThreadSchedStatus new_status);
+    [[nodiscard]] const std::vector<KSynchronizationObject*>& GetWaitObjectsForDebugging() const {
+        return wait_objects_for_debugging;
+    }
+
+    void SetMutexWaitAddressForDebugging(VAddr address) {
+        mutex_wait_address_for_debugging = address;
+    }
+
+    [[nodiscard]] VAddr GetMutexWaitAddressForDebugging() const {
+        return mutex_wait_address_for_debugging;
+    }
+
+    void AddWaiter(Thread* thread);
+
+    void RemoveWaiter(Thread* thread);
+
+    [[nodiscard]] Thread* RemoveWaiterByKey(s32* out_num_waiters, VAddr key);
+
+    [[nodiscard]] VAddr GetAddressKey() const {
+        return address_key;
+    }
+
+    [[nodiscard]] u32 GetAddressKeyValue() const {
+        return address_key_value;
+    }
+
+    [[nodiscard]] void SetAddressKey(VAddr key) {
+        address_key = key;
+    }
+
+    [[nodiscard]] void SetAddressKey(VAddr key, u32 val) {
+        address_key = key;
+        address_key_value = val;
+    }
+
+private:
+    static constexpr size_t PriorityInheritanceCountMax = 10;
+    union SyncObjectBuffer {
+        std::array<KSynchronizationObject*, Svc::ArgumentHandleCountMax> sync_objects;
+        std::array<Handle,
+                   Svc::ArgumentHandleCountMax*(sizeof(KSynchronizationObject*) / sizeof(Handle))>
+            handles;
+        constexpr SyncObjectBuffer() : sync_objects() {}
+    };
+    static_assert(sizeof(SyncObjectBuffer::sync_objects) == sizeof(SyncObjectBuffer::handles));
+
+    struct ConditionVariableComparator {
+        struct LightCompareType {
+            u64 cv_key{};
+            s32 priority{};
+
+            [[nodiscard]] constexpr u64 GetConditionVariableKey() const {
+                return cv_key;
+            }
+
+            [[nodiscard]] constexpr s32 GetPriority() const {
+                return priority;
+            }
+        };
+
+        template <typename T>
+        requires(
+            std::same_as<T, Thread> ||
+            std::same_as<T, LightCompareType>) static constexpr int Compare(const T& lhs,
+                                                                            const Thread& rhs) {
+            const uintptr_t l_key = lhs.GetConditionVariableKey();
+            const uintptr_t r_key = rhs.GetConditionVariableKey();
+
+            if (l_key < r_key) {
+                // Sort first by key
+                return -1;
+            } else if (l_key == r_key && lhs.GetPriority() < rhs.GetPriority()) {
+                // And then by priority.
+                return -1;
+            } else {
+                return 1;
+            }
+        }
+    };
+
+    Common::IntrusiveRedBlackTreeNode condvar_arbiter_tree_node{};
+
+    using ConditionVariableThreadTreeTraits =
+        Common::IntrusiveRedBlackTreeMemberTraitsDeferredAssert<&Thread::condvar_arbiter_tree_node>;
+    using ConditionVariableThreadTree =
+        ConditionVariableThreadTreeTraits::TreeType<ConditionVariableComparator>;
+
+public:
+    using ConditionVariableThreadTreeType = ConditionVariableThreadTree;
+
+    [[nodiscard]] uintptr_t GetConditionVariableKey() const {
+        return condvar_key;
+    }
+
+    [[nodiscard]] uintptr_t GetAddressArbiterKey() const {
+        return condvar_key;
+    }
+
+    void SetConditionVariable(ConditionVariableThreadTree* tree, VAddr address, uintptr_t cv_key,
+                              u32 value) {
+        condvar_tree = tree;
+        condvar_key = cv_key;
+        address_key = address;
+        address_key_value = value;
+    }
+
+    void ClearConditionVariable() {
+        condvar_tree = nullptr;
+    }
+
+    [[nodiscard]] bool IsWaitingForConditionVariable() const {
+        return condvar_tree != nullptr;
+    }
+
+    void SetAddressArbiter(ConditionVariableThreadTree* tree, uintptr_t address) {
+        condvar_tree = tree;
+        condvar_key = address;
+    }
+
+    void ClearAddressArbiter() {
+        condvar_tree = nullptr;
+    }
+
+    [[nodiscard]] bool IsWaitingForAddressArbiter() const {
+        return condvar_tree != nullptr;
+    }
+
+    [[nodiscard]] ConditionVariableThreadTree* GetConditionVariableTree() const {
+        return condvar_tree;
+    }
+
+    [[nodiscard]] bool HasWaiters() const {
+        return !waiter_list.empty();
+    }
+
+private:
     void AddSchedulingFlag(ThreadSchedFlags flag);
     void RemoveSchedulingFlag(ThreadSchedFlags flag);
-
-    void SetCurrentPriority(u32 new_priority);
+    void AddWaiterImpl(Thread* thread);
+    void RemoveWaiterImpl(Thread* thread);
+    static void RestorePriority(KernelCore& kernel, Thread* thread);
 
     Common::SpinLock context_guard{};
     ThreadContext32 context_32{};
     ThreadContext64 context_64{};
     std::shared_ptr<Common::Fiber> host_context{};
 
-    ThreadStatus status = ThreadStatus::Dormant;
-    u32 scheduling_state = 0;
+    ThreadState thread_state = ThreadState::Initialized;
 
     u64 thread_id = 0;
 
@@ -652,11 +682,11 @@ private:
     /// Nominal thread priority, as set by the emulated application.
     /// The nominal priority is the thread priority without priority
     /// inheritance taken into account.
-    u32 nominal_priority = 0;
+    s32 base_priority{};
 
     /// Current thread priority. This may change over the course of the
     /// thread's lifetime in order to facilitate priority inheritance.
-    u32 current_priority = 0;
+    s32 current_priority{};
 
     u64 total_cpu_time_ticks = 0; ///< Total CPU running ticks.
     s64 schedule_count{};
@@ -671,37 +701,24 @@ private:
     Process* owner_process;
 
     /// Objects that the thread is waiting on, in the same order as they were
-    /// passed to WaitSynchronization.
-    ThreadSynchronizationObjects* wait_objects;
+    /// passed to WaitSynchronization. This is used for debugging only.
+    std::vector<KSynchronizationObject*> wait_objects_for_debugging;
 
-    SynchronizationObject* signaling_object;
+    /// The current mutex wait address. This is used for debugging only.
+    VAddr mutex_wait_address_for_debugging{};
+
+    KSynchronizationObject* signaling_object;
     ResultCode signaling_result{RESULT_SUCCESS};
 
     /// List of threads that are waiting for a mutex that is held by this thread.
     MutexWaitingThreads wait_mutex_threads;
 
     /// Thread that owns the lock that this thread is waiting for.
-    std::shared_ptr<Thread> lock_owner;
-
-    /// If waiting on a ConditionVariable, this is the ConditionVariable address
-    VAddr condvar_wait_address = 0;
-    /// If waiting on a Mutex, this is the mutex address
-    VAddr mutex_wait_address = 0;
-    /// The handle used to wait for the mutex.
-    Handle wait_handle = 0;
-
-    /// If waiting for an AddressArbiter, this is the address being waited on.
-    VAddr arb_wait_address{0};
-    bool waiting_for_arbitration{};
+    Thread* lock_owner{};
 
     /// Handle used as userdata to reference this object when inserting into the CoreTiming queue.
     Handle global_handle = 0;
 
-    /// Callback for HLE Events
-    HLECallback hle_callback;
-    Handle hle_time_event;
-    SynchronizationObject* hle_object;
-
     KScheduler* scheduler = nullptr;
 
     std::array<QueueEntry, Core::Hardware::NUM_CPU_CORES> per_core_priority_queue_entry{};
@@ -714,7 +731,7 @@ private:
 
     u32 pausing_state = 0;
     bool is_running = false;
-    bool is_waiting_on_sync = false;
+    bool is_cancellable = false;
     bool is_sync_cancelled = false;
 
     bool is_continuous_on_svc = false;
@@ -725,6 +742,18 @@ private:
 
     bool was_running = false;
 
+    bool signaled{};
+
+    ConditionVariableThreadTree* condvar_tree{};
+    uintptr_t condvar_key{};
+    VAddr address_key{};
+    u32 address_key_value{};
+    s32 num_kernel_waiters{};
+
+    using WaiterList = boost::intrusive::list<Thread>;
+    WaiterList waiter_list{};
+    WaiterList pinned_waiter_list{};
+
     std::string name;
 };
 
diff --git a/src/core/hle/kernel/time_manager.cpp b/src/core/hle/kernel/time_manager.cpp
index 79628e2b4..832edd629 100755
--- a/src/core/hle/kernel/time_manager.cpp
+++ b/src/core/hle/kernel/time_manager.cpp
@@ -18,12 +18,10 @@ TimeManager::TimeManager(Core::System& system_) : system{system_} {
     time_manager_event_type = Core::Timing::CreateEvent(
         "Kernel::TimeManagerCallback",
         [this](std::uintptr_t thread_handle, std::chrono::nanoseconds) {
-            const KScopedSchedulerLock lock(system.Kernel());
-            const auto proper_handle = static_cast<Handle>(thread_handle);
-
             std::shared_ptr<Thread> thread;
             {
                 std::lock_guard lock{mutex};
+                const auto proper_handle = static_cast<Handle>(thread_handle);
                 if (cancelled_events[proper_handle]) {
                     return;
                 }
@@ -32,7 +30,7 @@ TimeManager::TimeManager(Core::System& system_) : system{system_} {
 
             if (thread) {
                 // Thread can be null if process has exited
-                thread->OnWakeUp();
+                thread->Wakeup();
             }
         });
 }
@@ -42,8 +40,7 @@ void TimeManager::ScheduleTimeEvent(Handle& event_handle, Thread* timetask, s64
     event_handle = timetask->GetGlobalHandle();
     if (nanoseconds > 0) {
         ASSERT(timetask);
-        ASSERT(timetask->GetStatus() != ThreadStatus::Ready);
-        ASSERT(timetask->GetStatus() != ThreadStatus::WaitMutex);
+        ASSERT(timetask->GetState() != ThreadState::Runnable);
         system.CoreTiming().ScheduleEvent(std::chrono::nanoseconds{nanoseconds},
                                           time_manager_event_type, event_handle);
     } else {
diff --git a/src/core/hle/service/nfp/nfp.cpp b/src/core/hle/service/nfp/nfp.cpp
index 5557da72e..641bcadea 100755
--- a/src/core/hle/service/nfp/nfp.cpp
+++ b/src/core/hle/service/nfp/nfp.cpp
@@ -190,12 +190,6 @@ private:
     void GetDeviceState(Kernel::HLERequestContext& ctx) {
         LOG_DEBUG(Service_NFP, "called");
 
-        auto nfc_event = nfp_interface.GetNFCEvent();
-        if (!nfc_event->ShouldWait(&ctx.GetThread()) && !has_attached_handle) {
-            device_state = DeviceState::TagFound;
-            nfc_event->Clear();
-        }
-
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.Push<u32>(static_cast<u32>(device_state));
diff --git a/src/core/hle/service/nvflinger/nvflinger.cpp b/src/core/hle/service/nvflinger/nvflinger.cpp
index 4b3581949..ceaa93d28 100755
--- a/src/core/hle/service/nvflinger/nvflinger.cpp
+++ b/src/core/hle/service/nvflinger/nvflinger.cpp
@@ -38,6 +38,10 @@ void NVFlinger::SplitVSync() {
     system.RegisterHostThread();
     std::string name = "yuzu:VSyncThread";
     MicroProfileOnThreadCreate(name.c_str());
+
+    // Cleanup
+    SCOPE_EXIT({ MicroProfileOnThreadExit(); });
+
     Common::SetCurrentThreadName(name.c_str());
     Common::SetCurrentThreadPriority(Common::ThreadPriority::High);
     s64 delay = 0;
diff --git a/src/core/hle/service/sm/sm.cpp b/src/core/hle/service/sm/sm.cpp
index 4da69f503..2b91a89d1 100755
--- a/src/core/hle/service/sm/sm.cpp
+++ b/src/core/hle/service/sm/sm.cpp
@@ -139,9 +139,6 @@ void SM::GetService(Kernel::HLERequestContext& ctx) {
         server_port->AppendPendingSession(server);
     }
 
-    // Wake the threads waiting on the ServerPort
-    server_port->Signal();
-
     LOG_DEBUG(Service_SM, "called service={} -> session={}", name, client->GetObjectId());
     IPC::ResponseBuilder rb{ctx, 2, 0, 1, IPC::ResponseBuilder::Flags::AlwaysMoveHandles};
     rb.Push(RESULT_SUCCESS);
diff --git a/src/input_common/sdl/sdl_impl.cpp b/src/input_common/sdl/sdl_impl.cpp
index 0b531f698..d32eb732a 100755
--- a/src/input_common/sdl/sdl_impl.cpp
+++ b/src/input_common/sdl/sdl_impl.cpp
@@ -1030,11 +1030,44 @@ public:
         }
         return {};
     }
-    [[nodiscard]] std::optional<Common::ParamPackage> FromEvent(const SDL_Event& event) const {
+    [[nodiscard]] std::optional<Common::ParamPackage> FromEvent(SDL_Event& event) {
         switch (event.type) {
         case SDL_JOYAXISMOTION:
-            if (std::abs(event.jaxis.value / 32767.0) < 0.5) {
+            if (!axis_memory.count(event.jaxis.which) ||
+                !axis_memory[event.jaxis.which].count(event.jaxis.axis)) {
+                axis_memory[event.jaxis.which][event.jaxis.axis] = event.jaxis.value;
+                axis_event_count[event.jaxis.which][event.jaxis.axis] = 1;
                 break;
+            } else {
+                axis_event_count[event.jaxis.which][event.jaxis.axis]++;
+                // The joystick and axis exist in our map if we take this branch, so no checks
+                // needed
+                if (std::abs(
+                        (event.jaxis.value - axis_memory[event.jaxis.which][event.jaxis.axis]) /
+                        32767.0) < 0.5) {
+                    break;
+                } else {
+                    if (axis_event_count[event.jaxis.which][event.jaxis.axis] == 2 &&
+                        IsAxisAtPole(event.jaxis.value) &&
+                        IsAxisAtPole(axis_memory[event.jaxis.which][event.jaxis.axis])) {
+                        // If we have exactly two events and both are near a pole, this is
+                        // likely a digital input masquerading as an analog axis; Instead of
+                        // trying to look at the direction the axis travelled, assume the first
+                        // event was press and the second was release; This should handle most
+                        // digital axes while deferring to the direction of travel for analog
+                        // axes
+                        event.jaxis.value = static_cast<Sint16>(
+                            std::copysign(32767, axis_memory[event.jaxis.which][event.jaxis.axis]));
+                    } else {
+                        // There are more than two events, so this is likely a true analog axis,
+                        // check the direction it travelled
+                        event.jaxis.value = static_cast<Sint16>(std::copysign(
+                            32767,
+                            event.jaxis.value - axis_memory[event.jaxis.which][event.jaxis.axis]));
+                    }
+                    axis_memory.clear();
+                    axis_event_count.clear();
+                }
             }
             [[fallthrough]];
         case SDL_JOYBUTTONUP:
@@ -1043,6 +1076,16 @@ public:
         }
         return std::nullopt;
     }
+
+private:
+    // Determine whether an axis value is close to an extreme or center
+    // Some controllers have a digital D-Pad as a pair of analog sticks, with 3 possible values per
+    // axis, which is why the center must be considered a pole
+    bool IsAxisAtPole(int16_t value) const {
+        return std::abs(value) >= 32767 || std::abs(value) < 327;
+    }
+    std::unordered_map<SDL_JoystickID, std::unordered_map<uint8_t, int16_t>> axis_memory;
+    std::unordered_map<SDL_JoystickID, std::unordered_map<uint8_t, uint32_t>> axis_event_count;
 };
 
 class SDLMotionPoller final : public SDLPoller {
diff --git a/src/yuzu/debugger/wait_tree.cpp b/src/yuzu/debugger/wait_tree.cpp
index 0925c10b4..3ded85720 100755
--- a/src/yuzu/debugger/wait_tree.cpp
+++ b/src/yuzu/debugger/wait_tree.cpp
@@ -14,10 +14,10 @@
 #include "core/core.h"
 #include "core/hle/kernel/handle_table.h"
 #include "core/hle/kernel/k_scheduler.h"
-#include "core/hle/kernel/mutex.h"
+#include "core/hle/kernel/k_synchronization_object.h"
 #include "core/hle/kernel/process.h"
 #include "core/hle/kernel/readable_event.h"
-#include "core/hle/kernel/synchronization_object.h"
+#include "core/hle/kernel/svc_common.h"
 #include "core/hle/kernel/thread.h"
 #include "core/memory.h"
 
@@ -116,7 +116,7 @@ QString WaitTreeText::GetText() const {
 WaitTreeMutexInfo::WaitTreeMutexInfo(VAddr mutex_address, const Kernel::HandleTable& handle_table)
     : mutex_address(mutex_address) {
     mutex_value = Core::System::GetInstance().Memory().Read32(mutex_address);
-    owner_handle = static_cast<Kernel::Handle>(mutex_value & Kernel::Mutex::MutexOwnerMask);
+    owner_handle = static_cast<Kernel::Handle>(mutex_value & Kernel::Svc::HandleWaitMask);
     owner = handle_table.Get<Kernel::Thread>(owner_handle);
 }
 
@@ -127,7 +127,7 @@ QString WaitTreeMutexInfo::GetText() const {
 }
 
 std::vector<std::unique_ptr<WaitTreeItem>> WaitTreeMutexInfo::GetChildren() const {
-    const bool has_waiters = (mutex_value & Kernel::Mutex::MutexHasWaitersFlag) != 0;
+    const bool has_waiters = (mutex_value & Kernel::Svc::HandleWaitMask) != 0;
 
     std::vector<std::unique_ptr<WaitTreeItem>> list;
     list.push_back(std::make_unique<WaitTreeText>(tr("has waiters: %1").arg(has_waiters)));
@@ -169,7 +169,8 @@ std::vector<std::unique_ptr<WaitTreeItem>> WaitTreeCallstack::GetChildren() cons
     return list;
 }
 
-WaitTreeSynchronizationObject::WaitTreeSynchronizationObject(const Kernel::SynchronizationObject& o)
+WaitTreeSynchronizationObject::WaitTreeSynchronizationObject(
+    const Kernel::KSynchronizationObject& o)
     : object(o) {}
 WaitTreeSynchronizationObject::~WaitTreeSynchronizationObject() = default;
 
@@ -188,7 +189,7 @@ QString WaitTreeSynchronizationObject::GetText() const {
 }
 
 std::unique_ptr<WaitTreeSynchronizationObject> WaitTreeSynchronizationObject::make(
-    const Kernel::SynchronizationObject& object) {
+    const Kernel::KSynchronizationObject& object) {
     switch (object.GetHandleType()) {
     case Kernel::HandleType::ReadableEvent:
         return std::make_unique<WaitTreeEvent>(static_cast<const Kernel::ReadableEvent&>(object));
@@ -202,7 +203,7 @@ std::unique_ptr<WaitTreeSynchronizationObject> WaitTreeSynchronizationObject::ma
 std::vector<std::unique_ptr<WaitTreeItem>> WaitTreeSynchronizationObject::GetChildren() const {
     std::vector<std::unique_ptr<WaitTreeItem>> list;
 
-    const auto& threads = object.GetWaitingThreads();
+    const auto& threads = object.GetWaitingThreadsForDebugging();
     if (threads.empty()) {
         list.push_back(std::make_unique<WaitTreeText>(tr("waited by no thread")));
     } else {
@@ -211,8 +212,8 @@ std::vector<std::unique_ptr<WaitTreeItem>> WaitTreeSynchronizationObject::GetChi
     return list;
 }
 
-WaitTreeObjectList::WaitTreeObjectList(
-    const std::vector<std::shared_ptr<Kernel::SynchronizationObject>>& list, bool w_all)
+WaitTreeObjectList::WaitTreeObjectList(const std::vector<Kernel::KSynchronizationObject*>& list,
+                                       bool w_all)
     : object_list(list), wait_all(w_all) {}
 
 WaitTreeObjectList::~WaitTreeObjectList() = default;
@@ -237,8 +238,8 @@ WaitTreeThread::~WaitTreeThread() = default;
 QString WaitTreeThread::GetText() const {
     const auto& thread = static_cast<const Kernel::Thread&>(object);
     QString status;
-    switch (thread.GetStatus()) {
-    case Kernel::ThreadStatus::Ready:
+    switch (thread.GetState()) {
+    case Kernel::ThreadState::Runnable:
         if (!thread.IsPaused()) {
             if (thread.WasRunning()) {
                 status = tr("running");
@@ -249,35 +250,14 @@ QString WaitTreeThread::GetText() const {
             status = tr("paused");
         }
         break;
-    case Kernel::ThreadStatus::Paused:
-        status = tr("paused");
+    case Kernel::ThreadState::Waiting:
+        status = tr("waiting");
         break;
-    case Kernel::ThreadStatus::WaitHLEEvent:
-        status = tr("waiting for HLE return");
+    case Kernel::ThreadState::Initialized:
+        status = tr("initialized");
         break;
-    case Kernel::ThreadStatus::WaitSleep:
-        status = tr("sleeping");
-        break;
-    case Kernel::ThreadStatus::WaitIPC:
-        status = tr("waiting for IPC reply");
-        break;
-    case Kernel::ThreadStatus::WaitSynch:
-        status = tr("waiting for objects");
-        break;
-    case Kernel::ThreadStatus::WaitMutex:
-        status = tr("waiting for mutex");
-        break;
-    case Kernel::ThreadStatus::WaitCondVar:
-        status = tr("waiting for condition variable");
-        break;
-    case Kernel::ThreadStatus::WaitArb:
-        status = tr("waiting for address arbiter");
-        break;
-    case Kernel::ThreadStatus::Dormant:
-        status = tr("dormant");
-        break;
-    case Kernel::ThreadStatus::Dead:
-        status = tr("dead");
+    case Kernel::ThreadState::Terminated:
+        status = tr("terminated");
         break;
     }
 
@@ -293,8 +273,8 @@ QColor WaitTreeThread::GetColor() const {
     const std::size_t color_index = IsDarkTheme() ? 1 : 0;
 
     const auto& thread = static_cast<const Kernel::Thread&>(object);
-    switch (thread.GetStatus()) {
-    case Kernel::ThreadStatus::Ready:
+    switch (thread.GetState()) {
+    case Kernel::ThreadState::Runnable:
         if (!thread.IsPaused()) {
             if (thread.WasRunning()) {
                 return QColor(WaitTreeColors[0][color_index]);
@@ -304,21 +284,11 @@ QColor WaitTreeThread::GetColor() const {
         } else {
             return QColor(WaitTreeColors[2][color_index]);
         }
-    case Kernel::ThreadStatus::Paused:
+    case Kernel::ThreadState::Waiting:
         return QColor(WaitTreeColors[3][color_index]);
-    case Kernel::ThreadStatus::WaitHLEEvent:
-    case Kernel::ThreadStatus::WaitIPC:
-        return QColor(WaitTreeColors[4][color_index]);
-    case Kernel::ThreadStatus::WaitSleep:
-        return QColor(WaitTreeColors[5][color_index]);
-    case Kernel::ThreadStatus::WaitSynch:
-    case Kernel::ThreadStatus::WaitMutex:
-    case Kernel::ThreadStatus::WaitCondVar:
-    case Kernel::ThreadStatus::WaitArb:
-        return QColor(WaitTreeColors[6][color_index]);
-    case Kernel::ThreadStatus::Dormant:
+    case Kernel::ThreadState::Initialized:
         return QColor(WaitTreeColors[7][color_index]);
-    case Kernel::ThreadStatus::Dead:
+    case Kernel::ThreadState::Terminated:
         return QColor(WaitTreeColors[8][color_index]);
     default:
         return WaitTreeItem::GetColor();
@@ -354,11 +324,11 @@ std::vector<std::unique_ptr<WaitTreeItem>> WaitTreeThread::GetChildren() const {
     list.push_back(std::make_unique<WaitTreeText>(tr("thread id = %1").arg(thread.GetThreadID())));
     list.push_back(std::make_unique<WaitTreeText>(tr("priority = %1(current) / %2(normal)")
                                                       .arg(thread.GetPriority())
-                                                      .arg(thread.GetNominalPriority())));
+                                                      .arg(thread.GetBasePriority())));
     list.push_back(std::make_unique<WaitTreeText>(
         tr("last running ticks = %1").arg(thread.GetLastScheduledTick())));
 
-    const VAddr mutex_wait_address = thread.GetMutexWaitAddress();
+    const VAddr mutex_wait_address = thread.GetMutexWaitAddressForDebugging();
     if (mutex_wait_address != 0) {
         const auto& handle_table = thread.GetOwnerProcess()->GetHandleTable();
         list.push_back(std::make_unique<WaitTreeMutexInfo>(mutex_wait_address, handle_table));
@@ -366,9 +336,9 @@ std::vector<std::unique_ptr<WaitTreeItem>> WaitTreeThread::GetChildren() const {
         list.push_back(std::make_unique<WaitTreeText>(tr("not waiting for mutex")));
     }
 
-    if (thread.GetStatus() == Kernel::ThreadStatus::WaitSynch) {
-        list.push_back(std::make_unique<WaitTreeObjectList>(thread.GetSynchronizationObjects(),
-                                                            thread.IsWaitingSync()));
+    if (thread.GetState() == Kernel::ThreadState::Waiting) {
+        list.push_back(std::make_unique<WaitTreeObjectList>(thread.GetWaitObjectsForDebugging(),
+                                                            thread.IsCancellable()));
     }
 
     list.push_back(std::make_unique<WaitTreeCallstack>(thread));
@@ -380,7 +350,7 @@ WaitTreeEvent::WaitTreeEvent(const Kernel::ReadableEvent& object)
     : WaitTreeSynchronizationObject(object) {}
 WaitTreeEvent::~WaitTreeEvent() = default;
 
-WaitTreeThreadList::WaitTreeThreadList(const std::vector<std::shared_ptr<Kernel::Thread>>& list)
+WaitTreeThreadList::WaitTreeThreadList(const std::vector<Kernel::Thread*>& list)
     : thread_list(list) {}
 WaitTreeThreadList::~WaitTreeThreadList() = default;
 
diff --git a/src/yuzu/debugger/wait_tree.h b/src/yuzu/debugger/wait_tree.h
index 8e3bc4b24..cf96911ea 100755
--- a/src/yuzu/debugger/wait_tree.h
+++ b/src/yuzu/debugger/wait_tree.h
@@ -18,8 +18,8 @@ class EmuThread;
 
 namespace Kernel {
 class HandleTable;
+class KSynchronizationObject;
 class ReadableEvent;
-class SynchronizationObject;
 class Thread;
 } // namespace Kernel
 
@@ -102,30 +102,29 @@ private:
 class WaitTreeSynchronizationObject : public WaitTreeExpandableItem {
     Q_OBJECT
 public:
-    explicit WaitTreeSynchronizationObject(const Kernel::SynchronizationObject& object);
+    explicit WaitTreeSynchronizationObject(const Kernel::KSynchronizationObject& object);
     ~WaitTreeSynchronizationObject() override;
 
     static std::unique_ptr<WaitTreeSynchronizationObject> make(
-        const Kernel::SynchronizationObject& object);
+        const Kernel::KSynchronizationObject& object);
     QString GetText() const override;
     std::vector<std::unique_ptr<WaitTreeItem>> GetChildren() const override;
 
 protected:
-    const Kernel::SynchronizationObject& object;
+    const Kernel::KSynchronizationObject& object;
 };
 
 class WaitTreeObjectList : public WaitTreeExpandableItem {
     Q_OBJECT
 public:
-    WaitTreeObjectList(const std::vector<std::shared_ptr<Kernel::SynchronizationObject>>& list,
-                       bool wait_all);
+    WaitTreeObjectList(const std::vector<Kernel::KSynchronizationObject*>& list, bool wait_all);
     ~WaitTreeObjectList() override;
 
     QString GetText() const override;
     std::vector<std::unique_ptr<WaitTreeItem>> GetChildren() const override;
 
 private:
-    const std::vector<std::shared_ptr<Kernel::SynchronizationObject>>& object_list;
+    const std::vector<Kernel::KSynchronizationObject*>& object_list;
     bool wait_all;
 };
 
@@ -150,14 +149,14 @@ public:
 class WaitTreeThreadList : public WaitTreeExpandableItem {
     Q_OBJECT
 public:
-    explicit WaitTreeThreadList(const std::vector<std::shared_ptr<Kernel::Thread>>& list);
+    explicit WaitTreeThreadList(const std::vector<Kernel::Thread*>& list);
     ~WaitTreeThreadList() override;
 
     QString GetText() const override;
     std::vector<std::unique_ptr<WaitTreeItem>> GetChildren() const override;
 
 private:
-    const std::vector<std::shared_ptr<Kernel::Thread>>& thread_list;
+    const std::vector<Kernel::Thread*>& thread_list;
 };
 
 class WaitTreeModel : public QAbstractItemModel {