fixups

Signed-off-by: lizzie <lizzie@eden-emu.dev>

src/audio_core/audio_core.cpp

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2022 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -8,10 +11,11 @@
 
 namespace AudioCore {
 
-AudioCore::AudioCore(Core::System& system) : audio_manager{std::make_unique<AudioManager>()} {
+AudioCore::AudioCore(Core::System& system) {
+    audio_manager.emplace();
     CreateSinks();
     // Must be created after the sinks
-    adsp = std::make_unique<ADSP::ADSP>(system, *output_sink);
+    adsp.emplace(system, *output_sink);
 }
 
 AudioCore ::~AudioCore() {

src/audio_core/audio_core.h

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2022 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -15,10 +18,7 @@ class System;
 
 namespace AudioCore {
 
-class AudioManager;
+/// @brief Main audio class, stored inside the core, and holding the audio manager, all sinks, and the ADSP.
-/**
- * Main audio class, stored inside the core, and holding the audio manager, all sinks, and the ADSP.
- */
 class AudioCore {
 public:
     explicit AudioCore(Core::System& system);
@@ -50,27 +50,22 @@ public:
      */
     Sink::Sink& GetInputSink();
 
-    /**
+    /// @brief Get the ADSP.
-     * Get the ADSP.
+    /// @return Ref to the ADSP.
-     *
-     * @return Ref to the ADSP.
-     */
     ADSP::ADSP& ADSP();
 
 private:
-    /**
+    /// @brief Create the sinks on startup.
-     * Create the sinks on startup.
-     */
     void CreateSinks();
 
     /// Main audio manager for audio in/out
-    std::unique_ptr<AudioManager> audio_manager;
+    std::optional<AudioManager> audio_manager;
     /// Sink used for audio renderer and audio out
     std::unique_ptr<Sink::Sink> output_sink;
     /// Sink used for audio input
     std::unique_ptr<Sink::Sink> input_sink;
     /// The ADSP in the sysmodule
-    std::unique_ptr<ADSP::ADSP> adsp;
+    std::optional<ADSP::ADSP> adsp;
 };
 
 } // namespace AudioCore

src/audio_core/audio_in_manager.cpp

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2022 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -41,8 +44,7 @@ void Manager::ReleaseSessionId(const size_t session_id) {
 Result Manager::LinkToManager() {
     std::scoped_lock l{mutex};
     if (!linked_to_manager) {
-        AudioManager& manager{system.AudioCore().GetAudioManager()};
+        system.AudioCore().GetAudioManager().SetInManager(std::bind(&Manager::BufferReleaseAndRegister, this));
-        manager.SetInManager(std::bind(&Manager::BufferReleaseAndRegister, this));
         linked_to_manager = true;
     }
 

src/audio_core/audio_manager.cpp

@@ -1,81 +1,77 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2022 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
 #include "audio_core/audio_manager.h"
+#include "common/thread.h"
 #include "core/core.h"
 #include "core/hle/service/audio/errors.h"
 
 namespace AudioCore {
 
 AudioManager::AudioManager() {
-    thread = std::jthread([this]() { ThreadFunc(); });
+    thread = std::jthread([this](std::stop_token stop_token) {
+        Common::SetCurrentThreadName("AudioManager");
+        std::unique_lock l{events.GetAudioEventLock()};
+        events.ClearEvents();
+        while (!stop_token.stop_requested()) {
+            const auto timed_out{events.Wait(l, std::chrono::seconds(2))};
+            if (events.CheckAudioEventSet(Event::Type::Max)) {
+                break;
+            }
+            for (size_t i = 0; i < buffer_events.size(); i++) {
+                const auto event_type = Event::Type(i);
+                if (events.CheckAudioEventSet(event_type) || timed_out) {
+                    if (buffer_events[i]) {
+                        buffer_events[i]();
+                    }
+                }
+                events.SetAudioEvent(event_type, false);
+            }
+        }
+    });
 }
 
 void AudioManager::Shutdown() {
-    running = false;
     events.SetAudioEvent(Event::Type::Max, true);
-    thread.join();
+    if (thread.joinable()) {
+        thread.request_stop();
+        thread.join();
+    }
 }
 
 Result AudioManager::SetOutManager(BufferEventFunc buffer_func) {
-    if (!running) {
+    if (thread.joinable()) {
-        return Service::Audio::ResultOperationFailed;
+        std::scoped_lock l{lock};
+        const auto index{events.GetManagerIndex(Event::Type::AudioOutManager)};
+        if (buffer_events[index] == nullptr) {
+            buffer_events[index] = std::move(buffer_func);
+            needs_update = true;
+            events.SetAudioEvent(Event::Type::AudioOutManager, true);
+        }
+        return ResultSuccess;
     }
-
+    return Service::Audio::ResultOperationFailed;
-    std::scoped_lock l{lock};
-
-    const auto index{events.GetManagerIndex(Event::Type::AudioOutManager)};
-    if (buffer_events[index] == nullptr) {
-        buffer_events[index] = std::move(buffer_func);
-        needs_update = true;
-        events.SetAudioEvent(Event::Type::AudioOutManager, true);
-    }
-    return ResultSuccess;
 }
 
 Result AudioManager::SetInManager(BufferEventFunc buffer_func) {
-    if (!running) {
+    if (thread.joinable()) {
-        return Service::Audio::ResultOperationFailed;
+        std::scoped_lock l{lock};
+        const auto index{events.GetManagerIndex(Event::Type::AudioInManager)};
+        if (buffer_events[index] == nullptr) {
+            buffer_events[index] = std::move(buffer_func);
+            needs_update = true;
+            events.SetAudioEvent(Event::Type::AudioInManager, true);
+        }
+        return ResultSuccess;
     }
-
+    return Service::Audio::ResultOperationFailed;
-    std::scoped_lock l{lock};
-
-    const auto index{events.GetManagerIndex(Event::Type::AudioInManager)};
-    if (buffer_events[index] == nullptr) {
-        buffer_events[index] = std::move(buffer_func);
-        needs_update = true;
-        events.SetAudioEvent(Event::Type::AudioInManager, true);
-    }
-    return ResultSuccess;
 }
 
 void AudioManager::SetEvent(const Event::Type type, const bool signalled) {
     events.SetAudioEvent(type, signalled);
 }
 
-void AudioManager::ThreadFunc() {
-    std::unique_lock l{events.GetAudioEventLock()};
-    events.ClearEvents();
-    running = true;
-
-    while (running) {
-        const auto timed_out{events.Wait(l, std::chrono::seconds(2))};
-
-        if (events.CheckAudioEventSet(Event::Type::Max)) {
-            break;
-        }
-
-        for (size_t i = 0; i < buffer_events.size(); i++) {
-            const auto event_type = static_cast<Event::Type>(i);
-
-            if (events.CheckAudioEventSet(event_type) || timed_out) {
-                if (buffer_events[i]) {
-                    buffer_events[i]();
-                }
-            }
-            events.SetAudioEvent(event_type, false);
-        }
-    }
-}
-
 } // namespace AudioCore

src/audio_core/audio_manager.h

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2022 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -66,13 +69,6 @@ public:
     void SetEvent(Event::Type type, bool signalled);
 
 private:
-    /**
-     * Main thread, waiting on a manager signal and calling the registered function.
-     */
-    void ThreadFunc();
-
-    /// Is the main thread running?
-    std::atomic<bool> running{};
     /// Unused
     bool needs_update{};
     /// Events to be set and signalled

src/audio_core/audio_out_manager.cpp

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2022 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -40,8 +43,7 @@ void Manager::ReleaseSessionId(const size_t session_id) {
 Result Manager::LinkToManager() {
     std::scoped_lock l{mutex};
     if (!linked_to_manager) {
-        AudioManager& manager{system.AudioCore().GetAudioManager()};
+        system.AudioCore().GetAudioManager().SetOutManager(std::bind(&Manager::BufferReleaseAndRegister, this));
-        manager.SetOutManager(std::bind(&Manager::BufferReleaseAndRegister, this));
         linked_to_manager = true;
     }
 

src/core/core.cpp

@@ -271,7 +271,7 @@ struct System::Impl {
 
     SystemResultStatus SetupForApplicationProcess(System& system, Frontend::EmuWindow& emu_window) {
         host1x_core.emplace(system);
-        gpu_core = VideoCore::CreateGPU(emu_window, system);
+        VideoCore::CreateGPU(gpu_core, emu_window, system);
         if (!gpu_core)
             return SystemResultStatus::ErrorVideoCore;
 
@@ -347,7 +347,7 @@ struct System::Impl {
 
         // Register with applet manager
         // All threads are started, begin main process execution, now that we're in the clear
-        applet_manager.CreateAndInsertByFrontendAppletParameters(std::move(process), params);
+        applet_manager.CreateAndInsertByFrontendAppletParameters(std::make_unique<Service::Process>(*std::move(process)), params);
 
         if (Settings::values.gamecard_inserted) {
             if (Settings::values.gamecard_current_game) {
@@ -391,10 +391,8 @@ struct System::Impl {
         is_powered_on = false;
         exit_locked = false;
         exit_requested = false;
-
+        if (gpu_core)
-        if (gpu_core != nullptr) {
             gpu_core->NotifyShutdown();
-        }
 
         stop_event.request_stop();
         core_timing.SyncPause(false);
@@ -478,6 +476,7 @@ struct System::Impl {
     std::optional<Memory::CheatEngine> cheat_engine;
     std::optional<Tools::Freezer> memory_freezer;
     std::optional<Tools::RenderdocAPI> renderdoc_api;
+    std::optional<Tegra::GPU> gpu_core;
 
     std::array<Core::GPUDirtyMemoryManager, Core::Hardware::NUM_CPU_CORES> gpu_dirty_memory_managers;
     std::vector<std::vector<u8>> user_channel;
@@ -492,7 +491,6 @@ struct System::Impl {
     std::unique_ptr<FileSys::ContentProviderUnion> content_provider;
     /// AppLoader used to load the current executing application
     std::unique_ptr<Loader::AppLoader> app_loader;
-    std::unique_ptr<Tegra::GPU> gpu_core;
     std::stop_source stop_event;
 
     mutable std::mutex suspend_guard;

src/core/cpu_manager.cpp

@@ -25,9 +25,11 @@ CpuManager::~CpuManager() = default;
 
 void CpuManager::Initialize() {
     num_cores = is_multicore ? Core::Hardware::NUM_CPU_CORES : 1;
-    gpu_barrier = std::make_unique<Common::Barrier>(num_cores + 1);
+    gpu_barrier.emplace(num_cores + 1);
     for (std::size_t core = 0; core < num_cores; core++)
-        core_data[core].host_thread = std::jthread([this, core](std::stop_token token) { RunThread(token, core); });
+        core_data[core].host_thread = std::jthread([this, core](std::stop_token token) {
+            RunThread(token, core);
+        });
 }
 
 void CpuManager::Shutdown() {
@@ -63,7 +65,7 @@ void CpuManager::HandleInterrupt() {
     auto& kernel = system.Kernel();
     auto core_index = kernel.CurrentPhysicalCoreIndex();
 
-    Kernel::KInterruptManager::HandleInterrupt(kernel, static_cast<s32>(core_index));
+    Kernel::KInterruptManager::HandleInterrupt(kernel, s32(core_index));
 }
 
 ///////////////////////////////////////////////////////////////////////////////

src/core/cpu_manager.h

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2018 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -86,22 +89,20 @@ private:
     void ShutdownThread();
     void RunThread(std::stop_token stop_token, std::size_t core);
 
+    static constexpr std::size_t max_cycle_runs = 5;
+
+    std::optional<Common::Barrier> gpu_barrier{};
     struct CoreData {
         std::shared_ptr<Common::Fiber> host_context;
         std::jthread host_thread;
     };
-
-    std::unique_ptr<Common::Barrier> gpu_barrier{};
     std::array<CoreData, Core::Hardware::NUM_CPU_CORES> core_data{};
-
+    System& system;
-    bool is_async_gpu{};
-    bool is_multicore{};
     std::atomic<std::size_t> current_core{};
     std::size_t idle_count{};
     std::size_t num_cores{};
-    static constexpr std::size_t max_cycle_runs = 5;
+    bool is_async_gpu{};
-
+    bool is_multicore{};
-    System& system;
 };
 
 } // namespace Core

src/core/hle/service/am/applet_manager.cpp

@@ -268,7 +268,7 @@ void AppletManager::SetWindowSystem(WindowSystem* window_system) {
 
     if (Settings::values.enable_overlay && m_window_system->GetOverlayDisplayApplet() == nullptr) {
         if (auto overlay_process = CreateProcess(m_system, static_cast<u64>(AppletProgramId::OverlayDisplay), 0, 0)) {
-            auto overlay_applet = std::make_shared<Applet>(m_system, std::move(overlay_process), false);
+            auto overlay_applet = std::make_shared<Applet>(m_system, std::make_unique<Service::Process>(*std::move(overlay_process)), false);
             overlay_applet->program_id = static_cast<u64>(AppletProgramId::OverlayDisplay);
             overlay_applet->applet_id = AppletId::OverlayDisplay;
             overlay_applet->type = AppletType::OverlayApplet;

src/core/hle/service/am/button_poller.cpp

@@ -1,9 +1,11 @@
-// SPDX-FileCopyrightText: Copyright 2025 Eden Emulator Project
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
 // SPDX-License-Identifier: GPL-3.0-or-later
 
 // SPDX-FileCopyrightText: Copyright 2024 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
+#include <thread>
+#include "common/thread.h"
 #include "core/core.h"
 #include "core/hle/service/am/am_types.h"
 #include "core/hle/service/am/button_poller.h"
@@ -34,16 +36,15 @@ ButtonPressDuration ClassifyPressDuration(std::chrono::steady_clock::time_point
 
 } // namespace
 
-ButtonPoller::ButtonPoller(Core::System& system, WindowSystem& window_system)
+ButtonPoller::ButtonPoller(Core::System& system, WindowSystem& window_system) {
-    : m_window_system(window_system) {
     // TODO: am reads this from the home button state in hid, which is controller-agnostic.
     Core::HID::ControllerUpdateCallback engine_callback{
-        .on_change =
+        .on_change = [this, &window_system](Core::HID::ControllerTriggerType type) {
-            [this](Core::HID::ControllerTriggerType type) {
+            if (type == Core::HID::ControllerTriggerType::Button) {
-                if (type == Core::HID::ControllerTriggerType::Button) {
+                std::unique_lock lk{m_mutex};
-                    this->OnButtonStateChanged();
+                OnButtonStateChanged(window_system);
-                }
+            }
-            },
+        },
         .is_npad_service = true,
     };
 
@@ -52,25 +53,35 @@ ButtonPoller::ButtonPoller(Core::System& system, WindowSystem& window_system)
     m_player1 = system.HIDCore().GetEmulatedController(Core::HID::NpadIdType::Player1);
     m_player1_key = m_player1->SetCallback(engine_callback);
 
-    m_thread = std::thread([this] { this->ThreadLoop(); });
+    m_thread = std::jthread([this, &window_system](std::stop_token stop_token) {
+        Common::SetCurrentThreadName("ButtonPoller");
+        while (!stop_token.stop_requested()) {
+            using namespace std::chrono_literals;
+            std::unique_lock lk{m_mutex};
+            m_cv.wait_for(lk, 50ms);
+            if (stop_token.stop_requested())
+                break;
+            OnButtonStateChanged(window_system);
+            std::this_thread::sleep_for(5ms);
+        }
+    });
 }
 
 ButtonPoller::~ButtonPoller() {
     m_handheld->DeleteCallback(m_handheld_key);
     m_player1->DeleteCallback(m_player1_key);
-    m_stop = true;
     m_cv.notify_all();
     if (m_thread.joinable()) {
+        m_thread.request_stop();
         m_thread.join();
     }
 }
 
-void ButtonPoller::OnButtonStateChanged() {
+void ButtonPoller::OnButtonStateChanged(WindowSystem& window_system) {
-    std::lock_guard lk{m_mutex};
+    auto const home_button = m_handheld->GetHomeButtons().home.Value()
-    const bool home_button =
+        || m_player1->GetHomeButtons().home.Value();
-        m_handheld->GetHomeButtons().home.Value() || m_player1->GetHomeButtons().home.Value();
+    auto const capture_button = m_handheld->GetCaptureButtons().capture.Value()
-    const bool capture_button = m_handheld->GetCaptureButtons().capture.Value() ||
+        || m_player1->GetCaptureButtons().capture.Value();
-                                m_player1->GetCaptureButtons().capture.Value();
 
     // Buttons pressed which were not previously pressed
     if (home_button && !m_home_button_press_start) {
@@ -90,7 +101,7 @@ void ButtonPoller::OnButtonStateChanged() {
     if (home_button && m_home_button_press_start && !m_home_button_long_sent) {
         const auto duration = ClassifyPressDuration(*m_home_button_press_start);
         if (duration != ButtonPressDuration::ShortPressing) {
-            m_window_system.OnSystemButtonPress(SystemButtonType::HomeButtonLongPressing);
+            window_system.OnSystemButtonPress(SystemButtonType::HomeButtonLongPressing);
             m_home_button_long_sent = true;
         }
      }
@@ -98,7 +109,7 @@ void ButtonPoller::OnButtonStateChanged() {
     if (capture_button && m_capture_button_press_start && !m_capture_button_long_sent) {
          const auto duration = ClassifyPressDuration(*m_capture_button_press_start);
          if (duration != ButtonPressDuration::ShortPressing) {
-             m_window_system.OnSystemButtonPress(SystemButtonType::CaptureButtonLongPressing);
+             window_system.OnSystemButtonPress(SystemButtonType::CaptureButtonLongPressing);
              m_capture_button_long_sent = true;
          }
      }
@@ -107,9 +118,8 @@ void ButtonPoller::OnButtonStateChanged() {
     if (!home_button && m_home_button_press_start) {
         if(!m_home_button_long_sent) {
             const auto duration = ClassifyPressDuration(*m_home_button_press_start);
-            m_window_system.OnSystemButtonPress(
+            window_system.OnSystemButtonPress(duration == ButtonPressDuration::ShortPressing
-                duration == ButtonPressDuration::ShortPressing ? SystemButtonType::HomeButtonShortPressing
+                ? SystemButtonType::HomeButtonShortPressing : SystemButtonType::HomeButtonLongPressing);
-                                                               : SystemButtonType::HomeButtonLongPressing);
         }
         m_home_button_press_start = std::nullopt;
         m_home_button_long_sent = false;
@@ -117,9 +127,8 @@ void ButtonPoller::OnButtonStateChanged() {
     if (!capture_button && m_capture_button_press_start) {
         if (!m_capture_button_long_sent) {
             const auto duration = ClassifyPressDuration(*m_capture_button_press_start);
-            m_window_system.OnSystemButtonPress(
+            window_system.OnSystemButtonPress(duration == ButtonPressDuration::ShortPressing
-                duration == ButtonPressDuration::ShortPressing ? SystemButtonType::CaptureButtonShortPressing
+                ? SystemButtonType::CaptureButtonShortPressing : SystemButtonType::CaptureButtonLongPressing);
-                                                               : SystemButtonType::CaptureButtonLongPressing);
         }
         m_capture_button_press_start = std::nullopt;
         m_capture_button_long_sent = false;
@@ -130,16 +139,4 @@ void ButtonPoller::OnButtonStateChanged() {
     // }
 }
 
-void ButtonPoller::ThreadLoop() {
-    using namespace std::chrono_literals;
-    std::unique_lock lk{m_mutex};
-    while (!m_stop) {
-        m_cv.wait_for(lk, 50ms);
-        if (m_stop) break;
-        lk.unlock();
-        OnButtonStateChanged();
-        lk.lock();
-    }
-}
-
 } // namespace Service::AM

src/core/hle/service/am/button_poller.h

@@ -1,4 +1,4 @@
-// SPDX-FileCopyrightText: Copyright 2025 Eden Emulator Project
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
 // SPDX-License-Identifier: GPL-3.0-or-later
 
 // SPDX-FileCopyrightText: Copyright 2024 yuzu Emulator Project
@@ -30,31 +30,23 @@ class ButtonPoller {
 public:
     explicit ButtonPoller(Core::System& system, WindowSystem& window_system);
     ~ButtonPoller();
+    void OnButtonStateChanged(WindowSystem& window_system);
 
 private:
-    void OnButtonStateChanged();
+    std::mutex m_mutex;
-    void ThreadLoop();
+    std::condition_variable m_cv;
-
+    std::jthread m_thread;
-private:
-    WindowSystem& m_window_system;
-
-    Core::HID::EmulatedController* m_handheld{};
-    int m_handheld_key{};
-    Core::HID::EmulatedController* m_player1{};
-    int m_player1_key{};
-
     std::optional<std::chrono::steady_clock::time_point> m_home_button_press_start{};
     std::optional<std::chrono::steady_clock::time_point> m_capture_button_press_start{};
     std::optional<std::chrono::steady_clock::time_point> m_power_button_press_start{};
 
-    bool m_home_button_long_sent{};
+    Core::HID::EmulatedController* m_handheld{};
-    bool m_capture_button_long_sent{};
+    Core::HID::EmulatedController* m_player1{};
-    bool m_power_button_long_sent{};
+    int32_t m_handheld_key{};
-
+    int32_t m_player1_key{};
-    std::thread m_thread;
+    bool m_home_button_long_sent : 1 = false;
-    std::atomic<bool> m_stop{false};
+    bool m_capture_button_long_sent : 1 = false;
-    std::condition_variable m_cv;
+    bool m_power_button_long_sent : 1 = false;
-    std::mutex m_mutex;
 };
 
 } // namespace Service::AM

src/core/hle/service/am/event_observer.cpp

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2024 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -15,12 +18,24 @@ enum class UserDataTag : u32 {
 };
 
 EventObserver::EventObserver(Core::System& system, WindowSystem& window_system)
-    : m_system(system), m_context(system, "am:EventObserver"), m_window_system(window_system),
+    : m_system(system), m_context(system, "am:EventObserver")
-      m_wakeup_event(m_context), m_wakeup_holder(m_wakeup_event.GetHandle()) {
+    , m_window_system(window_system)
+    , m_wakeup_event(m_context)
+    , m_wakeup_holder(m_wakeup_event.GetHandle())
+{
     m_window_system.SetEventObserver(this);
     m_wakeup_holder.SetUserData(static_cast<uintptr_t>(UserDataTag::WakeupEvent));
     m_wakeup_holder.LinkToMultiWait(std::addressof(m_multi_wait));
-    m_thread = std::thread([&] { this->ThreadFunc(); });
+    m_thread = std::thread([this] {
+        Common::SetCurrentThreadName("am:EventObserver");
+        while (true) {
+            auto* signaled_holder = this->WaitSignaled();
+            if (!signaled_holder) {
+                break;
+            }
+            this->Process(signaled_holder);
+        }
+    });
 }
 
 EventObserver::~EventObserver() {
@@ -146,17 +161,4 @@ void EventObserver::DestroyAppletProcessHolderLocked(ProcessHolder* holder) {
     delete holder;
 }
 
-void EventObserver::ThreadFunc() {
-    Common::SetCurrentThreadName("am:EventObserver");
-
-    while (true) {
-        auto* signaled_holder = this->WaitSignaled();
-        if (!signaled_holder) {
-            break;
-        }
-
-        this->Process(signaled_holder);
-    }
-}
-
 } // namespace Service::AM

src/core/hle/service/am/event_observer.h

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2024 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -41,9 +44,6 @@ private:
 private:
     void DestroyAppletProcessHolderLocked(ProcessHolder* holder);
 
-private:
-    void ThreadFunc();
-
 private:
     // System reference and context.
     Core::System& m_system;

src/core/hle/service/am/process_creation.cpp

@@ -1,6 +1,10 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2024 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
+#include <optional>
 #include "core/core.h"
 #include "core/file_sys/content_archive.h"
 #include "core/file_sys/nca_metadata.h"
@@ -16,7 +20,7 @@ namespace Service::AM {
 
 namespace {
 
-FileSys::StorageId GetStorageIdForFrontendSlot(
+[[nodiscard]] FileSys::StorageId GetStorageIdForFrontendSlot(
     std::optional<FileSys::ContentProviderUnionSlot> slot) {
     if (!slot.has_value()) {
         return FileSys::StorageId::None;
@@ -36,31 +40,23 @@ FileSys::StorageId GetStorageIdForFrontendSlot(
     }
 }
 
-std::unique_ptr<Process> CreateProcessImpl(std::unique_ptr<Loader::AppLoader>& out_loader,
+[[nodiscard]] inline std::optional<Process> CreateProcessImpl(std::unique_ptr<Loader::AppLoader>& out_loader, Loader::ResultStatus& out_load_result, Core::System& system, FileSys::VirtualFile file, u64 program_id, u64 program_index) {
-                                           Loader::ResultStatus& out_load_result,
-                                           Core::System& system, FileSys::VirtualFile file,
-                                           u64 program_id, u64 program_index) {
     // Get the appropriate loader to parse this NCA.
     out_loader = Loader::GetLoader(system, file, program_id, program_index);
-
     // Ensure we have a loader which can parse the NCA.
-    if (!out_loader) {
+    if (out_loader) {
-        return nullptr;
+        // Try to load the process.
+        auto process = std::make_optional<Process>(system);
+        if (process->Initialize(*out_loader, out_load_result)) {
+            return process;
+        }
     }
-
+    return std::nullopt;
-    // Try to load the process.
-    auto process = std::make_unique<Process>(system);
-    if (process->Initialize(*out_loader, out_load_result)) {
-        return process;
-    }
-
-    return nullptr;
 }
 
 } // Anonymous namespace
 
-std::unique_ptr<Process> CreateProcess(Core::System& system, u64 program_id,
+std::optional<Process> CreateProcess(Core::System& system, u64 program_id, u8 minimum_key_generation, u8 maximum_key_generation) {
-                                       u8 minimum_key_generation, u8 maximum_key_generation) {
     // Attempt to load program NCA.
     FileSys::VirtualFile nca_raw{};
 
@@ -70,7 +66,7 @@ std::unique_ptr<Process> CreateProcess(Core::System& system, u64 program_id,
 
     // Ensure we retrieved a program NCA.
     if (!nca_raw) {
-        return nullptr;
+        return std::nullopt;
     }
 
     // Ensure we have a suitable version.
@@ -79,9 +75,8 @@ std::unique_ptr<Process> CreateProcess(Core::System& system, u64 program_id,
         if (nca.GetStatus() == Loader::ResultStatus::Success &&
             (nca.GetKeyGeneration() < minimum_key_generation ||
              nca.GetKeyGeneration() > maximum_key_generation)) {
-            LOG_WARNING(Service_LDR, "Skipping program {:016X} with generation {}", program_id,
+            LOG_WARNING(Service_LDR, "Skipping program {:016X} with generation {}", program_id, nca.GetKeyGeneration());
-                        nca.GetKeyGeneration());
+            return std::nullopt;
-            return nullptr;
         }
     }
 
@@ -90,42 +85,32 @@ std::unique_ptr<Process> CreateProcess(Core::System& system, u64 program_id,
     return CreateProcessImpl(loader, status, system, nca_raw, program_id, 0);
 }
 
-std::unique_ptr<Process> CreateApplicationProcess(std::vector<u8>& out_control,
+std::optional<Process> CreateApplicationProcess(std::vector<u8>& out_control, std::unique_ptr<Loader::AppLoader>& out_loader, Loader::ResultStatus& out_load_result, Core::System& system, FileSys::VirtualFile file, u64 program_id, u64 program_index) {
-                                                  std::unique_ptr<Loader::AppLoader>& out_loader,
+    if (auto process = CreateProcessImpl(out_loader, out_load_result, system, file, program_id, program_index); process) {
-                                                  Loader::ResultStatus& out_load_result,
+        FileSys::NACP nacp;
-                                                  Core::System& system, FileSys::VirtualFile file,
+        if (out_loader->ReadControlData(nacp) == Loader::ResultStatus::Success) {
-                                                  u64 program_id, u64 program_index) {
+            out_control = nacp.GetRawBytes();
-    auto process =
+        } else {
-        CreateProcessImpl(out_loader, out_load_result, system, file, program_id, program_index);
+            out_control.resize(sizeof(FileSys::RawNACP));
-    if (!process) {
+            std::fill(out_control.begin(), out_control.end(), (u8) 0);
-        return nullptr;
+        }
+
+        auto& storage = system.GetContentProviderUnion();
+        Service::Glue::ApplicationLaunchProperty launch{};
+        launch.title_id = process->GetProgramId();
+
+        FileSys::PatchManager pm{launch.title_id, system.GetFileSystemController(), storage};
+        launch.version = pm.GetGameVersion().value_or(0);
+
+        // TODO(DarkLordZach): When FSController/Game Card Support is added, if
+        // current_process_game_card use correct StorageId
+        launch.base_game_storage_id = GetStorageIdForFrontendSlot(storage.GetSlotForEntry(launch.title_id, FileSys::ContentRecordType::Program));
+        launch.update_storage_id = GetStorageIdForFrontendSlot(storage.GetSlotForEntry(FileSys::GetUpdateTitleID(launch.title_id), FileSys::ContentRecordType::Program));
+
+        system.GetARPManager().Register(launch.title_id, launch, out_control);
+        return process;
     }
-
+    return std::nullopt;
-    FileSys::NACP nacp;
-    if (out_loader->ReadControlData(nacp) == Loader::ResultStatus::Success) {
-        out_control = nacp.GetRawBytes();
-    } else {
-        out_control.resize(sizeof(FileSys::RawNACP));
-        std::fill(out_control.begin(), out_control.end(), (u8) 0);
-    }
-
-    auto& storage = system.GetContentProviderUnion();
-    Service::Glue::ApplicationLaunchProperty launch{};
-    launch.title_id = process->GetProgramId();
-
-    FileSys::PatchManager pm{launch.title_id, system.GetFileSystemController(), storage};
-    launch.version = pm.GetGameVersion().value_or(0);
-
-    // TODO(DarkLordZach): When FSController/Game Card Support is added, if
-    // current_process_game_card use correct StorageId
-    launch.base_game_storage_id = GetStorageIdForFrontendSlot(
-        storage.GetSlotForEntry(launch.title_id, FileSys::ContentRecordType::Program));
-    launch.update_storage_id = GetStorageIdForFrontendSlot(storage.GetSlotForEntry(
-        FileSys::GetUpdateTitleID(launch.title_id), FileSys::ContentRecordType::Program));
-
-    system.GetARPManager().Register(launch.title_id, launch, out_control);
-
-    return process;
 }
 
 } // namespace Service::AM

src/core/hle/service/am/process_creation.h

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2024 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -24,12 +27,7 @@ class Process;
 
 namespace Service::AM {
 
-std::unique_ptr<Process> CreateProcess(Core::System& system, u64 program_id,
+std::optional<Process> CreateProcess(Core::System& system, u64 program_id, u8 minimum_key_generation, u8 maximum_key_generation);
-                                       u8 minimum_key_generation, u8 maximum_key_generation);
+std::optional<Process> CreateApplicationProcess(std::vector<u8>& out_control, std::unique_ptr<Loader::AppLoader>& out_loader, Loader::ResultStatus& out_load_result, Core::System& system, FileSys::VirtualFile file, u64 program_id, u64 program_index);
-std::unique_ptr<Process> CreateApplicationProcess(std::vector<u8>& out_control,
-                                                  std::unique_ptr<Loader::AppLoader>& out_loader,
-                                                  Loader::ResultStatus& out_load_result,
-                                                  Core::System& system, FileSys::VirtualFile file,
-                                                  u64 program_id, u64 program_index);
 
 } // namespace Service::AM

src/core/hle/service/am/service/application_creator.cpp

@@ -21,8 +21,7 @@ namespace Service::AM {
 
 namespace {
 
-Result CreateGuestApplication(SharedPointer<IApplicationAccessor>* out_application_accessor,
+Result CreateGuestApplication(SharedPointer<IApplicationAccessor>* out_application_accessor, Core::System& system, WindowSystem& window_system, u64 program_id) {
-                              Core::System& system, WindowSystem& window_system, u64 program_id) {
     FileSys::VirtualFile nca_raw{};
 
     // Get the program NCA from storage.
@@ -35,11 +34,10 @@ Result CreateGuestApplication(SharedPointer<IApplicationAccessor>* out_applicati
     std::vector<u8> control;
     std::unique_ptr<Loader::AppLoader> loader;
     Loader::ResultStatus result;
-    auto process =
+    auto process = CreateApplicationProcess(control, loader, result, system, nca_raw, program_id, 0);
-        CreateApplicationProcess(control, loader, result, system, nca_raw, program_id, 0);
+    R_UNLESS(process != std::nullopt, ResultUnknown);
-    R_UNLESS(process != nullptr, ResultUnknown);
 
-    const auto applet = std::make_shared<Applet>(system, std::move(process), true);
+    const auto applet = std::make_shared<Applet>(system, std::make_unique<Service::Process>(*std::move(process)), true);
     applet->program_id = program_id;
     applet->applet_id = AppletId::Application;
     applet->type = AppletType::Application;
@@ -47,8 +45,7 @@ Result CreateGuestApplication(SharedPointer<IApplicationAccessor>* out_applicati
 
     window_system.TrackApplet(applet, true);
 
-    *out_application_accessor =
+    *out_application_accessor = std::make_shared<IApplicationAccessor>(system, applet, window_system);
-        std::make_shared<IApplicationAccessor>(system, applet, window_system);
     R_SUCCEED();
 }
 
@@ -90,12 +87,10 @@ Result IApplicationCreator::CreateSystemApplication(
 
     std::vector<u8> control;
     std::unique_ptr<Loader::AppLoader> loader;
+    auto process = CreateProcess(system, application_id, 1, 22);
+    R_UNLESS(process != std::nullopt, ResultUnknown);
 
-    auto process =
+    const auto applet = std::make_shared<Applet>(system, std::make_unique<Service::Process>(*std::move(process)), true);
-        CreateProcess(system, application_id, 1, 22);
-    R_UNLESS(process != nullptr, ResultUnknown);
-
-    const auto applet = std::make_shared<Applet>(system, std::move(process), true);
     applet->program_id = application_id;
     applet->applet_id = AppletId::Starter;
     applet->type = AppletType::LibraryApplet;
@@ -103,8 +98,7 @@ Result IApplicationCreator::CreateSystemApplication(
 
     m_window_system.TrackApplet(applet, true);
 
-    *out_application_accessor =
+    *out_application_accessor = std::make_shared<IApplicationAccessor>(system, applet, m_window_system);
-        std::make_shared<IApplicationAccessor>(system, applet, m_window_system);
     Core::LaunchTimestampCache::SaveLaunchTimestamp(application_id);
     R_SUCCEED();
 }

src/core/hle/service/am/service/library_applet_creator.cpp

@@ -122,26 +122,23 @@ std::shared_ptr<ILibraryAppletAccessor> CreateGuestApplet(Core::System& system,
     };
 
     auto process = CreateProcess(system, program_id, Firmware1400, Firmware2200);
-    if (!process) {
+    if (process) {
-        // Couldn't initialize the guest process
+        const auto applet = std::make_shared<Applet>(system, std::make_unique<Service::Process>(*std::move(process)), false);
-        return {};
+        applet->program_id = program_id;
+        applet->applet_id = applet_id;
+        applet->type = AppletType::LibraryApplet;
+        applet->library_applet_mode = mode;
+        applet->window_visible = mode != LibraryAppletMode::AllForegroundInitiallyHidden;
+
+        auto broker = std::make_shared<AppletDataBroker>(system);
+        applet->caller_applet = caller_applet;
+        applet->caller_applet_broker = broker;
+        caller_applet->child_applets.push_back(applet);
+        window_system.TrackApplet(applet, false);
+        return std::make_shared<ILibraryAppletAccessor>(system, broker, applet);
     }
-
+    // Couldn't initialize the guest process
-    const auto applet = std::make_shared<Applet>(system, std::move(process), false);
+    return {};
-    applet->program_id = program_id;
-    applet->applet_id = applet_id;
-    applet->type = AppletType::LibraryApplet;
-    applet->library_applet_mode = mode;
-    applet->window_visible = mode != LibraryAppletMode::AllForegroundInitiallyHidden;
-
-    auto broker = std::make_shared<AppletDataBroker>(system);
-    applet->caller_applet = caller_applet;
-    applet->caller_applet_broker = broker;
-    caller_applet->child_applets.push_back(applet);
-
-    window_system.TrackApplet(applet, false);
-
-    return std::make_shared<ILibraryAppletAccessor>(system, broker, applet);
 }
 
 std::shared_ptr<ILibraryAppletAccessor> CreateFrontendApplet(Core::System& system,

src/core/hle/service/audio/audio_out_manager.cpp

@@ -1,4 +1,4 @@
-// SPDX-FileCopyrightText: Copyright 2025 Eden Emulator Project
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
 // SPDX-License-Identifier: GPL-3.0-or-later
 
 // SPDX-FileCopyrightText: Copyright 2018 yuzu Emulator Project
@@ -14,7 +14,9 @@ namespace Service::Audio {
 using namespace AudioCore::AudioOut;
 
 IAudioOutManager::IAudioOutManager(Core::System& system_)
-    : ServiceFramework{system_, "audout:u"}, impl{std::make_unique<Manager>(system_)} {
+    : ServiceFramework{system_, "audout:u"}
+    , impl(system_)
+{
     // clang-format off
     static const FunctionInfo functions[] = {
         {0, D<&IAudioOutManager::ListAudioOuts>, "ListAudioOuts"},

src/core/hle/service/audio/audio_out_manager.h

@@ -1,4 +1,4 @@
-// SPDX-FileCopyrightText: Copyright 2025 Eden Emulator Project
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
 // SPDX-License-Identifier: GPL-3.0-or-later
 
 // SPDX-FileCopyrightText: Copyright 2018 yuzu Emulator Project
@@ -43,7 +43,7 @@ private:
         AudioCore::AudioOut::AudioOutParameter parameter,
         InCopyHandle<Kernel::KProcess> process_handle, ClientAppletResourceUserId aruid);
 
-    std::unique_ptr<AudioCore::AudioOut::Manager> impl;
+    std::optional<AudioCore::AudioOut::Manager> impl;
 };
 
 } // namespace Service::Audio

src/core/hle/service/audio/audio_renderer_manager.cpp

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2018 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -15,7 +18,9 @@ namespace Service::Audio {
 using namespace AudioCore::Renderer;
 
 IAudioRendererManager::IAudioRendererManager(Core::System& system_)
-    : ServiceFramework{system_, "audren:u"}, impl{std::make_unique<Manager>(system_)} {
+    : ServiceFramework{system_, "audren:u"}
+    , impl(system_)
+{
     // clang-format off
     static const FunctionInfo functions[] = {
         {0, D<&IAudioRendererManager::OpenAudioRenderer>, "OpenAudioRenderer"},

src/core/hle/service/audio/audio_renderer_manager.h

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2018 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -30,7 +33,7 @@ private:
     Result GetAudioDeviceServiceWithRevisionInfo(Out<SharedPointer<IAudioDevice>> out_audio_device,
                                                  u32 revision, ClientAppletResourceUserId aruid);
 
-    std::unique_ptr<AudioCore::Renderer::Manager> impl;
+    std::optional<AudioCore::Renderer::Manager> impl;
     u32 num_audio_devices{0};
 };
 

src/core/hle/service/nvdrv/devices/nvhost_as_gpu.cpp

@@ -138,8 +138,7 @@ NvResult nvhost_as_gpu::AllocAsEx(IoctlAllocAsEx& params) {
         static_cast<u32>((vm.va_range_end - vm.va_range_split) >> vm.big_page_size_bits)};
     vm.big_page_allocator.emplace(start_big_pages, end_big_pages);
 
-    gmmu = std::make_shared<Tegra::MemoryManager>(system, max_big_page_bits, vm.va_range_split,
+    gmmu = std::make_unique<Tegra::MemoryManager>(system, max_big_page_bits, vm.va_range_split, vm.big_page_size_bits, VM::PAGE_SIZE_BITS);
-                                                  vm.big_page_size_bits, VM::PAGE_SIZE_BITS);
     system.GPU().InitAddressSpace(*gmmu);
     vm.initialised = true;
 
@@ -416,7 +415,7 @@ NvResult nvhost_as_gpu::BindChannel(IoctlBindChannel& params) {
     LOG_DEBUG(Service_NVDRV, "called, fd={:X}", params.fd);
 
     auto gpu_channel_device = module.GetDevice<nvhost_gpu>(params.fd);
-    gpu_channel_device->channel_state->memory_manager = gmmu;
+    gpu_channel_device->channel_state->memory_manager = gmmu.get();
     return NvResult::Success;
 }
 

src/core/hle/service/nvdrv/devices/nvhost_as_gpu.h

@@ -219,7 +219,7 @@ private:
 
         bool initialised{};
     } vm;
-    std::shared_ptr<Tegra::MemoryManager> gmmu;
+    std::unique_ptr<Tegra::MemoryManager> gmmu;
 };
 
 } // namespace Service::Nvidia::Devices

src/core/hle/service/os/process.cpp

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2024 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -10,14 +13,6 @@
 
 namespace Service {
 
-Process::Process(Core::System& system)
-    : m_system(system), m_process(), m_main_thread_priority(), m_main_thread_stack_size(),
-      m_process_started() {}
-
-Process::~Process() {
-    this->Finalize();
-}
-
 bool Process::Initialize(Loader::AppLoader& loader, Loader::ResultStatus& out_load_result) {
     // First, ensure we are not holding another process.
     this->Finalize();

src/core/hle/service/os/process.h

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2024 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -22,8 +25,8 @@ namespace Service {
 
 class Process {
 public:
-    explicit Process(Core::System& system);
+    inline explicit Process(Core::System& system) noexcept : m_system(system) {}
-    ~Process();
+    inline ~Process() { this->Finalize(); }
 
     bool Initialize(Loader::AppLoader& loader, Loader::ResultStatus& out_load_result);
     void Finalize();
@@ -50,8 +53,8 @@ public:
 private:
     Core::System& m_system;
     Kernel::KProcess* m_process{};
-    s32 m_main_thread_priority{};
     u64 m_main_thread_stack_size{};
+    s32 m_main_thread_priority{};
     bool m_process_started{};
 };
 

src/input_common/drivers/udp_client.cpp

@@ -13,6 +13,7 @@
 #include "common/param_package.h"
 #include "common/random.h"
 #include "common/settings.h"
+#include "common/thread.h"
 #include "input_common/drivers/udp_client.h"
 #include "input_common/helpers/udp_protocol.h"
 
@@ -131,6 +132,7 @@ private:
 };
 
 static void SocketLoop(Socket* socket) {
+    Common::SetCurrentThreadName("cemuhookWorker");
     socket->StartReceive();
     socket->StartSend(Socket::clock::now());
     socket->Loop();
@@ -326,9 +328,11 @@ void UDPClient::OnPadData(Response::PadData data, std::size_t client) {
 }
 
 void UDPClient::StartCommunication(std::size_t client, const std::string& host, u16 port) {
-    SocketCallback callback{[this](Response::Version version) { OnVersion(version); },
+    SocketCallback callback{
-                            [this](Response::PortInfo info) { OnPortInfo(info); },
+        [this](Response::Version version) { OnVersion(version); },
-                            [this, client](Response::PadData data) { OnPadData(data, client); }};
+        [this](Response::PortInfo info) { OnPortInfo(info); },
+        [this, client](Response::PadData data) { OnPadData(data, client); }
+    };
     LOG_INFO(Input, "Starting communication with UDP input server on {}:{}", host, port);
     clients[client].uuid = GetHostUUID(host);
     clients[client].host = host;
@@ -566,9 +570,7 @@ bool UDPClient::IsStickInverted(const Common::ParamPackage& params) {
     return true;
 }
 
-void TestCommunication(const std::string& host, u16 port,
+void TestCommunication(const std::string& host, u16 port, const std::function<void()>& success_callback, const std::function<void()>& failure_callback) {
-                       const std::function<void()>& success_callback,
-                       const std::function<void()>& failure_callback) {
     std::thread([=] {
         Common::Event success_event;
         SocketCallback callback{
@@ -601,40 +603,38 @@ CalibrationConfigurationJob::CalibrationConfigurationJob(
         u16 max_y{};
 
         Status current_status{Status::Initialized};
-        SocketCallback callback{[](Response::Version) {}, [](Response::PortInfo) {},
+        SocketCallback callback{[](Response::Version) {}, [](Response::PortInfo) {}, [&](Response::PadData data) {
-                                [&](Response::PadData data) {
+            constexpr u16 CALIBRATION_THRESHOLD = 100;
-                                    constexpr u16 CALIBRATION_THRESHOLD = 100;
 
-                                    if (current_status == Status::Initialized) {
+            if (current_status == Status::Initialized) {
-                                        // Receiving data means the communication is ready now
+                // Receiving data means the communication is ready now
-                                        current_status = Status::Ready;
+                current_status = Status::Ready;
-                                        status_callback(current_status);
+                status_callback(current_status);
-                                    }
+            }
-                                    if (data.touch[0].is_active == 0) {
+            if (data.touch[0].is_active == 0) {
-                                        return;
+                return;
-                                    }
+            }
-                                    LOG_DEBUG(Input, "Current touch: {} {}", data.touch[0].x,
+            LOG_DEBUG(Input, "Current touch: {} {}", data.touch[0].x, data.touch[0].y);
-                                              data.touch[0].y);
+            min_x = (std::min)(min_x, u16(data.touch[0].x));
-                                    min_x = (std::min)(min_x, static_cast<u16>(data.touch[0].x));
+            min_y = (std::min)(min_y, u16(data.touch[0].y));
-                                    min_y = (std::min)(min_y, static_cast<u16>(data.touch[0].y));
+            if (current_status == Status::Ready) {
-                                    if (current_status == Status::Ready) {
+                // First touch - min data (min_x/min_y)
-                                        // First touch - min data (min_x/min_y)
+                current_status = Status::Stage1Completed;
-                                        current_status = Status::Stage1Completed;
+                status_callback(current_status);
-                                        status_callback(current_status);
+            }
-                                    }
+            if (data.touch[0].x - min_x > CALIBRATION_THRESHOLD &&
-                                    if (data.touch[0].x - min_x > CALIBRATION_THRESHOLD &&
+                data.touch[0].y - min_y > CALIBRATION_THRESHOLD) {
-                                        data.touch[0].y - min_y > CALIBRATION_THRESHOLD) {
+                // Set the current position as max value and finishes
-                                        // Set the current position as max value and finishes
+                // configuration
-                                        // configuration
+                max_x = data.touch[0].x;
-                                        max_x = data.touch[0].x;
+                max_y = data.touch[0].y;
-                                        max_y = data.touch[0].y;
+                current_status = Status::Completed;
-                                        current_status = Status::Completed;
+                data_callback(min_x, min_y, max_x, max_y);
-                                        data_callback(min_x, min_y, max_x, max_y);
+                status_callback(current_status);
-                                        status_callback(current_status);
 
-                                        complete_event.Set();
+                complete_event.Set();
-                                    }
+            }
-                                }};
+        }};
         Socket socket{host, port, std::move(callback)};
         std::thread worker_thread{SocketLoop, &socket};
         complete_event.Wait();

src/video_core/CMakeLists.txt

@@ -334,7 +334,7 @@ if (YUZU_USE_EXTERNAL_FFMPEG)
     add_dependencies(video_core ffmpeg-build)
 endif()
 
-target_include_directories(video_core PRIVATE ${FFmpeg_INCLUDE_DIR})
+target_include_directories(video_core PUBLIC ${FFmpeg_INCLUDE_DIR})
 
 target_link_libraries(video_core PRIVATE ${FFmpeg_LIBRARIES})
 target_link_options(video_core PRIVATE ${FFmpeg_LDFLAGS})

src/video_core/cdma_pusher.cpp

@@ -20,8 +20,7 @@
 namespace Tegra {
 
 CDmaPusher::CDmaPusher(Host1x::Host1x& host1x_, s32 id)
-    : host_processor(host1x_)
+    : host1x{host1x_}
-    , host1x{host1x_}
     , current_class{ChClassId(id)}
 {
     thread = std::jthread([this](std::stop_token stop_token) {
@@ -99,7 +98,7 @@ void CDmaPusher::ExecuteCommand(u32 method, u32 arg) {
     switch (current_class) {
     case ChClassId::Control:
         LOG_TRACE(Service_NVDRV, "Class {} method {:#X} arg 0x{:X}", u32(current_class), method, arg);
-        host_processor.ProcessMethod(Host1x::Control::Method(method), arg);
+        host_processor.ProcessMethod(host1x, Host1x::Control::Method(method), arg);
         break;
     default:
         thi_regs.reg_array[method] = arg;

src/video_core/control/channel_state.cpp

@@ -17,28 +17,36 @@
 
 namespace Tegra::Control {
 
-ChannelState::ChannelState(s32 bind_id_) : bind_id{bind_id_}, initialized{} {}
+ChannelState::Payload::Payload(Core::System& system, MemoryManager& memory_manager, ChannelState& channel_state)
+    : maxwell_3d(memory_manager)
+    , fermi_2d(memory_manager)
+    , kepler_compute(memory_manager)
+    , maxwell_dma(memory_manager)
+    , kepler_memory(memory_manager)
+    , nv01_timer(memory_manager)
+    , dma_pusher(system, memory_manager, channel_state)
+{}
 
-void ChannelState::Init(Core::System& system, GPU& gpu, u64 program_id_) {
+ChannelState::ChannelState(s32 bind_id_)
+    : bind_id{bind_id_}
+{}
+
+void ChannelState::Init(Core::System& system, u64 program_id_) {
     ASSERT(memory_manager);
     program_id = program_id_;
-    dma_pusher.emplace(system, gpu, *memory_manager, *this);
+    payload.emplace(system, *memory_manager, *this);
-    maxwell_3d.emplace(system, *memory_manager);
-    fermi_2d.emplace(*memory_manager);
-    kepler_compute.emplace(system, *memory_manager);
-    maxwell_dma.emplace(system, *memory_manager);
-    kepler_memory.emplace(system, *memory_manager);
     initialized = true;
 }
 
 void ChannelState::BindRasterizer(VideoCore::RasterizerInterface* rasterizer) {
-    dma_pusher->BindRasterizer(rasterizer);
+    payload->dma_pusher.BindRasterizer(rasterizer);
     memory_manager->BindRasterizer(rasterizer);
-    maxwell_3d->BindRasterizer(rasterizer);
+    payload->maxwell_3d.BindRasterizer(rasterizer);
-    fermi_2d->BindRasterizer(rasterizer);
+    payload->fermi_2d.BindRasterizer(rasterizer);
-    kepler_memory->BindRasterizer(rasterizer);
+    payload->kepler_memory.BindRasterizer(rasterizer);
-    kepler_compute->BindRasterizer(rasterizer);
+    payload->kepler_compute.BindRasterizer(rasterizer);
-    maxwell_dma->BindRasterizer(rasterizer);
+    payload->maxwell_dma.BindRasterizer(rasterizer);
+    //payload->nv01_timer.BindRasterizer(rasterizer);
 }
 
 } // namespace Tegra::Control

src/video_core/control/channel_state.h

@@ -14,6 +14,7 @@
 #include "video_core/engines/kepler_compute.h"
 #include "video_core/engines/maxwell_3d.h"
 #include "video_core/engines/maxwell_dma.h"
+#include "video_core/engines/nv01_timer.h"
 #include "video_core/dma_pusher.h"
 
 namespace Core {
@@ -34,28 +35,33 @@ namespace Control {
 struct ChannelState {
     explicit ChannelState(s32 bind_id);
 
-    void Init(Core::System& system, GPU& gpu, u64 program_id);
+    void Init(Core::System& system, u64 program_id);
 
     void BindRasterizer(VideoCore::RasterizerInterface* rasterizer);
 
-    /// 3D engine
+    struct Payload {
-    std::optional<Engines::Maxwell3D> maxwell_3d;
+        explicit Payload(Core::System& system, MemoryManager& memory_manager, ChannelState& channel_state);
-    /// 2D engine
+
-    std::optional<Engines::Fermi2D> fermi_2d;
+        /// 3D engine
-    /// Compute engine
+        Engines::Maxwell3D maxwell_3d;
-    std::optional<Engines::KeplerCompute> kepler_compute;
+        /// 2D engine
-    /// DMA engine
+        Engines::Fermi2D fermi_2d;
-    std::optional<Engines::MaxwellDMA> maxwell_dma;
+        /// Compute engine
-    /// Inline memory engine
+        Engines::KeplerCompute kepler_compute;
-    std::optional<Engines::KeplerMemory> kepler_memory;
+        /// DMA engine
-    /// NV01 Timer
+        Engines::MaxwellDMA maxwell_dma;
-    std::optional<Engines::KeplerMemory> nv01_timer;
+        /// Inline memory engine
-    std::optional<DmaPusher> dma_pusher;
+        Engines::KeplerMemory kepler_memory;
-    std::shared_ptr<MemoryManager> memory_manager;
+        /// NV01 Timer
+        Engines::Nv01Timer nv01_timer;
+        DmaPusher dma_pusher;
+    };
+    std::optional<Payload> payload;
+    MemoryManager* memory_manager = nullptr;
 
     s32 bind_id = -1;
     u64 program_id = 0;
-    bool initialized{};
+    bool initialized = false;
 };
 
 } // namespace Control

src/video_core/control/channel_state_cache.cpp

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: 2022 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-3.0-or-later
 
@@ -6,8 +9,11 @@
 namespace VideoCommon {
 
 ChannelInfo::ChannelInfo(Tegra::Control::ChannelState& channel_state)
-    : maxwell3d{*channel_state.maxwell_3d}, kepler_compute{*channel_state.kepler_compute},
+    : maxwell3d{channel_state.payload->maxwell_3d}
-      gpu_memory{*channel_state.memory_manager}, program_id{channel_state.program_id} {}
+    , kepler_compute{channel_state.payload->kepler_compute}
+    , gpu_memory{*channel_state.memory_manager}
+    , program_id{channel_state.program_id}
+  {}
 
 template class VideoCommon::ChannelSetupCaches<VideoCommon::ChannelInfo>;
 

src/video_core/control/channel_state_cache.inc

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: 2022 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-3.0-or-later
 
@@ -41,7 +44,7 @@ void ChannelSetupCaches<P>::CreateChannel(struct Tegra::Control::ChannelState& c
     AddressSpaceRef new_gpu_mem_ref{
         .ref_count = 1,
         .storage_id = address_spaces.size(),
-        .gpu_memory = channel.memory_manager.get(),
+        .gpu_memory = channel.memory_manager,
     };
     address_spaces.emplace(channel.memory_manager->GetID(), new_gpu_mem_ref);
     OnGPUASRegister(channel.memory_manager->GetID());

src/video_core/control/scheduler.cpp

@@ -1,4 +1,4 @@
-// SPDX-FileCopyrightText: Copyright 2025 Eden Emulator Project
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
 // SPDX-License-Identifier: GPL-3.0-or-later
 
 // SPDX-FileCopyrightText: 2021 yuzu Emulator Project
@@ -12,11 +12,8 @@
 #include "video_core/gpu.h"
 
 namespace Tegra::Control {
-Scheduler::Scheduler(GPU& gpu_) : gpu{gpu_} {}
 
-Scheduler::~Scheduler() = default;
+void Scheduler::Push(GPU& gpu, s32 channel, CommandList&& entries) {
-
-void Scheduler::Push(s32 channel, CommandList&& entries) {
     std::shared_ptr<ChannelState> channel_state;
     {
         std::unique_lock lk(scheduling_guard);
@@ -27,8 +24,8 @@ void Scheduler::Push(s32 channel, CommandList&& entries) {
     }
     // Process commands outside the lock to reduce contention.
     // Multiple channels can prepare their commands in parallel.
-    channel_state->dma_pusher->Push(std::move(entries));
+    channel_state->payload->dma_pusher.Push(std::move(entries));
-    channel_state->dma_pusher->DispatchCalls();
+    channel_state->payload->dma_pusher.DispatchCalls();
 }
 
 void Scheduler::DeclareChannel(std::shared_ptr<ChannelState> new_channel) {

src/video_core/control/scheduler.h

@@ -22,17 +22,13 @@ struct ChannelState;
 
 class Scheduler {
 public:
-    explicit Scheduler(GPU& gpu_);
+    void Push(GPU& gpu, s32 channel, CommandList&& entries);
-    ~Scheduler();
-
-    void Push(s32 channel, CommandList&& entries);
 
     void DeclareChannel(std::shared_ptr<ChannelState> new_channel);
 
 private:
     ankerl::unordered_dense::map<s32, std::shared_ptr<ChannelState>> channels;
     std::mutex scheduling_guard;
-    GPU& gpu;
 };
 
 } // namespace Control

src/video_core/dma_pusher.cpp

@@ -12,37 +12,32 @@
 #include "video_core/guest_memory.h"
 #include "video_core/memory_manager.h"
 #include "video_core/rasterizer_interface.h"
-#include "video_core/texture_cache/util.h"
-
-#ifdef _MSC_VER
-#include <intrin.h>
-#endif
 
 namespace Tegra {
 
 constexpr u32 MacroRegistersStart = 0xE00;
 [[maybe_unused]] constexpr u32 ComputeInline = 0x6D;
 
-DmaPusher::DmaPusher(Core::System& system_, GPU& gpu_, MemoryManager& memory_manager_,
+DmaPusher::DmaPusher(Core::System& system_, MemoryManager& memory_manager_, Control::ChannelState& channel_state_)
-                     Control::ChannelState& channel_state_)
+    : system{system_}
-    : gpu{gpu_}, system{system_}, memory_manager{memory_manager_}, puller{gpu_, memory_manager_,
+    , memory_manager{memory_manager_}
-                                                                          *this, channel_state_}, signal_sync{false}, synced{true} {}
+    , channel_state{channel_state_}
+    , signal_sync{false}
+    , synced{false}
+{}
 
 DmaPusher::~DmaPusher() = default;
 
 void DmaPusher::DispatchCalls() {
-
     dma_pushbuffer_subindex = 0;
-
     dma_state.is_last_call = true;
-
     while (system.IsPoweredOn()) {
         if (!Step()) {
             break;
         }
     }
-    gpu.FlushCommands();
+    system.GPU().FlushCommands();
-    gpu.OnCommandListEnd();
+    system.GPU().OnCommandListEnd();
 }
 
 bool DmaPusher::Step() {
@@ -171,9 +166,9 @@ void DmaPusher::SetState(const CommandHeader& command_header) {
     dma_state.method_count = command_header.method_count;
 }
 
-void DmaPusher::CallMethod(u32 argument) const {
+void DmaPusher::CallMethod(u32 argument) {
     if (dma_state.method < non_puller_methods) {
-        puller.CallPullerMethod(Engines::Puller::MethodCall{
+        puller.CallPullerMethod(*this, Engines::Puller::MethodCall{
             dma_state.method,
             argument,
             dma_state.subchannel,
@@ -181,30 +176,29 @@ void DmaPusher::CallMethod(u32 argument) const {
         });
     } else {
         auto subchannel = subchannels[dma_state.subchannel];
-        if (subchannel->execution_mask[dma_state.method]) {
+        if (!subchannel->execution_mask[dma_state.method]) {
-            subchannel->ConsumeSink();
-            subchannel->current_dma_segment = dma_state.dma_get + dma_state.dma_word_offset;
-            subchannel->CallMethod(dma_state.method, argument, dma_state.is_last_call);
-        } else {
             subchannel->method_sink.emplace_back(dma_state.method, argument);
+        } else {
+            subchannel->ConsumeSink(system);
+            subchannel->current_dma_segment = dma_state.dma_get + dma_state.dma_word_offset;
+            subchannel->CallMethod(system, dma_state.method, argument, dma_state.is_last_call);
         }
     }
 }
 
-void DmaPusher::CallMultiMethod(const u32* base_start, u32 num_methods) const {
+void DmaPusher::CallMultiMethod(const u32* base_start, u32 num_methods) {
     if (dma_state.method < non_puller_methods) {
-        puller.CallMultiMethod(dma_state.method, dma_state.subchannel, base_start, num_methods, dma_state.method_count);
+        puller.CallMultiMethod(*this, dma_state.method, dma_state.subchannel, base_start, num_methods, dma_state.method_count);
     } else {
         auto subchannel = subchannels[dma_state.subchannel];
-        subchannel->ConsumeSink();
+        subchannel->ConsumeSink(system);
         subchannel->current_dma_segment = dma_state.dma_get + dma_state.dma_word_offset;
-        subchannel->CallMultiMethod(dma_state.method, base_start, num_methods, dma_state.method_count);
+        subchannel->CallMultiMethod(system, dma_state.method, base_start, num_methods, dma_state.method_count);
     }
 }
 
 void DmaPusher::BindRasterizer(VideoCore::RasterizerInterface* rasterizer_) {
     rasterizer = rasterizer_;
-    puller.BindRasterizer(rasterizer);
 }
 
 } // namespace Tegra

src/video_core/dma_pusher.h

@@ -1,4 +1,4 @@
-// SPDX-FileCopyrightText: Copyright 2025 Eden Emulator Project
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
 // SPDX-License-Identifier: GPL-3.0-or-later
 
 // SPDX-FileCopyrightText: Copyright 2021 yuzu Emulator Project
@@ -109,25 +109,21 @@ inline CommandHeader BuildCommandHeader(BufferMethods method, u32 arg_count, Sub
 struct CommandList final {
     CommandList() = default;
     explicit CommandList(std::size_t size) : command_lists(size) {}
-    explicit CommandList(
+    explicit CommandList(boost::container::small_vector<CommandHeader, 512>&& prefetch_command_list_)
-        boost::container::small_vector<CommandHeader, 512>&& prefetch_command_list_)
         : prefetch_command_list{std::move(prefetch_command_list_)} {}
 
     boost::container::small_vector<CommandListHeader, 512> command_lists;
     boost::container::small_vector<CommandHeader, 512> prefetch_command_list;
 };
 
-/**
+/// @brief The DmaPusher class implements DMA submission to FIFOs, providing an area of memory that the
- * The DmaPusher class implements DMA submission to FIFOs, providing an area of memory that the
+/// emulated app fills with commands and tells PFIFO to process. The pushbuffers are then assembled
- * emulated app fills with commands and tells PFIFO to process. The pushbuffers are then assembled
+/// into a "command stream" consisting of 32-bit words that make up "commands".
- * into a "command stream" consisting of 32-bit words that make up "commands".
+/// See https://envytools.readthedocs.io/en/latest/hw/fifo/dma-pusher.html#fifo-dma-pusher for
- * See https://envytools.readthedocs.io/en/latest/hw/fifo/dma-pusher.html#fifo-dma-pusher for
+/// details on this implementation.
- * details on this implementation.
- */
 class DmaPusher final {
 public:
-    explicit DmaPusher(Core::System& system_, GPU& gpu_, MemoryManager& memory_manager_,
+    explicit DmaPusher(Core::System& system_, MemoryManager& memory_manager_, Control::ChannelState& channel_state_);
-                       Control::ChannelState& channel_state_);
     ~DmaPusher();
 
     void Push(CommandList&& entries) {
@@ -136,8 +132,7 @@ public:
 
     void DispatchCalls();
 
-    void BindSubchannel(Engines::EngineInterface* engine, u32 subchannel_id,
+    void BindSubchannel(Engines::EngineInterface* engine, u32 subchannel_id, Engines::EngineTypes engine_type) {
-                        Engines::EngineTypes engine_type) {
         subchannels[subchannel_id] = engine;
         subchannel_type[subchannel_id] = engine_type;
     }
@@ -152,11 +147,11 @@ private:
 
     void SetState(const CommandHeader& command_header);
 
-    void CallMethod(u32 argument) const;
+    void CallMethod(u32 argument);
-    void CallMultiMethod(const u32* base_start, u32 num_methods) const;
+    void CallMultiMethod(const u32* base_start, u32 num_methods);
 
-    Common::ScratchBuffer<CommandHeader>
+public:
-        command_headers; ///< Buffer for list of commands fetched at once
+    Common::ScratchBuffer<CommandHeader> command_headers; ///< Buffer for list of commands fetched at once
 
     std::queue<CommandList> dma_pushbuffer; ///< Queue of command lists to be processed
     std::size_t dma_pushbuffer_subindex{};  ///< Index within a command list within the pushbuffer
@@ -172,24 +167,24 @@ private:
         bool is_last_call;
     };
 
+    Core::System& system;
+    MemoryManager& memory_manager;
+    Control::ChannelState& channel_state;
+
     DmaState dma_state{};
-    bool dma_increment_once{};
-
-    const bool ib_enable{true}; ///< IB mode enabled
-
     std::array<Engines::EngineInterface*, max_subchannels> subchannels{};
     std::array<Engines::EngineTypes, max_subchannels> subchannel_type;
 
-    GPU& gpu;
+    Engines::Puller puller;
-    Core::System& system;
-    MemoryManager& memory_manager;
-    mutable Engines::Puller puller;
-
-    VideoCore::RasterizerInterface* rasterizer;
-    bool signal_sync;
-    bool synced;
     std::mutex sync_mutex;
     std::condition_variable sync_cv;
+
+    VideoCore::RasterizerInterface* rasterizer = nullptr;
+
+    const bool ib_enable : 1 = true; ///< IB mode enabled
+    bool dma_increment_once : 1 = false;
+    bool signal_sync : 1 = false;
+    bool synced : 1 = false;
 };
 
 } // namespace Tegra

src/video_core/engines/engine_interface.h

@@ -12,6 +12,10 @@
 
 #include "common/common_types.h"
 
+namespace Core {
+class System;
+}
+
 namespace Tegra::Engines {
 
 enum class EngineTypes : u32 {
@@ -28,28 +32,25 @@ public:
     virtual ~EngineInterface() = default;
 
     /// Write the value to the register identified by method.
-    virtual void CallMethod(u32 method, u32 method_argument, bool is_last_call) = 0;
+    virtual void CallMethod(Core::System& system, u32 method, u32 method_argument, bool is_last_call) = 0;
 
     /// Write multiple values to the register identified by method.
-    virtual void CallMultiMethod(u32 method, const u32* base_start, u32 amount,
+    virtual void CallMultiMethod(Core::System& system, u32 method, const u32* base_start, u32 amount, u32 methods_pending) = 0;
-                                 u32 methods_pending) = 0;
 
-    void ConsumeSink() {
+    void ConsumeSink(Core::System& system) {
-        if (method_sink.empty()) {
+        if (!method_sink.empty()) {
-            return;
+            ConsumeSinkImpl(system);
         }
-        ConsumeSinkImpl();
     }
 
     std::bitset<(std::numeric_limits<u16>::max)()> execution_mask{};
     std::vector<std::pair<u32, u32>> method_sink{};
-    bool current_dirty{};
     GPUVAddr current_dma_segment;
-
+    bool current_dirty{};
 protected:
-    virtual void ConsumeSinkImpl() {
+    virtual void ConsumeSinkImpl(Core::System& system) {
         for (auto [method, value] : method_sink) {
-            CallMethod(method, value, true);
+            CallMethod(system, method, value, true);
         }
         method_sink.clear();
     }

src/video_core/engines/fermi_2d.cpp

@@ -36,9 +36,8 @@ void Fermi2D::BindRasterizer(VideoCore::RasterizerInterface* rasterizer_) {
     rasterizer = rasterizer_;
 }
 
-void Fermi2D::CallMethod(u32 method, u32 method_argument, bool is_last_call) {
+void Fermi2D::CallMethod(Core::System& system, u32 method, u32 method_argument, bool is_last_call) {
-    ASSERT_MSG(method < Regs::NUM_REGS,
+    ASSERT_MSG(method < Regs::NUM_REGS, "Invalid Fermi2D register, increase the size of the Regs structure");
-               "Invalid Fermi2D register, increase the size of the Regs structure");
     regs.reg_array[method] = method_argument;
 
     if (method == FERMI2D_REG_INDEX(pixels_from_memory.src_y0) + 1) {
@@ -46,13 +45,13 @@ void Fermi2D::CallMethod(u32 method, u32 method_argument, bool is_last_call) {
     }
 }
 
-void Fermi2D::CallMultiMethod(u32 method, const u32* base_start, u32 amount, u32 methods_pending) {
+void Fermi2D::CallMultiMethod(Core::System& system, u32 method, const u32* base_start, u32 amount, u32 methods_pending) {
     for (u32 i = 0; i < amount; ++i) {
-        CallMethod(method, base_start[i], methods_pending - i <= 1);
+        CallMethod(system, method, base_start[i], methods_pending - i <= 1);
     }
 }
 
-void Fermi2D::ConsumeSinkImpl() {
+void Fermi2D::ConsumeSinkImpl(Core::System& system) {
     for (auto [method, value] : method_sink) {
         regs.reg_array[method] = value;
     }
@@ -60,8 +59,7 @@ void Fermi2D::ConsumeSinkImpl() {
 }
 
 void Fermi2D::Blit() {
-    LOG_DEBUG(HW_GPU, "called. source address=0x{:x}, destination address=0x{:x}",
+    LOG_DEBUG(HW_GPU, "called. source address=0x{:x}, destination address=0x{:x}", regs.src.Address(), regs.dst.Address());
-              regs.src.Address(), regs.dst.Address());
 
     UNIMPLEMENTED_IF_MSG(regs.operation != Operation::SrcCopy, "Operation is not copy");
     UNIMPLEMENTED_IF_MSG(regs.src.layer != 0, "Source layer is not zero");

src/video_core/engines/fermi_2d.h

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2018 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -44,11 +47,10 @@ public:
     void BindRasterizer(VideoCore::RasterizerInterface* rasterizer);
 
     /// Write the value to the register identified by method.
-    void CallMethod(u32 method, u32 method_argument, bool is_last_call) override;
+    void CallMethod(Core::System& system, u32 method, u32 method_argument, bool is_last_call) override;
 
     /// Write multiple values to the register identified by method.
-    void CallMultiMethod(u32 method, const u32* base_start, u32 amount,
+    void CallMultiMethod(Core::System& system, u32 method, const u32* base_start, u32 amount, u32 methods_pending) override;
-                         u32 methods_pending) override;
 
     enum class Origin : u32 {
         Center = 0,
@@ -311,7 +313,7 @@ private:
     /// registers.
     void Blit();
 
-    void ConsumeSinkImpl() override;
+    void ConsumeSinkImpl(Core::System& system) override;
 };
 
 #define ASSERT_REG_POSITION(field_name, position)                                                  \

src/video_core/engines/kepler_compute.cpp

@@ -16,8 +16,10 @@
 
 namespace Tegra::Engines {
 
-KeplerCompute::KeplerCompute(Core::System& system_, MemoryManager& memory_manager_)
+KeplerCompute::KeplerCompute(MemoryManager& memory_manager_)
-    : system{system_}, memory_manager{memory_manager_}, upload_state{memory_manager, regs.upload} {
+    : memory_manager{memory_manager_}
+    , upload_state{memory_manager, regs.upload}
+{
     execution_mask.reset();
     execution_mask[KEPLER_COMPUTE_REG_INDEX(exec_upload)] = true;
     execution_mask[KEPLER_COMPUTE_REG_INDEX(data_upload)] = true;
@@ -31,16 +33,15 @@ void KeplerCompute::BindRasterizer(VideoCore::RasterizerInterface* rasterizer_)
     upload_state.BindRasterizer(rasterizer);
 }
 
-void KeplerCompute::ConsumeSinkImpl() {
+void KeplerCompute::ConsumeSinkImpl(Core::System& system) {
     for (auto [method, value] : method_sink) {
         regs.reg_array[method] = value;
     }
     method_sink.clear();
 }
 
-void KeplerCompute::CallMethod(u32 method, u32 method_argument, bool is_last_call) {
+void KeplerCompute::CallMethod(Core::System& system, u32 method, u32 method_argument, bool is_last_call) {
-    ASSERT_MSG(method < Regs::NUM_REGS,
+    ASSERT_MSG(method < Regs::NUM_REGS, "Invalid KeplerCompute register, increase the size of the Regs structure");
-               "Invalid KeplerCompute register, increase the size of the Regs structure");
 
     regs.reg_array[method] = method_argument;
 
@@ -78,8 +79,7 @@ void KeplerCompute::CallMethod(u32 method, u32 method_argument, bool is_last_cal
     }
 }
 
-void KeplerCompute::CallMultiMethod(u32 method, const u32* base_start, u32 amount,
+void KeplerCompute::CallMultiMethod(Core::System& system, u32 method, const u32* base_start, u32 amount, u32 methods_pending) {
-                                    u32 methods_pending) {
     switch (method) {
     case KEPLER_COMPUTE_REG_INDEX(data_upload):
         upload_address = current_dma_segment;
@@ -87,7 +87,7 @@ void KeplerCompute::CallMultiMethod(u32 method, const u32* base_start, u32 amoun
         return;
     default:
         for (u32 i = 0; i < amount; i++) {
-            CallMethod(method, base_start[i], methods_pending - i <= 1);
+            CallMethod(system, method, base_start[i], methods_pending - i <= 1);
         }
         break;
     }

src/video_core/engines/kepler_compute.h

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2018 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -42,7 +45,7 @@ namespace Tegra::Engines {
 
 class KeplerCompute final : public EngineInterface {
 public:
-    explicit KeplerCompute(Core::System& system, MemoryManager& memory_manager);
+    explicit KeplerCompute(MemoryManager& memory_manager);
     ~KeplerCompute();
 
     /// Binds a rasterizer to this engine.
@@ -199,11 +202,10 @@ public:
                   "KeplerCompute LaunchParams has wrong size");
 
     /// Write the value to the register identified by method.
-    void CallMethod(u32 method, u32 method_argument, bool is_last_call) override;
+    void CallMethod(Core::System& system, u32 method, u32 method_argument, bool is_last_call) override;
 
     /// Write multiple values to the register identified by method.
-    void CallMultiMethod(u32 method, const u32* base_start, u32 amount,
+    void CallMultiMethod(Core::System& system, u32 method, const u32* base_start, u32 amount, u32 methods_pending) override;
-                         u32 methods_pending) override;
 
     std::optional<GPUVAddr> GetIndirectComputeAddress() const {
         return indirect_compute;
@@ -212,7 +214,7 @@ public:
 private:
     void ProcessLaunch();
 
-    void ConsumeSinkImpl() override;
+    void ConsumeSinkImpl(Core::System& system) override;
 
     /// Retrieves information about a specific TIC entry from the TIC buffer.
     Texture::TICEntry GetTICEntry(u32 tic_index) const;
@@ -220,7 +222,6 @@ private:
     /// Retrieves information about a specific TSC entry from the TSC buffer.
     Texture::TSCEntry GetTSCEntry(u32 tsc_index) const;
 
-    Core::System& system;
     MemoryManager& memory_manager;
     VideoCore::RasterizerInterface* rasterizer = nullptr;
     Upload::State upload_state;

src/video_core/engines/kepler_memory.cpp

@@ -14,8 +14,9 @@
 
 namespace Tegra::Engines {
 
-KeplerMemory::KeplerMemory(Core::System& system_, MemoryManager& memory_manager)
+KeplerMemory::KeplerMemory(MemoryManager& memory_manager)
-    : system{system_}, upload_state{memory_manager, regs.upload} {}
+    : upload_state{memory_manager, regs.upload}
+{}
 
 KeplerMemory::~KeplerMemory() = default;
 
@@ -27,16 +28,15 @@ void KeplerMemory::BindRasterizer(VideoCore::RasterizerInterface* rasterizer_) {
     execution_mask[KEPLERMEMORY_REG_INDEX(data)] = true;
 }
 
-void KeplerMemory::ConsumeSinkImpl() {
+void KeplerMemory::ConsumeSinkImpl(Core::System& system) {
     for (auto [method, value] : method_sink) {
         regs.reg_array[method] = value;
     }
     method_sink.clear();
 }
 
-void KeplerMemory::CallMethod(u32 method, u32 method_argument, bool is_last_call) {
+void KeplerMemory::CallMethod(Core::System& system, u32 method, u32 method_argument, bool is_last_call) {
-    ASSERT_MSG(method < Regs::NUM_REGS,
+    ASSERT_MSG(method < Regs::NUM_REGS, "Invalid KeplerMemory register, increase the size of the Regs structure");
-               "Invalid KeplerMemory register, increase the size of the Regs structure");
 
     regs.reg_array[method] = method_argument;
 
@@ -52,15 +52,14 @@ void KeplerMemory::CallMethod(u32 method, u32 method_argument, bool is_last_call
     }
 }
 
-void KeplerMemory::CallMultiMethod(u32 method, const u32* base_start, u32 amount,
+void KeplerMemory::CallMultiMethod(Core::System& system, u32 method, const u32* base_start, u32 amount, u32 methods_pending) {
-                                   u32 methods_pending) {
     switch (method) {
     case KEPLERMEMORY_REG_INDEX(data):
         upload_state.ProcessData(base_start, amount);
         return;
     default:
         for (u32 i = 0; i < amount; i++) {
-            CallMethod(method, base_start[i], methods_pending - i <= 1);
+            CallMethod(system, method, base_start[i], methods_pending - i <= 1);
         }
         break;
     }

src/video_core/engines/kepler_memory.h

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2018 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -36,18 +39,17 @@ namespace Tegra::Engines {
 
 class KeplerMemory final : public EngineInterface {
 public:
-    explicit KeplerMemory(Core::System& system_, MemoryManager& memory_manager);
+    explicit KeplerMemory(MemoryManager& memory_manager);
     ~KeplerMemory() override;
 
     /// Binds a rasterizer to this engine.
     void BindRasterizer(VideoCore::RasterizerInterface* rasterizer);
 
     /// Write the value to the register identified by method.
-    void CallMethod(u32 method, u32 method_argument, bool is_last_call) override;
+    void CallMethod(Core::System& system, u32 method, u32 method_argument, bool is_last_call) override;
 
     /// Write multiple values to the register identified by method.
-    void CallMultiMethod(u32 method, const u32* base_start, u32 amount,
+    void CallMultiMethod(Core::System& system, u32 method, const u32* base_start, u32 amount, u32 methods_pending) override;
-                         u32 methods_pending) override;
 
     struct Regs {
         static constexpr size_t NUM_REGS = 0x7F;
@@ -73,9 +75,7 @@ public:
     } regs{};
 
 private:
-    void ConsumeSinkImpl() override;
+    void ConsumeSinkImpl(Core::System& system) override;
-
-    Core::System& system;
     Upload::State upload_state;
 };
 

src/video_core/engines/maxwell_3d.cpp

@@ -25,9 +25,8 @@ namespace Tegra::Engines {
 /// First register id that is actually a Macro call.
 constexpr u32 MacroRegistersStart = 0xE00;
 
-Maxwell3D::Maxwell3D(Core::System& system_, MemoryManager& memory_manager_)
+Maxwell3D::Maxwell3D(MemoryManager& memory_manager_)
     : draw_manager()
-    , system{system_}
     , memory_manager{memory_manager_}
 #ifdef ARCHITECTURE_x86_64
     , macro_engine(bool(Settings::values.disable_macro_jit))
@@ -201,11 +200,10 @@ bool Maxwell3D::IsMethodExecutable(u32 method) {
     }
 }
 
-void Maxwell3D::ProcessMacro(u32 method, const u32* base_start, u32 amount, bool is_last_call) {
+void Maxwell3D::ProcessMacro(Core::System& system, u32 method, const u32* base_start, u32 amount, bool is_last_call) {
     if (executing_macro == 0) {
         // A macro call must begin by writing the macro method's register, not its argument.
-        ASSERT_MSG((method % 2) == 0,
+        ASSERT((method % 2) == 0 && "Can't start macro execution by writing to the ARGS register");
-                   "Can't start macro execution by writing to the ARGS register");
         executing_macro = method;
     }
 
@@ -219,8 +217,8 @@ void Maxwell3D::ProcessMacro(u32 method, const u32* base_start, u32 amount, bool
 
     // Call the macro when there are no more parameters in the command buffer
     if (is_last_call) {
-        ConsumeSink();
+        ConsumeSink(system);
-        CallMacroMethod(executing_macro, macro_params);
+        CallMacroMethod(system, executing_macro, macro_params);
         macro_params.clear();
         macro_addresses.clear();
         macro_segments.clear();
@@ -287,7 +285,7 @@ u32 Maxwell3D::ProcessShadowRam(u32 method, u32 argument) {
     return argument;
 }
 
-void Maxwell3D::ConsumeSinkImpl() {
+void Maxwell3D::ConsumeSinkImpl(Core::System& system) {
     const auto control = shadow_state.shadow_ram_control;
     if (control == Regs::ShadowRamControl::Track || control == Regs::ShadowRamControl::TrackWithFilter) {
         for (auto [method, value] : method_sink) {
@@ -378,7 +376,7 @@ void Maxwell3D::ProcessMethodCall(u32 method, u32 argument, u32 nonshadow_argume
     }
 }
 
-void Maxwell3D::CallMacroMethod(u32 method, const std::vector<u32>& parameters) {
+void Maxwell3D::CallMacroMethod(Core::System& system, u32 method, const std::vector<u32>& parameters) {
     // Reset the current macro.
     executing_macro = 0;
 
@@ -387,11 +385,11 @@ void Maxwell3D::CallMacroMethod(u32 method, const std::vector<u32>& parameters)
         ((method - MacroRegistersStart) >> 1) % static_cast<u32>(macro_positions.size());
 
     // Execute the current macro.
-    macro_engine.Execute(*this, macro_positions[entry], parameters);
+    macro_engine.Execute(system, *this, macro_positions[entry], parameters);
     draw_manager.DrawDeferred(*this);
 }
 
-void Maxwell3D::CallMethod(u32 method, u32 method_argument, bool is_last_call) {
+void Maxwell3D::CallMethod(Core::System& system, u32 method, u32 method_argument, bool is_last_call) {
     // It is an error to write to a register other than the current macro's ARG register before
     // it has finished execution.
     if (executing_macro != 0) {
@@ -401,7 +399,7 @@ void Maxwell3D::CallMethod(u32 method, u32 method_argument, bool is_last_call) {
     // Methods after 0xE00 are special, they're actually triggers for some microcode that was
     // uploaded to the GPU during initialization.
     if (method >= MacroRegistersStart) {
-        ProcessMacro(method, &method_argument, 1, is_last_call);
+        ProcessMacro(system, method, &method_argument, 1, is_last_call);
         return;
     }
 
@@ -411,12 +409,11 @@ void Maxwell3D::CallMethod(u32 method, u32 method_argument, bool is_last_call) {
     ProcessMethodCall(method, argument, method_argument, is_last_call);
 }
 
-void Maxwell3D::CallMultiMethod(u32 method, const u32* base_start, u32 amount,
+void Maxwell3D::CallMultiMethod(Core::System& system, u32 method, const u32* base_start, u32 amount, u32 methods_pending) {
-                                u32 methods_pending) {
     // Methods after 0xE00 are special, they're actually triggers for some microcode that was
     // uploaded to the GPU during initialization.
     if (method >= MacroRegistersStart) {
-        ProcessMacro(method, base_start, amount, amount == methods_pending);
+        ProcessMacro(system, method, base_start, amount, amount == methods_pending);
         return;
     }
     switch (method) {
@@ -445,7 +442,7 @@ void Maxwell3D::CallMultiMethod(u32 method, const u32* base_start, u32 amount,
     }
     default:
         for (u32 i = 0; i < amount; i++) {
-            CallMethod(method, base_start[i], methods_pending - i <= 1);
+            CallMethod(system, method, base_start[i], methods_pending - i <= 1);
         }
         break;
     }
@@ -467,7 +464,7 @@ void Maxwell3D::ProcessFirmwareCall4() {
     regs.shadow_scratch[0] = 1;
 }
 
-void Maxwell3D::StampQueryResult(u64 payload, bool long_query) {
+void Maxwell3D::StampQueryResult(Core::System& system, u64 payload, bool long_query) {
     const GPUVAddr sequence_address{regs.report_semaphore.Address()};
     if (long_query) {
         memory_manager.Write<u64>(sequence_address + sizeof(u64), system.GPU().GetTicks());

src/video_core/engines/maxwell_3d.h

@@ -55,7 +55,7 @@ namespace Tegra::Engines {
 
 class Maxwell3D final : public EngineInterface {
 public:
-    explicit Maxwell3D(Core::System& system, MemoryManager& memory_manager);
+    explicit Maxwell3D(MemoryManager& memory_manager);
     ~Maxwell3D();
 
     /// Binds a rasterizer to this engine.
@@ -3129,11 +3129,10 @@ public:
     u32 GetRegisterValue(u32 method) const;
 
     /// Write the value to the register identified by method.
-    void CallMethod(u32 method, u32 method_argument, bool is_last_call) override;
+    void CallMethod(Core::System& system, u32 method, u32 method_argument, bool is_last_call) override;
 
     /// Write multiple values to the register identified by method.
-    void CallMultiMethod(u32 method, const u32* base_start, u32 amount,
+    void CallMultiMethod(Core::System& system, u32 method, const u32* base_start, u32 amount, u32 methods_pending) override;
-                         u32 methods_pending) override;
 
     bool ShouldExecute() const {
         return execute_on;
@@ -3190,13 +3189,13 @@ public:
 private:
     void InitializeRegisterDefaults();
 
-    void ProcessMacro(u32 method, const u32* base_start, u32 amount, bool is_last_call);
+    void ProcessMacro(Core::System& system, u32 method, const u32* base_start, u32 amount, bool is_last_call);
 
     u32 ProcessShadowRam(u32 method, u32 argument);
 
     void ProcessDirtyRegisters(u32 method, u32 argument);
 
-    void ConsumeSinkImpl() override;
+    void ConsumeSinkImpl(Core::System& system) override;
 
     void ProcessMethodCall(u32 method, u32 argument, u32 nonshadow_argument, bool is_last_call);
 
@@ -3212,7 +3211,7 @@ private:
      * @param method Method to call
      * @param parameters Arguments to the method call
      */
-    void CallMacroMethod(u32 method, const std::vector<u32>& parameters);
+    void CallMacroMethod(Core::System& system, u32 method, const std::vector<u32>& parameters);
 
     /// Handles writes to the macro uploading register.
     void ProcessMacroUpload(u32 data);
@@ -3227,7 +3226,7 @@ private:
     void ProcessQueryGet();
 
     /// Writes the query result accordingly.
-    void StampQueryResult(u64 payload, bool long_query);
+    void StampQueryResult(Core::System& system, u64 payload, bool long_query);
 
     /// Handles conditional rendering.
     void ProcessQueryCondition();
@@ -3242,7 +3241,6 @@ private:
 
     bool IsMethodExecutable(u32 method);
 
-    Core::System& system;
     MemoryManager& memory_manager;
 
     VideoCore::RasterizerInterface* rasterizer = nullptr;

src/video_core/engines/maxwell_dma.cpp

@@ -21,8 +21,9 @@ namespace Tegra::Engines {
 
 using namespace Texture;
 
-MaxwellDMA::MaxwellDMA(Core::System& system_, MemoryManager& memory_manager_)
+MaxwellDMA::MaxwellDMA(MemoryManager& memory_manager_)
-    : system{system_}, memory_manager{memory_manager_} {
+    : memory_manager{memory_manager_}
+{
     execution_mask.reset();
     execution_mask[offsetof(Regs, launch_dma) / sizeof(u32)] = true;
 }
@@ -33,14 +34,14 @@ void MaxwellDMA::BindRasterizer(VideoCore::RasterizerInterface* rasterizer_) {
     rasterizer = rasterizer_;
 }
 
-void MaxwellDMA::ConsumeSinkImpl() {
+void MaxwellDMA::ConsumeSinkImpl(Core::System& system) {
     for (auto [method, value] : method_sink) {
         regs.reg_array[method] = value;
     }
     method_sink.clear();
 }
 
-void MaxwellDMA::CallMethod(u32 method, u32 method_argument, bool is_last_call) {
+void MaxwellDMA::CallMethod(Core::System& system, u32 method, u32 method_argument, bool is_last_call) {
     ASSERT_MSG(method < NUM_REGS, "Invalid MaxwellDMA register");
 
     regs.reg_array[method] = method_argument;
@@ -50,16 +51,14 @@ void MaxwellDMA::CallMethod(u32 method, u32 method_argument, bool is_last_call)
     }
 }
 
-void MaxwellDMA::CallMultiMethod(u32 method, const u32* base_start, u32 amount,
+void MaxwellDMA::CallMultiMethod(Core::System& system, u32 method, const u32* base_start, u32 amount, u32 methods_pending) {
-                                 u32 methods_pending) {
     for (u32 i = 0; i < amount; ++i) {
-        CallMethod(method, base_start[i], methods_pending - i <= 1);
+        CallMethod(system, method, base_start[i], methods_pending - i <= 1);
     }
 }
 
 void MaxwellDMA::Launch() {
-    LOG_TRACE(Render_OpenGL, "DMA copy 0x{:x} -> 0x{:x}", static_cast<GPUVAddr>(regs.offset_in),
+    LOG_TRACE(Render_OpenGL, "DMA copy 0x{:x} -> 0x{:x}", static_cast<GPUVAddr>(regs.offset_in), GPUVAddr(regs.offset_out));
-              static_cast<GPUVAddr>(regs.offset_out));
 
     // TODO(Subv): Perform more research and implement all features of this engine.
     const LaunchDMA& launch = regs.launch_dma;

src/video_core/engines/maxwell_dma.h

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2018 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -238,15 +241,14 @@ public:
 
     void BindRasterizer(VideoCore::RasterizerInterface* rasterizer);
 
-    explicit MaxwellDMA(Core::System& system_, MemoryManager& memory_manager_);
+    explicit MaxwellDMA(MemoryManager& memory_manager_);
     ~MaxwellDMA() override;
 
     /// Write the value to the register identified by method.
-    void CallMethod(u32 method, u32 method_argument, bool is_last_call) override;
+    void CallMethod(Core::System& system, u32 method, u32 method_argument, bool is_last_call) override;
 
     /// Write multiple values to the register identified by method.
-    void CallMultiMethod(u32 method, const u32* base_start, u32 amount,
+    void CallMultiMethod(Core::System& system, u32 method, const u32* base_start, u32 amount, u32 methods_pending) override;
-                         u32 methods_pending) override;
 
 private:
     /// Performs the copy from the source buffer to the destination buffer as configured in the
@@ -261,9 +263,7 @@ private:
 
     void ReleaseSemaphore();
 
-    void ConsumeSinkImpl() override;
+    void ConsumeSinkImpl(Core::System& system) override;
-
-    Core::System& system;
 
     MemoryManager& memory_manager;
     VideoCore::RasterizerInterface* rasterizer = nullptr;

src/video_core/engines/nv01_timer.h

@@ -26,18 +26,16 @@ class MemoryManager;
 namespace Tegra::Engines {
 class Nv01Timer final : public EngineInterface {
 public:
-    explicit Nv01Timer(Core::System& system_, MemoryManager& memory_manager)
+    explicit Nv01Timer(MemoryManager& memory_manager) noexcept {}
-        : system{system_}
+    ~Nv01Timer() noexcept override {}
-    {}
-    ~Nv01Timer() override;
 
     /// Write the value to the register identified by method.
-    void CallMethod(u32 method, u32 method_argument, bool is_last_call) override {
+    void CallMethod(Core::System& system, u32 method, u32 method_argument, bool is_last_call) override {
         LOG_DEBUG(HW_GPU, "method={}, argument={}, is_last_call={}", method, method_argument, is_last_call);
     }
 
     /// Write multiple values to the register identified by method.
-    void CallMultiMethod(u32 method, const u32* base_start, u32 amount, u32 methods_pending) override {
+    void CallMultiMethod(Core::System& system, u32 method, const u32* base_start, u32 amount, u32 methods_pending) override {
         LOG_DEBUG(HW_GPU, "method={}, base_start={}, amount={}, pending={}", method, fmt::ptr(base_start), amount, methods_pending);
     }
 
@@ -46,7 +44,6 @@ public:
         INSERT_PADDING_BYTES_NOINIT(0x48);
     } regs{};
 private:
-    void ConsumeSinkImpl() override {}
+    void ConsumeSinkImpl(Core::System& system) override {}
-    Core::System& system;
 };
 }

src/video_core/engines/puller.cpp

@@ -22,37 +22,29 @@
 
 namespace Tegra::Engines {
 
-Puller::Puller(GPU& gpu_, MemoryManager& memory_manager_, DmaPusher& dma_pusher_,
+void Puller::ProcessBindMethod(DmaPusher& dma_pusher, const MethodCall& method_call) {
-               Control::ChannelState& channel_state_)
-    : gpu{gpu_}, memory_manager{memory_manager_}, dma_pusher{dma_pusher_}, channel_state{
-                                                                               channel_state_} {}
-
-Puller::~Puller() = default;
-
-void Puller::ProcessBindMethod(const MethodCall& method_call) {
     // Bind the current subchannel to the desired engine id.
-    LOG_DEBUG(HW_GPU, "Binding subchannel {} to engine {}", method_call.subchannel,
+    LOG_DEBUG(HW_GPU, "Binding subchannel {} to engine {}", method_call.subchannel, method_call.argument);
-              method_call.argument);
     const auto engine_id = static_cast<EngineID>(method_call.argument);
     bound_engines[method_call.subchannel] = engine_id;
     switch (engine_id) {
     case EngineID::FERMI_TWOD_A:
-        dma_pusher.BindSubchannel(&*channel_state.fermi_2d, method_call.subchannel, EngineTypes::Fermi2D);
+        dma_pusher.BindSubchannel(&dma_pusher.channel_state.payload->fermi_2d, method_call.subchannel, EngineTypes::Fermi2D);
         break;
     case EngineID::MAXWELL_B:
-        dma_pusher.BindSubchannel(&*channel_state.maxwell_3d, method_call.subchannel, EngineTypes::Maxwell3D);
+        dma_pusher.BindSubchannel(&dma_pusher.channel_state.payload->maxwell_3d, method_call.subchannel, EngineTypes::Maxwell3D);
         break;
     case EngineID::KEPLER_COMPUTE_B:
-        dma_pusher.BindSubchannel(&*channel_state.kepler_compute, method_call.subchannel, EngineTypes::KeplerCompute);
+        dma_pusher.BindSubchannel(&dma_pusher.channel_state.payload->kepler_compute, method_call.subchannel, EngineTypes::KeplerCompute);
         break;
     case EngineID::MAXWELL_DMA_COPY_A:
-        dma_pusher.BindSubchannel(&*channel_state.maxwell_dma, method_call.subchannel, EngineTypes::MaxwellDMA);
+        dma_pusher.BindSubchannel(&dma_pusher.channel_state.payload->maxwell_dma, method_call.subchannel, EngineTypes::MaxwellDMA);
         break;
     case EngineID::KEPLER_INLINE_TO_MEMORY_B:
-        dma_pusher.BindSubchannel(&*channel_state.kepler_memory, method_call.subchannel, EngineTypes::KeplerMemory);
+        dma_pusher.BindSubchannel(&dma_pusher.channel_state.payload->kepler_memory, method_call.subchannel, EngineTypes::KeplerMemory);
         break;
     case EngineID::NV01_TIMER:
-        dma_pusher.BindSubchannel(&*channel_state.nv01_timer, method_call.subchannel, EngineTypes::Nv01Timer);
+        dma_pusher.BindSubchannel(&dma_pusher.channel_state.payload->nv01_timer, method_call.subchannel, EngineTypes::Nv01Timer);
         break;
     default:
         UNIMPLEMENTED_MSG("Unimplemented engine {:04X}", engine_id);
@@ -60,15 +52,15 @@ void Puller::ProcessBindMethod(const MethodCall& method_call) {
     }
 }
 
-void Puller::ProcessFenceActionMethod() {
+void Puller::ProcessFenceActionMethod(DmaPusher& dma_pusher) {
     switch (regs.fence_action.op) {
     case Puller::FenceOperation::Acquire:
         // UNIMPLEMENTED_MSG("Channel Scheduling pending.");
         // WaitFence(regs.fence_action.syncpoint_id, regs.fence_value);
-        rasterizer->ReleaseFences();
+        dma_pusher.rasterizer->ReleaseFences();
         break;
     case Puller::FenceOperation::Increment:
-        rasterizer->SignalSyncPoint(regs.fence_action.syncpoint_id);
+        dma_pusher.rasterizer->SignalSyncPoint(regs.fence_action.syncpoint_id);
         break;
     default:
         UNIMPLEMENTED_MSG("Unimplemented operation {}", regs.fence_action.op.Value());
@@ -76,37 +68,35 @@ void Puller::ProcessFenceActionMethod() {
     }
 }
 
-void Puller::ProcessSemaphoreTriggerMethod() {
+void Puller::ProcessSemaphoreTriggerMethod(DmaPusher& dma_pusher) {
     const auto semaphoreOperationMask = 0xF;
-    const auto op =
+    const auto op = GpuSemaphoreOperation(regs.semaphore_trigger & semaphoreOperationMask);
-        static_cast<GpuSemaphoreOperation>(regs.semaphore_trigger & semaphoreOperationMask);
     if (op == GpuSemaphoreOperation::WriteLong) {
         const GPUVAddr sequence_address{regs.semaphore_address.SemaphoreAddress()};
         const u32 payload = regs.semaphore_sequence;
-        rasterizer->Query(sequence_address, VideoCommon::QueryType::Payload,
+        dma_pusher.rasterizer->Query(sequence_address, VideoCommon::QueryType::Payload, VideoCommon::QueryPropertiesFlags::HasTimeout, payload, 0);
-                          VideoCommon::QueryPropertiesFlags::HasTimeout, payload, 0);
     } else {
         do {
-            const u32 word{memory_manager.Read<u32>(regs.semaphore_address.SemaphoreAddress())};
+            const u32 word = dma_pusher.memory_manager.Read<u32>(regs.semaphore_address.SemaphoreAddress());
             regs.acquire_source = true;
             regs.acquire_value = regs.semaphore_sequence;
             if (op == GpuSemaphoreOperation::AcquireEqual) {
                 regs.acquire_active = true;
                 regs.acquire_mode = false;
                 if (word != regs.acquire_value) {
-                    rasterizer->ReleaseFences();
+                    dma_pusher.rasterizer->ReleaseFences();
                     continue;
                 }
             } else if (op == GpuSemaphoreOperation::AcquireGequal) {
                 regs.acquire_active = true;
                 regs.acquire_mode = true;
                 if (word < regs.acquire_value) {
-                    rasterizer->ReleaseFences();
+                    dma_pusher.rasterizer->ReleaseFences();
                     continue;
                 }
             } else if (op == GpuSemaphoreOperation::AcquireMask) {
                 if (word && regs.semaphore_sequence == 0) {
-                    rasterizer->ReleaseFences();
+                    dma_pusher.rasterizer->ReleaseFences();
                     continue;
                 }
             } else {
@@ -116,21 +106,20 @@ void Puller::ProcessSemaphoreTriggerMethod() {
     }
 }
 
-void Puller::ProcessSemaphoreRelease() {
+void Puller::ProcessSemaphoreRelease(DmaPusher& dma_pusher) {
     const GPUVAddr sequence_address{regs.semaphore_address.SemaphoreAddress()};
     const u32 payload = regs.semaphore_release;
-    rasterizer->Query(sequence_address, VideoCommon::QueryType::Payload,
+    dma_pusher.rasterizer->Query(sequence_address, VideoCommon::QueryType::Payload, VideoCommon::QueryPropertiesFlags::IsAFence, payload, 0);
-                      VideoCommon::QueryPropertiesFlags::IsAFence, payload, 0);
 }
 
-void Puller::ProcessSemaphoreAcquire() {
+void Puller::ProcessSemaphoreAcquire(DmaPusher& dma_pusher) {
-    u32 word = memory_manager.Read<u32>(regs.semaphore_address.SemaphoreAddress());
+    u32 word = dma_pusher.memory_manager.Read<u32>(regs.semaphore_address.SemaphoreAddress());
     const auto value = regs.semaphore_acquire;
     while (word != value) {
         regs.acquire_active = true;
         regs.acquire_value = value;
-        rasterizer->ReleaseFences();
+        dma_pusher.rasterizer->ReleaseFences();
-        word = memory_manager.Read<u32>(regs.semaphore_address.SemaphoreAddress());
+        word = dma_pusher.memory_manager.Read<u32>(regs.semaphore_address.SemaphoreAddress());
         // TODO(kemathe73) figure out how to do the acquire_timeout
         regs.acquire_mode = false;
         regs.acquire_source = false;
@@ -138,13 +127,13 @@ void Puller::ProcessSemaphoreAcquire() {
 }
 
 /// Calls a GPU puller method.
-void Puller::CallPullerMethod(const MethodCall& method_call) {
+void Puller::CallPullerMethod(DmaPusher& dma_pusher, const MethodCall& method_call) {
     regs.reg_array[method_call.method] = method_call.argument;
     const auto method = static_cast<BufferMethods>(method_call.method);
 
     switch (method) {
     case BufferMethods::BindObject: {
-        ProcessBindMethod(method_call);
+        ProcessBindMethod(dma_pusher, method_call);
         break;
     }
     case BufferMethods::Nop:
@@ -155,16 +144,16 @@ void Puller::CallPullerMethod(const MethodCall& method_call) {
     case BufferMethods::WrcacheFlush:
         break;
     case BufferMethods::RefCnt:
-        rasterizer->SignalReference();
+        dma_pusher.rasterizer->SignalReference();
         break;
     case BufferMethods::SyncpointOperation:
-        ProcessFenceActionMethod();
+        ProcessFenceActionMethod(dma_pusher);
         break;
     case BufferMethods::WaitForIdle:
-        rasterizer->WaitForIdle();
+        dma_pusher.rasterizer->WaitForIdle();
         break;
     case BufferMethods::SemaphoreOperation: {
-        ProcessSemaphoreTriggerMethod();
+        ProcessSemaphoreTriggerMethod(dma_pusher);
         break;
     }
     case BufferMethods::NonStallInterrupt: {
@@ -177,7 +166,7 @@ void Puller::CallPullerMethod(const MethodCall& method_call) {
     }
     case BufferMethods::MemOpB: {
         // Implement this better.
-        rasterizer->InvalidateGPUCache();
+        dma_pusher.rasterizer->InvalidateGPUCache();
         break;
     }
     case BufferMethods::MemOpC:
@@ -186,11 +175,11 @@ void Puller::CallPullerMethod(const MethodCall& method_call) {
         break;
     }
     case BufferMethods::SemaphoreAcquire: {
-        ProcessSemaphoreAcquire();
+        ProcessSemaphoreAcquire(dma_pusher);
         break;
     }
     case BufferMethods::SemaphoreRelease: {
-        ProcessSemaphoreRelease();
+        ProcessSemaphoreRelease(dma_pusher);
         break;
     }
     case BufferMethods::Yield: {
@@ -205,27 +194,26 @@ void Puller::CallPullerMethod(const MethodCall& method_call) {
 }
 
 /// Calls a GPU engine method.
-void Puller::CallEngineMethod(const MethodCall& method_call) {
+void Puller::CallEngineMethod(DmaPusher& dma_pusher, const MethodCall& method_call) {
     const EngineID engine = bound_engines[method_call.subchannel];
-
     switch (engine) {
     case EngineID::FERMI_TWOD_A:
-        channel_state.fermi_2d->CallMethod(method_call.method, method_call.argument, method_call.IsLastCall());
+        dma_pusher.channel_state.payload->fermi_2d.CallMethod(dma_pusher.system, method_call.method, method_call.argument, method_call.IsLastCall());
         break;
     case EngineID::MAXWELL_B:
-        channel_state.maxwell_3d->CallMethod(method_call.method, method_call.argument, method_call.IsLastCall());
+        dma_pusher.channel_state.payload->maxwell_3d.CallMethod(dma_pusher.system, method_call.method, method_call.argument, method_call.IsLastCall());
         break;
     case EngineID::KEPLER_COMPUTE_B:
-        channel_state.kepler_compute->CallMethod(method_call.method, method_call.argument, method_call.IsLastCall());
+        dma_pusher.channel_state.payload->kepler_compute.CallMethod(dma_pusher.system, method_call.method, method_call.argument, method_call.IsLastCall());
         break;
     case EngineID::MAXWELL_DMA_COPY_A:
-        channel_state.maxwell_dma->CallMethod(method_call.method, method_call.argument, method_call.IsLastCall());
+        dma_pusher.channel_state.payload->maxwell_dma.CallMethod(dma_pusher.system, method_call.method, method_call.argument, method_call.IsLastCall());
         break;
     case EngineID::KEPLER_INLINE_TO_MEMORY_B:
-        channel_state.kepler_memory->CallMethod(method_call.method, method_call.argument, method_call.IsLastCall());
+        dma_pusher.channel_state.payload->kepler_memory.CallMethod(dma_pusher.system, method_call.method, method_call.argument, method_call.IsLastCall());
         break;
     case EngineID::NV01_TIMER:
-        channel_state.nv01_timer->CallMethod(method_call.method, method_call.argument, method_call.IsLastCall());
+        dma_pusher.channel_state.payload->nv01_timer.CallMethod(dma_pusher.system, method_call.method, method_call.argument, method_call.IsLastCall());
         break;
     default:
         UNIMPLEMENTED_MSG("Unimplemented engine");
@@ -234,28 +222,26 @@ void Puller::CallEngineMethod(const MethodCall& method_call) {
 }
 
 /// Calls a GPU engine multivalue method.
-void Puller::CallEngineMultiMethod(u32 method, u32 subchannel, const u32* base_start, u32 amount,
+void Puller::CallEngineMultiMethod(DmaPusher& dma_pusher, u32 method, u32 subchannel, const u32* base_start, u32 amount, u32 methods_pending) {
-                                   u32 methods_pending) {
     const EngineID engine = bound_engines[subchannel];
-
     switch (engine) {
     case EngineID::FERMI_TWOD_A:
-        channel_state.fermi_2d->CallMultiMethod(method, base_start, amount, methods_pending);
+        dma_pusher.channel_state.payload->fermi_2d.CallMultiMethod(dma_pusher.system, method, base_start, amount, methods_pending);
         break;
     case EngineID::MAXWELL_B:
-        channel_state.maxwell_3d->CallMultiMethod(method, base_start, amount, methods_pending);
+        dma_pusher.channel_state.payload->maxwell_3d.CallMultiMethod(dma_pusher.system, method, base_start, amount, methods_pending);
         break;
     case EngineID::KEPLER_COMPUTE_B:
-        channel_state.kepler_compute->CallMultiMethod(method, base_start, amount, methods_pending);
+        dma_pusher.channel_state.payload->kepler_compute.CallMultiMethod(dma_pusher.system, method, base_start, amount, methods_pending);
         break;
     case EngineID::MAXWELL_DMA_COPY_A:
-        channel_state.maxwell_dma->CallMultiMethod(method, base_start, amount, methods_pending);
+        dma_pusher.channel_state.payload->maxwell_dma.CallMultiMethod(dma_pusher.system, method, base_start, amount, methods_pending);
         break;
     case EngineID::KEPLER_INLINE_TO_MEMORY_B:
-        channel_state.kepler_memory->CallMultiMethod(method, base_start, amount, methods_pending);
+        dma_pusher.channel_state.payload->kepler_memory.CallMultiMethod(dma_pusher.system, method, base_start, amount, methods_pending);
         break;
     case EngineID::NV01_TIMER:
-        channel_state.nv01_timer->CallMultiMethod(method, base_start, amount, methods_pending);
+        dma_pusher.channel_state.payload->nv01_timer.CallMultiMethod(dma_pusher.system, method, base_start, amount, methods_pending);
         break;
     default:
         UNIMPLEMENTED_MSG("Unimplemented engine");
@@ -264,31 +250,26 @@ void Puller::CallEngineMultiMethod(u32 method, u32 subchannel, const u32* base_s
 }
 
 /// Calls a GPU method.
-void Puller::CallMethod(const MethodCall& method_call) {
+void Puller::CallMethod(DmaPusher& dma_pusher, const MethodCall& method_call) {
-    LOG_TRACE(HW_GPU, "Processing method {:08X} on subchannel {}", method_call.method,
+    LOG_TRACE(HW_GPU, "Processing method {:08X} on subchannel {}", method_call.method, method_call.subchannel);
-              method_call.subchannel);
-
     ASSERT(method_call.subchannel < bound_engines.size());
 
-    if (ExecuteMethodOnEngine(method_call.method)) {
+    if (ExecuteMethodOnEngine(dma_pusher, method_call.method)) {
-        CallEngineMethod(method_call);
+        CallEngineMethod(dma_pusher, method_call);
     } else {
-        CallPullerMethod(method_call);
+        CallPullerMethod(dma_pusher, method_call);
     }
 }
 
 /// Calls a GPU multivalue method.
-void Puller::CallMultiMethod(u32 method, u32 subchannel, const u32* base_start, u32 amount,
+void Puller::CallMultiMethod(DmaPusher& dma_pusher, u32 method, u32 subchannel, const u32* base_start, u32 amount, u32 methods_pending) {
-                             u32 methods_pending) {
     LOG_TRACE(HW_GPU, "Processing method {:08X} on subchannel {}", method, subchannel);
-
     ASSERT(subchannel < bound_engines.size());
-
+    if (ExecuteMethodOnEngine(dma_pusher, method)) {
-    if (ExecuteMethodOnEngine(method)) {
+        CallEngineMultiMethod(dma_pusher, method, subchannel, base_start, amount, methods_pending);
-        CallEngineMultiMethod(method, subchannel, base_start, amount, methods_pending);
     } else {
         for (u32 i = 0; i < amount; i++) {
-            CallPullerMethod(MethodCall{
+            CallPullerMethod(dma_pusher, MethodCall{
                 method,
                 base_start[i],
                 subchannel,
@@ -298,13 +279,9 @@ void Puller::CallMultiMethod(u32 method, u32 subchannel, const u32* base_start,
     }
 }
 
-void Puller::BindRasterizer(VideoCore::RasterizerInterface* rasterizer_) {
-    rasterizer = rasterizer_;
-}
-
 /// Determines where the method should be executed.
-[[nodiscard]] bool Puller::ExecuteMethodOnEngine(u32 method) {
+[[nodiscard]] bool Puller::ExecuteMethodOnEngine(DmaPusher& dma_pusher, u32 method) {
-    const auto buffer_method = static_cast<BufferMethods>(method);
+    const auto buffer_method = BufferMethods(method);
     return buffer_method >= BufferMethods::NonPullerMethods;
 }
 

src/video_core/engines/puller.h

@@ -70,32 +70,13 @@ public:
         BitField<8, 24, u32> syncpoint_id;
     };
 
-    explicit Puller(GPU& gpu_, MemoryManager& memory_manager_, DmaPusher& dma_pusher,
+    void CallMethod(DmaPusher& dma_pusher, const MethodCall& method_call);
-                    Control::ChannelState& channel_state);
+    void CallMultiMethod(DmaPusher& dma_pusher, u32 method, u32 subchannel, const u32* base_start, u32 amount, u32 methods_pending);
-    ~Puller();
+    void BindRasterizer(DmaPusher& dma_pusher, VideoCore::RasterizerInterface* rasterizer);
-
+    void CallPullerMethod(DmaPusher& dma_pusher, const MethodCall& method_call);
-    void CallMethod(const MethodCall& method_call);
+    void CallEngineMethod(DmaPusher& dma_pusher, const MethodCall& method_call);
-
+    void CallEngineMultiMethod(DmaPusher& dma_pusher, u32 method, u32 subchannel, const u32* base_start, u32 amount, u32 methods_pending);
-    void CallMultiMethod(u32 method, u32 subchannel, const u32* base_start, u32 amount,
-                         u32 methods_pending);
-
-    void BindRasterizer(VideoCore::RasterizerInterface* rasterizer);
-
-    void CallPullerMethod(const MethodCall& method_call);
-
-    void CallEngineMethod(const MethodCall& method_call);
-
-    void CallEngineMultiMethod(u32 method, u32 subchannel, const u32* base_start, u32 amount,
-                               u32 methods_pending);
-
 private:
-    Tegra::GPU& gpu;
-
-    MemoryManager& memory_manager;
-    DmaPusher& dma_pusher;
-    Control::ChannelState& channel_state;
-    VideoCore::RasterizerInterface* rasterizer = nullptr;
-
     static constexpr std::size_t NUM_REGS = 0x800;
     struct Regs {
         static constexpr size_t NUM_REGS = 0x40;
@@ -139,12 +120,12 @@ private:
         };
     } regs{};
 
-    void ProcessBindMethod(const MethodCall& method_call);
+    void ProcessBindMethod(DmaPusher& dma_pusher, const MethodCall& method_call);
-    void ProcessFenceActionMethod();
+    void ProcessFenceActionMethod(DmaPusher& dma_pusher);
-    void ProcessSemaphoreAcquire();
+    void ProcessSemaphoreAcquire(DmaPusher& dma_pusher);
-    void ProcessSemaphoreRelease();
+    void ProcessSemaphoreRelease(DmaPusher& dma_pusher);
-    void ProcessSemaphoreTriggerMethod();
+    void ProcessSemaphoreTriggerMethod(DmaPusher& dma_pusher);
-    [[nodiscard]] bool ExecuteMethodOnEngine(u32 method);
+    [[nodiscard]] bool ExecuteMethodOnEngine(DmaPusher& dma_pusher, u32 method);
 
     /// Mapping of command subchannels to their bound engine ids
     std::array<EngineID, 8> bound_engines{};
@@ -157,8 +138,7 @@ private:
     };
 
 #define ASSERT_REG_POSITION(field_name, position)                                                  \
-    static_assert(offsetof(Regs, field_name) == position * 4,                                      \
+    static_assert(offsetof(Regs, field_name) == position * 4, "Field " #field_name " has invalid position")
-                  "Field " #field_name " has invalid position")
 
     ASSERT_REG_POSITION(semaphore_address, 0x4);
     ASSERT_REG_POSITION(semaphore_sequence, 0x6);

src/video_core/gpu.cpp

@@ -40,30 +40,31 @@
 namespace Tegra {
 
 struct GPU::Impl {
-    explicit Impl(GPU& gpu_, Core::System& system_, bool is_async_, bool use_nvdec_)
+    explicit Impl(Core::System& system_, bool is_async_, bool use_nvdec_)
-        : gpu{gpu_}, system{system_}, host1x{system.Host1x()}, use_nvdec{use_nvdec_},
+        : system{system_}
-          shader_notify{std::make_unique<VideoCore::ShaderNotify>()}, is_async{is_async_},
+        , use_nvdec{use_nvdec_}
-          gpu_thread{system_, is_async_}, scheduler{std::make_unique<Control::Scheduler>(gpu)} {}
+        , shader_notify()
+        , is_async{is_async_}
+        , gpu_thread{system_}
+    {}
 
     ~Impl() = default;
 
     std::shared_ptr<Control::ChannelState> CreateChannel(s32 channel_id) {
         auto channel_state = std::make_shared<Tegra::Control::ChannelState>(channel_id);
         channels.emplace(channel_id, channel_state);
-        scheduler->DeclareChannel(channel_state);
+        scheduler.DeclareChannel(channel_state);
         return channel_state;
     }
 
     void BindChannel(s32 channel_id) {
-        if (bound_channel == channel_id) {
+        if (bound_channel != channel_id) {
-            return;
+            auto it = channels.find(channel_id);
+            ASSERT(it != channels.end());
+            bound_channel = channel_id;
+            current_channel = it->second.get();
+            renderer->ReadRasterizer()->BindChannel(*current_channel);
         }
-        auto it = channels.find(channel_id);
-        ASSERT(it != channels.end());
-        bound_channel = channel_id;
-        current_channel = it->second.get();
-
-        rasterizer->BindChannel(*current_channel);
     }
 
     std::shared_ptr<Control::ChannelState> AllocateChannel() {
@@ -71,13 +72,13 @@ struct GPU::Impl {
     }
 
     void InitChannel(Control::ChannelState& to_init, u64 program_id) {
-        to_init.Init(system, gpu, program_id);
+        to_init.Init(system, program_id);
-        to_init.BindRasterizer(rasterizer);
+        to_init.BindRasterizer(renderer->ReadRasterizer());
-        rasterizer->InitializeChannel(to_init);
+        renderer->ReadRasterizer()->InitializeChannel(to_init);
     }
 
     void InitAddressSpace(Tegra::MemoryManager& memory_manager) {
-        memory_manager.BindRasterizer(rasterizer);
+        memory_manager.BindRasterizer(renderer->ReadRasterizer());
     }
 
     void ReleaseChannel(Control::ChannelState& to_release) {
@@ -87,26 +88,26 @@ struct GPU::Impl {
     /// Binds a renderer to the GPU.
     void BindRenderer(std::unique_ptr<VideoCore::RendererBase> renderer_) {
         renderer = std::move(renderer_);
-        rasterizer = renderer->ReadRasterizer();
+        system.Host1x().memory_manager.BindInterface(renderer->ReadRasterizer());
-        host1x.MemoryManager().BindInterface(rasterizer);
+        system.Host1x().gmmu_manager.BindRasterizer(renderer->ReadRasterizer());
-        host1x.gmmu_manager.BindRasterizer(rasterizer);
     }
 
     /// Flush all current written commands into the host GPU for execution.
     void FlushCommands() {
-        rasterizer->FlushCommands();
+        renderer->ReadRasterizer()->FlushCommands();
     }
 
     /// Synchronizes CPU writes with Host GPU memory.
     void InvalidateGPUCache() {
-        std::function<void(PAddr, size_t)> callback_writes(
+        std::function<void(PAddr, size_t)> callback_writes([this](PAddr address, size_t size) {
-            [this](PAddr address, size_t size) { rasterizer->OnCacheInvalidation(address, size); });
+            renderer->ReadRasterizer()->OnCacheInvalidation(address, size);
+        });
         system.GatherGPUDirtyMemory(callback_writes);
     }
 
     /// Signal the ending of command list.
     void OnCommandListEnd() {
-        rasterizer->ReleaseFences(false);
+        renderer->ReadRasterizer()->ReleaseFences(false);
         Settings::UpdateGPUAccuracy();
     }
 
@@ -143,62 +144,6 @@ struct GPU::Impl {
         }
     }
 
-    /// Returns a reference to the Maxwell3D GPU engine.
-    [[nodiscard]] Engines::Maxwell3D& Maxwell3D() {
-        ASSERT(current_channel);
-        return *current_channel->maxwell_3d;
-    }
-
-    /// Returns a const reference to the Maxwell3D GPU engine.
-    [[nodiscard]] const Engines::Maxwell3D& Maxwell3D() const {
-        ASSERT(current_channel);
-        return *current_channel->maxwell_3d;
-    }
-
-    /// Returns a reference to the KeplerCompute GPU engine.
-    [[nodiscard]] Engines::KeplerCompute& KeplerCompute() {
-        ASSERT(current_channel);
-        return *current_channel->kepler_compute;
-    }
-
-    /// Returns a reference to the KeplerCompute GPU engine.
-    [[nodiscard]] const Engines::KeplerCompute& KeplerCompute() const {
-        ASSERT(current_channel);
-        return *current_channel->kepler_compute;
-    }
-
-    /// Returns a reference to the GPU DMA pusher.
-    [[nodiscard]] Tegra::DmaPusher& DmaPusher() {
-        ASSERT(current_channel);
-        return *current_channel->dma_pusher;
-    }
-
-    /// Returns a const reference to the GPU DMA pusher.
-    [[nodiscard]] const Tegra::DmaPusher& DmaPusher() const {
-        ASSERT(current_channel);
-        return *current_channel->dma_pusher;
-    }
-
-    /// Returns a reference to the underlying renderer.
-    [[nodiscard]] VideoCore::RendererBase& Renderer() {
-        return *renderer;
-    }
-
-    /// Returns a const reference to the underlying renderer.
-    [[nodiscard]] const VideoCore::RendererBase& Renderer() const {
-        return *renderer;
-    }
-
-    /// Returns a reference to the shader notifier.
-    [[nodiscard]] VideoCore::ShaderNotify& ShaderNotify() {
-        return *shader_notify;
-    }
-
-    /// Returns a const reference to the shader notifier.
-    [[nodiscard]] const VideoCore::ShaderNotify& ShaderNotify() const {
-        return *shader_notify;
-    }
-
     [[nodiscard]] u64 GetTicks() const {
         u64 gpu_tick = system.CoreTiming().GetGPUTicks();
         Settings::GpuOverclock overclock = Settings::values.fast_gpu_time.GetValue();
@@ -210,14 +155,6 @@ struct GPU::Impl {
         return gpu_tick;
     }
 
-    [[nodiscard]] bool IsAsync() const {
-        return is_async;
-    }
-
-    [[nodiscard]] bool UseNvdec() const {
-        return use_nvdec;
-    }
-
     void RendererFrameEndNotify() {
         system.GetPerfStats().EndGameFrame();
     }
@@ -227,7 +164,7 @@ struct GPU::Impl {
     /// core timing events.
     void Start() {
         Settings::UpdateGPUAccuracy();
-        gpu_thread.StartThread(*renderer, renderer->Context(), *scheduler);
+        gpu_thread.StartThread(*renderer, renderer->Context(), scheduler);
     }
 
     void NotifyShutdown() {
@@ -251,25 +188,24 @@ struct GPU::Impl {
 
     /// Push GPU command entries to be processed
     void PushGPUEntries(s32 channel, Tegra::CommandList&& entries) {
-        gpu_thread.SubmitList(channel, std::move(entries));
+        gpu_thread.SubmitList(channel, std::move(entries), is_async);
     }
 
     /// Notify rasterizer that any caches of the specified region should be flushed to Switch memory
     void FlushRegion(DAddr addr, u64 size) {
-        gpu_thread.FlushRegion(addr, size);
+        gpu_thread.FlushRegion(addr, size, is_async);
     }
 
     VideoCore::RasterizerDownloadArea OnCPURead(DAddr addr, u64 size) {
-        auto raster_area = rasterizer->GetFlushArea(addr, size);
+        auto raster_area = renderer->ReadRasterizer()->GetFlushArea(addr, size);
         if (raster_area.preemtive) {
             return raster_area;
         }
         raster_area.preemtive = true;
         const u64 fence = RequestSyncOperation([this, &raster_area]() {
-            rasterizer->FlushRegion(raster_area.start_address,
+            renderer->ReadRasterizer()->FlushRegion(raster_area.start_address, raster_area.end_address - raster_area.start_address);
-                                    raster_area.end_address - raster_area.start_address);
         });
-        gpu_thread.TickGPU();
+        gpu_thread.TickGPU(is_async);
         WaitForSyncOperation(fence);
         return raster_area;
     }
@@ -280,16 +216,15 @@ struct GPU::Impl {
     }
 
     bool OnCPUWrite(DAddr addr, u64 size) {
-        return rasterizer->OnCPUWrite(addr, size);
+        return renderer->ReadRasterizer()->OnCPUWrite(addr, size);
     }
 
     /// Notify rasterizer that any caches of the specified region should be flushed and invalidated
     void FlushAndInvalidateRegion(DAddr addr, u64 size) {
-        gpu_thread.FlushAndInvalidateRegion(addr, size);
+        gpu_thread.FlushAndInvalidateRegion(addr, size, is_async);
     }
 
-    void RequestComposite(std::vector<Tegra::FramebufferConfig>&& layers,
+    void RequestComposite(std::vector<Tegra::FramebufferConfig>&& layers, std::vector<Service::Nvidia::NvFence>&& fences) {
-                          std::vector<Service::Nvidia::NvFence>&& fences) {
         size_t num_fences{fences.size()};
         size_t current_request_counter{};
         {
@@ -304,7 +239,7 @@ struct GPU::Impl {
             }
         }
         const auto wait_fence = RequestSyncOperation([this, current_request_counter, &layers, &fences, num_fences] {
-            auto& syncpoint_manager = host1x.GetSyncpointManager();
+            auto& syncpoint_manager = system.Host1x().GetSyncpointManager();
             if (num_fences == 0) {
                 renderer->Composite(layers);
             }
@@ -322,7 +257,7 @@ struct GPU::Impl {
                 syncpoint_manager.RegisterGuestAction(fences[i].id, fences[i].value, executer);
             }
         });
-        gpu_thread.TickGPU();
+        gpu_thread.TickGPU(is_async);
         WaitForSyncOperation(wait_fence);
     }
 
@@ -331,23 +266,20 @@ struct GPU::Impl {
 
         const auto wait_fence =
             RequestSyncOperation([&] { out = renderer->GetAppletCaptureBuffer(); });
-        gpu_thread.TickGPU();
+        gpu_thread.TickGPU(is_async);
         WaitForSyncOperation(wait_fence);
 
         return out;
     }
 
-    GPU& gpu;
     Core::System& system;
-    Host1x::Host1x& host1x;
 
     std::unique_ptr<VideoCore::RendererBase> renderer;
-    VideoCore::RasterizerInterface* rasterizer = nullptr;
     const bool use_nvdec;
 
     s32 new_channel_id{1};
     /// Shader build notifier
-    std::unique_ptr<VideoCore::ShaderNotify> shader_notify;
+    VideoCore::ShaderNotify shader_notify;
     /// When true, we are about to shut down emulation session, so terminate outstanding tasks
     std::atomic_bool shutting_down{};
 
@@ -371,7 +303,7 @@ struct GPU::Impl {
     VideoCommon::GPUThread::ThreadManager gpu_thread;
     std::unique_ptr<Core::Frontend::GraphicsContext> cpu_context;
 
-    std::unique_ptr<Tegra::Control::Scheduler> scheduler;
+    Tegra::Control::Scheduler scheduler;
     ankerl::unordered_dense::map<s32, std::shared_ptr<Tegra::Control::ChannelState>> channels;
     Tegra::Control::ChannelState* current_channel;
     s32 bound_channel{-1};
@@ -382,7 +314,8 @@ struct GPU::Impl {
 };
 
 GPU::GPU(Core::System& system, bool is_async, bool use_nvdec)
-    : impl{std::make_unique<Impl>(*this, system, is_async, use_nvdec)} {}
+    : impl{std::make_unique<Impl>(system, is_async, use_nvdec)}
+{}
 
 GPU::~GPU() = default;
 
@@ -423,8 +356,9 @@ void GPU::OnCommandListEnd() {
 }
 
 u64 GPU::RequestFlush(DAddr addr, std::size_t size) {
-    return impl->RequestSyncOperation(
+    return impl->RequestSyncOperation([this, addr, size]() {
-        [this, addr, size]() { impl->rasterizer->FlushRegion(addr, size); });
+        impl->renderer->ReadRasterizer()->FlushRegion(addr, size);
+    });
 }
 
 u64 GPU::CurrentSyncRequestFence() const {
@@ -441,52 +375,52 @@ void GPU::TickWork() {
 
 /// Gets a mutable reference to the Host1x interface
 Host1x::Host1x& GPU::Host1x() {
-    return impl->host1x;
+    return impl->system.Host1x();
 }
 
 /// Gets an immutable reference to the Host1x interface.
 const Host1x::Host1x& GPU::Host1x() const {
-    return impl->host1x;
+    return impl->system.Host1x();
 }
 
 Engines::Maxwell3D& GPU::Maxwell3D() {
-    return impl->Maxwell3D();
+    return impl->current_channel->payload->maxwell_3d;
 }
 
 const Engines::Maxwell3D& GPU::Maxwell3D() const {
-    return impl->Maxwell3D();
+    return impl->current_channel->payload->maxwell_3d;
 }
 
 Engines::KeplerCompute& GPU::KeplerCompute() {
-    return impl->KeplerCompute();
+    return impl->current_channel->payload->kepler_compute;
 }
 
 const Engines::KeplerCompute& GPU::KeplerCompute() const {
-    return impl->KeplerCompute();
+    return impl->current_channel->payload->kepler_compute;
 }
 
 Tegra::DmaPusher& GPU::DmaPusher() {
-    return impl->DmaPusher();
+    return impl->current_channel->payload->dma_pusher;
 }
 
 const Tegra::DmaPusher& GPU::DmaPusher() const {
-    return impl->DmaPusher();
+    return impl->current_channel->payload->dma_pusher;
 }
 
 VideoCore::RendererBase& GPU::Renderer() {
-    return impl->Renderer();
+    return *impl->renderer;
 }
 
 const VideoCore::RendererBase& GPU::Renderer() const {
-    return impl->Renderer();
+    return *impl->renderer;
 }
 
 VideoCore::ShaderNotify& GPU::ShaderNotify() {
-    return impl->ShaderNotify();
+    return impl->shader_notify;
 }
 
 const VideoCore::ShaderNotify& GPU::ShaderNotify() const {
-    return impl->ShaderNotify();
+    return impl->shader_notify;
 }
 
 void GPU::RequestComposite(std::vector<Tegra::FramebufferConfig>&& layers,
@@ -503,11 +437,11 @@ u64 GPU::GetTicks() const {
 }
 
 bool GPU::IsAsync() const {
-    return impl->IsAsync();
+    return impl->is_async;
 }
 
 bool GPU::UseNvdec() const {
-    return impl->UseNvdec();
+    return impl->use_nvdec;
 }
 
 void GPU::RendererFrameEndNotify() {

src/video_core/gpu_thread.cpp

@@ -19,8 +19,9 @@
 
 namespace VideoCommon::GPUThread {
 
-ThreadManager::ThreadManager(Core::System& system_, bool is_async_)
+ThreadManager::ThreadManager(Core::System& system_)
-    : system{system_}, is_async{is_async_} {}
+    : system{system_}
+{}
 
 ThreadManager::~ThreadManager() = default;
 
@@ -39,7 +40,7 @@ void ThreadManager::StartThread(VideoCore::RendererBase& renderer, Core::Fronten
                 break;
             }
             if (auto* submit_list = std::get_if<SubmitListCommand>(&next.data)) {
-                scheduler.Push(submit_list->channel, std::move(submit_list->entries));
+                scheduler.Push(system.GPU(), submit_list->channel, std::move(submit_list->entries));
             } else if (std::holds_alternative<GPUTickCommand>(next.data)) {
                 system.GPU().TickWork();
             } else if (const auto* flush = std::get_if<FlushRegionCommand>(&next.data)) {
@@ -60,41 +61,40 @@ void ThreadManager::StartThread(VideoCore::RendererBase& renderer, Core::Fronten
     });
 }
 
-void ThreadManager::SubmitList(s32 channel, Tegra::CommandList&& entries) {
+void ThreadManager::SubmitList(s32 channel, Tegra::CommandList&& entries, bool is_async) {
-    PushCommand(SubmitListCommand(channel, std::move(entries)));
+    PushCommand(SubmitListCommand(channel, std::move(entries)), false, is_async);
 }
 
-void ThreadManager::FlushRegion(DAddr addr, u64 size) {
+void ThreadManager::FlushRegion(DAddr addr, u64 size, bool is_async) {
     if (!is_async) {
         // Always flush with synchronous GPU mode
-        PushCommand(FlushRegionCommand(addr, size));
+        PushCommand(FlushRegionCommand(addr, size), false, is_async);
     }
-    return;
 }
 
-void ThreadManager::TickGPU() {
+void ThreadManager::TickGPU(bool is_async) {
-    PushCommand(GPUTickCommand());
+    PushCommand(GPUTickCommand(), false, is_async);
 }
 
 void ThreadManager::InvalidateRegion(DAddr addr, u64 size) {
     rasterizer->OnCacheInvalidation(addr, size);
 }
 
-void ThreadManager::FlushAndInvalidateRegion(DAddr addr, u64 size) {
+void ThreadManager::FlushAndInvalidateRegion(DAddr addr, u64 size, bool is_async) {
     if (Settings::IsGPULevelHigh()) {
         if (!is_async) {
-            PushCommand(FlushRegionCommand(addr, size));
+            PushCommand(FlushRegionCommand(addr, size), false, is_async);
         } else {
             auto& gpu = system.GPU();
             const u64 fence = gpu.RequestFlush(addr, size);
-            TickGPU();
+            TickGPU(is_async);
             gpu.WaitForSyncOperation(fence);
         }
     }
     rasterizer->OnCacheInvalidation(addr, size);
 }
 
-u64 ThreadManager::PushCommand(CommandData&& command_data, bool block) {
+u64 ThreadManager::PushCommand(CommandData&& command_data, bool block, bool is_async) {
     if (!is_async) {
         // In synchronous GPU mode, block the caller until the command has executed
         block = true;

src/video_core/gpu_thread.h

@@ -15,6 +15,7 @@
 
 #include "common/bounded_threadsafe_queue.h"
 #include "common/polyfill_thread.h"
+#include "video_core/dma_pusher.h"
 #include "video_core/framebuffer_config.h"
 
 namespace Tegra {
@@ -103,7 +104,7 @@ struct SynchState final {
 /// Class used to manage the GPU thread
 class ThreadManager final {
 public:
-    explicit ThreadManager(Core::System& system_, bool is_async_);
+    explicit ThreadManager(Core::System& system_);
     ~ThreadManager();
 
     /// Creates and starts the GPU thread.
@@ -111,27 +112,25 @@ public:
                      Tegra::Control::Scheduler& scheduler);
 
     /// Push GPU command entries to be processed
-    void SubmitList(s32 channel, Tegra::CommandList&& entries);
+    void SubmitList(s32 channel, Tegra::CommandList&& entries, bool is_async);
 
     /// Notify rasterizer that any caches of the specified region should be flushed to Switch memory
-    void FlushRegion(DAddr addr, u64 size);
+    void FlushRegion(DAddr addr, u64 size, bool is_async);
 
     /// Notify rasterizer that any caches of the specified region should be invalidated
     void InvalidateRegion(DAddr addr, u64 size);
 
     /// Notify rasterizer that any caches of the specified region should be flushed and invalidated
-    void FlushAndInvalidateRegion(DAddr addr, u64 size);
+    void FlushAndInvalidateRegion(DAddr addr, u64 size, bool is_async);
 
-    void TickGPU();
+    void TickGPU(bool is_async);
 
 private:
     /// Pushes a command to be executed by the GPU thread
-    u64 PushCommand(CommandData&& command_data, bool block = false);
+    u64 PushCommand(CommandData&& command_data, bool block, bool is_async);
 
     Core::System& system;
-    const bool is_async;
     VideoCore::RasterizerInterface* rasterizer = nullptr;
-
     SynchState state;
     std::jthread thread;
 };

src/video_core/host1x/control.cpp

@@ -1,4 +1,4 @@
-// SPDX-FileCopyrightText: Copyright 2025 Eden Emulator Project
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
 // SPDX-License-Identifier: GPL-3.0-or-later
 
 // SPDX-FileCopyrightText: 2021 yuzu Emulator Project
@@ -10,26 +10,22 @@
 
 namespace Tegra::Host1x {
 
-Control::Control(Host1x& host1x_) : host1x(host1x_) {}
+void Control::ProcessMethod(Host1x& host1x, Method method, u32 argument) {
-
-Control::~Control() = default;
-
-void Control::ProcessMethod(Method method, u32 argument) {
     switch (method) {
     case Method::LoadSyncptPayload32:
         syncpoint_value = argument;
         break;
     case Method::WaitSyncpt:
     case Method::WaitSyncpt32:
-        Execute(argument);
+        Execute(host1x, argument);
         break;
     default:
-        UNIMPLEMENTED_MSG("Control method {:#X}", static_cast<u32>(method));
+        UNIMPLEMENTED_MSG("Control method {:#X}", u32(method));
         break;
     }
 }
 
-void Control::Execute(u32 data) {
+void Control::Execute(Host1x& host1x, u32 data) {
     LOG_TRACE(Service_NVDRV, "Control wait syncpt {} value {}", data, syncpoint_value);
     host1x.GetSyncpointManager().WaitHost(data, syncpoint_value);
 }

src/video_core/host1x/control.h

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: 2021 yuzu Emulator Project
 // SPDX-FileCopyrightText: 2021 Skyline Team and Contributors
 // SPDX-License-Identifier: GPL-3.0-or-later
@@ -19,17 +22,11 @@ public:
         WaitSyncpt32 = 0x50,
     };
 
-    explicit Control(Host1x& host1x);
-    ~Control();
-
     /// Writes the method into the state, Invoke Execute() if encountered
-    void ProcessMethod(Method method, u32 argument);
+    void ProcessMethod(Host1x& host1x, Method method, u32 argument);
-
-private:
     /// For Host1x, execute is waiting on a syncpoint previously written into the state
-    void Execute(u32 data);
+    void Execute(Host1x& host1x, u32 data);
 
-    Host1x& host1x;
     u32 syncpoint_value{};
 };
 

src/video_core/host1x/host1x.cpp

@@ -27,22 +27,22 @@ void Host1x::StartDevice(s32 fd, ChannelType type, u32 syncpt) {
 #ifdef YUZU_LEGACY
         std::call_once(nvdec_first_init, []() {std::this_thread::sleep_for(std::chrono::milliseconds{500});}); // HACK: For Astroneer
 #endif
-        devices[fd] = std::make_unique<Tegra::Host1x::Nvdec>(*this, fd, syncpt);
+        devices[fd].emplace<Tegra::Host1x::Nvdec>(*this, fd, syncpt);
         break;
     case ChannelType::VIC:
 #ifdef YUZU_LEGACY
         std::call_once(vic_first_init, []() {std::this_thread::sleep_for(std::chrono::milliseconds{500});}); // HACK: For Astroneer
 #endif
-        devices[fd] = std::make_unique<Tegra::Host1x::Vic>(*this, fd, syncpt);
+        devices[fd].emplace<Tegra::Host1x::Vic>(*this, fd, syncpt);
         break;
     default:
-        LOG_ERROR(HW_GPU, "Unimplemented host1x device {}", static_cast<u32>(type));
+        LOG_ERROR(HW_GPU, "Unimplemented host1x device {}", u32(type));
         break;
     }
 }
 
 void Host1x::StopDevice(s32 fd, ChannelType type) {
-    devices.erase(fd);
+    devices[fd].emplace<std::monostate>();
 }
 
 } // namespace Tegra::Host1x

src/video_core/host1x/host1x.h

@@ -8,10 +8,14 @@
 
 #include <ankerl/unordered_dense.h>
 #include <unordered_map>
-#include <queue>
+#include <variant>
 
 #include "common/common_types.h"
 
+// fd types?
+#include "video_core/host1x/nvdec.h"
+#include "video_core/host1x/vic.h"
+
 #include "common/address_space.h"
 #include "video_core/cdma_pusher.h"
 #include "video_core/host1x/gpu_device_memory_manager.h"
@@ -31,118 +35,90 @@ class Nvdec;
 
 class FrameQueue {
 public:
+    struct FrameDevice {
+        std::deque<std::pair<u64, std::shared_ptr<FFmpeg::Frame>>> m_presentation_order;
+        std::unordered_map<u64, std::shared_ptr<FFmpeg::Frame>> m_decode_order;
+    };
+
     void Open(s32 fd) {
         std::scoped_lock l{m_mutex};
-        m_presentation_order.insert({fd, {}});
+        m_frame_devices.insert_or_assign(fd, FrameDevice{});
-        m_decode_order.insert({fd, {}});
     }
 
     void Close(s32 fd) {
         std::scoped_lock l{m_mutex};
-        m_presentation_order.erase(fd);
+        m_frame_devices.erase(fd);
-        m_decode_order.erase(fd);
     }
 
     s32 VicFindNvdecFdFromOffset(u64 search_offset) {
         std::scoped_lock l{m_mutex};
-        for (auto& map : m_presentation_order) {
+        for (auto const& [fd, dev] : m_frame_devices) {
-            for (auto& [offset, frame] : map.second) {
+            for (auto const& [offset, frame] : dev.m_presentation_order)
-                if (offset == search_offset) {
+                if (offset == search_offset)
-                    return map.first;
+                    return fd;
-                }
+            for (auto const& [offset, frame] : dev.m_decode_order)
-            }
+                if (offset == search_offset)
-        }
+                    return fd;
-        for (auto& map : m_decode_order) {
-            for (auto& [offset, frame] : map.second) {
-                if (offset == search_offset) {
-                    return map.first;
-                }
-            }
         }
         return -1;
     }
 
     void PushPresentOrder(s32 fd, u64 offset, std::shared_ptr<FFmpeg::Frame>&& frame) {
         std::scoped_lock l{m_mutex};
-        auto map = m_presentation_order.find(fd);
+        if (auto const it = m_frame_devices.find(fd); it != m_frame_devices.end()) {
-        if (map == m_presentation_order.end()) {
+            if (it->second.m_presentation_order.size() >= MAX_PRESENT_QUEUE)
-            return;
+                it->second.m_presentation_order.pop_front();
+            it->second.m_presentation_order.emplace_back(offset, std::move(frame));
         }
-
-        if (map->second.size() >= MAX_PRESENT_QUEUE) {
-            map->second.pop_front();
-        }
-
-        map->second.emplace_back(offset, std::move(frame));
     }
 
     void PushDecodeOrder(s32 fd, u64 offset, std::shared_ptr<FFmpeg::Frame>&& frame) {
         std::scoped_lock l{m_mutex};
-        auto map = m_decode_order.find(fd);
+        if (auto const it = m_frame_devices.find(fd); it != m_frame_devices.end()) {
-        if (map == m_decode_order.end()) {
+            it->second.m_decode_order.insert_or_assign(offset, std::move(frame));
-            return;
+            if (it->second.m_decode_order.size() > MAX_DECODE_MAP) {
-        }
+                auto it2 = it->second.m_decode_order.begin();
-
+                std::advance(it2, it->second.m_decode_order.size() - MAX_DECODE_MAP);
-        map->second.insert_or_assign(offset, std::move(frame));
+                it->second.m_decode_order.erase(it->second.m_decode_order.begin(), it2);
-
+            }
-        if (map->second.size() > MAX_DECODE_MAP) {
-            auto it = map->second.begin();
-            std::advance(it, map->second.size() - MAX_DECODE_MAP);
-            map->second.erase(map->second.begin(), it);
         }
     }
 
     std::shared_ptr<FFmpeg::Frame> GetFrame(s32 fd, u64 offset) {
-        if (fd == -1) {
+        if (fd != -1) {
-            return {};
+            std::scoped_lock l{m_mutex};
+            if (auto const it = m_frame_devices.find(fd); it != m_frame_devices.end()) {
+                if (it->second.m_presentation_order.size() > 0)
+                    return GetPresentOrderLocked(fd);
+                if (it->second.m_decode_order.size() > 0)
+                    return GetDecodeOrderLocked(fd, offset);
+            }
         }
-
-        std::scoped_lock l{m_mutex};
-
-        auto present_map = m_presentation_order.find(fd);
-        if (present_map != m_presentation_order.end() && !present_map->second.empty()) {
-            return GetPresentOrderLocked(fd);
-        }
-
-        auto decode_map = m_decode_order.find(fd);
-        if (decode_map != m_decode_order.end() && !decode_map->second.empty()) {
-            return GetDecodeOrderLocked(fd, offset);
-        }
-
         return {};
     }
 
 private:
     std::shared_ptr<FFmpeg::Frame> GetPresentOrderLocked(s32 fd) {
-        auto map = m_presentation_order.find(fd);
+        if (auto const it = m_frame_devices.find(fd); it != m_frame_devices.end()) {
-        if (map == m_presentation_order.end() || map->second.empty()) {
+            auto frame = std::move(it->second.m_presentation_order.front().second);
-            return {};
+            it->second.m_presentation_order.pop_front();
+            return frame;
         }
-
+        return {};
-        auto frame = std::move(map->second.front().second);
-        map->second.pop_front();
-        return frame;
     }
 
     std::shared_ptr<FFmpeg::Frame> GetDecodeOrderLocked(s32 fd, u64 offset) {
-        auto map = m_decode_order.find(fd);
+        if (auto const it = m_frame_devices.find(fd); it != m_frame_devices.end()) {
-        if (map == m_decode_order.end() || map->second.empty()) {
+            if (auto const it2 = it->second.m_decode_order.find(offset); it2 != it->second.m_decode_order.end()) {
-            return {};
+                // TODO: this "mapped" prevents us from fully embracing ankerl
+                return std::move(it->second.m_decode_order.extract(it2).mapped());
+            }
         }
-
+        return {};
-        auto it = map->second.find(offset);
-        if (it == map->second.end()) {
-            return {};
-        }
-        // TODO: this "mapped" prevents us from fully embracing ankerl
-        return std::move(map->second.extract(it).mapped());
     }
 
-    using FramePtr = std::shared_ptr<FFmpeg::Frame>;
-
     std::mutex m_mutex{};
-    ankerl::unordered_dense::map<s32, std::deque<std::pair<u64, FramePtr>>> m_presentation_order;
+    ankerl::unordered_dense::map<s32, FrameDevice> m_frame_devices;
-    ankerl::unordered_dense::map<s32, std::unordered_map<u64, FramePtr>> m_decode_order;
 
     static constexpr size_t MAX_PRESENT_QUEUE = 100;
     static constexpr size_t MAX_DECODE_MAP = 200;
@@ -196,11 +172,11 @@ public:
     void StopDevice(s32 fd, ChannelType type);
 
     void PushEntries(s32 fd, ChCommandHeaderList&& entries) {
-        auto it = devices.find(fd);
+        if (auto const nvdec = std::get_if<Tegra::Host1x::Nvdec>(&devices[fd])) {
-        if (it == devices.end()) {
+            nvdec->PushEntries(std::move(entries));
-            return;
+        } else if (auto const vic = std::get_if<Tegra::Host1x::Vic>(&devices[fd])) {
+            vic->PushEntries(std::move(entries));
         }
-        it->second->PushEntries(std::move(entries));
     }
 
     Core::System& system;
@@ -209,7 +185,11 @@ public:
     Tegra::MemoryManager gmmu_manager;
     Common::FlatAllocator<u32, 0, 32> allocator;
     FrameQueue frame_queue;
-    ankerl::unordered_dense::map<s32, std::unique_ptr<CDmaPusher>> devices;
+    std::array<std::variant<
+        std::monostate,
+        Tegra::Host1x::Nvdec,
+        Tegra::Host1x::Vic
+    >, 1024> devices;
 #ifdef YUZU_LEGACY
     std::once_flag nvdec_first_init;
     std::once_flag vic_first_init;

src/video_core/host1x/vic.cpp

@@ -407,15 +407,15 @@ void Vic::ReadInterlacedY8__V8U8_N420(const SlotStruct& slot, std::span<const Pl
         };
 
         switch (slot.config.deinterlace_mode) {
-        case DXVAHD_DEINTERLACE_MODE_PRIVATE::WEAVE:
+        case DxvhadDeinterlaceModePrivate::Weave:
             // Due to the fact that we do not write to memory in nvdec, we cannot use Weave as it
             // relies on the previous frame.
             DecodeBobField();
             break;
-        case DXVAHD_DEINTERLACE_MODE_PRIVATE::BOB_FIELD:
+        case DxvhadDeinterlaceModePrivate::BobField:
             DecodeBobField();
             break;
-        case DXVAHD_DEINTERLACE_MODE_PRIVATE::DISI1:
+        case DxvhadDeinterlaceModePrivate::Disi1:
             // Due to the fact that we do not write to memory in nvdec, we cannot use DISI1 as it
             // relies on previous/next frames.
             DecodeBobField();
@@ -431,13 +431,13 @@ void Vic::ReadInterlacedY8__V8U8_N420(const SlotStruct& slot, std::span<const Pl
 
 void Vic::ReadY8__V8U8_N420(const SlotStruct& slot, std::span<const PlaneOffsets> offsets, std::shared_ptr<const FFmpeg::Frame> frame, bool planar) noexcept {
     switch (slot.config.frame_format) {
-    case DXVAHD_FRAME_FORMAT::PROGRESSIVE:
+    case DxvhadFrameFormat::Progressive:
         ReadProgressiveY8__V8U8_N420(slot, offsets, std::move(frame), planar, false);
         break;
-    case DXVAHD_FRAME_FORMAT::TOP_FIELD:
+    case DxvhadFrameFormat::TopField:
         ReadInterlacedY8__V8U8_N420(slot, offsets, std::move(frame), planar, true);
         break;
-    case DXVAHD_FRAME_FORMAT::BOTTOM_FIELD:
+    case DxvhadFrameFormat::BottomField:
         ReadInterlacedY8__V8U8_N420(slot, offsets, std::move(frame), planar, false);
         break;
     default:
@@ -860,7 +860,7 @@ void Vic::WriteY8__V8U8_N420(const OutputSurfaceConfig& output_surface_config) n
     };
 
     switch (output_surface_config.out_block_kind) {
-    case BLK_KIND::GENERIC_16Bx2: {
+    case BlkKind::Generic_16Bx2: {
         u32 const block_height = u32(output_surface_config.out_block_height);
         auto const out_luma_swizzle_size = Texture::CalculateSize(true, BytesPerPixel, out_luma_width, out_luma_height, 1, block_height, 0);
         auto const out_chroma_swizzle_size = Texture::CalculateSize(true, BytesPerPixel * 2, out_chroma_width, out_chroma_height, 1, block_height, 0);
@@ -889,7 +889,7 @@ void Vic::WriteY8__V8U8_N420(const OutputSurfaceConfig& output_surface_config) n
             Texture::SwizzleTexture(out_chroma, chroma_scratch, BytesPerPixel, out_chroma_width, out_chroma_height, 1, block_height, 0, 1);
         }
     } break;
-    case BLK_KIND::PITCH: {
+    case BlkKind::Pitch: {
         LOG_TRACE(HW_GPU, "Writing Y8__V8U8_N420 swizzled frame\n"
             "\tinput surface {}x{} stride {} size {:#X}\n"
             "\toutput   luma {}x{} stride {} size {:#X} block height {} swizzled size 0x{:X}\n",
@@ -1032,7 +1032,7 @@ void Vic::WriteABGR(const OutputSurfaceConfig& output_surface_config, VideoPixel
     };
 
     switch (output_surface_config.out_block_kind) {
-    case BLK_KIND::GENERIC_16Bx2: {
+    case BlkKind::Generic_16Bx2: {
         const u32 block_height = u32(output_surface_config.out_block_height);
         auto const out_swizzle_size = Texture::CalculateSize(true, BytesPerPixel, out_luma_width, out_luma_height, 1, block_height, 0);
         LOG_TRACE(HW_GPU, "Writing ABGR swizzled frame\n"
@@ -1051,7 +1051,7 @@ void Vic::WriteABGR(const OutputSurfaceConfig& output_surface_config, VideoPixel
             Texture::SwizzleTexture(out_luma, luma_scratch, BytesPerPixel, out_luma_width, out_luma_height, 1, block_height, 0, 1);
         }
     } break;
-    case BLK_KIND::PITCH: {
+    case BlkKind::Pitch: {
         LOG_TRACE(HW_GPU, "Writing ABGR pitch frame\n"
             "\tinput surface {}x{} stride {} size {:#X}"
             "\toutput surface {}x{} stride {} size {:#X}",

src/video_core/host1x/vic.h

@@ -6,16 +6,12 @@
 
 #pragma once
 
-#include <condition_variable>
-#include <functional>
 #include <memory>
-#include <mutex>
-#include <thread>
 
 #include "common/common_types.h"
 #include "common/scratch_buffer.h"
 #include "video_core/cdma_pusher.h"
-#include "video_core/host1x/host1x.h"
+#include "video_core/host1x/ffmpeg.h"
 
 namespace Tegra::Host1x {
 class Host1x;
@@ -138,52 +134,53 @@ enum SurfaceIndex : u32 {
     CombinedMotion = 7,
 };
 
-enum class DXVAHD_ALPHA_FILL_MODE : u32 {
+// Note: these will inevitably collide with Win32 defines if you use their UPPER_SNAKE_CASE naming
-    OPAQUE = 0,
+enum class DxvhadAlphaFillMode : u32 {
-    BACKGROUND = 1,
+    Opaque = 0,
-    DESTINATION = 2,
+    Background = 1,
-    SOURCE_STREAM = 3,
+    Destination = 2,
-    COMPOSITED = 4,
+    SourceStream = 3,
-    SOURCE_ALPHA = 5,
+    Composited = 4,
+    SourceAlpha = 5,
 };
 
-enum class DXVAHD_FRAME_FORMAT : u64 {
+enum class DxvhadFrameFormat : u64 {
-    PROGRESSIVE = 0,
+    Progressive = 0,
-    INTERLACED_TOP_FIELD_FIRST = 1,
+    InterlacedTopFieldFirst = 1,
-    INTERLACED_BOTTOM_FIELD_FIRST = 2,
+    InterlacedBottomFieldFirst = 2,
-    TOP_FIELD = 3,
+    TopField = 3,
-    BOTTOM_FIELD = 4,
+    BottomField = 4,
-    SUBPIC_PROGRESSIVE = 5,
+    SubpicProgressive = 5,
-    SUBPIC_INTERLACED_TOP_FIELD_FIRST = 6,
+    SubpicInterlacedTopFieldFirst = 6,
-    SUBPIC_INTERLACED_BOTTOM_FIELD_FIRST = 7,
+    SubpicInterlacedBottomFieldFirst = 7,
-    SUBPIC_TOP_FIELD = 8,
+    SubpicTopField = 8,
-    SUBPIC_BOTTOM_FIELD = 9,
+    SubpicBottomField = 9,
-    TOP_FIELD_CHROMA_BOTTOM = 10,
+    TopFieldChromaBottom = 10,
-    BOTTOM_FIELD_CHROMA_TOP = 11,
+    BottomFieldChromaTop = 11,
-    SUBPIC_TOP_FIELD_CHROMA_BOTTOM = 12,
+    SubpicTopFieldChromaBottom = 12,
-    SUBPIC_BOTTOM_FIELD_CHROMA_TOP = 13,
+    SubpicBottomFieldChromaTop = 13,
 };
 
-enum class DXVAHD_DEINTERLACE_MODE_PRIVATE : u64 {
+enum class DxvhadDeinterlaceModePrivate : u64 {
-    WEAVE = 0,
+    Weave = 0,
-    BOB_FIELD = 1,
+    BobField = 1,
-    BOB = 2,
+    Bob = 2,
-    NEWBOB = 3,
+    Newbob = 3,
-    DISI1 = 4,
+    Disi1 = 4,
-    WEAVE_LUMA_BOB_FIELD_CHROMA = 5,
+    WeaveLumaBobFieldChroma = 5,
-    MAX = 0xF,
+    Max = 0xF,
 };
 
-enum class BLK_KIND {
+enum class BlkKind {
-    PITCH = 0,
+    Pitch = 0,
-    GENERIC_16Bx2 = 1,
+    Generic_16Bx2 = 1,
     // These are unsupported in the vic
-    BL_NAIVE = 2,
+    BlNaive = 2,
-    BL_KEPLER_XBAR_RAW = 3,
+    BlKeplerXbarRaw = 3,
-    VP2_TILED = 15,
+    Vp2Tiled = 15,
 };
 
-enum class BLEND_SRCFACTC : u32 {
+enum class BlendSrcFactC : u32 {
     K1 = 0,
     K1_TIMES_DST = 1,
     NEG_K1_TIMES_DST = 2,
@@ -191,7 +188,7 @@ enum class BLEND_SRCFACTC : u32 {
     ZERO = 4,
 };
 
-enum class BLEND_DSTFACTC : u32 {
+enum class BlendDstFactC : u32 {
     K1 = 0,
     K2 = 1,
     K1_TIMES_DST = 2,
@@ -201,7 +198,7 @@ enum class BLEND_DSTFACTC : u32 {
     ONE = 6,
 };
 
-enum class BLEND_SRCFACTA : u32 {
+enum class BlendSrcFactA : u32 {
     K1 = 0,
     K2 = 1,
     NEG_K1_TIMES_DST = 2,
@@ -209,7 +206,7 @@ enum class BLEND_SRCFACTA : u32 {
     MAX = 7,
 };
 
-enum class BLEND_DSTFACTA : u32 {
+enum class BlendDstFactA : u32 {
     K2 = 0,
     NEG_K1_TIMES_SRC = 1,
     ZERO = 2,
@@ -232,7 +229,7 @@ static_assert(sizeof(PipeConfig) == 0x10, "PipeConfig has the wrong size!");
 
 struct OutputConfig {
     union {
-        BitField<0, 3, DXVAHD_ALPHA_FILL_MODE> alpha_fill_mode;
+        BitField<0, 3, DxvhadAlphaFillMode> alpha_fill_mode;
         BitField<3, 3, u64> alpha_fill_slot;
         BitField<6, 10, u64> background_a;
         BitField<16, 10, u64> background_r;
@@ -265,7 +262,7 @@ struct OutputSurfaceConfig {
         BitField<0, 7, VideoPixelFormat> out_pixel_format;
         BitField<7, 2, u32> out_chroma_loc_horiz;
         BitField<9, 2, u32> out_chroma_loc_vert;
-        BitField<11, 4, BLK_KIND> out_block_kind;
+        BitField<11, 4, BlkKind> out_block_kind;
         BitField<15, 4, u32> out_block_height; // in gobs, log2
         BitField<19, 3, u32> reserved0;
         BitField<22, 10, u32> reserved1;
@@ -365,7 +362,7 @@ struct SlotConfig {
         BitField<14, 1, u64> prev_prev_motion_field_enable;
         BitField<15, 1, u64> combined_motion_field_enable;
 
-        BitField<16, 4, DXVAHD_FRAME_FORMAT> frame_format;
+        BitField<16, 4, DxvhadFrameFormat> frame_format;
         BitField<20, 2, u64> filter_length_y; // 0: 1-tap, 1: 2-tap, 2: 5-tap, 3: 10-tap
         BitField<22, 2, u64> filter_length_x;
         BitField<24, 12, u64> panoramic;
@@ -377,7 +374,7 @@ struct SlotConfig {
         BitField<10, 10, u64> filter_detail;
         BitField<20, 10, u64> chroma_noise;
         BitField<30, 10, u64> chroma_detail;
-        BitField<40, 4, DXVAHD_DEINTERLACE_MODE_PRIVATE> deinterlace_mode;
+        BitField<40, 4, DxvhadDeinterlaceModePrivate> deinterlace_mode;
         BitField<44, 3, u64> motion_accumulation_weight;
         BitField<47, 11, u64> noise_iir;
         BitField<58, 4, u64> light_level;
@@ -484,13 +481,13 @@ struct BlendingSlotStruct {
         BitField<26, 6, u32> reserved1;
     };
     union {
-        BitField<0, 3, BLEND_SRCFACTC> src_factor_color_match_select;
+        BitField<0, 3, BlendSrcFactC> src_factor_color_match_select;
         BitField<3, 1, u32> reserved2;
-        BitField<4, 3, BLEND_DSTFACTC> dst_factor_color_match_select;
+        BitField<4, 3, BlendDstFactC> dst_factor_color_match_select;
         BitField<7, 1, u32> reserved3;
-        BitField<8, 3, BLEND_SRCFACTA> src_factor_a_match_select;
+        BitField<8, 3, BlendSrcFactA> src_factor_a_match_select;
         BitField<11, 1, u32> reserved4;
-        BitField<12, 3, BLEND_DSTFACTA> dst_factor_a_match_select;
+        BitField<12, 3, BlendDstFactA> dst_factor_a_match_select;
         BitField<15, 1, u32> reserved5;
         BitField<16, 4, u32> reserved6;
         BitField<20, 4, u32> reserved7;
@@ -624,8 +621,8 @@ private:
     VicRegisters regs{};
 
     Common::ScratchBuffer<u8> swizzle_scratch;
-    Common::ScratchBuffer<Pixel> output_surface;
+    Common::ScratchBuffer<Tegra::Host1x::Pixel> output_surface;
-    Common::ScratchBuffer<Pixel> slot_surface;
+    Common::ScratchBuffer<Tegra::Host1x::Pixel> slot_surface;
     Common::ScratchBuffer<u8> luma_scratch;
     Common::ScratchBuffer<u8> chroma_scratch;
 

src/video_core/macro.cpp

@@ -64,10 +64,10 @@ bool IsTopologySafe(Maxwell3D::Regs::PrimitiveTopology topology) {
 
 } // Anonymous namespace
 
-void HLE_DrawArraysIndirect::Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
+void HLE_DrawArraysIndirect::Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
     auto topology = static_cast<Maxwell3D::Regs::PrimitiveTopology>(parameters[0]);
     if (!maxwell3d.AnyParametersDirty() || !IsTopologySafe(topology)) {
-        Fallback(maxwell3d, parameters);
+        Fallback(system, maxwell3d, parameters);
         return;
     }
 
@@ -93,7 +93,7 @@ void HLE_DrawArraysIndirect::Execute(Engines::Maxwell3D& maxwell3d, std::span<co
         maxwell3d.replace_table.clear();
     }
 }
-void HLE_DrawArraysIndirect::Fallback(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters) {
+void HLE_DrawArraysIndirect::Fallback(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters) {
     SCOPE_EXIT {
         if (extended) {
             maxwell3d.engine_state = Maxwell3D::EngineHint::None;
@@ -123,10 +123,10 @@ void HLE_DrawArraysIndirect::Fallback(Engines::Maxwell3D& maxwell3d, std::span<c
     }
 }
 
-void HLE_DrawIndexedIndirect::Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
+void HLE_DrawIndexedIndirect::Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
     auto topology = static_cast<Maxwell3D::Regs::PrimitiveTopology>(parameters[0]);
     if (!maxwell3d.AnyParametersDirty() || !IsTopologySafe(topology)) {
-        Fallback(maxwell3d, parameters);
+        Fallback(system, maxwell3d, parameters);
         return;
     }
 
@@ -161,7 +161,7 @@ void HLE_DrawIndexedIndirect::Execute(Engines::Maxwell3D& maxwell3d, std::span<c
         maxwell3d.replace_table.clear();
     }
 }
-void HLE_DrawIndexedIndirect::Fallback(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters) {
+void HLE_DrawIndexedIndirect::Fallback(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters) {
     maxwell3d.RefreshParameters();
     const u32 instance_count = (maxwell3d.GetRegisterValue(0xD1B) & parameters[2]);
     const u32 element_base = parameters[4];
@@ -184,7 +184,7 @@ void HLE_DrawIndexedIndirect::Fallback(Engines::Maxwell3D& maxwell3d, std::span<
         maxwell3d.replace_table.clear();
     }
 }
-void HLE_MultiLayerClear::Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
+void HLE_MultiLayerClear::Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
     maxwell3d.RefreshParameters();
     ASSERT(parameters.size() == 1);
 
@@ -196,47 +196,44 @@ void HLE_MultiLayerClear::Execute(Engines::Maxwell3D& maxwell3d, std::span<const
     maxwell3d.regs.clear_surface.raw = clear_params.raw;
     maxwell3d.draw_manager.Clear(maxwell3d, num_layers);
 }
-void HLE_MultiDrawIndexedIndirectCount::Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
+void HLE_MultiDrawIndexedIndirectCount::Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
     const auto topology = Maxwell3D::Regs::PrimitiveTopology(parameters[2]);
-    if (!IsTopologySafe(topology)) {
+    if (IsTopologySafe(topology)) {
-        Fallback(maxwell3d, parameters);
+        const u32 start_indirect = parameters[0];
-        return;
+        const u32 end_indirect = parameters[1];
+        if (start_indirect >= end_indirect) {
+            // Nothing to do.
+            return;
+        }
+        const u32 padding = parameters[3]; // padding is in words
+        // size of each indirect segment
+        const u32 indirect_words = 5 + padding;
+        const u32 stride = indirect_words * sizeof(u32);
+        const std::size_t draw_count = end_indirect - start_indirect;
+        const u32 estimate = u32(maxwell3d.EstimateIndexBufferSize());
+        maxwell3d.dirty.flags[VideoCommon::Dirty::IndexBuffer] = true;
+        auto& params = maxwell3d.draw_manager.indirect_state;
+        params.is_byte_count = false;
+        params.is_indexed = true;
+        params.include_count = true;
+        params.count_start_address = maxwell3d.GetMacroAddress(4);
+        params.indirect_start_address = maxwell3d.GetMacroAddress(5);
+        params.buffer_size = stride * draw_count;
+        params.max_draw_counts = draw_count;
+        params.stride = stride;
+        maxwell3d.dirty.flags[VideoCommon::Dirty::IndexBuffer] = true;
+        maxwell3d.engine_state = Maxwell3D::EngineHint::OnHLEMacro;
+        maxwell3d.SetHLEReplacementAttributeType(0, 0x640, Maxwell3D::HLEReplacementAttributeType::BaseVertex);
+        maxwell3d.SetHLEReplacementAttributeType(0, 0x644, Maxwell3D::HLEReplacementAttributeType::BaseInstance);
+        maxwell3d.SetHLEReplacementAttributeType(0, 0x648, Maxwell3D::HLEReplacementAttributeType::DrawID);
+        maxwell3d.draw_manager.DrawIndexedIndirect(maxwell3d, topology, 0, estimate);
+        maxwell3d.engine_state = Maxwell3D::EngineHint::None;
+        maxwell3d.replace_table.clear();
+    } else {
+        Fallback(system, maxwell3d, parameters);
     }
-
-    const u32 start_indirect = parameters[0];
-    const u32 end_indirect = parameters[1];
-    if (start_indirect >= end_indirect) {
-        // Nothing to do.
-        return;
-    }
-
-    const u32 padding = parameters[3]; // padding is in words
-
-    // size of each indirect segment
-    const u32 indirect_words = 5 + padding;
-    const u32 stride = indirect_words * sizeof(u32);
-    const std::size_t draw_count = end_indirect - start_indirect;
-    const u32 estimate = static_cast<u32>(maxwell3d.EstimateIndexBufferSize());
-    maxwell3d.dirty.flags[VideoCommon::Dirty::IndexBuffer] = true;
-    auto& params = maxwell3d.draw_manager.indirect_state;
-    params.is_byte_count = false;
-    params.is_indexed = true;
-    params.include_count = true;
-    params.count_start_address = maxwell3d.GetMacroAddress(4);
-    params.indirect_start_address = maxwell3d.GetMacroAddress(5);
-    params.buffer_size = stride * draw_count;
-    params.max_draw_counts = draw_count;
-    params.stride = stride;
-    maxwell3d.dirty.flags[VideoCommon::Dirty::IndexBuffer] = true;
-    maxwell3d.engine_state = Maxwell3D::EngineHint::OnHLEMacro;
-    maxwell3d.SetHLEReplacementAttributeType(0, 0x640, Maxwell3D::HLEReplacementAttributeType::BaseVertex);
-    maxwell3d.SetHLEReplacementAttributeType(0, 0x644, Maxwell3D::HLEReplacementAttributeType::BaseInstance);
-    maxwell3d.SetHLEReplacementAttributeType(0, 0x648, Maxwell3D::HLEReplacementAttributeType::DrawID);
-    maxwell3d.draw_manager.DrawIndexedIndirect(maxwell3d, topology, 0, estimate);
-    maxwell3d.engine_state = Maxwell3D::EngineHint::None;
-    maxwell3d.replace_table.clear();
 }
-void HLE_MultiDrawIndexedIndirectCount::Fallback(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters) {
+void HLE_MultiDrawIndexedIndirectCount::Fallback(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters) {
     SCOPE_EXIT {
         // Clean everything.
         maxwell3d.regs.vertex_id_base = 0x0;
@@ -250,7 +247,7 @@ void HLE_MultiDrawIndexedIndirectCount::Fallback(Engines::Maxwell3D& maxwell3d,
         // Nothing to do.
         return;
     }
-    const auto topology = static_cast<Maxwell3D::Regs::PrimitiveTopology>(parameters[2]);
+    const auto topology = Maxwell3D::Regs::PrimitiveTopology(parameters[2]);
     const u32 padding = parameters[3];
     const std::size_t max_draws = parameters[4];
     const u32 indirect_words = 5 + padding;
@@ -265,41 +262,41 @@ void HLE_MultiDrawIndexedIndirectCount::Fallback(Engines::Maxwell3D& maxwell3d,
         maxwell3d.engine_state = Maxwell3D::EngineHint::OnHLEMacro;
         maxwell3d.SetHLEReplacementAttributeType(0, 0x640, Maxwell3D::HLEReplacementAttributeType::BaseVertex);
         maxwell3d.SetHLEReplacementAttributeType(0, 0x644, Maxwell3D::HLEReplacementAttributeType::BaseInstance);
-        maxwell3d.CallMethod(0x8e3, 0x648, true);
+        maxwell3d.CallMethod(system, 0x8e3, 0x648, true);
-        maxwell3d.CallMethod(0x8e4, static_cast<u32>(index), true);
+        maxwell3d.CallMethod(system, 0x8e4, u32(index), true);
         maxwell3d.dirty.flags[VideoCommon::Dirty::IndexBuffer] = true;
         maxwell3d.draw_manager.DrawIndex(maxwell3d, topology, parameters[base + 2], parameters[base], base_vertex, base_instance, parameters[base + 1]);
     }
 }
-void HLE_DrawIndirectByteCount::Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
+void HLE_DrawIndirectByteCount::Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
     const bool force = maxwell3d.Rasterizer().HasDrawTransformFeedback();
-    if (!force) {
+    if (force) {
-        Fallback(maxwell3d, parameters);
+        auto topology = Maxwell3D::Regs::PrimitiveTopology(parameters[0] & 0xFFFFU);
-        return;
+        auto& params = maxwell3d.draw_manager.indirect_state;
+        params.is_byte_count = true;
+        params.is_indexed = false;
+        params.include_count = false;
+        params.count_start_address = 0;
+        params.indirect_start_address = maxwell3d.GetMacroAddress(2);
+        params.buffer_size = 4;
+        params.max_draw_counts = 1;
+        params.stride = parameters[1];
+        maxwell3d.regs.draw.begin = parameters[0];
+        maxwell3d.regs.draw_auto_stride = parameters[1];
+        maxwell3d.regs.draw_auto_byte_count = parameters[2];
+        maxwell3d.draw_manager.DrawArrayIndirect(maxwell3d, topology);
+    } else {
+        Fallback(system, maxwell3d, parameters);
     }
-    auto topology = Maxwell3D::Regs::PrimitiveTopology(parameters[0] & 0xFFFFU);
-    auto& params = maxwell3d.draw_manager.indirect_state;
-    params.is_byte_count = true;
-    params.is_indexed = false;
-    params.include_count = false;
-    params.count_start_address = 0;
-    params.indirect_start_address = maxwell3d.GetMacroAddress(2);
-    params.buffer_size = 4;
-    params.max_draw_counts = 1;
-    params.stride = parameters[1];
-    maxwell3d.regs.draw.begin = parameters[0];
-    maxwell3d.regs.draw_auto_stride = parameters[1];
-    maxwell3d.regs.draw_auto_byte_count = parameters[2];
-    maxwell3d.draw_manager.DrawArrayIndirect(maxwell3d, topology);
 }
-void HLE_DrawIndirectByteCount::Fallback(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters) {
+void HLE_DrawIndirectByteCount::Fallback(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters) {
     maxwell3d.RefreshParameters();
     maxwell3d.regs.draw.begin = parameters[0];
     maxwell3d.regs.draw_auto_stride = parameters[1];
     maxwell3d.regs.draw_auto_byte_count = parameters[2];
     maxwell3d.draw_manager.DrawArray(maxwell3d, maxwell3d.regs.draw.topology, 0, maxwell3d.regs.draw_auto_byte_count / maxwell3d.regs.draw_auto_stride, 0, 1);
 }
-void HLE_C713C83D8F63CCF3::Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
+void HLE_C713C83D8F63CCF3::Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
     maxwell3d.RefreshParameters();
     const u32 offset = (parameters[0] & 0x3FFFFFFF) << 2;
     const u32 address = maxwell3d.regs.shadow_scratch[24];
@@ -309,7 +306,7 @@ void HLE_C713C83D8F63CCF3::Execute(Engines::Maxwell3D& maxwell3d, std::span<cons
     const_buffer.address_low = address << 8;
     const_buffer.offset = offset;
 }
-void HLE_D7333D26E0A93EDE::Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
+void HLE_D7333D26E0A93EDE::Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
     maxwell3d.RefreshParameters();
     const size_t index = parameters[0];
     const u32 address = maxwell3d.regs.shadow_scratch[42 + index];
@@ -319,7 +316,7 @@ void HLE_D7333D26E0A93EDE::Execute(Engines::Maxwell3D& maxwell3d, std::span<cons
     const_buffer.address_high = (address >> 24) & 0xFF;
     const_buffer.address_low = address << 8;
 }
-void HLE_BindShader::Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
+void HLE_BindShader::Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
     maxwell3d.RefreshParameters();
     auto& regs = maxwell3d.regs;
     const u32 index = parameters[0];
@@ -343,7 +340,7 @@ void HLE_BindShader::Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32>
     bind_group.raw_config = 0x11;
     maxwell3d.ProcessCBBind(bind_group_id);
 }
-void HLE_SetRasterBoundingBox::Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
+void HLE_SetRasterBoundingBox::Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
     maxwell3d.RefreshParameters();
     const u32 raster_mode = parameters[0];
     auto& regs = maxwell3d.regs;
@@ -352,7 +349,7 @@ void HLE_SetRasterBoundingBox::Execute(Engines::Maxwell3D& maxwell3d, std::span<
     regs.raster_bounding_box.raw = raster_mode & 0xFFFFF00F;
     regs.raster_bounding_box.pad.Assign(scratch_data & raster_enabled);
 }
-void HLE_ClearConstBuffer::Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
+void HLE_ClearConstBuffer::Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
     static constexpr std::array<u32, 0x7000> zeroes{}; //must be bigger than either 7000 or 5F00
     maxwell3d.RefreshParameters();
     auto& regs = maxwell3d.regs;
@@ -362,7 +359,7 @@ void HLE_ClearConstBuffer::Execute(Engines::Maxwell3D& maxwell3d, std::span<cons
     regs.const_buffer.offset = 0;
     maxwell3d.ProcessCBMultiData(zeroes.data(), parameters[2] * 4);
 }
-void HLE_ClearMemory::Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
+void HLE_ClearMemory::Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
     maxwell3d.RefreshParameters();
     const u32 needed_memory = parameters[2] / sizeof(u32);
     if (needed_memory > zero_memory.size()) {
@@ -373,10 +370,10 @@ void HLE_ClearMemory::Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32
     regs.upload.line_count = 1;
     regs.upload.dest.address_high = parameters[0];
     regs.upload.dest.address_low = parameters[1];
-    maxwell3d.CallMethod(size_t(MAXWELL3D_REG_INDEX(launch_dma)), 0x1011, true);
+    maxwell3d.CallMethod(system, size_t(MAXWELL3D_REG_INDEX(launch_dma)), 0x1011, true);
-    maxwell3d.CallMultiMethod(size_t(MAXWELL3D_REG_INDEX(inline_data)), zero_memory.data(), needed_memory, needed_memory);
+    maxwell3d.CallMultiMethod(system, size_t(MAXWELL3D_REG_INDEX(inline_data)), zero_memory.data(), needed_memory, needed_memory);
 }
-void HLE_TransformFeedbackSetup::Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
+void HLE_TransformFeedbackSetup::Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method) {
     maxwell3d.RefreshParameters();
     auto& regs = maxwell3d.regs;
     regs.transform_feedback_enabled = 1;
@@ -388,8 +385,8 @@ void HLE_TransformFeedbackSetup::Execute(Engines::Maxwell3D& maxwell3d, std::spa
     regs.upload.line_count = 1;
     regs.upload.dest.address_high = parameters[0];
     regs.upload.dest.address_low = parameters[1];
-    maxwell3d.CallMethod(size_t(MAXWELL3D_REG_INDEX(launch_dma)), 0x1011, true);
+    maxwell3d.CallMethod(system, size_t(MAXWELL3D_REG_INDEX(launch_dma)), 0x1011, true);
-    maxwell3d.CallMethod(size_t(MAXWELL3D_REG_INDEX(inline_data)), regs.transform_feedback.controls[0].stride, true);
+    maxwell3d.CallMethod(system, size_t(MAXWELL3D_REG_INDEX(inline_data)), regs.transform_feedback.controls[0].stride, true);
     maxwell3d.Rasterizer().RegisterTransformFeedback(regs.upload.dest.Address());
 }
 
@@ -429,7 +426,7 @@ void HLE_TransformFeedbackSetup::Execute(Engines::Maxwell3D& maxwell3d, std::spa
     }
 }
 
-void MacroInterpreterImpl::Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> params, u32 method) {
+void MacroInterpreterImpl::Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> params, u32 method) {
     Reset();
 
     registers[1] = params[0];
@@ -439,7 +436,7 @@ void MacroInterpreterImpl::Execute(Engines::Maxwell3D& maxwell3d, std::span<cons
     // Execute the code until we hit an exit condition.
     bool keep_executing = true;
     while (keep_executing) {
-        keep_executing = Step(maxwell3d, false);
+        keep_executing = Step(system, maxwell3d, false);
     }
 
     // Assert the the macro used all the input parameters
@@ -462,7 +459,7 @@ void MacroInterpreterImpl::Reset() {
 /// @brief Executes a single macro instruction located at the current program counter. Returns whether
 /// the interpreter should keep running.
 /// @param is_delay_slot Whether the current step is being executed due to a delay slot in a previous instruction.
-bool MacroInterpreterImpl::Step(Engines::Maxwell3D& maxwell3d, bool is_delay_slot) {
+bool MacroInterpreterImpl::Step(Core::System& system, Engines::Maxwell3D& maxwell3d, bool is_delay_slot) {
     u32 base_address = pc;
 
     Macro::Opcode opcode = GetOpcode();
@@ -478,11 +475,11 @@ bool MacroInterpreterImpl::Step(Engines::Maxwell3D& maxwell3d, bool is_delay_slo
     switch (opcode.operation) {
     case Macro::Operation::ALU: {
         u32 result = GetALUResult(opcode.alu_operation, GetRegister(opcode.src_a), GetRegister(opcode.src_b));
-        ProcessResult(maxwell3d, opcode.result_operation, opcode.dst, result);
+        ProcessResult(system, maxwell3d, opcode.result_operation, opcode.dst, result);
         break;
     }
     case Macro::Operation::AddImmediate: {
-        ProcessResult(maxwell3d, opcode.result_operation, opcode.dst, GetRegister(opcode.src_a) + opcode.immediate);
+        ProcessResult(system, maxwell3d, opcode.result_operation, opcode.dst, GetRegister(opcode.src_a) + opcode.immediate);
         break;
     }
     case Macro::Operation::ExtractInsert: {
@@ -492,7 +489,7 @@ bool MacroInterpreterImpl::Step(Engines::Maxwell3D& maxwell3d, bool is_delay_slo
         src = (src >> opcode.bf_src_bit) & opcode.GetBitfieldMask();
         dst &= ~(opcode.GetBitfieldMask() << opcode.bf_dst_bit);
         dst |= src << opcode.bf_dst_bit;
-        ProcessResult(maxwell3d, opcode.result_operation, opcode.dst, dst);
+        ProcessResult(system, maxwell3d, opcode.result_operation, opcode.dst, dst);
         break;
     }
     case Macro::Operation::ExtractShiftLeftImmediate: {
@@ -501,7 +498,7 @@ bool MacroInterpreterImpl::Step(Engines::Maxwell3D& maxwell3d, bool is_delay_slo
 
         u32 result = ((src >> dst) & opcode.GetBitfieldMask()) << opcode.bf_dst_bit;
 
-        ProcessResult(maxwell3d, opcode.result_operation, opcode.dst, result);
+        ProcessResult(system, maxwell3d, opcode.result_operation, opcode.dst, result);
         break;
     }
     case Macro::Operation::ExtractShiftLeftRegister: {
@@ -510,12 +507,12 @@ bool MacroInterpreterImpl::Step(Engines::Maxwell3D& maxwell3d, bool is_delay_slo
 
         u32 result = ((src >> opcode.bf_src_bit) & opcode.GetBitfieldMask()) << dst;
 
-        ProcessResult(maxwell3d, opcode.result_operation, opcode.dst, result);
+        ProcessResult(system, maxwell3d, opcode.result_operation, opcode.dst, result);
         break;
     }
     case Macro::Operation::Read: {
         u32 result = Read(maxwell3d, GetRegister(opcode.src_a) + opcode.immediate);
-        ProcessResult(maxwell3d, opcode.result_operation, opcode.dst, result);
+        ProcessResult(system, maxwell3d, opcode.result_operation, opcode.dst, result);
         break;
     }
     case Macro::Operation::Branch: {
@@ -531,7 +528,7 @@ bool MacroInterpreterImpl::Step(Engines::Maxwell3D& maxwell3d, bool is_delay_slo
 
             delayed_pc = base_address + opcode.GetBranchTarget();
             // Execute one more instruction due to the delay slot.
-            return Step(maxwell3d, true);
+            return Step(system, maxwell3d, true);
         }
         break;
     }
@@ -544,7 +541,7 @@ bool MacroInterpreterImpl::Step(Engines::Maxwell3D& maxwell3d, bool is_delay_slo
     // cause an exit if it's executed inside a delay slot.
     if (opcode.is_exit && !is_delay_slot) {
         // Exit has a delay slot, execute the next instruction
-        Step(maxwell3d, true);
+        Step(system, maxwell3d, true);
         return false;
     }
     return true;
@@ -591,7 +588,7 @@ u32 MacroInterpreterImpl::GetALUResult(Macro::ALUOperation operation, u32 src_a,
 }
 
 /// Performs the result operation on the input result and stores it in the specified register (if necessary).
-void MacroInterpreterImpl::ProcessResult(Engines::Maxwell3D& maxwell3d, Macro::ResultOperation operation, u32 reg, u32 result) {
+void MacroInterpreterImpl::ProcessResult(Core::System& system, Engines::Maxwell3D& maxwell3d, Macro::ResultOperation operation, u32 reg, u32 result) {
     switch (operation) {
     case Macro::ResultOperation::IgnoreAndFetch:
         // Fetch parameter and ignore result.
@@ -609,12 +606,12 @@ void MacroInterpreterImpl::ProcessResult(Engines::Maxwell3D& maxwell3d, Macro::R
     case Macro::ResultOperation::FetchAndSend:
         // Fetch parameter and send result.
         SetRegister(reg, FetchParameter());
-        Send(maxwell3d, result);
+        Send(system, maxwell3d, result);
         break;
     case Macro::ResultOperation::MoveAndSend:
         // Move and send result.
         SetRegister(reg, result);
-        Send(maxwell3d, result);
+        Send(system, maxwell3d, result);
         break;
     case Macro::ResultOperation::FetchAndSetMethod:
         // Fetch parameter and use result as Method Address.
@@ -625,13 +622,13 @@ void MacroInterpreterImpl::ProcessResult(Engines::Maxwell3D& maxwell3d, Macro::R
         // Move result and use as Method Address, then fetch and send parameter.
         SetRegister(reg, result);
         SetMethodAddress(result);
-        Send(maxwell3d, FetchParameter());
+        Send(system, maxwell3d, FetchParameter());
         break;
     case Macro::ResultOperation::MoveAndSetMethodSend:
         // Move result and use as Method Address, then send bits 12:17 of result.
         SetRegister(reg, result);
         SetMethodAddress(result);
-        Send(maxwell3d, (result >> 12) & 0b111111);
+        Send(system, maxwell3d, (result >> 12) & 0b111111);
         break;
     default:
         UNIMPLEMENTED_MSG("Unimplemented result operation {}", operation);
@@ -672,8 +669,8 @@ void MacroInterpreterImpl::SetRegister(u32 register_id, u32 value) {
 }
 
 /// Calls a GPU Engine method with the input parameter.
-void MacroInterpreterImpl::Send(Engines::Maxwell3D& maxwell3d, u32 value) {
+void MacroInterpreterImpl::Send(Core::System& system, Engines::Maxwell3D& maxwell3d, u32 value) {
-    maxwell3d.CallMethod(method_address.address, value, true);
+    maxwell3d.CallMethod(system, method_address.address, value, true);
     // Increment the method address by the method increment.
     method_address.address.Assign(method_address.address.Value() + method_address.increment.Value());
 }
@@ -724,34 +721,35 @@ static const auto default_cg_mode = nullptr; //Allow RWE
 #endif
 
 struct MacroJITx64Impl final : public Xbyak::CodeGenerator, public DynamicCachedMacro {
-    explicit MacroJITx64Impl(std::span<const u32> code_)
+    explicit MacroJITx64Impl(Core::System& system, std::span<const u32> code_)
         : Xbyak::CodeGenerator(MAX_CODE_SIZE, default_cg_mode)
         , code{code_}
     {
-        Compile();
+        Compile(system);
     }
 
-    void Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, u32 method) override;
+    void Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, u32 method) override;
 
-    void Compile_ALU(Macro::Opcode opcode);
+    void Compile_ALU(Core::System& system, Macro::Opcode opcode);
-    void Compile_AddImmediate(Macro::Opcode opcode);
+    void Compile_AddImmediate(Core::System& system, Macro::Opcode opcode);
-    void Compile_ExtractInsert(Macro::Opcode opcode);
+    void Compile_ExtractInsert(Core::System& system, Macro::Opcode opcode);
-    void Compile_ExtractShiftLeftImmediate(Macro::Opcode opcode);
+    void Compile_ExtractShiftLeftImmediate(Core::System& system, Macro::Opcode opcode);
-    void Compile_ExtractShiftLeftRegister(Macro::Opcode opcode);
+    void Compile_ExtractShiftLeftRegister(Core::System& system, Macro::Opcode opcode);
-    void Compile_Read(Macro::Opcode opcode);
+    void Compile_Read(Core::System& system, Macro::Opcode opcode);
     void Compile_Branch(Macro::Opcode opcode);
 
     void Optimizer_ScanFlags();
-    void Compile();
+    void Compile(Core::System& system);
-    bool Compile_NextInstruction();
+    bool Compile_NextInstruction(Core::System& system);
     Xbyak::Reg32 Compile_FetchParameter();
     Xbyak::Reg32 Compile_GetRegister(u32 index, Xbyak::Reg32 dst);
-    void Compile_ProcessResult(Macro::ResultOperation operation, u32 reg);
+    void Compile_ProcessResult(Core::System& system, Macro::ResultOperation operation, u32 reg);
-    void Compile_Send(Xbyak::Reg32 value);
+    void Compile_Send(Core::System& system, Xbyak::Reg32 value);
     Macro::Opcode GetOpCode() const;
 
     struct JITState {
-        Engines::Maxwell3D* maxwell3d{};
+        Engines::Maxwell3D* maxwell3d = nullptr;
+        Core::System* system = nullptr;
         std::array<u32, Macro::NUM_MACRO_REGISTERS> registers{};
         u32 carry_flag{};
     };
@@ -777,15 +775,16 @@ struct MacroJITx64Impl final : public Xbyak::CodeGenerator, public DynamicCached
     std::span<const u32> code;
 };
 
-void MacroJITx64Impl::Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, u32 method) {
+void MacroJITx64Impl::Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, u32 method) {
     ASSERT_OR_EXECUTE(program != nullptr, { return; });
     JITState state{};
     state.maxwell3d = &maxwell3d;
+    state.system = &system;
     state.registers = {};
     program(&state, parameters.data(), parameters.data() + parameters.size());
 }
 
-void MacroJITx64Impl::Compile_ALU(Macro::Opcode opcode) {
+void MacroJITx64Impl::Compile_ALU(Core::System& system, Macro::Opcode opcode) {
     const bool is_a_zero = opcode.src_a == 0;
     const bool is_b_zero = opcode.src_b == 0;
     const bool valid_operation = !is_a_zero && !is_b_zero;
@@ -902,10 +901,10 @@ void MacroJITx64Impl::Compile_ALU(Macro::Opcode opcode) {
         UNIMPLEMENTED_MSG("Unimplemented ALU operation {}", opcode.alu_operation.Value());
         break;
     }
-    Compile_ProcessResult(opcode.result_operation, opcode.dst);
+    Compile_ProcessResult(system, opcode.result_operation, opcode.dst);
 }
 
-void MacroJITx64Impl::Compile_AddImmediate(Macro::Opcode opcode) {
+void MacroJITx64Impl::Compile_AddImmediate(Core::System& system, Macro::Opcode opcode) {
     if (optimizer.skip_dummy_addimmediate) {
         // Games tend to use this as an exit instruction placeholder. It's to encode an instruction
         // without doing anything. In our case we can just not emit anything.
@@ -940,10 +939,10 @@ void MacroJITx64Impl::Compile_AddImmediate(Macro::Opcode opcode) {
             sub(result, opcode.immediate * -1);
         }
     }
-    Compile_ProcessResult(opcode.result_operation, opcode.dst);
+    Compile_ProcessResult(system, opcode.result_operation, opcode.dst);
 }
 
-void MacroJITx64Impl::Compile_ExtractInsert(Macro::Opcode opcode) {
+void MacroJITx64Impl::Compile_ExtractInsert(Core::System& system, Macro::Opcode opcode) {
     auto dst = Compile_GetRegister(opcode.src_a, RESULT);
     auto src = Compile_GetRegister(opcode.src_b, eax);
 
@@ -954,10 +953,10 @@ void MacroJITx64Impl::Compile_ExtractInsert(Macro::Opcode opcode) {
     shl(src, opcode.bf_dst_bit);
     or_(dst, src);
 
-    Compile_ProcessResult(opcode.result_operation, opcode.dst);
+    Compile_ProcessResult(system, opcode.result_operation, opcode.dst);
 }
 
-void MacroJITx64Impl::Compile_ExtractShiftLeftImmediate(Macro::Opcode opcode) {
+void MacroJITx64Impl::Compile_ExtractShiftLeftImmediate(Core::System& system, Macro::Opcode opcode) {
     const auto dst = Compile_GetRegister(opcode.src_a, ecx);
     const auto src = Compile_GetRegister(opcode.src_b, RESULT);
 
@@ -965,10 +964,10 @@ void MacroJITx64Impl::Compile_ExtractShiftLeftImmediate(Macro::Opcode opcode) {
     and_(src, opcode.GetBitfieldMask());
     shl(src, opcode.bf_dst_bit);
 
-    Compile_ProcessResult(opcode.result_operation, opcode.dst);
+    Compile_ProcessResult(system, opcode.result_operation, opcode.dst);
 }
 
-void MacroJITx64Impl::Compile_ExtractShiftLeftRegister(Macro::Opcode opcode) {
+void MacroJITx64Impl::Compile_ExtractShiftLeftRegister(Core::System& system, Macro::Opcode opcode) {
     const auto dst = Compile_GetRegister(opcode.src_a, ecx);
     const auto src = Compile_GetRegister(opcode.src_b, RESULT);
 
@@ -976,10 +975,10 @@ void MacroJITx64Impl::Compile_ExtractShiftLeftRegister(Macro::Opcode opcode) {
     and_(src, opcode.GetBitfieldMask());
     shl(src, dst.cvt8());
 
-    Compile_ProcessResult(opcode.result_operation, opcode.dst);
+    Compile_ProcessResult(system, opcode.result_operation, opcode.dst);
 }
 
-void MacroJITx64Impl::Compile_Read(Macro::Opcode opcode) {
+void MacroJITx64Impl::Compile_Read(Core::System& system, Macro::Opcode opcode) {
     if (optimizer.zero_reg_skip && opcode.src_a == 0) {
         if (opcode.immediate == 0) {
             xor_(RESULT, RESULT);
@@ -1005,23 +1004,21 @@ void MacroJITx64Impl::Compile_Read(Macro::Opcode opcode) {
         int3();
         L(pass_range_check);
     }
-    mov(rax, qword[STATE]);
+    mov(rax, qword[STATE + offsetof(JITState, maxwell3d)]);
-    mov(RESULT,
+    mov(RESULT, dword[rax + offsetof(Engines::Maxwell3D, regs) + offsetof(Engines::Maxwell3D::Regs, reg_array) + RESULT.cvt64() * sizeof(u32)]);
-        dword[rax + offsetof(Engines::Maxwell3D, regs) +
+    Compile_ProcessResult(system, opcode.result_operation, opcode.dst);
-              offsetof(Engines::Maxwell3D::Regs, reg_array) + RESULT.cvt64() * sizeof(u32)]);
-
-    Compile_ProcessResult(opcode.result_operation, opcode.dst);
 }
 
-static void MacroJIT_SendThunk(Engines::Maxwell3D* maxwell3d, Macro::MethodAddress method_address, u32 value) {
+static void MacroJIT_SendThunk(Core::System* system, Engines::Maxwell3D* maxwell3d, Macro::MethodAddress method_address, u32 value) {
-    maxwell3d->CallMethod(method_address.address, value, true);
+    maxwell3d->CallMethod(*system, method_address.address, value, true);
 }
 
-void MacroJITx64Impl::Compile_Send(Xbyak::Reg32 value) {
+void MacroJITx64Impl::Compile_Send(Core::System& system, Xbyak::Reg32 value) {
     Common::X64::ABI_PushRegistersAndAdjustStack(*this, PersistentCallerSavedRegs(), 0);
-    mov(Common::X64::ABI_PARAM1, qword[STATE]);
+    mov(Common::X64::ABI_PARAM1, qword[STATE + offsetof(JITState, system)]);
-    mov(Common::X64::ABI_PARAM2.cvt32(), METHOD_ADDRESS);
+    mov(Common::X64::ABI_PARAM2, qword[STATE + offsetof(JITState, maxwell3d)]);
-    mov(Common::X64::ABI_PARAM3.cvt32(), value);
+    mov(Common::X64::ABI_PARAM3.cvt32(), METHOD_ADDRESS);
+    mov(Common::X64::ABI_PARAM4.cvt32(), value);
     Common::X64::CallFarFunction(*this, &MacroJIT_SendThunk);
     Common::X64::ABI_PopRegistersAndAdjustStack(*this, PersistentCallerSavedRegs(), 0);
 
@@ -1045,9 +1042,8 @@ void MacroJITx64Impl::Compile_Send(Xbyak::Reg32 value) {
 }
 
 void MacroJITx64Impl::Compile_Branch(Macro::Opcode opcode) {
-    ASSERT_MSG(!is_delay_slot, "Executing a branch in a delay slot is not valid");
+    ASSERT(!is_delay_slot && "Executing a branch in a delay slot is not valid");
-    const s32 jump_address =
+    const s32 jump_address = s32(pc) + s32(opcode.GetBranchTarget() / sizeof(s32));
-        static_cast<s32>(pc) + static_cast<s32>(opcode.GetBranchTarget() / sizeof(s32));
 
     Xbyak::Label end;
     auto value = Compile_GetRegister(opcode.src_a, eax);
@@ -1116,7 +1112,7 @@ void MacroJITx64Impl::Optimizer_ScanFlags() {
     }
 }
 
-void MacroJITx64Impl::Compile() {
+void MacroJITx64Impl::Compile(Core::System& system) {
     labels.fill(Xbyak::Label());
 
     Common::X64::ABI_PushRegistersAndAdjustStack(*this, Common::X64::ABI_ALL_CALLEE_SAVED, 8);
@@ -1156,7 +1152,7 @@ void MacroJITx64Impl::Compile() {
             next_opcode = {};
         }
         pc = i;
-        Compile_NextInstruction();
+        Compile_NextInstruction(system);
     }
 
     L(end_of_code);
@@ -1167,7 +1163,7 @@ void MacroJITx64Impl::Compile() {
     program = getCode<ProgramType>();
 }
 
-bool MacroJITx64Impl::Compile_NextInstruction() {
+bool MacroJITx64Impl::Compile_NextInstruction(Core::System& system) {
     const auto opcode = GetOpCode();
     if (labels[pc].getAddress()) {
         return false;
@@ -1177,22 +1173,22 @@ bool MacroJITx64Impl::Compile_NextInstruction() {
 
     switch (opcode.operation) {
     case Macro::Operation::ALU:
-        Compile_ALU(opcode);
+        Compile_ALU(system, opcode);
         break;
     case Macro::Operation::AddImmediate:
-        Compile_AddImmediate(opcode);
+        Compile_AddImmediate(system, opcode);
         break;
     case Macro::Operation::ExtractInsert:
-        Compile_ExtractInsert(opcode);
+        Compile_ExtractInsert(system, opcode);
         break;
     case Macro::Operation::ExtractShiftLeftImmediate:
-        Compile_ExtractShiftLeftImmediate(opcode);
+        Compile_ExtractShiftLeftImmediate(system, opcode);
         break;
     case Macro::Operation::ExtractShiftLeftRegister:
-        Compile_ExtractShiftLeftRegister(opcode);
+        Compile_ExtractShiftLeftRegister(system, opcode);
         break;
     case Macro::Operation::Read:
-        Compile_Read(opcode);
+        Compile_Read(system, opcode);
         break;
     case Macro::Operation::Branch:
         Compile_Branch(opcode);
@@ -1264,7 +1260,7 @@ Xbyak::Reg32 MacroJITx64Impl::Compile_GetRegister(u32 index, Xbyak::Reg32 dst) {
     return dst;
 }
 
-void MacroJITx64Impl::Compile_ProcessResult(Macro::ResultOperation operation, u32 reg) {
+void MacroJITx64Impl::Compile_ProcessResult(Core::System& system, Macro::ResultOperation operation, u32 reg) {
     const auto SetRegister = [this](u32 reg_index, const Xbyak::Reg32& result) {
         // Register 0 is supposed to always return 0. NOP is implemented as a store to the zero
         // register.
@@ -1289,12 +1285,12 @@ void MacroJITx64Impl::Compile_ProcessResult(Macro::ResultOperation operation, u3
     case Macro::ResultOperation::FetchAndSend:
         // Fetch parameter and send result.
         SetRegister(reg, Compile_FetchParameter());
-        Compile_Send(RESULT);
+        Compile_Send(system, RESULT);
         break;
     case Macro::ResultOperation::MoveAndSend:
         // Move and send result.
         SetRegister(reg, RESULT);
-        Compile_Send(RESULT);
+        Compile_Send(system, RESULT);
         break;
     case Macro::ResultOperation::FetchAndSetMethod:
         // Fetch parameter and use result as Method Address.
@@ -1305,7 +1301,7 @@ void MacroJITx64Impl::Compile_ProcessResult(Macro::ResultOperation operation, u3
         // Move result and use as Method Address, then fetch and send parameter.
         SetRegister(reg, RESULT);
         SetMethodAddress(RESULT);
-        Compile_Send(Compile_FetchParameter());
+        Compile_Send(system, Compile_FetchParameter());
         break;
     case Macro::ResultOperation::MoveAndSetMethodSend:
         // Move result and use as Method Address, then send bits 12:17 of result.
@@ -1313,7 +1309,7 @@ void MacroJITx64Impl::Compile_ProcessResult(Macro::ResultOperation operation, u3
         SetMethodAddress(RESULT);
         shr(RESULT, 12);
         and_(RESULT, 0b111111);
-        Compile_Send(RESULT);
+        Compile_Send(system, RESULT);
         break;
     default:
         UNIMPLEMENTED_MSG("Unimplemented macro operation {}", operation);
@@ -1354,36 +1350,36 @@ static void Dump(u64 hash, std::span<const u32> code, bool decompiled = false) {
     macro_file.write(reinterpret_cast<const char*>(code.data()), code.size_bytes());
 }
 
-void MacroEngine::Execute(Engines::Maxwell3D& maxwell3d, u32 method, std::span<const u32> parameters) {
+void MacroEngine::Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, u32 method, std::span<const u32> parameters) {
-    auto const execute_variant = [&maxwell3d, &parameters, method](AnyCachedMacro& acm) {
+    auto const execute_variant = [&system, &maxwell3d, &parameters, method](AnyCachedMacro& acm) {
         if (auto a = std::get_if<HLE_DrawArraysIndirect>(&acm))
-            a->Execute(maxwell3d, parameters, method);
+            a->Execute(system, maxwell3d, parameters, method);
         if (auto a = std::get_if<HLE_DrawIndexedIndirect>(&acm))
-            a->Execute(maxwell3d, parameters, method);
+            a->Execute(system, maxwell3d, parameters, method);
         if (auto a = std::get_if<HLE_MultiDrawIndexedIndirectCount>(&acm))
-            a->Execute(maxwell3d, parameters, method);
+            a->Execute(system, maxwell3d, parameters, method);
         if (auto a = std::get_if<HLE_MultiLayerClear>(&acm))
-            a->Execute(maxwell3d, parameters, method);
+            a->Execute(system, maxwell3d, parameters, method);
         if (auto a = std::get_if<HLE_C713C83D8F63CCF3>(&acm))
-            a->Execute(maxwell3d, parameters, method);
+            a->Execute(system, maxwell3d, parameters, method);
         if (auto a = std::get_if<HLE_D7333D26E0A93EDE>(&acm))
-            a->Execute(maxwell3d, parameters, method);
+            a->Execute(system, maxwell3d, parameters, method);
         if (auto a = std::get_if<HLE_BindShader>(&acm))
-            a->Execute(maxwell3d, parameters, method);
+            a->Execute(system, maxwell3d, parameters, method);
         if (auto a = std::get_if<HLE_SetRasterBoundingBox>(&acm))
-            a->Execute(maxwell3d, parameters, method);
+            a->Execute(system, maxwell3d, parameters, method);
         if (auto a = std::get_if<HLE_ClearConstBuffer>(&acm))
-            a->Execute(maxwell3d, parameters, method);
+            a->Execute(system, maxwell3d, parameters, method);
         if (auto a = std::get_if<HLE_ClearMemory>(&acm))
-            a->Execute(maxwell3d, parameters, method);
+            a->Execute(system, maxwell3d, parameters, method);
         if (auto a = std::get_if<HLE_TransformFeedbackSetup>(&acm))
-            a->Execute(maxwell3d, parameters, method);
+            a->Execute(system, maxwell3d, parameters, method);
         if (auto a = std::get_if<HLE_DrawIndirectByteCount>(&acm))
-            a->Execute(maxwell3d, parameters, method);
+            a->Execute(system, maxwell3d, parameters, method);
         if (auto a = std::get_if<MacroInterpreterImpl>(&acm))
-            a->Execute(maxwell3d, parameters, method);
+            a->Execute(system, maxwell3d, parameters, method);
         if (auto a = std::get_if<std::unique_ptr<DynamicCachedMacro>>(&acm))
-            a->get()->Execute(maxwell3d, parameters, method);
+            a->get()->Execute(system, maxwell3d, parameters, method);
     };
     if (auto const it = macro_cache.find(method); it != macro_cache.end()) {
         auto& ci = it->second;
@@ -1414,9 +1410,9 @@ void MacroEngine::Execute(Engines::Maxwell3D& maxwell3d, u32 method, std::span<c
             code.resize(macro_cached.size() - rebased_method);
             std::memcpy(code.data(), macro_cached.data() + rebased_method, code.size() * sizeof(u32));
             ci.hash = Common::HashValue(code);
-            ci.program = Compile(maxwell3d, code);
+            ci.program = Compile(system, maxwell3d, code);
         } else {
-            ci.program = Compile(maxwell3d, macro_code->second);
+            ci.program = Compile(system, maxwell3d, macro_code->second);
             ci.hash = Common::HashValue(macro_code->second);
         }
         if (CanBeHLEProgram(ci.hash) && !Settings::values.disable_macro_hle) {
@@ -1431,10 +1427,10 @@ void MacroEngine::Execute(Engines::Maxwell3D& maxwell3d, u32 method, std::span<c
     }
 }
 
-AnyCachedMacro MacroEngine::Compile(Engines::Maxwell3D& maxwell3d, std::span<const u32> code) {
+AnyCachedMacro MacroEngine::Compile(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> code) {
 #ifdef ARCHITECTURE_x86_64
     if (!is_interpreted)
-        return std::make_unique<MacroJITx64Impl>(code);
+        return std::make_unique<MacroJITx64Impl>(system, code);
 #endif
     return MacroInterpreterImpl(code);
 }

src/video_core/macro.h

@@ -14,6 +14,10 @@
 #include "common/bit_field.h"
 #include "common/common_types.h"
 
+namespace Core {
+class System;
+}
+
 namespace Tegra {
 
 namespace Engines {
@@ -106,61 +110,61 @@ struct HLEMacro {
 /// also assigning the base vertex/instance.
 struct HLE_DrawArraysIndirect final {
     HLE_DrawArraysIndirect(bool extended_) noexcept : extended{extended_} {}
-    void Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
+    void Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
-    void Fallback(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters);
+    void Fallback(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters);
     bool extended;
 };
 /// @note: these macros have two versions, a normal and extended version, with the extended version
 /// also assigning the base vertex/instance.
 struct HLE_DrawIndexedIndirect final {
     explicit HLE_DrawIndexedIndirect(bool extended_) noexcept : extended{extended_} {}
-    void Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
+    void Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
-    void Fallback(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters);
+    void Fallback(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters);
     bool extended;
 };
 struct HLE_MultiLayerClear final {
-    void Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
+    void Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
 };
 struct HLE_MultiDrawIndexedIndirectCount final {
-    void Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
+    void Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
-    void Fallback(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters);
+    void Fallback(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters);
 };
 struct HLE_DrawIndirectByteCount final {
-    void Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
+    void Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
-    void Fallback(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters);
+    void Fallback(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters);
 };
 struct HLE_C713C83D8F63CCF3 final {
-    void Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
+    void Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
 };
 struct HLE_D7333D26E0A93EDE final {
-    void Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
+    void Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
 };
 struct HLE_BindShader final {
-    void Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
+    void Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
 };
 struct HLE_SetRasterBoundingBox final {
-    void Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
+    void Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
 };
 struct HLE_ClearConstBuffer final {
     HLE_ClearConstBuffer(size_t base_size_) noexcept : base_size{base_size_} {}
-    void Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
+    void Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
     size_t base_size;
 };
 struct HLE_ClearMemory final {
-    void Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
+    void Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
     std::vector<u32> zero_memory;
 };
 struct HLE_TransformFeedbackSetup final {
-    void Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
+    void Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, [[maybe_unused]] u32 method);
 };
 struct MacroInterpreterImpl final {
     MacroInterpreterImpl() {}
     MacroInterpreterImpl(std::span<const u32> code_) : code{code_} {}
-    void Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> params, u32 method);
+    void Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> params, u32 method);
     void Reset();
-    bool Step(Engines::Maxwell3D& maxwell3d, bool is_delay_slot);
+    bool Step(Core::System& system, Engines::Maxwell3D& maxwell3d, bool is_delay_slot);
     u32 GetALUResult(Macro::ALUOperation operation, u32 src_a, u32 src_b);
-    void ProcessResult(Engines::Maxwell3D& maxwell3d, Macro::ResultOperation operation, u32 reg, u32 result);
+    void ProcessResult(Core::System& system, Engines::Maxwell3D& maxwell3d, Macro::ResultOperation operation, u32 reg, u32 result);
     bool EvaluateBranchCondition(Macro::BranchCondition cond, u32 value) const;
     Macro::Opcode GetOpcode() const;
     u32 GetRegister(u32 register_id) const;
@@ -169,7 +173,7 @@ struct MacroInterpreterImpl final {
     [[nodiscard]] inline void SetMethodAddress(u32 address) noexcept {
         method_address.raw = address;
     }
-    void Send(Engines::Maxwell3D& maxwell3d, u32 value);
+    void Send(Core::System& system, Engines::Maxwell3D& maxwell3d, u32 value);
     u32 Read(Engines::Maxwell3D& maxwell3d, u32 method) const;
     u32 FetchParameter();
     /// General purpose macro registers.
@@ -192,7 +196,7 @@ struct DynamicCachedMacro {
     /// Executes the macro code with the specified input parameters.
     /// @param parameters The parameters of the macro
     /// @param method     The method to execute
-    virtual void Execute(Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, u32 method) = 0;
+    virtual void Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> parameters, u32 method) = 0;
 };
 
 using AnyCachedMacro = std::variant<
@@ -227,8 +231,8 @@ struct MacroEngine {
         uploaded_macro_code.erase(method);
     }
     // Compiles the macro if its not in the cache, and executes the compiled macro
-    void Execute(Engines::Maxwell3D& maxwell3d, u32 method, std::span<const u32> parameters);
+    void Execute(Core::System& system, Engines::Maxwell3D& maxwell3d, u32 method, std::span<const u32> parameters);
-    AnyCachedMacro Compile(Engines::Maxwell3D& maxwell3d, std::span<const u32> code);
+    AnyCachedMacro Compile(Core::System& system, Engines::Maxwell3D& maxwell3d, std::span<const u32> code);
     struct CacheInfo {
         AnyCachedMacro program;
         u64 hash{};

src/video_core/renderer_opengl/gl_state_tracker.cpp

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2019 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -204,7 +207,7 @@ void SetupDirtyMisc(Tables& tables) {
 } // Anonymous namespace
 
 void StateTracker::SetupTables(Tegra::Control::ChannelState& channel_state) {
-    auto& tables{channel_state.maxwell_3d->dirty.tables};
+    auto& tables{channel_state.payload->maxwell_3d.dirty.tables};
     SetupDirtyFlags(tables);
     SetupDirtyColorMasks(tables);
     SetupDirtyViewports(tables);
@@ -231,7 +234,7 @@ void StateTracker::SetupTables(Tegra::Control::ChannelState& channel_state) {
 }
 
 void StateTracker::ChangeChannel(Tegra::Control::ChannelState& channel_state) {
-    flags = &channel_state.maxwell_3d->dirty.flags;
+    flags = &channel_state.payload->maxwell_3d.dirty.flags;
 }
 
 void StateTracker::InvalidateState() {

src/video_core/renderer_vulkan/vk_query_cache.cpp

@@ -126,16 +126,14 @@ public:
         current_query = nullptr;
         amend_value = 0;
         accumulation_value = 0;
-        queries_prefix_scan_pass = std::make_unique<QueriesPrefixScanPass>(
+        queries_prefix_scan_pass.emplace(device, scheduler, descriptor_pool, compute_pass_descriptor_queue);
-            device, scheduler, descriptor_pool, compute_pass_descriptor_queue);
 
         const VkBufferCreateInfo buffer_ci = {
             .sType = VK_STRUCTURE_TYPE_BUFFER_CREATE_INFO,
             .pNext = nullptr,
             .flags = 0,
             .size = 8,
-            .usage = VK_BUFFER_USAGE_TRANSFER_DST_BIT | VK_BUFFER_USAGE_TRANSFER_SRC_BIT |
+            .usage = VK_BUFFER_USAGE_TRANSFER_DST_BIT | VK_BUFFER_USAGE_TRANSFER_SRC_BIT | VK_BUFFER_USAGE_STORAGE_BUFFER_BIT,
-                     VK_BUFFER_USAGE_STORAGE_BUFFER_BIT,
             .sharingMode = VK_SHARING_MODE_EXCLUSIVE,
             .queueFamilyIndexCount = 0,
             .pQueueFamilyIndices = nullptr,
@@ -592,8 +590,7 @@ private:
     VideoCommon::HostQueryBase* current_query;
     bool has_started{};
     std::mutex flush_guard;
-
+    std::optional<QueriesPrefixScanPass> queries_prefix_scan_pass;
-    std::unique_ptr<QueriesPrefixScanPass> queries_prefix_scan_pass;
 };
 
 // Transform feedback queries
@@ -1266,41 +1263,23 @@ private:
 } // namespace
 
 struct QueryCacheRuntimeImpl {
-    QueryCacheRuntimeImpl(QueryCacheRuntime& runtime, VideoCore::RasterizerInterface* rasterizer_,
+    QueryCacheRuntimeImpl(QueryCacheRuntime& runtime, VideoCore::RasterizerInterface* rasterizer_, Tegra::MaxwellDeviceMemoryManager& device_memory_, Vulkan::BufferCache& buffer_cache_, const Device& device_, const MemoryAllocator& memory_allocator_, Scheduler& scheduler_, StagingBufferPool& staging_pool_, ComputePassDescriptorQueue& compute_pass_descriptor_queue, DescriptorPool& descriptor_pool, TextureCache& texture_cache_)
-                          Tegra::MaxwellDeviceMemoryManager& device_memory_,
+        : rasterizer{rasterizer_}, device_memory{device_memory_}, buffer_cache{buffer_cache_}
-                          Vulkan::BufferCache& buffer_cache_, const Device& device_,
+        , device{device_}, memory_allocator{memory_allocator_}, scheduler{scheduler_}
-                          const MemoryAllocator& memory_allocator_, Scheduler& scheduler_,
+        , staging_pool{staging_pool_}, guest_streamer(0, runtime)
-                          StagingBufferPool& staging_pool_,
+        , sample_streamer(size_t(QueryType::ZPassPixelCount64), runtime, rasterizer, texture_cache_, device, scheduler, memory_allocator, compute_pass_descriptor_queue, descriptor_pool)
-                          ComputePassDescriptorQueue& compute_pass_descriptor_queue,
+        , tfb_streamer(size_t(QueryType::StreamingByteCount), runtime, device, scheduler, memory_allocator, staging_pool)
-                          DescriptorPool& descriptor_pool, TextureCache& texture_cache_)
+        , primitives_succeeded_streamer(size_t(QueryType::StreamingPrimitivesSucceeded), runtime, tfb_streamer, device_memory_)
-        : rasterizer{rasterizer_}, device_memory{device_memory_}, buffer_cache{buffer_cache_},
+        , primitives_needed_minus_succeeded_streamer(size_t(QueryType::StreamingPrimitivesNeededMinusSucceeded), runtime, 0u)
-          device{device_}, memory_allocator{memory_allocator_}, scheduler{scheduler_},
+        , hcr_setup{}, hcr_is_set{}, is_hcr_running{}, maxwell3d{} {
-          staging_pool{staging_pool_}, guest_streamer(0, runtime),
-          sample_streamer(static_cast<size_t>(QueryType::ZPassPixelCount64), runtime, rasterizer,
-                          texture_cache_, device, scheduler, memory_allocator,
-                          compute_pass_descriptor_queue, descriptor_pool),
-          tfb_streamer(static_cast<size_t>(QueryType::StreamingByteCount), runtime, device,
-                       scheduler, memory_allocator, staging_pool),
-          primitives_succeeded_streamer(
-              static_cast<size_t>(QueryType::StreamingPrimitivesSucceeded), runtime, tfb_streamer,
-              device_memory_),
-          primitives_needed_minus_succeeded_streamer(
-              static_cast<size_t>(QueryType::StreamingPrimitivesNeededMinusSucceeded), runtime, 0u),
-          hcr_setup{}, hcr_is_set{}, is_hcr_running{}, maxwell3d{} {
 
         hcr_setup.sType = VK_STRUCTURE_TYPE_CONDITIONAL_RENDERING_BEGIN_INFO_EXT;
         hcr_setup.pNext = nullptr;
         hcr_setup.flags = 0;
 
-        const bool has_conditional_rendering = device.IsExtConditionalRendering();
+        VkBufferUsageFlags hcr_buffer_usage = VK_BUFFER_USAGE_TRANSFER_DST_BIT | VK_BUFFER_USAGE_STORAGE_BUFFER_BIT;
-        if (has_conditional_rendering) {
+        if (device.IsExtConditionalRendering()) {
-            conditional_resolve_pass = std::make_unique<ConditionalRenderingResolvePass>(
+            conditional_resolve_pass.emplace(device, scheduler, descriptor_pool, compute_pass_descriptor_queue);
-                device, scheduler, descriptor_pool, compute_pass_descriptor_queue);
-        }
-
-        VkBufferUsageFlags hcr_buffer_usage =
-            VK_BUFFER_USAGE_TRANSFER_DST_BIT | VK_BUFFER_USAGE_STORAGE_BUFFER_BIT;
-        if (has_conditional_rendering) {
             hcr_buffer_usage |= VK_BUFFER_USAGE_CONDITIONAL_RENDERING_BIT_EXT;
         }
 
@@ -1339,7 +1318,7 @@ struct QueryCacheRuntimeImpl {
     std::vector<std::vector<VkBufferCopy>> copies_setup;
 
     // Host conditional rendering data
-    std::unique_ptr<ConditionalRenderingResolvePass> conditional_resolve_pass;
+    std::optional<ConditionalRenderingResolvePass> conditional_resolve_pass;
     vk::Buffer hcr_resolve_buffer;
     VkConditionalRenderingBeginInfoEXT hcr_setup;
     VkBuffer hcr_buffer;
@@ -1351,13 +1330,7 @@ struct QueryCacheRuntimeImpl {
     Maxwell3D* maxwell3d;
 };
 
-QueryCacheRuntime::QueryCacheRuntime(VideoCore::RasterizerInterface* rasterizer,
+QueryCacheRuntime::QueryCacheRuntime(VideoCore::RasterizerInterface* rasterizer, Tegra::MaxwellDeviceMemoryManager& device_memory_, Vulkan::BufferCache& buffer_cache_, const Device& device_, const MemoryAllocator& memory_allocator_, Scheduler& scheduler_, StagingBufferPool& staging_pool_, ComputePassDescriptorQueue& compute_pass_descriptor_queue, DescriptorPool& descriptor_pool, TextureCache& texture_cache_) {
-                                     Tegra::MaxwellDeviceMemoryManager& device_memory_,
-                                     Vulkan::BufferCache& buffer_cache_, const Device& device_,
-                                     const MemoryAllocator& memory_allocator_,
-                                     Scheduler& scheduler_, StagingBufferPool& staging_pool_,
-                                     ComputePassDescriptorQueue& compute_pass_descriptor_queue,
-                                     DescriptorPool& descriptor_pool, TextureCache& texture_cache_) {
     impl = std::make_unique<QueryCacheRuntimeImpl>(
         *this, rasterizer, device_memory_, buffer_cache_, device_, memory_allocator_, scheduler_,
         staging_pool_, compute_pass_descriptor_queue, descriptor_pool, texture_cache_);
@@ -1686,8 +1659,7 @@ void QueryCacheRuntime::SyncValues(std::span<SyncValuesType> values, VkBuffer ba
     }
 
     impl->scheduler.RequestOutsideRenderPassOperationContext();
-    impl->scheduler.Record([src_buffer, dst_buffers = std::move(impl->buffers_to_upload_to),
+    impl->scheduler.Record([src_buffer, dst_buffers = std::move(impl->buffers_to_upload_to), vk_copies = std::move(impl->copies_setup)](vk::CommandBuffer cmdbuf) {
-                            vk_copies = std::move(impl->copies_setup)](vk::CommandBuffer cmdbuf) {
         size_t size = dst_buffers.size();
         for (size_t i = 0; i < size; i++) {
             cmdbuf.CopyBuffer(src_buffer, dst_buffers[i].first, vk_copies[i]);

src/video_core/renderer_vulkan/vk_state_tracker.cpp

@@ -235,7 +235,7 @@ void SetupRasterModes(Maxwell3D::DirtyState::Tables &tables) {
 } // Anonymous namespace
 
 void StateTracker::SetupTables(Tegra::Control::ChannelState& channel_state) {
-    auto& tables{channel_state.maxwell_3d->dirty.tables};
+    auto& tables{channel_state.payload->maxwell_3d.dirty.tables};
     SetupDirtyFlags(tables);
     SetupDirtyViewports(tables);
     SetupDirtyScissors(tables);
@@ -258,7 +258,7 @@ void StateTracker::SetupTables(Tegra::Control::ChannelState& channel_state) {
 }
 
 void StateTracker::ChangeChannel(Tegra::Control::ChannelState& channel_state) {
-    flags = &channel_state.maxwell_3d->dirty.flags;
+    flags = &channel_state.payload->maxwell_3d.dirty.flags;
 }
 
 void StateTracker::InvalidateState() {

src/video_core/renderer_vulkan/vk_update_descriptor.h

@@ -1,10 +1,13 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2019 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
 #pragma once
 
 #include <array>
-
+#include <variant>
 #include "video_core/vulkan_common/vulkan_wrapper.h"
 
 namespace Vulkan {
@@ -12,20 +15,15 @@ namespace Vulkan {
 class Device;
 class Scheduler;
 
-struct DescriptorUpdateEntry {
+union DescriptorUpdateEntry {
-    struct Empty {};
-
     DescriptorUpdateEntry() = default;
     DescriptorUpdateEntry(VkDescriptorImageInfo image_) : image{image_} {}
     DescriptorUpdateEntry(VkDescriptorBufferInfo buffer_) : buffer{buffer_} {}
     DescriptorUpdateEntry(VkBufferView texel_buffer_) : texel_buffer{texel_buffer_} {}
-
+    std::monostate empty{};
-    union {
+    VkDescriptorImageInfo image;
-        Empty empty{};
+    VkDescriptorBufferInfo buffer;
-        VkDescriptorImageInfo image;
+    VkBufferView texel_buffer;
-        VkDescriptorBufferInfo buffer;
-        VkBufferView texel_buffer;
-    };
 };
 
 class UpdateDescriptorQueue final {

src/video_core/video_core.cpp

@@ -45,23 +45,23 @@ std::unique_ptr<VideoCore::RendererBase> CreateRenderer(Core::System& system, Co
 
 namespace VideoCore {
 
-std::unique_ptr<Tegra::GPU> CreateGPU(Core::Frontend::EmuWindow& emu_window, Core::System& system) {
+/// @brief Creates an emulated GPU instance using the given system context.
+void CreateGPU(std::optional<Tegra::GPU>& gpu, Core::Frontend::EmuWindow& emu_window, Core::System& system) {
     Settings::UpdateRescalingInfo();
 
     const auto nvdec_value = Settings::values.nvdec_emulation.GetValue();
     const bool use_nvdec = nvdec_value != Settings::NvdecEmulation::Off;
     const bool use_async = Settings::values.use_asynchronous_gpu_emulation.GetValue();
-    auto gpu = std::make_unique<Tegra::GPU>(system, use_async, use_nvdec);
+    gpu.emplace(system, use_async, use_nvdec);
     auto context = emu_window.CreateSharedContext();
     auto scope = context->Acquire();
     try {
         auto renderer = CreateRenderer(system, emu_window, *gpu, std::move(context));
         gpu->BindRenderer(std::move(renderer));
-        return gpu;
     } catch (const std::runtime_error& exception) {
         scope.Cancel();
         LOG_ERROR(HW_GPU, "Failed to initialize GPU: {}", exception.what());
-        return nullptr;
+        gpu.reset();
     }
 }
 

src/video_core/video_core.h

@@ -1,9 +1,13 @@
+// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: 2014 Citra Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
 #pragma once
 
 #include <memory>
+#include <optional>
 
 namespace Core {
 class System;
@@ -20,8 +24,6 @@ class GPU;
 namespace VideoCore {
 
 class RendererBase;
-
+void CreateGPU(std::optional<Tegra::GPU>& gpu, Core::Frontend::EmuWindow& emu_window, Core::System& system);
-/// Creates an emulated GPU instance using the given system context.
-std::unique_ptr<Tegra::GPU> CreateGPU(Core::Frontend::EmuWindow& emu_window, Core::System& system);
 
 } // namespace VideoCore