Make configure dialog non-blocking

Currently, when the configure dialog is open, it prevents inputs to the main
window. This is annoying if you are trying to change your keybinds as you have
to constantly reopen the configure dialog to test your keys.

To make the window non-blocking, we use .show() instead of .exec() on the
configure dialog, and have a slot handle when the confirmation buttons (ok,
cancel, x) are pressed.

However, entering the configuration dialog disables inputs to prevent sending
inputs to the running game while changing inputs. We modify this by preventing
inputs from being sent to the game only when focus is not on the main menu. The
steps are as follows:
- Determine focus in/focus out on main window with QGuiApplication::focusWindowChanged
- Send focus in/focus out signal to ConfigureDialog slot
- ConfigureDialog slot calls EnableConfiguration/DisableConfiguration for ConfigureInput
- ConfigureInput calls EnableConfiguration/DisableConfiguration for ConfigureInputPlayer

Added:
- Signal FocusIn/FocusOut for GMainWindow
- EnableConfiguration()/DisableConfiguration() for ConfigureInputPlayer and
  ConfigureInput to control whether inputs to the game should be blocked.
  ConfigureInput::EnableConfiguration/DisableConfiguration simply forwards the
  call to ConfigureInputPlayer.

.github/ISSUE_TEMPLATE/blank_issue_template.yml

@@ -0,0 +1,10 @@
+name: New Issue (Developers Only)
+description: A blank issue template for developers only. If you are not a developer, do not use this issue template. Your issue WILL BE CLOSED if you do not use the appropriate issue template.
+body:
+  - type: markdown
+    attributes:
+      value: |
+        **If you are not a developer, do not use this issue template. Your issue WILL BE CLOSED if you do not use the appropriate issue template.**
+  - type: textarea
+    attributes:
+      label: "Issue"

.github/ISSUE_TEMPLATE/bug-report-feature-request.md

@@ -1,39 +0,0 @@
----
-name: Bug Report / Feature Request
-about: Tech support does not belong here. You should only file an issue here if you think you have experienced an actual bug with yuzu or you are requesting a feature you believe would make yuzu better.
-title: ''
-labels: ''
-assignees: ''
-
----
-
-<!---
-Please keep in mind yuzu is EXPERIMENTAL SOFTWARE.
-
-Please read the FAQ:
-https://yuzu-emu.org/wiki/faq/
-
-THIS IS NOT A SUPPORT FORUM, FOR SUPPORT GO TO:
-https://community.citra-emu.org/
-
-If the FAQ does not answer your question, please go to:
-https://community.citra-emu.org/
-
-When submitting an issue, please check the following:
-
-- You have read the above.
-- You have provided the version (commit hash) of yuzu you are using.
-- You have provided sufficient detail for the issue to be reproduced.
-- You have provided system specs (if relevant).
-- Please also provide:
-  - For any issues, a log file
-  - For crashes, a backtrace.
-  - For graphical issues, comparison screenshots with real hardware.
-  - For emulation inaccuracies, a test-case (if able).
-
--->
-
-
-
-
-

.github/ISSUE_TEMPLATE/bug_report.yml

@@ -0,0 +1,64 @@
+name: Bug Report
+description: File a bug report
+body:
+  - type: markdown
+    attributes:
+      value: Tech support does not belong here. You should only file an issue here if you think you have experienced an actual bug with yuzu.
+  - type: checkboxes
+    attributes:
+      label: Is there an existing issue for this?
+      description: Please search to see if an issue already exists for the bug you encountered.
+      options:
+      - label: I have searched the existing issues
+        required: true
+  - type: input
+    attributes:
+      label: Affected Build(s)
+      description: List the affected build(s) that this issue applies to.
+      placeholder: Mainline 1234 / Early Access 1234
+    validations:
+      required: true
+  - type: textarea
+    id: issue-desc
+    attributes:
+      label: Description of Issue
+      description: A brief description of the issue encountered along with any images and/or videos.
+    validations:
+      required: true
+  - type: textarea
+    id: expected-behavior
+    attributes:
+      label: Expected Behavior
+      description: A brief description of how it is expected to work along with any images and/or videos.
+    validations:
+      required: true
+  - type: textarea
+    id: reproduction-steps
+    attributes:
+      label: Reproduction Steps
+      description: A brief explanation of how to reproduce this issue. If possible, provide a save file to aid in reproducing the issue.
+    validations:
+      required: true
+  - type: textarea
+    id: log
+    attributes:
+      label: Log File
+      description: A log file will help our developers to better diagnose and fix the issue.
+    validations:
+      required: true
+  - type: textarea
+    id: system-config
+    attributes:
+      label: System Configuration
+      placeholder: |
+        CPU: Intel i5-10400 / AMD Ryzen 5 3600
+        GPU/Driver: NVIDIA GeForce GTX 1060 (Driver 512.95)
+        RAM: 16GB DDR4-3200
+        OS: Windows 11 22H2 (Build 22621.819)
+      value: |
+        CPU:
+        GPU/Driver:
+        RAM:
+        OS:
+    validations:
+      required: true

.github/ISSUE_TEMPLATE/feature_request.yml

@@ -0,0 +1,28 @@
+name: Feature Request
+description: File a feature request
+labels: "request"
+body:
+  - type: markdown
+    attributes:
+      value: Tech support does not belong here. You should only file an issue here if you are requesting a feature you believe would make yuzu better.
+  - type: checkboxes
+    attributes:
+      label: Is there an existing issue for this?
+      description: Please search to see if an issue already exists for the feature you are requesting.
+      options:
+      - label: I have searched the existing issues
+        required: true
+  - type: textarea
+    id: what-feature
+    attributes:
+      label: What feature are you suggesting?
+      description: A brief description of the requested feature.
+    validations:
+      required: true
+  - type: textarea
+    id: why-feature
+    attributes:
+      label: Why would this feature be useful?
+      description: A brief description of why this feature would make yuzu better.
+    validations:
+      required: true

.reuse/dep5

@@ -132,10 +132,6 @@ Files: vcpkg.json
 Copyright: 2022 yuzu Emulator Project
 License: GPL-3.0-or-later
 
-Files: .github/ISSUE_TEMPLATE/config.yml
-Copyright: 2020 tgsm <doodrabbit@hotmail.com>
-License: GPL-2.0-or-later
-
-Files: .github/ISSUE_TEMPLATE/bug-report-feature-request.md
-Copyright: 2016 MerryMage
+Files: .github/ISSUE_TEMPLATE/*
+Copyright: 2022 yuzu Emulator Project
 License: GPL-2.0-or-later

externals/CMakeLists.txt

@@ -92,10 +92,14 @@ endif()
 add_subdirectory(sirit)
 
 if (ENABLE_WEB_SERVICE)
-    find_package(OpenSSL 1.1)
-    if (OPENSSL_FOUND)
-        set(OPENSSL_LIBRARIES OpenSSL::SSL OpenSSL::Crypto)
-    else()
+    if (NOT WIN32)
+        find_package(OpenSSL 1.1)
+        if (OPENSSL_FOUND)
+            set(OPENSSL_LIBRARIES OpenSSL::SSL OpenSSL::Crypto)
+        endif()
+    endif()
+
+    if (WIN32 OR NOT OPENSSL_FOUND)
         # LibreSSL
         set(LIBRESSL_SKIP_INSTALL ON CACHE BOOL "")
         set(OPENSSLDIR "/etc/ssl/")

externals/microprofile/microprofileui.h

@@ -845,8 +845,6 @@ inline void MicroProfileDrawDetailedBars(uint32_t nWidth, uint32_t nHeight, int
     MicroProfile& S = *MicroProfileGet();
     MP_DEBUG_DUMP_RANGE();
     int nY = nBaseY - UI.nOffsetY;
-    int64_t nNumBoxes = 0;
-    int64_t nNumLines = 0;
 
     uint32_t nFrameNext = (S.nFrameCurrent+1) % MICROPROFILE_MAX_FRAME_HISTORY;
     MicroProfileFrameState* pFrameCurrent = &S.Frames[S.nFrameCurrent];
@@ -1149,7 +1147,6 @@ inline void MicroProfileDrawDetailedBars(uint32_t nWidth, uint32_t nHeight, int
                                 }
                             }
 #endif
-                            ++nNumBoxes;
                         }
                         else
                         {
@@ -1165,7 +1162,6 @@ inline void MicroProfileDrawDetailedBars(uint32_t nWidth, uint32_t nHeight, int
                                 }
                                 nLinesDrawn[nStackPos] = nLineX;
                                 MicroProfileDrawLineVertical(nLineX, fYStart + 0.5f, fYEnd + 0.5f, nColor|UI.nOpacityForeground);
-                                ++nNumLines;
                             }
                         }
                         nStackPos--;

src/audio_core/renderer/command/command_buffer.cpp

@@ -460,21 +460,23 @@ void CommandBuffer::GenerateDeviceSinkCommand(const s32 node_id, const s16 buffe
 
     cmd.session_id = session_id;
 
+    cmd.input_count = parameter.input_count;
+    s16 max_input{0};
+    for (u32 i = 0; i < parameter.input_count; i++) {
+        cmd.inputs[i] = buffer_offset + parameter.inputs[i];
+        max_input = std::max(max_input, cmd.inputs[i]);
+    }
+
     if (state.upsampler_info != nullptr) {
         const auto size_{state.upsampler_info->sample_count * parameter.input_count};
         const auto size_bytes{size_ * sizeof(s32)};
         const auto addr{memory_pool->Translate(state.upsampler_info->samples_pos, size_bytes)};
         cmd.sample_buffer = {reinterpret_cast<s32*>(addr),
-                             parameter.input_count * state.upsampler_info->sample_count};
+                             (max_input + 1) * state.upsampler_info->sample_count};
     } else {
         cmd.sample_buffer = samples_buffer;
     }
 
-    cmd.input_count = parameter.input_count;
-    for (u32 i = 0; i < parameter.input_count; i++) {
-        cmd.inputs[i] = buffer_offset + parameter.inputs[i];
-    }
-
     GenerateEnd<DeviceSinkCommand>(cmd);
 }
 

src/audio_core/renderer/performance/performance_manager.cpp

@@ -26,6 +26,7 @@ void PerformanceManager::CreateImpl(const size_t version) {
         impl = std::make_unique<
             PerformanceManagerImpl<PerformanceVersion::Version1, PerformanceFrameHeaderVersion1,
                                    PerformanceEntryVersion1, PerformanceDetailVersion1>>();
+        break;
     }
 }
 

src/common/CMakeLists.txt

@@ -34,6 +34,8 @@ add_library(common STATIC
     bit_util.h
     cityhash.cpp
     cityhash.h
+    cache_management.cpp
+    cache_management.h
     common_funcs.h
     common_types.h
     concepts.h

src/common/atomic_helpers.h

@@ -156,6 +156,7 @@ AE_FORCEINLINE void compiler_fence(memory_order order) AE_NO_TSAN {
         break;
     default:
         assert(false);
+        break;
     }
 }
 

src/common/cache_management.cpp

@@ -0,0 +1,60 @@
+// SPDX-FileCopyrightText: Copyright 2022 yuzu Emulator Project
+// SPDX-License-Identifier: GPL-2.0-or-later
+
+#include <cstring>
+
+#include "alignment.h"
+#include "cache_management.h"
+#include "common_types.h"
+
+namespace Common {
+
+#if defined(ARCHITECTURE_x86_64)
+
+// Most cache operations are no-ops on x86
+
+void DataCacheLineCleanByVAToPoU(void* start, size_t size) {}
+void DataCacheLineCleanAndInvalidateByVAToPoC(void* start, size_t size) {}
+void DataCacheLineCleanByVAToPoC(void* start, size_t size) {}
+void DataCacheZeroByVA(void* start, size_t size) {
+    std::memset(start, 0, size);
+}
+
+#elif defined(ARCHITECTURE_arm64)
+
+// BS/DminLine is log2(cache size in words), we want size in bytes
+#define EXTRACT_DMINLINE(ctr_el0) (1 << ((((ctr_el0) >> 16) & 0xf) + 2))
+#define EXTRACT_BS(dczid_el0) (1 << (((dczid_el0)&0xf) + 2))
+
+#define DEFINE_DC_OP(op_name, function_name)                                                       \
+    void function_name(void* start, size_t size) {                                                 \
+        size_t ctr_el0;                                                                            \
+        asm volatile("mrs %[ctr_el0], ctr_el0\n\t" : [ctr_el0] "=r"(ctr_el0));                     \
+        size_t cacheline_size = EXTRACT_DMINLINE(ctr_el0);                                         \
+        uintptr_t va_start = reinterpret_cast<uintptr_t>(start);                                   \
+        uintptr_t va_end = va_start + size;                                                        \
+        for (uintptr_t va = va_start; va < va_end; va += cacheline_size) {                         \
+            asm volatile("dc " #op_name ", %[va]\n\t" : : [va] "r"(va) : "memory");                \
+        }                                                                                          \
+    }
+
+#define DEFINE_DC_OP_DCZID(op_name, function_name)                                                 \
+    void function_name(void* start, size_t size) {                                                 \
+        size_t dczid_el0;                                                                          \
+        asm volatile("mrs %[dczid_el0], dczid_el0\n\t" : [dczid_el0] "=r"(dczid_el0));             \
+        size_t cacheline_size = EXTRACT_BS(dczid_el0);                                             \
+        uintptr_t va_start = reinterpret_cast<uintptr_t>(start);                                   \
+        uintptr_t va_end = va_start + size;                                                        \
+        for (uintptr_t va = va_start; va < va_end; va += cacheline_size) {                         \
+            asm volatile("dc " #op_name ", %[va]\n\t" : : [va] "r"(va) : "memory");                \
+        }                                                                                          \
+    }
+
+DEFINE_DC_OP(cvau, DataCacheLineCleanByVAToPoU);
+DEFINE_DC_OP(civac, DataCacheLineCleanAndInvalidateByVAToPoC);
+DEFINE_DC_OP(cvac, DataCacheLineCleanByVAToPoC);
+DEFINE_DC_OP_DCZID(zva, DataCacheZeroByVA);
+
+#endif
+
+} // namespace Common

src/common/cache_management.h

@@ -0,0 +1,27 @@
+// SPDX-FileCopyrightText: Copyright 2022 yuzu Emulator Project
+// SPDX-License-Identifier: GPL-2.0-or-later
+
+#pragma once
+
+#include "stdlib.h"
+
+namespace Common {
+
+// Data cache instructions enabled at EL0 by SCTLR_EL1.UCI.
+// VA = virtual address
+// PoC = point of coherency
+// PoU = point of unification
+
+// dc cvau
+void DataCacheLineCleanByVAToPoU(void* start, size_t size);
+
+// dc civac
+void DataCacheLineCleanAndInvalidateByVAToPoC(void* start, size_t size);
+
+// dc cvac
+void DataCacheLineCleanByVAToPoC(void* start, size_t size);
+
+// dc zva
+void DataCacheZeroByVA(void* start, size_t size);
+
+} // namespace Common

src/core/CMakeLists.txt

@@ -530,6 +530,11 @@ add_library(core STATIC
     hle/service/ncm/ncm.h
     hle/service/nfc/nfc.cpp
     hle/service/nfc/nfc.h
+    hle/service/nfc/nfc_device.cpp
+    hle/service/nfc/nfc_device.h
+    hle/service/nfc/nfc_result.h
+    hle/service/nfc/nfc_user.cpp
+    hle/service/nfc/nfc_user.h
     hle/service/nfp/amiibo_crypto.cpp
     hle/service/nfp/amiibo_crypto.h
     hle/service/nfp/nfp.cpp

src/core/arm/dynarmic/arm_dynarmic_64.cpp

@@ -348,7 +348,6 @@ std::shared_ptr<Dynarmic::A64::Jit> ARM_Dynarmic_64::MakeJit(Common::PageTable*
         if (Settings::values.cpu_accuracy.GetValue() == Settings::CPUAccuracy::Auto) {
             config.unsafe_optimizations = true;
             config.optimizations |= Dynarmic::OptimizationFlag::Unsafe_UnfuseFMA;
-            config.optimizations |= Dynarmic::OptimizationFlag::Unsafe_InaccurateNaN;
             config.fastmem_address_space_bits = 64;
             config.optimizations |= Dynarmic::OptimizationFlag::Unsafe_IgnoreGlobalMonitor;
         }

src/core/debugger/debugger.cpp

@@ -27,12 +27,21 @@ static void AsyncReceiveInto(Readable& r, Buffer& buffer, Callback&& c) {
                 const u8* buffer_start = reinterpret_cast<const u8*>(&buffer);
                 std::span<const u8> received_data{buffer_start, buffer_start + bytes_read};
                 c(received_data);
+                AsyncReceiveInto(r, buffer, c);
             }
-
-            AsyncReceiveInto(r, buffer, c);
         });
 }
 
+template <typename Callback>
+static void AsyncAccept(boost::asio::ip::tcp::acceptor& acceptor, Callback&& c) {
+    acceptor.async_accept([&, c](const boost::system::error_code& error, auto&& peer_socket) {
+        if (!error.failed()) {
+            c(peer_socket);
+            AsyncAccept(acceptor, c);
+        }
+    });
+}
+
 template <typename Readable, typename Buffer>
 static std::span<const u8> ReceiveInto(Readable& r, Buffer& buffer) {
     static_assert(std::is_trivial_v<Buffer>);
@@ -59,9 +68,7 @@ namespace Core {
 
 class DebuggerImpl : public DebuggerBackend {
 public:
-    explicit DebuggerImpl(Core::System& system_, u16 port)
-        : system{system_}, signal_pipe{io_context}, client_socket{io_context} {
-        frontend = std::make_unique<GDBStub>(*this, system);
+    explicit DebuggerImpl(Core::System& system_, u16 port) : system{system_} {
         InitializeServer(port);
     }
 
@@ -70,39 +77,42 @@ public:
     }
 
     bool SignalDebugger(SignalInfo signal_info) {
-        {
-            std::scoped_lock lk{connection_lock};
+        std::scoped_lock lk{connection_lock};
 
-            if (stopped) {
-                // Do not notify the debugger about another event.
-                // It should be ignored.
-                return false;
-            }
-
-            // Set up the state.
-            stopped = true;
-            info = signal_info;
+        if (stopped || !state) {
+            // Do not notify the debugger about another event.
+            // It should be ignored.
+            return false;
         }
 
+        // Set up the state.
+        stopped = true;
+        state->info = signal_info;
+
         // Write a single byte into the pipe to wake up the debug interface.
-        boost::asio::write(signal_pipe, boost::asio::buffer(&stopped, sizeof(stopped)));
+        boost::asio::write(state->signal_pipe, boost::asio::buffer(&stopped, sizeof(stopped)));
+
         return true;
     }
 
+    // These functions are callbacks from the frontend, and the lock will be held.
+    // There is no need to relock it.
+
     std::span<const u8> ReadFromClient() override {
-        return ReceiveInto(client_socket, client_data);
+        return ReceiveInto(state->client_socket, state->client_data);
     }
 
     void WriteToClient(std::span<const u8> data) override {
-        boost::asio::write(client_socket, boost::asio::buffer(data.data(), data.size_bytes()));
+        boost::asio::write(state->client_socket,
+                           boost::asio::buffer(data.data(), data.size_bytes()));
     }
 
     void SetActiveThread(Kernel::KThread* thread) override {
-        active_thread = thread;
+        state->active_thread = thread;
     }
 
     Kernel::KThread* GetActiveThread() override {
-        return active_thread;
+        return state->active_thread;
     }
 
 private:
@@ -113,65 +123,78 @@ private:
 
         // Run the connection thread.
         connection_thread = std::jthread([&, port](std::stop_token stop_token) {
+            Common::SetCurrentThreadName("Debugger");
+
             try {
                 // Initialize the listening socket and accept a new client.
                 tcp::endpoint endpoint{boost::asio::ip::address_v4::any(), port};
                 tcp::acceptor acceptor{io_context, endpoint};
 
-                acceptor.async_accept(client_socket, [](const auto&) {});
-                io_context.run_one();
-                io_context.restart();
+                AsyncAccept(acceptor, [&](auto&& peer) { AcceptConnection(std::move(peer)); });
 
-                if (stop_token.stop_requested()) {
-                    return;
+                while (!stop_token.stop_requested() && io_context.run()) {
                 }
-
-                ThreadLoop(stop_token);
             } catch (const std::exception& ex) {
                 LOG_CRITICAL(Debug_GDBStub, "Stopping server: {}", ex.what());
             }
         });
     }
 
+    void AcceptConnection(boost::asio::ip::tcp::socket&& peer) {
+        LOG_INFO(Debug_GDBStub, "Accepting new peer connection");
+
+        std::scoped_lock lk{connection_lock};
+
+        // Ensure everything is stopped.
+        PauseEmulation();
+
+        // Set up the new frontend.
+        frontend = std::make_unique<GDBStub>(*this, system);
+
+        // Set the new state. This will tear down any existing state.
+        state = ConnectionState{
+            .client_socket{std::move(peer)},
+            .signal_pipe{io_context},
+            .info{},
+            .active_thread{},
+            .client_data{},
+            .pipe_data{},
+        };
+
+        // Set up the client signals for new data.
+        AsyncReceiveInto(state->signal_pipe, state->pipe_data, [&](auto d) { PipeData(d); });
+        AsyncReceiveInto(state->client_socket, state->client_data, [&](auto d) { ClientData(d); });
+
+        // Set the active thread.
+        UpdateActiveThread();
+
+        // Set up the frontend.
+        frontend->Connected();
+    }
+
     void ShutdownServer() {
         connection_thread.request_stop();
         io_context.stop();
         connection_thread.join();
     }
 
-    void ThreadLoop(std::stop_token stop_token) {
-        Common::SetCurrentThreadName("Debugger");
-
-        // Set up the client signals for new data.
-        AsyncReceiveInto(signal_pipe, pipe_data, [&](auto d) { PipeData(d); });
-        AsyncReceiveInto(client_socket, client_data, [&](auto d) { ClientData(d); });
-
-        // Set the active thread.
-        UpdateActiveThread();
-
-        // Set up the frontend.
-        frontend->Connected();
-
-        // Main event loop.
-        while (!stop_token.stop_requested() && io_context.run()) {
-        }
-    }
-
     void PipeData(std::span<const u8> data) {
-        switch (info.type) {
+        std::scoped_lock lk{connection_lock};
+
+        switch (state->info.type) {
         case SignalType::Stopped:
         case SignalType::Watchpoint:
             // Stop emulation.
             PauseEmulation();
 
             // Notify the client.
-            active_thread = info.thread;
+            state->active_thread = state->info.thread;
             UpdateActiveThread();
 
-            if (info.type == SignalType::Watchpoint) {
-                frontend->Watchpoint(active_thread, *info.watchpoint);
+            if (state->info.type == SignalType::Watchpoint) {
+                frontend->Watchpoint(state->active_thread, *state->info.watchpoint);
             } else {
-                frontend->Stopped(active_thread);
+                frontend->Stopped(state->active_thread);
             }
 
             break;
@@ -179,8 +202,8 @@ private:
             frontend->ShuttingDown();
 
             // Wait for emulation to shut down gracefully now.
-            signal_pipe.close();
-            client_socket.shutdown(boost::asio::socket_base::shutdown_both);
+            state->signal_pipe.close();
+            state->client_socket.shutdown(boost::asio::socket_base::shutdown_both);
             LOG_INFO(Debug_GDBStub, "Shut down server");
 
             break;
@@ -188,17 +211,16 @@ private:
     }
 
     void ClientData(std::span<const u8> data) {
+        std::scoped_lock lk{connection_lock};
+
         const auto actions{frontend->ClientData(data)};
         for (const auto action : actions) {
             switch (action) {
             case DebuggerAction::Interrupt: {
-                {
-                    std::scoped_lock lk{connection_lock};
-                    stopped = true;
-                }
+                stopped = true;
                 PauseEmulation();
                 UpdateActiveThread();
-                frontend->Stopped(active_thread);
+                frontend->Stopped(state->active_thread);
                 break;
             }
             case DebuggerAction::Continue:
@@ -206,15 +228,15 @@ private:
                 break;
             case DebuggerAction::StepThreadUnlocked:
                 MarkResumed([&] {
-                    active_thread->SetStepState(Kernel::StepState::StepPending);
-                    active_thread->Resume(Kernel::SuspendType::Debug);
-                    ResumeEmulation(active_thread);
+                    state->active_thread->SetStepState(Kernel::StepState::StepPending);
+                    state->active_thread->Resume(Kernel::SuspendType::Debug);
+                    ResumeEmulation(state->active_thread);
                 });
                 break;
             case DebuggerAction::StepThreadLocked: {
                 MarkResumed([&] {
-                    active_thread->SetStepState(Kernel::StepState::StepPending);
-                    active_thread->Resume(Kernel::SuspendType::Debug);
+                    state->active_thread->SetStepState(Kernel::StepState::StepPending);
+                    state->active_thread->Resume(Kernel::SuspendType::Debug);
                 });
                 break;
             }
@@ -254,15 +276,14 @@ private:
     template <typename Callback>
     void MarkResumed(Callback&& cb) {
         Kernel::KScopedSchedulerLock sl{system.Kernel()};
-        std::scoped_lock cl{connection_lock};
         stopped = false;
         cb();
     }
 
     void UpdateActiveThread() {
         const auto& threads{ThreadList()};
-        if (std::find(threads.begin(), threads.end(), active_thread) == threads.end()) {
-            active_thread = threads[0];
+        if (std::find(threads.begin(), threads.end(), state->active_thread) == threads.end()) {
+            state->active_thread = threads[0];
         }
     }
 
@@ -274,18 +295,22 @@ private:
     System& system;
     std::unique_ptr<DebuggerFrontend> frontend;
 
+    boost::asio::io_context io_context;
     std::jthread connection_thread;
     std::mutex connection_lock;
-    boost::asio::io_context io_context;
-    boost::process::async_pipe signal_pipe;
-    boost::asio::ip::tcp::socket client_socket;
 
-    SignalInfo info;
-    Kernel::KThread* active_thread;
-    bool pipe_data;
-    bool stopped;
+    struct ConnectionState {
+        boost::asio::ip::tcp::socket client_socket;
+        boost::process::async_pipe signal_pipe;
 
-    std::array<u8, 4096> client_data;
+        SignalInfo info;
+        Kernel::KThread* active_thread;
+        std::array<u8, 4096> client_data;
+        bool pipe_data;
+    };
+
+    std::optional<ConnectionState> state{};
+    bool stopped{};
 };
 
 Debugger::Debugger(Core::System& system, u16 port) {

src/core/debugger/gdbstub.cpp

@@ -606,6 +606,8 @@ void GDBStub::HandleQuery(std::string_view command) {
     } else if (command.starts_with("StartNoAckMode")) {
         no_ack = true;
         SendReply(GDB_STUB_REPLY_OK);
+    } else if (command.starts_with("Rcmd,")) {
+        HandleRcmd(Common::HexStringToVector(command.substr(5), false));
     } else {
         SendReply(GDB_STUB_REPLY_EMPTY);
     }
@@ -645,6 +647,155 @@ void GDBStub::HandleVCont(std::string_view command, std::vector<DebuggerAction>&
     }
 }
 
+constexpr std::array<std::pair<const char*, Kernel::Svc::MemoryState>, 22> MemoryStateNames{{
+    {"----- Free -----", Kernel::Svc::MemoryState::Free},
+    {"Io              ", Kernel::Svc::MemoryState::Io},
+    {"Static          ", Kernel::Svc::MemoryState::Static},
+    {"Code            ", Kernel::Svc::MemoryState::Code},
+    {"CodeData        ", Kernel::Svc::MemoryState::CodeData},
+    {"Normal          ", Kernel::Svc::MemoryState::Normal},
+    {"Shared          ", Kernel::Svc::MemoryState::Shared},
+    {"AliasCode       ", Kernel::Svc::MemoryState::AliasCode},
+    {"AliasCodeData   ", Kernel::Svc::MemoryState::AliasCodeData},
+    {"Ipc             ", Kernel::Svc::MemoryState::Ipc},
+    {"Stack           ", Kernel::Svc::MemoryState::Stack},
+    {"ThreadLocal     ", Kernel::Svc::MemoryState::ThreadLocal},
+    {"Transfered      ", Kernel::Svc::MemoryState::Transfered},
+    {"SharedTransfered", Kernel::Svc::MemoryState::SharedTransfered},
+    {"SharedCode      ", Kernel::Svc::MemoryState::SharedCode},
+    {"Inaccessible    ", Kernel::Svc::MemoryState::Inaccessible},
+    {"NonSecureIpc    ", Kernel::Svc::MemoryState::NonSecureIpc},
+    {"NonDeviceIpc    ", Kernel::Svc::MemoryState::NonDeviceIpc},
+    {"Kernel          ", Kernel::Svc::MemoryState::Kernel},
+    {"GeneratedCode   ", Kernel::Svc::MemoryState::GeneratedCode},
+    {"CodeOut         ", Kernel::Svc::MemoryState::CodeOut},
+    {"Coverage        ", Kernel::Svc::MemoryState::Coverage},
+}};
+
+static constexpr const char* GetMemoryStateName(Kernel::Svc::MemoryState state) {
+    for (size_t i = 0; i < MemoryStateNames.size(); i++) {
+        if (std::get<1>(MemoryStateNames[i]) == state) {
+            return std::get<0>(MemoryStateNames[i]);
+        }
+    }
+    return "Unknown         ";
+}
+
+static constexpr const char* GetMemoryPermissionString(const Kernel::Svc::MemoryInfo& info) {
+    if (info.state == Kernel::Svc::MemoryState::Free) {
+        return "   ";
+    }
+
+    switch (info.permission) {
+    case Kernel::Svc::MemoryPermission::ReadExecute:
+        return "r-x";
+    case Kernel::Svc::MemoryPermission::Read:
+        return "r--";
+    case Kernel::Svc::MemoryPermission::ReadWrite:
+        return "rw-";
+    default:
+        return "---";
+    }
+}
+
+static VAddr GetModuleEnd(Kernel::KPageTable& page_table, VAddr base) {
+    Kernel::Svc::MemoryInfo mem_info;
+    VAddr cur_addr{base};
+
+    // Expect: r-x Code (.text)
+    mem_info = page_table.QueryInfo(cur_addr).GetSvcMemoryInfo();
+    cur_addr = mem_info.base_address + mem_info.size;
+    if (mem_info.state != Kernel::Svc::MemoryState::Code ||
+        mem_info.permission != Kernel::Svc::MemoryPermission::ReadExecute) {
+        return cur_addr - 1;
+    }
+
+    // Expect: r-- Code (.rodata)
+    mem_info = page_table.QueryInfo(cur_addr).GetSvcMemoryInfo();
+    cur_addr = mem_info.base_address + mem_info.size;
+    if (mem_info.state != Kernel::Svc::MemoryState::Code ||
+        mem_info.permission != Kernel::Svc::MemoryPermission::Read) {
+        return cur_addr - 1;
+    }
+
+    // Expect: rw- CodeData (.data)
+    mem_info = page_table.QueryInfo(cur_addr).GetSvcMemoryInfo();
+    cur_addr = mem_info.base_address + mem_info.size;
+    return cur_addr - 1;
+}
+
+void GDBStub::HandleRcmd(const std::vector<u8>& command) {
+    std::string_view command_str{reinterpret_cast<const char*>(&command[0]), command.size()};
+    std::string reply;
+
+    auto* process = system.CurrentProcess();
+    auto& page_table = process->PageTable();
+
+    if (command_str == "get info") {
+        Loader::AppLoader::Modules modules;
+        system.GetAppLoader().ReadNSOModules(modules);
+
+        reply = fmt::format("Process:     {:#x} ({})\n"
+                            "Program Id:  {:#018x}\n",
+                            process->GetProcessID(), process->GetName(), process->GetProgramID());
+        reply +=
+            fmt::format("Layout:\n"
+                        "  Alias: {:#012x} - {:#012x}\n"
+                        "  Heap:  {:#012x} - {:#012x}\n"
+                        "  Aslr:  {:#012x} - {:#012x}\n"
+                        "  Stack: {:#012x} - {:#012x}\n"
+                        "Modules:\n",
+                        page_table.GetAliasRegionStart(), page_table.GetAliasRegionEnd(),
+                        page_table.GetHeapRegionStart(), page_table.GetHeapRegionEnd(),
+                        page_table.GetAliasCodeRegionStart(), page_table.GetAliasCodeRegionEnd(),
+                        page_table.GetStackRegionStart(), page_table.GetStackRegionEnd());
+
+        for (const auto& [vaddr, name] : modules) {
+            reply += fmt::format("  {:#012x} - {:#012x} {}\n", vaddr,
+                                 GetModuleEnd(page_table, vaddr), name);
+        }
+    } else if (command_str == "get mappings") {
+        reply = "Mappings:\n";
+        VAddr cur_addr = 0;
+
+        while (true) {
+            using MemoryAttribute = Kernel::Svc::MemoryAttribute;
+
+            auto mem_info = page_table.QueryInfo(cur_addr).GetSvcMemoryInfo();
+
+            if (mem_info.state != Kernel::Svc::MemoryState::Inaccessible ||
+                mem_info.base_address + mem_info.size - 1 != std::numeric_limits<u64>::max()) {
+                const char* state = GetMemoryStateName(mem_info.state);
+                const char* perm = GetMemoryPermissionString(mem_info);
+
+                const char l = True(mem_info.attribute & MemoryAttribute::Locked) ? 'L' : '-';
+                const char i = True(mem_info.attribute & MemoryAttribute::IpcLocked) ? 'I' : '-';
+                const char d = True(mem_info.attribute & MemoryAttribute::DeviceShared) ? 'D' : '-';
+                const char u = True(mem_info.attribute & MemoryAttribute::Uncached) ? 'U' : '-';
+
+                reply +=
+                    fmt::format("  {:#012x} - {:#012x} {} {} {}{}{}{} [{}, {}]\n",
+                                mem_info.base_address, mem_info.base_address + mem_info.size - 1,
+                                perm, state, l, i, d, u, mem_info.ipc_count, mem_info.device_count);
+            }
+
+            const uintptr_t next_address = mem_info.base_address + mem_info.size;
+            if (next_address <= cur_addr) {
+                break;
+            }
+
+            cur_addr = next_address;
+        }
+    } else if (command_str == "help") {
+        reply = "Commands:\n  get info\n  get mappings\n";
+    } else {
+        reply = "Unknown command.\nCommands:\n  get info\n  get mappings\n";
+    }
+
+    std::span<const u8> reply_span{reinterpret_cast<u8*>(&reply.front()), reply.size()};
+    SendReply(Common::HexToString(reply_span, false));
+}
+
 Kernel::KThread* GDBStub::GetThreadByID(u64 thread_id) {
     const auto& threads{system.GlobalSchedulerContext().GetThreadList()};
     for (auto* thread : threads) {

src/core/debugger/gdbstub.h

@@ -32,6 +32,7 @@ private:
     void ExecuteCommand(std::string_view packet, std::vector<DebuggerAction>& actions);
     void HandleVCont(std::string_view command, std::vector<DebuggerAction>& actions);
     void HandleQuery(std::string_view command);
+    void HandleRcmd(const std::vector<u8>& command);
     void HandleBreakpointInsert(std::string_view command);
     void HandleBreakpointRemove(std::string_view command);
     std::vector<char>::const_iterator CommandEnd() const;

src/core/hid/emulated_console.cpp

@@ -19,27 +19,26 @@ void EmulatedConsole::ReloadFromSettings() {
 }
 
 void EmulatedConsole::SetTouchParams() {
-    // TODO(german77): Support any number of fingers
     std::size_t index = 0;
 
-    // Hardcode mouse, touchscreen and cemuhook parameters
+    // We can't use mouse as touch if native mouse is enabled
     if (!Settings::values.mouse_enabled) {
-        // We can't use mouse as touch if native mouse is enabled
         touch_params[index++] = Common::ParamPackage{"engine:mouse,axis_x:10,axis_y:11,button:0"};
     }
 
     touch_params[index++] =
-        Common::ParamPackage{"engine:touch,axis_x:0,axis_y:1,button:0,touch_id:0"};
+        Common::ParamPackage{"engine:cemuhookudp,axis_x:17,axis_y:18,button:65536"};
     touch_params[index++] =
-        Common::ParamPackage{"engine:touch,axis_x:2,axis_y:3,button:1,touch_id:1"};
-    touch_params[index++] =
-        Common::ParamPackage{"engine:touch,axis_x:4,axis_y:5,button:2,touch_id:2"};
-    touch_params[index++] =
-        Common::ParamPackage{"engine:touch,axis_x:6,axis_y:7,button:3,touch_id:3"};
-    touch_params[index++] =
-        Common::ParamPackage{"engine:cemuhookudp,axis_x:17,axis_y:18,button:65536,touch_id:0"};
-    touch_params[index++] =
-        Common::ParamPackage{"engine:cemuhookudp,axis_x:19,axis_y:20,button:131072,touch_id:1"};
+        Common::ParamPackage{"engine:cemuhookudp,axis_x:19,axis_y:20,button:131072"};
+
+    for (int i = 0; i < static_cast<int>(MaxActiveTouchInputs); i++) {
+        Common::ParamPackage touchscreen_param{};
+        touchscreen_param.Set("engine", "touch");
+        touchscreen_param.Set("axis_x", i * 2);
+        touchscreen_param.Set("axis_y", (i * 2) + 1);
+        touchscreen_param.Set("button", i);
+        touch_params[index++] = touchscreen_param;
+    }
 
     const auto button_index =
         static_cast<u64>(Settings::values.touch_from_button_map_index.GetValue());
@@ -47,7 +46,7 @@ void EmulatedConsole::SetTouchParams() {
 
     // Map the rest of the fingers from touch from button configuration
     for (const auto& config_entry : touch_buttons) {
-        if (index >= touch_params.size()) {
+        if (index >= MaxTouchDevices) {
             continue;
         }
         Common::ParamPackage params{config_entry};
@@ -60,7 +59,6 @@ void EmulatedConsole::SetTouchParams() {
         touch_button_params.Set("button", params.Serialize());
         touch_button_params.Set("x", x);
         touch_button_params.Set("y", y);
-        touch_button_params.Set("touch_id", static_cast<int>(index));
         touch_params[index] = touch_button_params;
         index++;
     }
@@ -178,12 +176,38 @@ void EmulatedConsole::SetMotion(const Common::Input::CallbackStatus& callback) {
 }
 
 void EmulatedConsole::SetTouch(const Common::Input::CallbackStatus& callback, std::size_t index) {
-    if (index >= console.touch_values.size()) {
+    if (index >= MaxTouchDevices) {
         return;
     }
     std::unique_lock lock{mutex};
 
-    console.touch_values[index] = TransformToTouch(callback);
+    const auto touch_input = TransformToTouch(callback);
+    auto touch_index = GetIndexFromFingerId(index);
+    bool is_new_input = false;
+
+    if (!touch_index.has_value() && touch_input.pressed.value) {
+        touch_index = GetNextFreeIndex();
+        is_new_input = true;
+    }
+
+    // No free entries or invalid state. Ignore input
+    if (!touch_index.has_value()) {
+        return;
+    }
+
+    auto& touch_value = console.touch_values[touch_index.value()];
+
+    if (is_new_input) {
+        touch_value.pressed.value = true;
+        touch_value.id = static_cast<u32>(index);
+    }
+
+    touch_value.x = touch_input.x;
+    touch_value.y = touch_input.y;
+
+    if (!touch_input.pressed.value) {
+        touch_value.pressed.value = false;
+    }
 
     if (is_configuring) {
         lock.unlock();
@@ -191,11 +215,15 @@ void EmulatedConsole::SetTouch(const Common::Input::CallbackStatus& callback, st
         return;
     }
 
-    // TODO(german77): Remap touch id in sequential order
-    console.touch_state[index] = {
-        .position = {console.touch_values[index].x.value, console.touch_values[index].y.value},
-        .id = static_cast<u32>(console.touch_values[index].id),
-        .pressed = console.touch_values[index].pressed.value,
+    // Touch outside allowed range. Ignore input
+    if (touch_index.value() >= MaxActiveTouchInputs) {
+        return;
+    }
+
+    console.touch_state[touch_index.value()] = {
+        .position = {touch_value.x.value, touch_value.y.value},
+        .id = static_cast<u32>(touch_index.value()),
+        .pressed = touch_input.pressed.value,
     };
 
     lock.unlock();
@@ -222,6 +250,28 @@ TouchFingerState EmulatedConsole::GetTouch() const {
     return console.touch_state;
 }
 
+std::optional<std::size_t> EmulatedConsole::GetIndexFromFingerId(std::size_t finger_id) const {
+    for (std::size_t index = 0; index < MaxTouchDevices; ++index) {
+        const auto& finger = console.touch_values[index];
+        if (!finger.pressed.value) {
+            continue;
+        }
+        if (finger.id == static_cast<int>(finger_id)) {
+            return index;
+        }
+    }
+    return std::nullopt;
+}
+
+std::optional<std::size_t> EmulatedConsole::GetNextFreeIndex() const {
+    for (std::size_t index = 0; index < MaxTouchDevices; ++index) {
+        if (!console.touch_values[index].pressed.value) {
+            return index;
+        }
+    }
+    return std::nullopt;
+}
+
 void EmulatedConsole::TriggerOnChange(ConsoleTriggerType type) {
     std::scoped_lock lock{callback_mutex};
     for (const auto& poller_pair : callback_list) {

src/core/hid/emulated_console.h

@@ -7,6 +7,7 @@
 #include <functional>
 #include <memory>
 #include <mutex>
+#include <optional>
 #include <unordered_map>
 
 #include "common/common_funcs.h"
@@ -20,6 +21,8 @@
 #include "core/hid/motion_input.h"
 
 namespace Core::HID {
+static constexpr std::size_t MaxTouchDevices = 32;
+static constexpr std::size_t MaxActiveTouchInputs = 16;
 
 struct ConsoleMotionInfo {
     Common::Input::MotionStatus raw_status{};
@@ -27,13 +30,13 @@ struct ConsoleMotionInfo {
 };
 
 using ConsoleMotionDevices = std::unique_ptr<Common::Input::InputDevice>;
-using TouchDevices = std::array<std::unique_ptr<Common::Input::InputDevice>, 16>;
+using TouchDevices = std::array<std::unique_ptr<Common::Input::InputDevice>, MaxTouchDevices>;
 
 using ConsoleMotionParams = Common::ParamPackage;
-using TouchParams = std::array<Common::ParamPackage, 16>;
+using TouchParams = std::array<Common::ParamPackage, MaxTouchDevices>;
 
 using ConsoleMotionValues = ConsoleMotionInfo;
-using TouchValues = std::array<Common::Input::TouchStatus, 16>;
+using TouchValues = std::array<Common::Input::TouchStatus, MaxTouchDevices>;
 
 struct TouchFinger {
     u64 last_touch{};
@@ -55,7 +58,7 @@ struct ConsoleMotion {
     bool is_at_rest{};
 };
 
-using TouchFingerState = std::array<TouchFinger, 16>;
+using TouchFingerState = std::array<TouchFinger, MaxActiveTouchInputs>;
 
 struct ConsoleStatus {
     // Data from input_common
@@ -166,6 +169,10 @@ private:
      */
     void SetTouch(const Common::Input::CallbackStatus& callback, std::size_t index);
 
+    std::optional<std::size_t> GetIndexFromFingerId(std::size_t finger_id) const;
+
+    std::optional<std::size_t> GetNextFreeIndex() const;
+
     /**
      * Triggers a callback that something has changed on the console status
      * @param type Input type of the event to trigger

src/core/hid/input_converter.cpp

@@ -200,9 +200,6 @@ Common::Input::TouchStatus TransformToTouch(const Common::Input::CallbackStatus&
     x = std::clamp(x, 0.0f, 1.0f);
     y = std::clamp(y, 0.0f, 1.0f);
 
-    // Limit id to maximum number of fingers
-    status.id = std::clamp(status.id, 0, 16);
-
     if (status.pressed.inverted) {
         status.pressed.value = !status.pressed.value;
     }

src/core/hle/kernel/hle_ipc.h

@@ -199,7 +199,7 @@ public:
     ~HLERequestContext();
 
     /// Returns a pointer to the IPC command buffer for this request.
-    u32* CommandBuffer() {
+    [[nodiscard]] u32* CommandBuffer() {
         return cmd_buf.data();
     }
 
@@ -207,7 +207,7 @@ public:
      * Returns the session through which this request was made. This can be used as a map key to
      * access per-client data on services.
      */
-    Kernel::KServerSession* Session() {
+    [[nodiscard]] Kernel::KServerSession* Session() {
         return server_session;
     }
 
@@ -217,61 +217,61 @@ public:
     /// Writes data from this context back to the requesting process/thread.
     Result WriteToOutgoingCommandBuffer(KThread& requesting_thread);
 
-    u32_le GetHipcCommand() const {
+    [[nodiscard]] u32_le GetHipcCommand() const {
         return command;
     }
 
-    u32_le GetTipcCommand() const {
+    [[nodiscard]] u32_le GetTipcCommand() const {
         return static_cast<u32_le>(command_header->type.Value()) -
                static_cast<u32_le>(IPC::CommandType::TIPC_CommandRegion);
     }
 
-    u32_le GetCommand() const {
+    [[nodiscard]] u32_le GetCommand() const {
         return command_header->IsTipc() ? GetTipcCommand() : GetHipcCommand();
     }
 
-    bool IsTipc() const {
+    [[nodiscard]] bool IsTipc() const {
         return command_header->IsTipc();
     }
 
-    IPC::CommandType GetCommandType() const {
+    [[nodiscard]] IPC::CommandType GetCommandType() const {
         return command_header->type;
     }
 
-    u64 GetPID() const {
+    [[nodiscard]] u64 GetPID() const {
         return pid;
     }
 
-    u32 GetDataPayloadOffset() const {
+    [[nodiscard]] u32 GetDataPayloadOffset() const {
         return data_payload_offset;
     }
 
-    const std::vector<IPC::BufferDescriptorX>& BufferDescriptorX() const {
+    [[nodiscard]] const std::vector<IPC::BufferDescriptorX>& BufferDescriptorX() const {
         return buffer_x_desciptors;
     }
 
-    const std::vector<IPC::BufferDescriptorABW>& BufferDescriptorA() const {
+    [[nodiscard]] const std::vector<IPC::BufferDescriptorABW>& BufferDescriptorA() const {
         return buffer_a_desciptors;
     }
 
-    const std::vector<IPC::BufferDescriptorABW>& BufferDescriptorB() const {
+    [[nodiscard]] const std::vector<IPC::BufferDescriptorABW>& BufferDescriptorB() const {
         return buffer_b_desciptors;
     }
 
-    const std::vector<IPC::BufferDescriptorC>& BufferDescriptorC() const {
+    [[nodiscard]] const std::vector<IPC::BufferDescriptorC>& BufferDescriptorC() const {
         return buffer_c_desciptors;
     }
 
-    const IPC::DomainMessageHeader& GetDomainMessageHeader() const {
+    [[nodiscard]] const IPC::DomainMessageHeader& GetDomainMessageHeader() const {
         return domain_message_header.value();
     }
 
-    bool HasDomainMessageHeader() const {
+    [[nodiscard]] bool HasDomainMessageHeader() const {
         return domain_message_header.has_value();
     }
 
     /// Helper function to read a buffer using the appropriate buffer descriptor
-    std::vector<u8> ReadBuffer(std::size_t buffer_index = 0) const;
+    [[nodiscard]] std::vector<u8> ReadBuffer(std::size_t buffer_index = 0) const;
 
     /// Helper function to write a buffer using the appropriate buffer descriptor
     std::size_t WriteBuffer(const void* buffer, std::size_t size,
@@ -308,22 +308,34 @@ public:
     }
 
     /// Helper function to get the size of the input buffer
-    std::size_t GetReadBufferSize(std::size_t buffer_index = 0) const;
+    [[nodiscard]] std::size_t GetReadBufferSize(std::size_t buffer_index = 0) const;
 
     /// Helper function to get the size of the output buffer
-    std::size_t GetWriteBufferSize(std::size_t buffer_index = 0) const;
+    [[nodiscard]] std::size_t GetWriteBufferSize(std::size_t buffer_index = 0) const;
+
+    /// Helper function to derive the number of elements able to be contained in the read buffer
+    template <typename T>
+    [[nodiscard]] std::size_t GetReadBufferNumElements(std::size_t buffer_index = 0) const {
+        return GetReadBufferSize(buffer_index) / sizeof(T);
+    }
+
+    /// Helper function to derive the number of elements able to be contained in the write buffer
+    template <typename T>
+    [[nodiscard]] std::size_t GetWriteBufferNumElements(std::size_t buffer_index = 0) const {
+        return GetWriteBufferSize(buffer_index) / sizeof(T);
+    }
 
     /// Helper function to test whether the input buffer at buffer_index can be read
-    bool CanReadBuffer(std::size_t buffer_index = 0) const;
+    [[nodiscard]] bool CanReadBuffer(std::size_t buffer_index = 0) const;
 
     /// Helper function to test whether the output buffer at buffer_index can be written
-    bool CanWriteBuffer(std::size_t buffer_index = 0) const;
+    [[nodiscard]] bool CanWriteBuffer(std::size_t buffer_index = 0) const;
 
-    Handle GetCopyHandle(std::size_t index) const {
+    [[nodiscard]] Handle GetCopyHandle(std::size_t index) const {
         return incoming_copy_handles.at(index);
     }
 
-    Handle GetMoveHandle(std::size_t index) const {
+    [[nodiscard]] Handle GetMoveHandle(std::size_t index) const {
         return incoming_move_handles.at(index);
     }
 
@@ -348,13 +360,13 @@ public:
         manager = manager_;
     }
 
-    std::string Description() const;
+    [[nodiscard]] std::string Description() const;
 
-    KThread& GetThread() {
+    [[nodiscard]] KThread& GetThread() {
         return *thread;
     }
 
-    std::shared_ptr<SessionRequestManager> GetManager() const {
+    [[nodiscard]] std::shared_ptr<SessionRequestManager> GetManager() const {
         return manager.lock();
     }
 

src/core/hle/kernel/init/init_slab_setup.cpp

@@ -243,6 +243,7 @@ void InitializeSlabHeaps(Core::System& system, KMemoryLayout& memory_layout) {
             // If we somehow get an invalid type, abort.
         default:
             ASSERT_MSG(false, "Unknown slab type: {}", slab_types[i]);
+            break;
         }
 
         // If we've hit the end of a gap, free it.

src/core/hle/kernel/k_handle_table.cpp

@@ -2,6 +2,7 @@
 // SPDX-License-Identifier: GPL-2.0-or-later
 
 #include "core/hle/kernel/k_handle_table.h"
+#include "core/hle/kernel/k_process.h"
 
 namespace Kernel {
 
@@ -82,6 +83,22 @@ Result KHandleTable::Add(Handle* out_handle, KAutoObject* obj) {
     R_SUCCEED();
 }
 
+KScopedAutoObject<KAutoObject> KHandleTable::GetObjectForIpc(Handle handle,
+                                                             KThread* cur_thread) const {
+    // Handle pseudo-handles.
+    ASSERT(cur_thread != nullptr);
+    if (handle == Svc::PseudoHandle::CurrentProcess) {
+        auto* const cur_process = cur_thread->GetOwnerProcess();
+        ASSERT(cur_process != nullptr);
+        return cur_process;
+    }
+    if (handle == Svc::PseudoHandle::CurrentThread) {
+        return cur_thread;
+    }
+
+    return GetObjectForIpcWithoutPseudoHandle(handle);
+}
+
 Result KHandleTable::Reserve(Handle* out_handle) {
     KScopedDisableDispatch dd{m_kernel};
     KScopedSpinLock lk(m_lock);

src/core/hle/kernel/k_handle_table.h

@@ -113,21 +113,7 @@ public:
         return this->GetObjectImpl(handle);
     }
 
-    KScopedAutoObject<KAutoObject> GetObjectForIpc(Handle handle, KThread* cur_thread) const {
-        // Handle pseudo-handles.
-        ASSERT(cur_thread != nullptr);
-        if (handle == Svc::PseudoHandle::CurrentProcess) {
-            auto* const cur_process =
-                static_cast<KAutoObject*>(static_cast<void*>(cur_thread->GetOwnerProcess()));
-            ASSERT(cur_process != nullptr);
-            return cur_process;
-        }
-        if (handle == Svc::PseudoHandle::CurrentThread) {
-            return static_cast<KAutoObject*>(cur_thread);
-        }
-
-        return GetObjectForIpcWithoutPseudoHandle(handle);
-    }
+    KScopedAutoObject<KAutoObject> GetObjectForIpc(Handle handle, KThread* cur_thread) const;
 
     KScopedAutoObject<KAutoObject> GetObjectByIndex(Handle* out_handle, size_t index) const {
         KScopedDisableDispatch dd{m_kernel};

src/core/hle/kernel/k_page_table.cpp

@@ -2301,6 +2301,7 @@ Result KPageTable::SetProcessMemoryPermission(VAddr addr, size_t size,
             break;
         default:
             ASSERT(false);
+            break;
         }
     }
 
@@ -2803,6 +2804,7 @@ Result KPageTable::Operate(VAddr addr, size_t num_pages, const KPageGroup& page_
             break;
         default:
             ASSERT(false);
+            break;
         }
 
         addr += size;
@@ -2838,6 +2840,7 @@ Result KPageTable::Operate(VAddr addr, size_t num_pages, KMemoryPermission perm,
         break;
     default:
         ASSERT(false);
+        break;
     }
     R_SUCCEED();
 }

src/core/hle/kernel/k_page_table.h

@@ -320,6 +320,9 @@ public:
     constexpr VAddr GetAliasCodeRegionStart() const {
         return m_alias_code_region_start;
     }
+    constexpr VAddr GetAliasCodeRegionEnd() const {
+        return m_alias_code_region_end;
+    }
     constexpr VAddr GetAliasCodeRegionSize() const {
         return m_alias_code_region_end - m_alias_code_region_start;
     }

src/core/hle/kernel/k_process.cpp

@@ -395,6 +395,7 @@ Result KProcess::LoadFromMetadata(const FileSys::ProgramMetadata& metadata, std:
 
     default:
         ASSERT(false);
+        break;
     }
 
     // Create TLS region

src/core/hle/kernel/svc.cpp

@@ -2701,14 +2701,24 @@ static Result GetThreadList(Core::System& system, u32* out_num_threads, VAddr ou
     return ResultSuccess;
 }
 
-static Result FlushProcessDataCache32([[maybe_unused]] Core::System& system,
-                                      [[maybe_unused]] Handle handle, [[maybe_unused]] u32 address,
-                                      [[maybe_unused]] u32 size) {
-    // Note(Blinkhawk): For emulation purposes of the data cache this is mostly a no-op,
-    // as all emulation is done in the same cache level in host architecture, thus data cache
-    // does not need flushing.
-    LOG_DEBUG(Kernel_SVC, "called");
-    return ResultSuccess;
+static Result FlushProcessDataCache32(Core::System& system, Handle process_handle, u64 address,
+                                      u64 size) {
+    // Validate address/size.
+    R_UNLESS(size > 0, ResultInvalidSize);
+    R_UNLESS(address == static_cast<uintptr_t>(address), ResultInvalidCurrentMemory);
+    R_UNLESS(size == static_cast<size_t>(size), ResultInvalidCurrentMemory);
+
+    // Get the process from its handle.
+    KScopedAutoObject process =
+        system.Kernel().CurrentProcess()->GetHandleTable().GetObject<KProcess>(process_handle);
+    R_UNLESS(process.IsNotNull(), ResultInvalidHandle);
+
+    // Verify the region is within range.
+    auto& page_table = process->PageTable();
+    R_UNLESS(page_table.Contains(address, size), ResultInvalidCurrentMemory);
+
+    // Perform the operation.
+    R_RETURN(system.Memory().FlushDataCache(*process, address, size));
 }
 
 namespace {

src/core/hle/kernel/svc_wrap.h

@@ -722,4 +722,12 @@ void SvcWrap32(Core::System& system) {
     FuncReturn(system, retval);
 }
 
+// Used by Invalidate/Store/FlushProcessDataCache32
+template <Result func(Core::System&, Handle, u64, u64)>
+void SvcWrap32(Core::System& system) {
+    const u64 address = (Param(system, 3) << 32) | Param(system, 2);
+    const u64 size = (Param(system, 4) << 32) | Param(system, 1);
+    FuncReturn32(system, func(system, Param32(system, 0), address, size).raw);
+}
+
 } // namespace Kernel

src/core/hle/result.h

@@ -28,30 +28,49 @@ enum class ErrorModule : u32 {
     Loader = 9,
     CMIF = 10,
     HIPC = 11,
+    TMA = 12,
+    DMNT = 13,
+    GDS = 14,
     PM = 15,
     NS = 16,
+    BSDSockets = 17,
     HTC = 18,
+    TSC = 19,
     NCMContent = 20,
     SM = 21,
     RO = 22,
+    GC = 23,
     SDMMC = 24,
     OVLN = 25,
     SPL = 26,
+    Socket = 27,
+    HTCLOW = 29,
+    DDSF = 30,
+    HTCFS = 31,
+    Async = 32,
+    Util = 33,
+    TIPC = 35,
+    ANIF = 37,
     ETHC = 100,
     I2C = 101,
     GPIO = 102,
     UART = 103,
+    CPAD = 104,
     Settings = 105,
+    FTM = 106,
     WLAN = 107,
     XCD = 108,
+    TMP451 = 109,
     NIFM = 110,
     Hwopus = 111,
+    LSM6DS3 = 112,
     Bluetooth = 113,
     VI = 114,
     NFP = 115,
     Time = 116,
     FGM = 117,
     OE = 118,
+    BH1730FVC = 119,
     PCIe = 120,
     Friends = 121,
     BCAT = 122,
@@ -65,7 +84,7 @@ enum class ErrorModule : u32 {
     AHID = 130,
     Qlaunch = 132,
     PCV = 133,
-    OMM = 134,
+    USBPD = 134,
     BPC = 135,
     PSM = 136,
     NIM = 137,
@@ -75,18 +94,22 @@ enum class ErrorModule : u32 {
     NSD = 141,
     PCTL = 142,
     BTM = 143,
+    LA = 144,
     ETicket = 145,
     NGC = 146,
     ERPT = 147,
     APM = 148,
+    CEC = 149,
     Profiler = 150,
     ErrorUpload = 151,
+    LIDBE = 152,
     Audio = 153,
     NPNS = 154,
     NPNSHTTPSTREAM = 155,
     ARP = 157,
     SWKBD = 158,
     BOOT = 159,
+    NetDiag = 160,
     NFCMifare = 161,
     UserlandAssert = 162,
     Fatal = 163,
@@ -94,17 +117,68 @@ enum class ErrorModule : u32 {
     SPSM = 165,
     BGTC = 167,
     UserlandCrash = 168,
+    SASBUS = 169,
+    PI = 170,
+    AudioCtrl = 172,
+    LBL = 173,
+    JIT = 175,
+    HDCP = 176,
+    OMM = 177,
+    PDM = 178,
+    OLSC = 179,
     SREPO = 180,
     Dauth = 181,
+    STDFU = 182,
+    DBG = 183,
+    DHCPS = 186,
+    SPI = 187,
+    AVM = 188,
+    PWM = 189,
+    RTC = 191,
+    Regulator = 192,
+    LED = 193,
+    SIO = 195,
+    PCM = 196,
+    CLKRST = 197,
+    POWCTL = 198,
+    AudioOld = 201,
     HID = 202,
     LDN = 203,
+    CS = 204,
     Irsensor = 205,
     Capture = 206,
     Manu = 208,
     ATK = 209,
+    WEB = 210,
+    LCS = 211,
     GRC = 212,
+    Repair = 213,
+    Album = 214,
+    RID = 215,
     Migration = 216,
     MigrationLdcServ = 217,
+    HIDBUS = 218,
+    ENS = 219,
+    WebSocket = 223,
+    DCDMTP = 227,
+    PGL = 228,
+    Notification = 229,
+    INS = 230,
+    LP2P = 231,
+    RCD = 232,
+    LCM40607 = 233,
+    PRC = 235,
+    TMAHTC = 237,
+    ECTX = 238,
+    MNPP = 239,
+    HSHL = 240,
+    CAPMTP = 242,
+    DP2HDMI = 244,
+    Cradle = 245,
+    SProfile = 246,
+    NDRM = 250,
+    TSPM = 499,
+    DevMenu = 500,
     GeneralWebApplet = 800,
     WifiWebAuthApplet = 809,
     WhitelistedApplet = 810,

src/core/hle/service/am/applets/applet_error.cpp

@@ -144,6 +144,7 @@ void Error::Initialize() {
         break;
     default:
         UNIMPLEMENTED_MSG("Unimplemented LibAppletError mode={:02X}!", mode);
+        break;
     }
 }
 

src/core/hle/service/am/applets/applet_general_backend.cpp

@@ -129,6 +129,7 @@ void Auth::Execute() {
     }
     default:
         unimplemented_log();
+        break;
     }
 }
 
@@ -192,6 +193,7 @@ void PhotoViewer::Execute() {
         break;
     default:
         UNIMPLEMENTED_MSG("Unimplemented PhotoViewer applet mode={:02X}!", mode);
+        break;
     }
 }
 

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

@@ -122,10 +122,10 @@ private:
     }
 
     void GetReleasedAudioInBuffer(Kernel::HLERequestContext& ctx) {
-        auto write_buffer_size = ctx.GetWriteBufferSize() / sizeof(u64);
-        std::vector<u64> released_buffers(write_buffer_size, 0);
+        const auto write_buffer_size = ctx.GetWriteBufferNumElements<u64>();
+        std::vector<u64> released_buffers(write_buffer_size);
 
-        auto count = impl->GetReleasedBuffers(released_buffers);
+        const auto count = impl->GetReleasedBuffers(released_buffers);
 
         [[maybe_unused]] std::string tags{};
         for (u32 i = 0; i < count; i++) {
@@ -228,7 +228,7 @@ void AudInU::ListAudioIns(Kernel::HLERequestContext& ctx) {
     LOG_DEBUG(Service_Audio, "called");
 
     const auto write_count =
-        static_cast<u32>(ctx.GetWriteBufferSize() / sizeof(AudioDevice::AudioDeviceName));
+        static_cast<u32>(ctx.GetWriteBufferNumElements<AudioDevice::AudioDeviceName>());
     std::vector<AudioDevice::AudioDeviceName> device_names{};
 
     u32 out_count{0};
@@ -248,7 +248,7 @@ void AudInU::ListAudioInsAutoFiltered(Kernel::HLERequestContext& ctx) {
     LOG_DEBUG(Service_Audio, "called");
 
     const auto write_count =
-        static_cast<u32>(ctx.GetWriteBufferSize() / sizeof(AudioDevice::AudioDeviceName));
+        static_cast<u32>(ctx.GetWriteBufferNumElements<AudioDevice::AudioDeviceName>());
     std::vector<AudioDevice::AudioDeviceName> device_names{};
 
     u32 out_count{0};

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

@@ -129,16 +129,16 @@ private:
     }
 
     void GetReleasedAudioOutBuffers(Kernel::HLERequestContext& ctx) {
-        auto write_buffer_size = ctx.GetWriteBufferSize() / sizeof(u64);
-        std::vector<u64> released_buffers(write_buffer_size, 0);
+        const auto write_buffer_size = ctx.GetWriteBufferNumElements<u64>();
+        std::vector<u64> released_buffers(write_buffer_size);
 
-        auto count = impl->GetReleasedBuffers(released_buffers);
+        const auto count = impl->GetReleasedBuffers(released_buffers);
 
         [[maybe_unused]] std::string tags{};
         for (u32 i = 0; i < count; i++) {
             tags += fmt::format("{:08X}, ", released_buffers[i]);
         }
-        [[maybe_unused]] auto sessionid{impl->GetSystem().GetSessionId()};
+        [[maybe_unused]] const auto sessionid{impl->GetSystem().GetSessionId()};
         LOG_TRACE(Service_Audio, "called. Session {} released {} buffers: {}", sessionid, count,
                   tags);
 
@@ -244,7 +244,7 @@ void AudOutU::ListAudioOuts(Kernel::HLERequestContext& ctx) {
     std::scoped_lock l{impl->mutex};
 
     const auto write_count =
-        static_cast<u32>(ctx.GetWriteBufferSize() / sizeof(AudioDevice::AudioDeviceName));
+        static_cast<u32>(ctx.GetWriteBufferNumElements<AudioDevice::AudioDeviceName>());
     std::vector<AudioDevice::AudioDeviceName> device_names{};
     if (write_count > 0) {
         device_names.emplace_back("DeviceOut");

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

@@ -274,7 +274,7 @@ public:
 
 private:
     void ListAudioDeviceName(Kernel::HLERequestContext& ctx) {
-        const size_t in_count = ctx.GetWriteBufferSize() / sizeof(AudioDevice::AudioDeviceName);
+        const size_t in_count = ctx.GetWriteBufferNumElements<AudioDevice::AudioDeviceName>();
 
         std::vector<AudioDevice::AudioDeviceName> out_names{};
 
@@ -335,7 +335,7 @@ private:
     }
 
     void GetActiveAudioDeviceName(Kernel::HLERequestContext& ctx) {
-        const auto write_size = ctx.GetWriteBufferSize() / sizeof(char);
+        const auto write_size = ctx.GetWriteBufferSize();
         std::string out_name{"AudioTvOutput"};
 
         LOG_DEBUG(Service_Audio, "(STUBBED) called. Name={}", out_name);
@@ -387,7 +387,7 @@ private:
     }
 
     void ListAudioOutputDeviceName(Kernel::HLERequestContext& ctx) {
-        const size_t in_count = ctx.GetWriteBufferSize() / sizeof(AudioDevice::AudioDeviceName);
+        const size_t in_count = ctx.GetWriteBufferNumElements<AudioDevice::AudioDeviceName>();
 
         std::vector<AudioDevice::AudioDeviceName> out_names{};
 

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

@@ -68,7 +68,7 @@ private:
                                  ExtraBehavior extra_behavior) {
         u32 consumed = 0;
         u32 sample_count = 0;
-        std::vector<opus_int16> samples(ctx.GetWriteBufferSize() / sizeof(opus_int16));
+        std::vector<opus_int16> samples(ctx.GetWriteBufferNumElements<opus_int16>());
 
         if (extra_behavior == ExtraBehavior::ResetContext) {
             ResetDecoderContext();

src/core/hle/service/bcat/bcat_module.cpp

@@ -443,7 +443,7 @@ private:
     }
 
     void Read(Kernel::HLERequestContext& ctx) {
-        auto write_size = ctx.GetWriteBufferSize() / sizeof(DeliveryCacheDirectoryEntry);
+        auto write_size = ctx.GetWriteBufferNumElements<DeliveryCacheDirectoryEntry>();
 
         LOG_DEBUG(Service_BCAT, "called, write_size={:016X}", write_size);
 
@@ -533,7 +533,7 @@ private:
     }
 
     void EnumerateDeliveryCacheDirectory(Kernel::HLERequestContext& ctx) {
-        auto size = ctx.GetWriteBufferSize() / sizeof(DirectoryName);
+        auto size = ctx.GetWriteBufferNumElements<DirectoryName>();
 
         LOG_DEBUG(Service_BCAT, "called, size={:016X}", size);
 

src/core/hle/service/es/es.cpp

@@ -192,12 +192,10 @@ private:
     }
 
     void ListCommonTicketRightsIds(Kernel::HLERequestContext& ctx) {
-        u32 out_entries;
-        if (keys.GetCommonTickets().empty())
-            out_entries = 0;
-        else
-            out_entries = static_cast<u32>(ctx.GetWriteBufferSize() / sizeof(u128));
-
+        size_t out_entries = 0;
+        if (!keys.GetCommonTickets().empty()) {
+            out_entries = ctx.GetWriteBufferNumElements<u128>();
+        }
         LOG_DEBUG(Service_ETicket, "called, entries={:016X}", out_entries);
 
         keys.PopulateTickets();
@@ -206,20 +204,19 @@ private:
         std::transform(tickets.begin(), tickets.end(), std::back_inserter(ids),
                        [](const auto& pair) { return pair.first; });
 
-        out_entries = static_cast<u32>(std::min<std::size_t>(ids.size(), out_entries));
+        out_entries = std::min(ids.size(), out_entries);
         ctx.WriteBuffer(ids.data(), out_entries * sizeof(u128));
 
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(ResultSuccess);
-        rb.Push<u32>(out_entries);
+        rb.Push<u32>(static_cast<u32>(out_entries));
     }
 
     void ListPersonalizedTicketRightsIds(Kernel::HLERequestContext& ctx) {
-        u32 out_entries;
-        if (keys.GetPersonalizedTickets().empty())
-            out_entries = 0;
-        else
-            out_entries = static_cast<u32>(ctx.GetWriteBufferSize() / sizeof(u128));
+        size_t out_entries = 0;
+        if (!keys.GetPersonalizedTickets().empty()) {
+            out_entries = ctx.GetWriteBufferNumElements<u128>();
+        }
 
         LOG_DEBUG(Service_ETicket, "called, entries={:016X}", out_entries);
 
@@ -229,12 +226,12 @@ private:
         std::transform(tickets.begin(), tickets.end(), std::back_inserter(ids),
                        [](const auto& pair) { return pair.first; });
 
-        out_entries = static_cast<u32>(std::min<std::size_t>(ids.size(), out_entries));
+        out_entries = std::min(ids.size(), out_entries);
         ctx.WriteBuffer(ids.data(), out_entries * sizeof(u128));
 
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(ResultSuccess);
-        rb.Push<u32>(out_entries);
+        rb.Push<u32>(static_cast<u32>(out_entries));
     }
 
     void GetCommonTicketSize(Kernel::HLERequestContext& ctx) {

src/core/hle/service/filesystem/fsp_srv.cpp

@@ -277,7 +277,7 @@ private:
         LOG_DEBUG(Service_FS, "called.");
 
         // Calculate how many entries we can fit in the output buffer
-        const u64 count_entries = ctx.GetWriteBufferSize() / sizeof(FileSys::Entry);
+        const u64 count_entries = ctx.GetWriteBufferNumElements<FileSys::Entry>();
 
         // Cap at total number of entries.
         const u64 actual_entries = std::min(count_entries, entries.size() - next_entry_index);
@@ -543,7 +543,7 @@ public:
         LOG_DEBUG(Service_FS, "called");
 
         // Calculate how many entries we can fit in the output buffer
-        const u64 count_entries = ctx.GetWriteBufferSize() / sizeof(SaveDataInfo);
+        const u64 count_entries = ctx.GetWriteBufferNumElements<SaveDataInfo>();
 
         // Cap at total number of entries.
         const u64 actual_entries = std::min(count_entries, info.size() - next_entry_index);

src/core/hle/service/hid/hid.cpp

@@ -36,8 +36,9 @@ namespace Service::HID {
 
 // Updating period for each HID device.
 // Period time is obtained by measuring the number of samples in a second on HW using a homebrew
-// Correct pad_update_ns is 4ms this is overclocked to lower input lag
-constexpr auto pad_update_ns = std::chrono::nanoseconds{1 * 1000 * 1000}; // (1ms, 1000Hz)
+// Correct npad_update_ns is 4ms this is overclocked to lower input lag
+constexpr auto npad_update_ns = std::chrono::nanoseconds{1 * 1000 * 1000};    // (1ms, 1000Hz)
+constexpr auto default_update_ns = std::chrono::nanoseconds{4 * 1000 * 1000}; // (4ms, 1000Hz)
 constexpr auto mouse_keyboard_update_ns = std::chrono::nanoseconds{8 * 1000 * 1000}; // (8ms, 125Hz)
 constexpr auto motion_update_ns = std::chrono::nanoseconds{5 * 1000 * 1000};         // (5ms, 200Hz)
 
@@ -75,8 +76,16 @@ IAppletResource::IAppletResource(Core::System& system_,
     GetController<Controller_Stubbed>(HidController::UniquePad).SetCommonHeaderOffset(0x5A00);
 
     // Register update callbacks
-    pad_update_event = Core::Timing::CreateEvent(
+    npad_update_event = Core::Timing::CreateEvent(
         "HID::UpdatePadCallback",
+        [this](std::uintptr_t user_data, s64 time,
+               std::chrono::nanoseconds ns_late) -> std::optional<std::chrono::nanoseconds> {
+            const auto guard = LockService();
+            UpdateNpad(user_data, ns_late);
+            return std::nullopt;
+        });
+    default_update_event = Core::Timing::CreateEvent(
+        "HID::UpdateDefaultCallback",
         [this](std::uintptr_t user_data, s64 time,
                std::chrono::nanoseconds ns_late) -> std::optional<std::chrono::nanoseconds> {
             const auto guard = LockService();
@@ -100,7 +109,9 @@ IAppletResource::IAppletResource(Core::System& system_,
             return std::nullopt;
         });
 
-    system.CoreTiming().ScheduleLoopingEvent(pad_update_ns, pad_update_ns, pad_update_event);
+    system.CoreTiming().ScheduleLoopingEvent(npad_update_ns, npad_update_ns, npad_update_event);
+    system.CoreTiming().ScheduleLoopingEvent(default_update_ns, default_update_ns,
+                                             default_update_event);
     system.CoreTiming().ScheduleLoopingEvent(mouse_keyboard_update_ns, mouse_keyboard_update_ns,
                                              mouse_keyboard_update_event);
     system.CoreTiming().ScheduleLoopingEvent(motion_update_ns, motion_update_ns,
@@ -118,7 +129,8 @@ void IAppletResource::DeactivateController(HidController controller) {
 }
 
 IAppletResource::~IAppletResource() {
-    system.CoreTiming().UnscheduleEvent(pad_update_event, 0);
+    system.CoreTiming().UnscheduleEvent(npad_update_event, 0);
+    system.CoreTiming().UnscheduleEvent(default_update_event, 0);
     system.CoreTiming().UnscheduleEvent(mouse_keyboard_update_event, 0);
     system.CoreTiming().UnscheduleEvent(motion_update_event, 0);
 }
@@ -144,10 +156,20 @@ void IAppletResource::UpdateControllers(std::uintptr_t user_data,
         if (controller == controllers[static_cast<size_t>(HidController::Mouse)]) {
             continue;
         }
+        // Npad has it's own update event
+        if (controller == controllers[static_cast<size_t>(HidController::NPad)]) {
+            continue;
+        }
         controller->OnUpdate(core_timing);
     }
 }
 
+void IAppletResource::UpdateNpad(std::uintptr_t user_data, std::chrono::nanoseconds ns_late) {
+    auto& core_timing = system.CoreTiming();
+
+    controllers[static_cast<size_t>(HidController::NPad)]->OnUpdate(core_timing);
+}
+
 void IAppletResource::UpdateMouseKeyboard(std::uintptr_t user_data,
                                           std::chrono::nanoseconds ns_late) {
     auto& core_timing = system.CoreTiming();

src/core/hle/service/hid/hid.h

@@ -71,12 +71,14 @@ private:
 
     void GetSharedMemoryHandle(Kernel::HLERequestContext& ctx);
     void UpdateControllers(std::uintptr_t user_data, std::chrono::nanoseconds ns_late);
+    void UpdateNpad(std::uintptr_t user_data, std::chrono::nanoseconds ns_late);
     void UpdateMouseKeyboard(std::uintptr_t user_data, std::chrono::nanoseconds ns_late);
     void UpdateMotion(std::uintptr_t user_data, std::chrono::nanoseconds ns_late);
 
     KernelHelpers::ServiceContext& service_context;
 
-    std::shared_ptr<Core::Timing::EventType> pad_update_event;
+    std::shared_ptr<Core::Timing::EventType> npad_update_event;
+    std::shared_ptr<Core::Timing::EventType> default_update_event;
     std::shared_ptr<Core::Timing::EventType> mouse_keyboard_update_event;
     std::shared_ptr<Core::Timing::EventType> motion_update_event;
 

src/core/hle/service/ldn/ldn.cpp

@@ -292,7 +292,7 @@ public:
 
     void GetNetworkInfoLatestUpdate(Kernel::HLERequestContext& ctx) {
         const std::size_t network_buffer_size = ctx.GetWriteBufferSize(0);
-        const std::size_t node_buffer_count = ctx.GetWriteBufferSize(1) / sizeof(NodeLatestUpdate);
+        const std::size_t node_buffer_count = ctx.GetWriteBufferNumElements<NodeLatestUpdate>(1);
 
         if (node_buffer_count == 0 || network_buffer_size != sizeof(NetworkInfo)) {
             LOG_ERROR(Service_LDN, "Invalid buffer, size = {}, count = {}", network_buffer_size,
@@ -333,7 +333,7 @@ public:
         const auto channel{rp.PopEnum<WifiChannel>()};
         const auto scan_filter{rp.PopRaw<ScanFilter>()};
 
-        const std::size_t network_info_size = ctx.GetWriteBufferSize() / sizeof(NetworkInfo);
+        const std::size_t network_info_size = ctx.GetWriteBufferNumElements<NetworkInfo>();
 
         if (network_info_size == 0) {
             LOG_ERROR(Service_LDN, "Invalid buffer size {}", network_info_size);

src/core/hle/service/nfc/nfc.cpp

@@ -7,6 +7,7 @@
 #include "common/settings.h"
 #include "core/hle/ipc_helpers.h"
 #include "core/hle/service/nfc/nfc.h"
+#include "core/hle/service/nfc/nfc_user.h"
 #include "core/hle/service/service.h"
 #include "core/hle/service/sm/sm.h"
 
@@ -97,76 +98,6 @@ private:
     }
 };
 
-class IUser final : public ServiceFramework<IUser> {
-public:
-    explicit IUser(Core::System& system_) : ServiceFramework{system_, "NFC::IUser"} {
-        // clang-format off
-        static const FunctionInfo functions[] = {
-            {0, &IUser::InitializeOld, "InitializeOld"},
-            {1, &IUser::FinalizeOld, "FinalizeOld"},
-            {2, &IUser::GetStateOld, "GetStateOld"},
-            {3, &IUser::IsNfcEnabledOld, "IsNfcEnabledOld"},
-            {400, &IUser::InitializeOld, "Initialize"},
-            {401, &IUser::FinalizeOld, "Finalize"},
-            {402, &IUser::GetStateOld, "GetState"},
-            {403, &IUser::IsNfcEnabledOld, "IsNfcEnabled"},
-            {404, nullptr, "ListDevices"},
-            {405, nullptr, "GetDeviceState"},
-            {406, nullptr, "GetNpadId"},
-            {407, nullptr, "AttachAvailabilityChangeEvent"},
-            {408, nullptr, "StartDetection"},
-            {409, nullptr, "StopDetection"},
-            {410, nullptr, "GetTagInfo"},
-            {411, nullptr, "AttachActivateEvent"},
-            {412, nullptr, "AttachDeactivateEvent"},
-            {1000, nullptr, "ReadMifare"},
-            {1001, nullptr, "WriteMifare"},
-            {1300, nullptr, "SendCommandByPassThrough"},
-            {1301, nullptr, "KeepPassThroughSession"},
-            {1302, nullptr, "ReleasePassThroughSession"},
-        };
-        // clang-format on
-
-        RegisterHandlers(functions);
-    }
-
-private:
-    enum class NfcStates : u32 {
-        Finalized = 6,
-    };
-
-    void InitializeOld(Kernel::HLERequestContext& ctx) {
-        LOG_DEBUG(Service_NFC, "called");
-
-        IPC::ResponseBuilder rb{ctx, 2, 0};
-        rb.Push(ResultSuccess);
-        // We don't deal with hardware initialization so we can just stub this.
-    }
-
-    void IsNfcEnabledOld(Kernel::HLERequestContext& ctx) {
-        LOG_DEBUG(Service_NFC, "IsNfcEnabledOld");
-
-        IPC::ResponseBuilder rb{ctx, 3};
-        rb.Push(ResultSuccess);
-        rb.PushRaw<u8>(true);
-    }
-
-    void GetStateOld(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_NFC, "(STUBBED) called");
-
-        IPC::ResponseBuilder rb{ctx, 3};
-        rb.Push(ResultSuccess);
-        rb.PushEnum(NfcStates::Finalized); // TODO(ogniK): Figure out if this matches nfp
-    }
-
-    void FinalizeOld(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_NFC, "(STUBBED) called");
-
-        IPC::ResponseBuilder rb{ctx, 2};
-        rb.Push(ResultSuccess);
-    }
-};
-
 class NFC_U final : public ServiceFramework<NFC_U> {
 public:
     explicit NFC_U(Core::System& system_) : ServiceFramework{system_, "nfc:user"} {

src/core/hle/service/nfc/nfc_device.cpp

@@ -0,0 +1,197 @@
+// SPDX-FileCopyrightText: Copyright 2022 yuzu Emulator Project
+// SPDX-License-Identifier: GPL-2.0-or-later
+
+#include "common/input.h"
+#include "common/logging/log.h"
+#include "core/core.h"
+#include "core/hid/emulated_controller.h"
+#include "core/hid/hid_core.h"
+#include "core/hid/hid_types.h"
+#include "core/hle/ipc_helpers.h"
+#include "core/hle/kernel/k_event.h"
+#include "core/hle/service/nfc/nfc_device.h"
+#include "core/hle/service/nfc/nfc_result.h"
+#include "core/hle/service/nfc/nfc_user.h"
+
+namespace Service::NFC {
+NfcDevice::NfcDevice(Core::HID::NpadIdType npad_id_, Core::System& system_,
+                     KernelHelpers::ServiceContext& service_context_,
+                     Kernel::KEvent* availability_change_event_)
+    : npad_id{npad_id_}, system{system_}, service_context{service_context_},
+      availability_change_event{availability_change_event_} {
+    activate_event = service_context.CreateEvent("IUser:NFCActivateEvent");
+    deactivate_event = service_context.CreateEvent("IUser:NFCDeactivateEvent");
+    npad_device = system.HIDCore().GetEmulatedController(npad_id);
+
+    Core::HID::ControllerUpdateCallback engine_callback{
+        .on_change = [this](Core::HID::ControllerTriggerType type) { NpadUpdate(type); },
+        .is_npad_service = false,
+    };
+    is_controller_set = true;
+    callback_key = npad_device->SetCallback(engine_callback);
+}
+
+NfcDevice::~NfcDevice() {
+    activate_event->Close();
+    deactivate_event->Close();
+    if (!is_controller_set) {
+        return;
+    }
+    npad_device->DeleteCallback(callback_key);
+    is_controller_set = false;
+};
+
+void NfcDevice::NpadUpdate(Core::HID::ControllerTriggerType type) {
+    if (type == Core::HID::ControllerTriggerType::Connected ||
+        type == Core::HID::ControllerTriggerType::Disconnected) {
+        availability_change_event->Signal();
+        return;
+    }
+
+    if (type != Core::HID::ControllerTriggerType::Nfc) {
+        return;
+    }
+
+    if (!npad_device->IsConnected()) {
+        return;
+    }
+
+    const auto nfc_status = npad_device->GetNfc();
+    switch (nfc_status.state) {
+    case Common::Input::NfcState::NewAmiibo:
+        LoadNfcTag(nfc_status.data);
+        break;
+    case Common::Input::NfcState::AmiiboRemoved:
+        if (device_state != NFP::DeviceState::SearchingForTag) {
+            CloseNfcTag();
+        }
+        break;
+    default:
+        break;
+    }
+}
+
+bool NfcDevice::LoadNfcTag(std::span<const u8> data) {
+    if (device_state != NFP::DeviceState::SearchingForTag) {
+        LOG_ERROR(Service_NFC, "Game is not looking for nfc tag, current state {}", device_state);
+        return false;
+    }
+
+    if (data.size() != sizeof(NFP::EncryptedNTAG215File)) {
+        LOG_ERROR(Service_NFC, "Not an amiibo, size={}", data.size());
+        return false;
+    }
+
+    memcpy(&encrypted_tag_data, data.data(), sizeof(NFP::EncryptedNTAG215File));
+
+    device_state = NFP::DeviceState::TagFound;
+    deactivate_event->GetReadableEvent().Clear();
+    activate_event->Signal();
+    return true;
+}
+
+void NfcDevice::CloseNfcTag() {
+    LOG_INFO(Service_NFC, "Remove nfc tag");
+
+    device_state = NFP::DeviceState::TagRemoved;
+    encrypted_tag_data = {};
+    activate_event->GetReadableEvent().Clear();
+    deactivate_event->Signal();
+}
+
+Kernel::KReadableEvent& NfcDevice::GetActivateEvent() const {
+    return activate_event->GetReadableEvent();
+}
+
+Kernel::KReadableEvent& NfcDevice::GetDeactivateEvent() const {
+    return deactivate_event->GetReadableEvent();
+}
+
+void NfcDevice::Initialize() {
+    device_state =
+        npad_device->HasNfc() ? NFP::DeviceState::Initialized : NFP::DeviceState::Unavailable;
+    encrypted_tag_data = {};
+}
+
+void NfcDevice::Finalize() {
+    if (device_state == NFP::DeviceState::SearchingForTag ||
+        device_state == NFP::DeviceState::TagRemoved) {
+        StopDetection();
+    }
+    device_state = NFP::DeviceState::Unavailable;
+}
+
+Result NfcDevice::StartDetection(s32 protocol_) {
+    if (device_state != NFP::DeviceState::Initialized &&
+        device_state != NFP::DeviceState::TagRemoved) {
+        LOG_ERROR(Service_NFC, "Wrong device state {}", device_state);
+        return WrongDeviceState;
+    }
+
+    if (!npad_device->SetPollingMode(Common::Input::PollingMode::NFC)) {
+        LOG_ERROR(Service_NFC, "Nfc not supported");
+        return NfcDisabled;
+    }
+
+    device_state = NFP::DeviceState::SearchingForTag;
+    protocol = protocol_;
+    return ResultSuccess;
+}
+
+Result NfcDevice::StopDetection() {
+    npad_device->SetPollingMode(Common::Input::PollingMode::Active);
+
+    if (device_state == NFP::DeviceState::Initialized) {
+        return ResultSuccess;
+    }
+
+    if (device_state == NFP::DeviceState::TagFound ||
+        device_state == NFP::DeviceState::TagMounted) {
+        CloseNfcTag();
+        return ResultSuccess;
+    }
+    if (device_state == NFP::DeviceState::SearchingForTag ||
+        device_state == NFP::DeviceState::TagRemoved) {
+        device_state = NFP::DeviceState::Initialized;
+        return ResultSuccess;
+    }
+
+    LOG_ERROR(Service_NFC, "Wrong device state {}", device_state);
+    return WrongDeviceState;
+}
+
+Result NfcDevice::GetTagInfo(NFP::TagInfo& tag_info) const {
+    if (device_state != NFP::DeviceState::TagFound &&
+        device_state != NFP::DeviceState::TagMounted) {
+        LOG_ERROR(Service_NFC, "Wrong device state {}", device_state);
+        if (device_state == NFP::DeviceState::TagRemoved) {
+            return TagRemoved;
+        }
+        return WrongDeviceState;
+    }
+
+    // Protocol and tag type may change here
+    tag_info = {
+        .uuid = encrypted_tag_data.uuid.uid,
+        .uuid_length = static_cast<u8>(encrypted_tag_data.uuid.uid.size()),
+        .protocol = NFP::TagProtocol::TypeA,
+        .tag_type = NFP::TagType::Type2,
+    };
+
+    return ResultSuccess;
+}
+
+u64 NfcDevice::GetHandle() const {
+    // Generate a handle based of the npad id
+    return static_cast<u64>(npad_id);
+}
+
+NFP::DeviceState NfcDevice::GetCurrentState() const {
+    return device_state;
+}
+
+Core::HID::NpadIdType NfcDevice::GetNpadId() const {
+    return npad_id;
+}
+
+} // namespace Service::NFC

src/core/hle/service/nfc/nfc_device.h

@@ -0,0 +1,70 @@
+// SPDX-FileCopyrightText: Copyright 2022 yuzu Emulator Project
+// SPDX-License-Identifier: GPL-2.0-or-later
+
+#pragma once
+
+#include "common/common_types.h"
+#include "core/hle/service/kernel_helpers.h"
+#include "core/hle/service/nfp/nfp_types.h"
+#include "core/hle/service/service.h"
+
+namespace Kernel {
+class KEvent;
+class KReadableEvent;
+} // namespace Kernel
+
+namespace Core {
+class System;
+} // namespace Core
+
+namespace Core::HID {
+class EmulatedController;
+enum class ControllerTriggerType;
+enum class NpadIdType : u32;
+} // namespace Core::HID
+
+namespace Service::NFC {
+class NfcDevice {
+public:
+    NfcDevice(Core::HID::NpadIdType npad_id_, Core::System& system_,
+              KernelHelpers::ServiceContext& service_context_,
+              Kernel::KEvent* availability_change_event_);
+    ~NfcDevice();
+
+    void Initialize();
+    void Finalize();
+
+    Result StartDetection(s32 protocol_);
+    Result StopDetection();
+
+    Result GetTagInfo(NFP::TagInfo& tag_info) const;
+
+    u64 GetHandle() const;
+    NFP::DeviceState GetCurrentState() const;
+    Core::HID::NpadIdType GetNpadId() const;
+
+    Kernel::KReadableEvent& GetActivateEvent() const;
+    Kernel::KReadableEvent& GetDeactivateEvent() const;
+
+private:
+    void NpadUpdate(Core::HID::ControllerTriggerType type);
+    bool LoadNfcTag(std::span<const u8> data);
+    void CloseNfcTag();
+
+    bool is_controller_set{};
+    int callback_key;
+    const Core::HID::NpadIdType npad_id;
+    Core::System& system;
+    Core::HID::EmulatedController* npad_device = nullptr;
+    KernelHelpers::ServiceContext& service_context;
+    Kernel::KEvent* activate_event = nullptr;
+    Kernel::KEvent* deactivate_event = nullptr;
+    Kernel::KEvent* availability_change_event = nullptr;
+
+    s32 protocol{};
+    NFP::DeviceState device_state{NFP::DeviceState::Unavailable};
+
+    NFP::EncryptedNTAG215File encrypted_tag_data{};
+};
+
+} // namespace Service::NFC

src/core/hle/service/nfc/nfc_result.h

@@ -0,0 +1,17 @@
+// SPDX-FileCopyrightText: Copyright 2022 yuzu Emulator Project
+// SPDX-License-Identifier: GPL-2.0-or-later
+
+#pragma once
+
+#include "core/hle/result.h"
+
+namespace Service::NFC {
+
+constexpr Result DeviceNotFound(ErrorModule::NFC, 64);
+constexpr Result InvalidArgument(ErrorModule::NFC, 65);
+constexpr Result WrongDeviceState(ErrorModule::NFC, 73);
+constexpr Result NfcDisabled(ErrorModule::NFC, 80);
+constexpr Result TagRemoved(ErrorModule::NFC, 97);
+constexpr Result CorruptedData(ErrorModule::NFC, 144);
+
+} // namespace Service::NFC

src/core/hle/service/nfc/nfc_user.cpp

@@ -0,0 +1,365 @@
+// SPDX-FileCopyrightText: Copyright 2022 yuzu Emulator Project
+// SPDX-License-Identifier: GPL-2.0-or-later
+
+#include "common/logging/log.h"
+#include "core/core.h"
+#include "core/hid/hid_types.h"
+#include "core/hle/ipc_helpers.h"
+#include "core/hle/kernel/k_event.h"
+#include "core/hle/service/nfc/nfc_device.h"
+#include "core/hle/service/nfc/nfc_result.h"
+#include "core/hle/service/nfc/nfc_user.h"
+#include "core/hle/service/time/clock_types.h"
+
+namespace Service::NFC {
+
+IUser::IUser(Core::System& system_)
+    : ServiceFramework{system_, "NFC::IUser"}, service_context{system_, service_name} {
+    static const FunctionInfo functions[] = {
+        {0, &IUser::Initialize, "InitializeOld"},
+        {1, &IUser::Finalize, "FinalizeOld"},
+        {2, &IUser::GetState, "GetStateOld"},
+        {3, &IUser::IsNfcEnabled, "IsNfcEnabledOld"},
+        {400, &IUser::Initialize, "Initialize"},
+        {401, &IUser::Finalize, "Finalize"},
+        {402, &IUser::GetState, "GetState"},
+        {403, &IUser::IsNfcEnabled, "IsNfcEnabled"},
+        {404, &IUser::ListDevices, "ListDevices"},
+        {405, &IUser::GetDeviceState, "GetDeviceState"},
+        {406, &IUser::GetNpadId, "GetNpadId"},
+        {407, &IUser::AttachAvailabilityChangeEvent, "AttachAvailabilityChangeEvent"},
+        {408, &IUser::StartDetection, "StartDetection"},
+        {409, &IUser::StopDetection, "StopDetection"},
+        {410, &IUser::GetTagInfo, "GetTagInfo"},
+        {411, &IUser::AttachActivateEvent, "AttachActivateEvent"},
+        {412, &IUser::AttachDeactivateEvent, "AttachDeactivateEvent"},
+        {1000, nullptr, "ReadMifare"},
+        {1001, nullptr, "WriteMifare"},
+        {1300, &IUser::SendCommandByPassThrough, "SendCommandByPassThrough"},
+        {1301, nullptr, "KeepPassThroughSession"},
+        {1302, nullptr, "ReleasePassThroughSession"},
+    };
+    RegisterHandlers(functions);
+
+    availability_change_event = service_context.CreateEvent("IUser:AvailabilityChangeEvent");
+
+    for (u32 device_index = 0; device_index < 10; device_index++) {
+        devices[device_index] =
+            std::make_shared<NfcDevice>(Core::HID::IndexToNpadIdType(device_index), system,
+                                        service_context, availability_change_event);
+    }
+}
+
+IUser ::~IUser() {
+    availability_change_event->Close();
+}
+
+void IUser::Initialize(Kernel::HLERequestContext& ctx) {
+    LOG_INFO(Service_NFC, "called");
+
+    state = State::Initialized;
+
+    for (auto& device : devices) {
+        device->Initialize();
+    }
+
+    IPC::ResponseBuilder rb{ctx, 2, 0};
+    rb.Push(ResultSuccess);
+}
+
+void IUser::Finalize(Kernel::HLERequestContext& ctx) {
+    LOG_INFO(Service_NFC, "called");
+
+    state = State::NonInitialized;
+
+    for (auto& device : devices) {
+        device->Finalize();
+    }
+
+    IPC::ResponseBuilder rb{ctx, 2};
+    rb.Push(ResultSuccess);
+}
+
+void IUser::GetState(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_NFC, "called");
+
+    IPC::ResponseBuilder rb{ctx, 3};
+    rb.Push(ResultSuccess);
+    rb.PushEnum(state);
+}
+
+void IUser::IsNfcEnabled(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_NFC, "called");
+
+    IPC::ResponseBuilder rb{ctx, 3};
+    rb.Push(ResultSuccess);
+    rb.Push(state != State::NonInitialized);
+}
+
+void IUser::ListDevices(Kernel::HLERequestContext& ctx) {
+    LOG_INFO(Service_NFC, "called");
+
+    if (state == State::NonInitialized) {
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(NfcDisabled);
+        return;
+    }
+
+    if (!ctx.CanWriteBuffer()) {
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(InvalidArgument);
+        return;
+    }
+
+    if (ctx.GetWriteBufferSize() == 0) {
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(InvalidArgument);
+        return;
+    }
+
+    std::vector<u64> nfp_devices;
+    const std::size_t max_allowed_devices = ctx.GetWriteBufferNumElements<u64>();
+
+    for (auto& device : devices) {
+        if (nfp_devices.size() >= max_allowed_devices) {
+            continue;
+        }
+        if (device->GetCurrentState() != NFP::DeviceState::Unavailable) {
+            nfp_devices.push_back(device->GetHandle());
+        }
+    }
+
+    if (nfp_devices.empty()) {
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(DeviceNotFound);
+        return;
+    }
+
+    ctx.WriteBuffer(nfp_devices);
+
+    IPC::ResponseBuilder rb{ctx, 3};
+    rb.Push(ResultSuccess);
+    rb.Push(static_cast<s32>(nfp_devices.size()));
+}
+
+void IUser::GetDeviceState(Kernel::HLERequestContext& ctx) {
+    IPC::RequestParser rp{ctx};
+    const auto device_handle{rp.Pop<u64>()};
+    LOG_DEBUG(Service_NFC, "called, device_handle={}", device_handle);
+
+    auto device = GetNfcDevice(device_handle);
+
+    if (!device.has_value()) {
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(DeviceNotFound);
+        return;
+    }
+
+    IPC::ResponseBuilder rb{ctx, 3};
+    rb.Push(ResultSuccess);
+    rb.PushEnum(device.value()->GetCurrentState());
+}
+
+void IUser::GetNpadId(Kernel::HLERequestContext& ctx) {
+    IPC::RequestParser rp{ctx};
+    const auto device_handle{rp.Pop<u64>()};
+    LOG_DEBUG(Service_NFC, "called, device_handle={}", device_handle);
+
+    if (state == State::NonInitialized) {
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(NfcDisabled);
+        return;
+    }
+
+    auto device = GetNfcDevice(device_handle);
+
+    if (!device.has_value()) {
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(DeviceNotFound);
+        return;
+    }
+
+    IPC::ResponseBuilder rb{ctx, 3};
+    rb.Push(ResultSuccess);
+    rb.PushEnum(device.value()->GetNpadId());
+}
+
+void IUser::AttachAvailabilityChangeEvent(Kernel::HLERequestContext& ctx) {
+    LOG_INFO(Service_NFC, "called");
+
+    if (state == State::NonInitialized) {
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(NfcDisabled);
+        return;
+    }
+
+    IPC::ResponseBuilder rb{ctx, 2, 1};
+    rb.Push(ResultSuccess);
+    rb.PushCopyObjects(availability_change_event->GetReadableEvent());
+}
+
+void IUser::StartDetection(Kernel::HLERequestContext& ctx) {
+    IPC::RequestParser rp{ctx};
+    const auto device_handle{rp.Pop<u64>()};
+    const auto nfp_protocol{rp.Pop<s32>()};
+    LOG_INFO(Service_NFC, "called, device_handle={}, nfp_protocol={}", device_handle, nfp_protocol);
+
+    if (state == State::NonInitialized) {
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(NfcDisabled);
+        return;
+    }
+
+    auto device = GetNfcDevice(device_handle);
+
+    if (!device.has_value()) {
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(DeviceNotFound);
+        return;
+    }
+
+    const auto result = device.value()->StartDetection(nfp_protocol);
+    IPC::ResponseBuilder rb{ctx, 2};
+    rb.Push(result);
+}
+
+void IUser::StopDetection(Kernel::HLERequestContext& ctx) {
+    IPC::RequestParser rp{ctx};
+    const auto device_handle{rp.Pop<u64>()};
+    LOG_INFO(Service_NFC, "called, device_handle={}", device_handle);
+
+    if (state == State::NonInitialized) {
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(NfcDisabled);
+        return;
+    }
+
+    auto device = GetNfcDevice(device_handle);
+
+    if (!device.has_value()) {
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(DeviceNotFound);
+        return;
+    }
+
+    const auto result = device.value()->StopDetection();
+    IPC::ResponseBuilder rb{ctx, 2};
+    rb.Push(result);
+}
+
+void IUser::GetTagInfo(Kernel::HLERequestContext& ctx) {
+    IPC::RequestParser rp{ctx};
+    const auto device_handle{rp.Pop<u64>()};
+    LOG_INFO(Service_NFC, "called, device_handle={}", device_handle);
+
+    if (state == State::NonInitialized) {
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(NfcDisabled);
+        return;
+    }
+
+    auto device = GetNfcDevice(device_handle);
+
+    if (!device.has_value()) {
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(DeviceNotFound);
+        return;
+    }
+
+    NFP::TagInfo tag_info{};
+    const auto result = device.value()->GetTagInfo(tag_info);
+    ctx.WriteBuffer(tag_info);
+    IPC::ResponseBuilder rb{ctx, 2};
+    rb.Push(result);
+}
+
+void IUser::AttachActivateEvent(Kernel::HLERequestContext& ctx) {
+    IPC::RequestParser rp{ctx};
+    const auto device_handle{rp.Pop<u64>()};
+    LOG_DEBUG(Service_NFC, "called, device_handle={}", device_handle);
+
+    if (state == State::NonInitialized) {
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(NfcDisabled);
+        return;
+    }
+
+    auto device = GetNfcDevice(device_handle);
+
+    if (!device.has_value()) {
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(DeviceNotFound);
+        return;
+    }
+
+    IPC::ResponseBuilder rb{ctx, 2, 1};
+    rb.Push(ResultSuccess);
+    rb.PushCopyObjects(device.value()->GetActivateEvent());
+}
+
+void IUser::AttachDeactivateEvent(Kernel::HLERequestContext& ctx) {
+    IPC::RequestParser rp{ctx};
+    const auto device_handle{rp.Pop<u64>()};
+    LOG_DEBUG(Service_NFC, "called, device_handle={}", device_handle);
+
+    if (state == State::NonInitialized) {
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(NfcDisabled);
+        return;
+    }
+
+    auto device = GetNfcDevice(device_handle);
+
+    if (!device.has_value()) {
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(DeviceNotFound);
+        return;
+    }
+
+    IPC::ResponseBuilder rb{ctx, 2, 1};
+    rb.Push(ResultSuccess);
+    rb.PushCopyObjects(device.value()->GetDeactivateEvent());
+}
+
+void IUser::SendCommandByPassThrough(Kernel::HLERequestContext& ctx) {
+    IPC::RequestParser rp{ctx};
+    const auto device_handle{rp.Pop<u64>()};
+    const auto timeout{rp.PopRaw<Time::Clock::TimeSpanType>()};
+    const auto command_data{ctx.ReadBuffer()};
+
+    LOG_INFO(Service_NFC, "(STUBBED) called, device_handle={}, timeout={}, data_size={}",
+             device_handle, timeout.ToSeconds(), command_data.size());
+
+    if (state == State::NonInitialized) {
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(NfcDisabled);
+        return;
+    }
+
+    auto device = GetNfcDevice(device_handle);
+
+    if (!device.has_value()) {
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(DeviceNotFound);
+        return;
+    }
+
+    std::vector<u8> out_data(1);
+    // TODO: Request data from nfc device
+    ctx.WriteBuffer(out_data);
+
+    IPC::ResponseBuilder rb{ctx, 3};
+    rb.Push(ResultSuccess);
+    rb.Push(static_cast<u32>(out_data.size()));
+}
+
+std::optional<std::shared_ptr<NfcDevice>> IUser::GetNfcDevice(u64 handle) {
+    for (auto& device : devices) {
+        if (device->GetHandle() == handle) {
+            return device;
+        }
+    }
+    return std::nullopt;
+}
+
+} // namespace Service::NFC

src/core/hle/service/nfc/nfc_user.h

@@ -0,0 +1,52 @@
+// SPDX-FileCopyrightText: Copyright 2022 yuzu Emulator Project
+// SPDX-License-Identifier: GPL-2.0-or-later
+
+#pragma once
+
+#include <array>
+#include <memory>
+#include <optional>
+
+#include "core/hle/service/kernel_helpers.h"
+#include "core/hle/service/service.h"
+
+namespace Service::NFC {
+class NfcDevice;
+
+class IUser final : public ServiceFramework<IUser> {
+public:
+    explicit IUser(Core::System& system_);
+    ~IUser();
+
+private:
+    enum class State : u32 {
+        NonInitialized,
+        Initialized,
+    };
+
+    void Initialize(Kernel::HLERequestContext& ctx);
+    void Finalize(Kernel::HLERequestContext& ctx);
+    void GetState(Kernel::HLERequestContext& ctx);
+    void IsNfcEnabled(Kernel::HLERequestContext& ctx);
+    void ListDevices(Kernel::HLERequestContext& ctx);
+    void GetDeviceState(Kernel::HLERequestContext& ctx);
+    void GetNpadId(Kernel::HLERequestContext& ctx);
+    void AttachAvailabilityChangeEvent(Kernel::HLERequestContext& ctx);
+    void StartDetection(Kernel::HLERequestContext& ctx);
+    void StopDetection(Kernel::HLERequestContext& ctx);
+    void GetTagInfo(Kernel::HLERequestContext& ctx);
+    void AttachActivateEvent(Kernel::HLERequestContext& ctx);
+    void AttachDeactivateEvent(Kernel::HLERequestContext& ctx);
+    void SendCommandByPassThrough(Kernel::HLERequestContext& ctx);
+
+    std::optional<std::shared_ptr<NfcDevice>> GetNfcDevice(u64 handle);
+
+    KernelHelpers::ServiceContext service_context;
+
+    std::array<std::shared_ptr<NfcDevice>, 10> devices{};
+
+    State state{State::NonInitialized};
+    Kernel::KEvent* availability_change_event;
+};
+
+} // namespace Service::NFC

src/core/hle/service/nfp/amiibo_crypto.cpp

@@ -12,7 +12,6 @@
 #include "common/fs/fs.h"
 #include "common/fs/path_util.h"
 #include "common/logging/log.h"
-#include "core/hle/service/mii/mii_manager.h"
 #include "core/hle/service/nfp/amiibo_crypto.h"
 
 namespace Service::NFP::AmiiboCrypto {

src/core/hle/service/nfp/nfp_device.cpp

@@ -2,10 +2,7 @@
 // SPDX-License-Identifier: GPL-2.0-or-later
 
 #include <array>
-#include <atomic>
 
-#include "common/fs/file.h"
-#include "common/fs/path_util.h"
 #include "common/input.h"
 #include "common/logging/log.h"
 #include "common/string_util.h"
@@ -19,7 +16,6 @@
 #include "core/hle/service/mii/mii_manager.h"
 #include "core/hle/service/mii/types.h"
 #include "core/hle/service/nfp/amiibo_crypto.h"
-#include "core/hle/service/nfp/nfp.h"
 #include "core/hle/service/nfp/nfp_device.h"
 #include "core/hle/service/nfp/nfp_result.h"
 #include "core/hle/service/nfp/nfp_user.h"
@@ -49,6 +45,8 @@ NfpDevice::NfpDevice(Core::HID::NpadIdType npad_id_, Core::System& system_,
 }
 
 NfpDevice::~NfpDevice() {
+    activate_event->Close();
+    deactivate_event->Close();
     if (!is_controller_set) {
         return;
     }

src/core/hle/service/nfp/nfp_device.h

@@ -3,11 +3,10 @@
 
 #pragma once
 
-#include <array>
 #include <span>
 #include <vector>
 
-#include "common/common_funcs.h"
+#include "common/common_types.h"
 #include "core/hle/service/kernel_helpers.h"
 #include "core/hle/service/nfp/nfp_types.h"
 #include "core/hle/service/service.h"

src/core/hle/service/nfp/nfp_user.cpp

@@ -1,9 +1,6 @@
 // SPDX-FileCopyrightText: Copyright 2018 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
-#include <array>
-#include <atomic>
-
 #include "common/logging/log.h"
 #include "core/core.h"
 #include "core/hid/hid_types.h"
@@ -55,8 +52,12 @@ IUser::IUser(Core::System& system_)
     }
 }
 
+IUser ::~IUser() {
+    availability_change_event->Close();
+}
+
 void IUser::Initialize(Kernel::HLERequestContext& ctx) {
-    LOG_INFO(Service_NFC, "called");
+    LOG_INFO(Service_NFP, "called");
 
     state = State::Initialized;
 
@@ -64,7 +65,7 @@ void IUser::Initialize(Kernel::HLERequestContext& ctx) {
         device->Initialize();
     }
 
-    IPC::ResponseBuilder rb{ctx, 2, 0};
+    IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(ResultSuccess);
 }
 
@@ -103,9 +104,9 @@ void IUser::ListDevices(Kernel::HLERequestContext& ctx) {
     }
 
     std::vector<u64> nfp_devices;
-    const std::size_t max_allowed_devices = ctx.GetWriteBufferSize() / sizeof(u64);
+    const std::size_t max_allowed_devices = ctx.GetWriteBufferNumElements<u64>();
 
-    for (auto& device : devices) {
+    for (const auto& device : devices) {
         if (nfp_devices.size() >= max_allowed_devices) {
             continue;
         }
@@ -114,7 +115,7 @@ void IUser::ListDevices(Kernel::HLERequestContext& ctx) {
         }
     }
 
-    if (nfp_devices.size() == 0) {
+    if (nfp_devices.empty()) {
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(DeviceNotFound);
         return;
@@ -551,9 +552,9 @@ void IUser::AttachDeactivateEvent(Kernel::HLERequestContext& ctx) {
 }
 
 void IUser::GetState(Kernel::HLERequestContext& ctx) {
-    LOG_DEBUG(Service_NFC, "called");
+    LOG_DEBUG(Service_NFP, "called");
 
-    IPC::ResponseBuilder rb{ctx, 3, 0};
+    IPC::ResponseBuilder rb{ctx, 3};
     rb.Push(ResultSuccess);
     rb.PushEnum(state);
 }

src/core/hle/service/nfp/nfp_user.h

@@ -3,6 +3,10 @@
 
 #pragma once
 
+#include <array>
+#include <memory>
+#include <optional>
+
 #include "core/hle/service/kernel_helpers.h"
 #include "core/hle/service/service.h"
 
@@ -12,6 +16,7 @@ class NfpDevice;
 class IUser final : public ServiceFramework<IUser> {
 public:
     explicit IUser(Core::System& system_);
+    ~IUser();
 
 private:
     enum class State : u32 {

src/core/hle/service/ns/iplatform_service_manager.cpp

@@ -279,13 +279,10 @@ void IPlatformServiceManager::GetSharedFontInOrderOfPriority(Kernel::HLERequestC
         font_sizes.push_back(region.size);
     }
 
-    // Resize buffers if game requests smaller size output.
-    font_codes.resize(
-        std::min<std::size_t>(font_codes.size(), ctx.GetWriteBufferSize(0) / sizeof(u32)));
-    font_offsets.resize(
-        std::min<std::size_t>(font_offsets.size(), ctx.GetWriteBufferSize(1) / sizeof(u32)));
-    font_sizes.resize(
-        std::min<std::size_t>(font_sizes.size(), ctx.GetWriteBufferSize(2) / sizeof(u32)));
+    // Resize buffers if game requests smaller size output
+    font_codes.resize(std::min(font_codes.size(), ctx.GetWriteBufferNumElements<u32>(0)));
+    font_offsets.resize(std::min(font_offsets.size(), ctx.GetWriteBufferNumElements<u32>(1)));
+    font_sizes.resize(std::min(font_sizes.size(), ctx.GetWriteBufferNumElements<u32>(2)));
 
     ctx.WriteBuffer(font_codes, 0);
     ctx.WriteBuffer(font_offsets, 1);

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

@@ -300,11 +300,10 @@ Kernel::KEvent* nvhost_ctrl_gpu::QueryEvent(u32 event_id) {
         return error_notifier_event;
     case 2:
         return unknown_event;
-    default: {
+    default:
         LOG_CRITICAL(Service_NVDRV, "Unknown Ctrl GPU Event {}", event_id);
+        return nullptr;
     }
-    }
-    return nullptr;
 }
 
 } // namespace Service::Nvidia::Devices

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

@@ -364,11 +364,10 @@ Kernel::KEvent* nvhost_gpu::QueryEvent(u32 event_id) {
         return sm_exception_breakpoint_pause_report_event;
     case 3:
         return error_notifier_event;
-    default: {
+    default:
         LOG_CRITICAL(Service_NVDRV, "Unknown Ctrl GPU Event {}", event_id);
+        return nullptr;
     }
-    }
-    return nullptr;
 }
 
 } // namespace Service::Nvidia::Devices

src/core/hle/service/nvflinger/buffer_queue_core.cpp

@@ -23,15 +23,17 @@ void BufferQueueCore::NotifyShutdown() {
 }
 
 void BufferQueueCore::SignalDequeueCondition() {
+    dequeue_possible.store(true);
     dequeue_condition.notify_all();
 }
 
-bool BufferQueueCore::WaitForDequeueCondition() {
+bool BufferQueueCore::WaitForDequeueCondition(std::unique_lock<std::mutex>& lk) {
     if (is_shutting_down) {
         return false;
     }
 
-    dequeue_condition.wait(mutex);
+    dequeue_condition.wait(lk, [&] { return dequeue_possible.load(); });
+    dequeue_possible.store(false);
 
     return true;
 }

src/core/hle/service/nvflinger/buffer_queue_core.h

@@ -38,7 +38,7 @@ public:
 
 private:
     void SignalDequeueCondition();
-    bool WaitForDequeueCondition();
+    bool WaitForDequeueCondition(std::unique_lock<std::mutex>& lk);
 
     s32 GetMinUndequeuedBufferCountLocked(bool async) const;
     s32 GetMinMaxBufferCountLocked(bool async) const;
@@ -60,7 +60,8 @@ private:
     BufferQueueDefs::SlotsType slots{};
     std::vector<BufferItem> queue;
     s32 override_max_buffer_count{};
-    mutable std::condition_variable_any dequeue_condition;
+    std::condition_variable dequeue_condition;
+    std::atomic<bool> dequeue_possible{};
     const bool use_async_buffer{}; // This is always disabled on HOS
     bool dequeue_buffer_cannot_block{};
     PixelFormat default_buffer_format{PixelFormat::Rgba8888};

src/core/hle/service/nvflinger/buffer_queue_producer.cpp

@@ -121,8 +121,8 @@ Status BufferQueueProducer::SetBufferCount(s32 buffer_count) {
     return Status::NoError;
 }
 
-Status BufferQueueProducer::WaitForFreeSlotThenRelock(bool async, s32* found,
-                                                      Status* return_flags) const {
+Status BufferQueueProducer::WaitForFreeSlotThenRelock(bool async, s32* found, Status* return_flags,
+                                                      std::unique_lock<std::mutex>& lk) const {
     bool try_again = true;
 
     while (try_again) {
@@ -214,7 +214,7 @@ Status BufferQueueProducer::WaitForFreeSlotThenRelock(bool async, s32* found,
                 return Status::WouldBlock;
             }
 
-            if (!core->WaitForDequeueCondition()) {
+            if (!core->WaitForDequeueCondition(lk)) {
                 // We are no longer running
                 return Status::NoError;
             }
@@ -237,7 +237,7 @@ Status BufferQueueProducer::DequeueBuffer(s32* out_slot, Fence* out_fence, bool
     Status return_flags = Status::NoError;
     bool attached_by_consumer = false;
     {
-        std::scoped_lock lock{core->mutex};
+        std::unique_lock lock{core->mutex};
         core->WaitWhileAllocatingLocked();
 
         if (format == PixelFormat::NoFormat) {
@@ -248,7 +248,7 @@ Status BufferQueueProducer::DequeueBuffer(s32* out_slot, Fence* out_fence, bool
         usage |= core->consumer_usage_bit;
 
         s32 found{};
-        Status status = WaitForFreeSlotThenRelock(async, &found, &return_flags);
+        Status status = WaitForFreeSlotThenRelock(async, &found, &return_flags, lock);
         if (status != Status::NoError) {
             return status;
         }
@@ -400,13 +400,13 @@ Status BufferQueueProducer::AttachBuffer(s32* out_slot,
         return Status::BadValue;
     }
 
-    std::scoped_lock lock{core->mutex};
+    std::unique_lock lock{core->mutex};
     core->WaitWhileAllocatingLocked();
 
     Status return_flags = Status::NoError;
     s32 found{};
 
-    const auto status = WaitForFreeSlotThenRelock(false, &found, &return_flags);
+    const auto status = WaitForFreeSlotThenRelock(false, &found, &return_flags, lock);
     if (status != Status::NoError) {
         return status;
     }

src/core/hle/service/nvflinger/buffer_queue_producer.h

@@ -70,7 +70,8 @@ public:
 private:
     BufferQueueProducer(const BufferQueueProducer&) = delete;
 
-    Status WaitForFreeSlotThenRelock(bool async, s32* found, Status* return_flags) const;
+    Status WaitForFreeSlotThenRelock(bool async, s32* found, Status* return_flags,
+                                     std::unique_lock<std::mutex>& lk) const;
 
     Kernel::KEvent* buffer_wait_event{};
     Service::KernelHelpers::ServiceContext& service_context;

src/core/hle/service/service.cpp

@@ -228,6 +228,7 @@ Result ServiceFrameworkBase::HandleSyncRequest(Kernel::KServerSession& session,
         }
 
         UNIMPLEMENTED_MSG("command_type={}", ctx.GetCommandType());
+        break;
     }
 
     // If emulation was shutdown, we are closing service threads, do not write the response back to

src/core/hle/service/set/set.cpp

@@ -83,7 +83,7 @@ void PushResponseLanguageCode(Kernel::HLERequestContext& ctx, std::size_t num_la
 }
 
 void GetAvailableLanguageCodesImpl(Kernel::HLERequestContext& ctx, std::size_t max_entries) {
-    const std::size_t requested_amount = ctx.GetWriteBufferSize() / sizeof(LanguageCode);
+    const std::size_t requested_amount = ctx.GetWriteBufferNumElements<LanguageCode>();
     const std::size_t max_amount = std::min(requested_amount, max_entries);
     const std::size_t copy_amount = std::min(available_language_codes.size(), max_amount);
     const std::size_t copy_size = copy_amount * sizeof(LanguageCode);

src/core/hle/service/time/time_zone_manager.cpp

@@ -280,6 +280,7 @@ static constexpr int TransitionTime(int year, Rule rule, int offset) {
     }
     default:
         ASSERT(false);
+        break;
     }
     return value + rule.transition_time + offset;
 }

src/core/memory.cpp

@@ -6,6 +6,7 @@
 
 #include "common/assert.h"
 #include "common/atomic_ops.h"
+#include "common/cache_management.h"
 #include "common/common_types.h"
 #include "common/logging/log.h"
 #include "common/page_table.h"
@@ -329,6 +330,55 @@ struct Memory::Impl {
             });
     }
 
+    template <typename Callback>
+    Result PerformCacheOperation(const Kernel::KProcess& process, VAddr dest_addr, std::size_t size,
+                                 Callback&& cb) {
+        class InvalidMemoryException : public std::exception {};
+
+        try {
+            WalkBlock(
+                process, dest_addr, size,
+                [&](const std::size_t block_size, const VAddr current_vaddr) {
+                    LOG_ERROR(HW_Memory, "Unmapped cache maintenance @ {:#018X}", current_vaddr);
+                    throw InvalidMemoryException();
+                },
+                [&](const std::size_t block_size, u8* const host_ptr) { cb(block_size, host_ptr); },
+                [&](const VAddr current_vaddr, const std::size_t block_size, u8* const host_ptr) {
+                    system.GPU().FlushRegion(current_vaddr, block_size);
+                    cb(block_size, host_ptr);
+                },
+                [](const std::size_t block_size) {});
+        } catch (InvalidMemoryException&) {
+            return Kernel::ResultInvalidCurrentMemory;
+        }
+
+        return ResultSuccess;
+    }
+
+    Result InvalidateDataCache(const Kernel::KProcess& process, VAddr dest_addr, std::size_t size) {
+        auto perform = [&](const std::size_t block_size, u8* const host_ptr) {
+            // Do nothing; this operation (dc ivac) cannot be supported
+            // from EL0
+        };
+        return PerformCacheOperation(process, dest_addr, size, perform);
+    }
+
+    Result StoreDataCache(const Kernel::KProcess& process, VAddr dest_addr, std::size_t size) {
+        auto perform = [&](const std::size_t block_size, u8* const host_ptr) {
+            // dc cvac: Store to point of coherency
+            Common::DataCacheLineCleanByVAToPoC(host_ptr, block_size);
+        };
+        return PerformCacheOperation(process, dest_addr, size, perform);
+    }
+
+    Result FlushDataCache(const Kernel::KProcess& process, VAddr dest_addr, std::size_t size) {
+        auto perform = [&](const std::size_t block_size, u8* const host_ptr) {
+            // dc civac: Store to point of coherency, and invalidate from cache
+            Common::DataCacheLineCleanAndInvalidateByVAToPoC(host_ptr, block_size);
+        };
+        return PerformCacheOperation(process, dest_addr, size, perform);
+    }
+
     void MarkRegionDebug(VAddr vaddr, u64 size, bool debug) {
         if (vaddr == 0) {
             return;
@@ -786,6 +836,21 @@ void Memory::ZeroBlock(const Kernel::KProcess& process, VAddr dest_addr, const s
     impl->ZeroBlock(process, dest_addr, size);
 }
 
+Result Memory::InvalidateDataCache(const Kernel::KProcess& process, VAddr dest_addr,
+                                   const std::size_t size) {
+    return impl->InvalidateDataCache(process, dest_addr, size);
+}
+
+Result Memory::StoreDataCache(const Kernel::KProcess& process, VAddr dest_addr,
+                              const std::size_t size) {
+    return impl->StoreDataCache(process, dest_addr, size);
+}
+
+Result Memory::FlushDataCache(const Kernel::KProcess& process, VAddr dest_addr,
+                              const std::size_t size) {
+    return impl->FlushDataCache(process, dest_addr, size);
+}
+
 void Memory::RasterizerMarkRegionCached(VAddr vaddr, u64 size, bool cached) {
     impl->RasterizerMarkRegionCached(vaddr, size, cached);
 }

src/core/memory.h

@@ -7,6 +7,7 @@
 #include <memory>
 #include <string>
 #include "common/common_types.h"
+#include "core/hle/result.h"
 
 namespace Common {
 struct PageTable;
@@ -449,6 +450,39 @@ public:
      */
     void ZeroBlock(const Kernel::KProcess& process, VAddr dest_addr, std::size_t size);
 
+    /**
+     * Invalidates a range of bytes within the current process' address space at the specified
+     * virtual address.
+     *
+     * @param process   The process that will have data invalidated within its address space.
+     * @param dest_addr The destination virtual address to invalidate the data from.
+     * @param size      The size of the range to invalidate, in bytes.
+     *
+     */
+    Result InvalidateDataCache(const Kernel::KProcess& process, VAddr dest_addr, std::size_t size);
+
+    /**
+     * Stores a range of bytes within the current process' address space at the specified
+     * virtual address.
+     *
+     * @param process   The process that will have data stored within its address space.
+     * @param dest_addr The destination virtual address to store the data from.
+     * @param size      The size of the range to store, in bytes.
+     *
+     */
+    Result StoreDataCache(const Kernel::KProcess& process, VAddr dest_addr, std::size_t size);
+
+    /**
+     * Flushes a range of bytes within the current process' address space at the specified
+     * virtual address.
+     *
+     * @param process   The process that will have data flushed within its address space.
+     * @param dest_addr The destination virtual address to flush the data from.
+     * @param size      The size of the range to flush, in bytes.
+     *
+     */
+    Result FlushDataCache(const Kernel::KProcess& process, VAddr dest_addr, std::size_t size);
+
     /**
      * Marks each page within the specified address range as cached or uncached.
      *

src/core/memory/dmnt_cheat_vm.cpp

@@ -761,7 +761,7 @@ void DmntCheatVm::Execute(const CheatProcessMetadata& metadata) {
             u64 src_address =
                 GetCheatProcessAddress(metadata, begin_cond->mem_type, begin_cond->rel_address);
             u64 src_value = 0;
-            switch (store_static->bit_width) {
+            switch (begin_cond->bit_width) {
             case 1:
             case 2:
             case 4:

src/input_common/helpers/touch_from_buttons.cpp

@@ -10,8 +10,8 @@ namespace InputCommon {
 class TouchFromButtonDevice final : public Common::Input::InputDevice {
 public:
     using Button = std::unique_ptr<Common::Input::InputDevice>;
-    TouchFromButtonDevice(Button button_, int touch_id_, float x_, float y_)
-        : button(std::move(button_)), touch_id(touch_id_), x(x_), y(y_) {
+    TouchFromButtonDevice(Button button_, float x_, float y_)
+        : button(std::move(button_)), x(x_), y(y_) {
         last_button_value = false;
         button->SetCallback({
             .on_change =
@@ -34,7 +34,6 @@ public:
             .pressed = button_status,
             .x = {},
             .y = {},
-            .id = touch_id,
         };
         status.x.properties = properties;
         status.y.properties = properties;
@@ -62,7 +61,6 @@ public:
 private:
     Button button;
     bool last_button_value;
-    const int touch_id;
     const float x;
     const float y;
     const Common::Input::AnalogProperties properties{0.0f, 1.0f, 0.5f, 0.0f, false};
@@ -73,10 +71,9 @@ std::unique_ptr<Common::Input::InputDevice> TouchFromButton::Create(
     const std::string null_engine = Common::ParamPackage{{"engine", "null"}}.Serialize();
     auto button = Common::Input::CreateDeviceFromString<Common::Input::InputDevice>(
         params.Get("button", null_engine));
-    const auto touch_id = params.Get("touch_id", 0);
     const float x = params.Get("x", 0.0f) / 1280.0f;
     const float y = params.Get("y", 0.0f) / 720.0f;
-    return std::make_unique<TouchFromButtonDevice>(std::move(button), touch_id, x, y);
+    return std::make_unique<TouchFromButtonDevice>(std::move(button), x, y);
 }
 
 } // namespace InputCommon

src/input_common/input_poller.cpp

@@ -229,13 +229,12 @@ private:
 
 class InputFromTouch final : public Common::Input::InputDevice {
 public:
-    explicit InputFromTouch(PadIdentifier identifier_, int touch_id_, int button_, bool toggle_,
-                            bool inverted_, int axis_x_, int axis_y_,
-                            Common::Input::AnalogProperties properties_x_,
+    explicit InputFromTouch(PadIdentifier identifier_, int button_, bool toggle_, bool inverted_,
+                            int axis_x_, int axis_y_, Common::Input::AnalogProperties properties_x_,
                             Common::Input::AnalogProperties properties_y_,
                             InputEngine* input_engine_)
-        : identifier(identifier_), touch_id(touch_id_), button(button_), toggle(toggle_),
-          inverted(inverted_), axis_x(axis_x_), axis_y(axis_y_), properties_x(properties_x_),
+        : identifier(identifier_), button(button_), toggle(toggle_), inverted(inverted_),
+          axis_x(axis_x_), axis_y(axis_y_), properties_x(properties_x_),
           properties_y(properties_y_), input_engine(input_engine_) {
         UpdateCallback engine_callback{[this]() { OnChange(); }};
         const InputIdentifier button_input_identifier{
@@ -271,8 +270,7 @@ public:
     }
 
     Common::Input::TouchStatus GetStatus() const {
-        Common::Input::TouchStatus status;
-        status.id = touch_id;
+        Common::Input::TouchStatus status{};
         status.pressed = {
             .value = input_engine->GetButton(identifier, button),
             .inverted = inverted,
@@ -307,7 +305,6 @@ public:
 
 private:
     const PadIdentifier identifier;
-    const int touch_id;
     const int button;
     const bool toggle;
     const bool inverted;
@@ -919,7 +916,6 @@ std::unique_ptr<Common::Input::InputDevice> InputFactory::CreateTriggerDevice(
 
 std::unique_ptr<Common::Input::InputDevice> InputFactory::CreateTouchDevice(
     const Common::ParamPackage& params) {
-    const auto touch_id = params.Get("touch_id", 0);
     const auto deadzone = std::clamp(params.Get("deadzone", 0.0f), 0.0f, 1.0f);
     const auto range = std::clamp(params.Get("range", 1.0f), 0.25f, 1.50f);
     const auto threshold = std::clamp(params.Get("threshold", 0.5f), 0.0f, 1.0f);
@@ -954,8 +950,8 @@ std::unique_ptr<Common::Input::InputDevice> InputFactory::CreateTouchDevice(
     input_engine->PreSetAxis(identifier, axis_x);
     input_engine->PreSetAxis(identifier, axis_y);
     input_engine->PreSetButton(identifier, button);
-    return std::make_unique<InputFromTouch>(identifier, touch_id, button, toggle, inverted, axis_x,
-                                            axis_y, properties_x, properties_y, input_engine.get());
+    return std::make_unique<InputFromTouch>(identifier, button, toggle, inverted, axis_x, axis_y,
+                                            properties_x, properties_y, input_engine.get());
 }
 
 std::unique_ptr<Common::Input::InputDevice> InputFactory::CreateBatteryDevice(

src/shader_recompiler/backend/glasm/emit_glasm.cpp

@@ -320,6 +320,7 @@ void SetupOptions(const IR::Program& program, const Profile& profile,
     }
     if (stage == Stage::Fragment) {
         header += "OPTION ARB_draw_buffers;";
+        header += "OPTION ARB_fragment_layer_viewport;";
     }
 }
 

src/shader_recompiler/backend/glasm/emit_glasm_context_get_set.cpp

@@ -104,6 +104,9 @@ void EmitGetAttribute(EmitContext& ctx, IR::Inst& inst, IR::Attribute attr, Scal
     case IR::Attribute::PrimitiveId:
         ctx.Add("MOV.F {}.x,primitive.id;", inst);
         break;
+    case IR::Attribute::Layer:
+        ctx.Add("MOV.F {}.x,fragment.layer;", inst);
+        break;
     case IR::Attribute::PositionX:
     case IR::Attribute::PositionY:
     case IR::Attribute::PositionZ:

src/shader_recompiler/backend/glsl/emit_glsl_context_get_set.cpp

@@ -205,6 +205,9 @@ void EmitGetAttribute(EmitContext& ctx, IR::Inst& inst, IR::Attribute attr,
     case IR::Attribute::PrimitiveId:
         ctx.AddF32("{}=itof(gl_PrimitiveID);", inst);
         break;
+    case IR::Attribute::Layer:
+        ctx.AddF32("{}=itof(gl_Layer);", inst);
+        break;
     case IR::Attribute::PositionX:
     case IR::Attribute::PositionY:
     case IR::Attribute::PositionZ:

src/shader_recompiler/backend/spirv/emit_spirv_context_get_set.cpp

@@ -315,6 +315,8 @@ Id EmitGetAttribute(EmitContext& ctx, IR::Attribute attr, Id vertex) {
     switch (attr) {
     case IR::Attribute::PrimitiveId:
         return ctx.OpBitcast(ctx.F32[1], ctx.OpLoad(ctx.U32[1], ctx.primitive_id));
+    case IR::Attribute::Layer:
+        return ctx.OpBitcast(ctx.F32[1], ctx.OpLoad(ctx.U32[1], ctx.layer));
     case IR::Attribute::PositionX:
     case IR::Attribute::PositionY:
     case IR::Attribute::PositionZ:

src/shader_recompiler/backend/spirv/spirv_emit_context.cpp

@@ -1359,6 +1359,11 @@ void EmitContext::DefineInputs(const IR::Program& program) {
     if (loads[IR::Attribute::PrimitiveId]) {
         primitive_id = DefineInput(*this, U32[1], false, spv::BuiltIn::PrimitiveId);
     }
+    if (loads[IR::Attribute::Layer]) {
+        AddCapability(spv::Capability::Geometry);
+        layer = DefineInput(*this, U32[1], false, spv::BuiltIn::Layer);
+        Decorate(layer, spv::Decoration::Flat);
+    }
     if (loads.AnyComponent(IR::Attribute::PositionX)) {
         const bool is_fragment{stage != Stage::Fragment};
         const spv::BuiltIn built_in{is_fragment ? spv::BuiltIn::Position : spv::BuiltIn::FragCoord};

src/video_core/CMakeLists.txt

@@ -28,6 +28,10 @@ add_library(video_core STATIC
     dirty_flags.h
     dma_pusher.cpp
     dma_pusher.h
+    engines/sw_blitter/blitter.cpp
+    engines/sw_blitter/blitter.h
+    engines/sw_blitter/converter.cpp
+    engines/sw_blitter/converter.h
     engines/const_buffer_info.h
     engines/engine_interface.h
     engines/engine_upload.cpp

src/video_core/buffer_cache/buffer_cache.h

@@ -992,7 +992,20 @@ void BufferCache<P>::BindHostIndexBuffer() {
     TouchBuffer(buffer, index_buffer.buffer_id);
     const u32 offset = buffer.Offset(index_buffer.cpu_addr);
     const u32 size = index_buffer.size;
-    SynchronizeBuffer(buffer, index_buffer.cpu_addr, size);
+    if (maxwell3d->inline_index_draw_indexes.size()) {
+        if constexpr (USE_MEMORY_MAPS) {
+            auto upload_staging = runtime.UploadStagingBuffer(size);
+            std::array<BufferCopy, 1> copies{
+                {BufferCopy{.src_offset = upload_staging.offset, .dst_offset = 0, .size = size}}};
+            std::memcpy(upload_staging.mapped_span.data(),
+                        maxwell3d->inline_index_draw_indexes.data(), size);
+            runtime.CopyBuffer(buffer, upload_staging.buffer, copies);
+        } else {
+            buffer.ImmediateUpload(0, maxwell3d->inline_index_draw_indexes);
+        }
+    } else {
+        SynchronizeBuffer(buffer, index_buffer.cpu_addr, size);
+    }
     if constexpr (HAS_FULL_INDEX_AND_PRIMITIVE_SUPPORT) {
         const u32 new_offset = offset + maxwell3d->regs.index_buffer.first *
                                             maxwell3d->regs.index_buffer.FormatSizeInBytes();
@@ -1275,7 +1288,15 @@ void BufferCache<P>::UpdateIndexBuffer() {
     }
     flags[Dirty::IndexBuffer] = false;
     last_index_count = index_array.count;
-
+    if (maxwell3d->inline_index_draw_indexes.size()) {
+        auto inline_index_size = static_cast<u32>(maxwell3d->inline_index_draw_indexes.size());
+        index_buffer = Binding{
+            .cpu_addr = 0,
+            .size = inline_index_size,
+            .buffer_id = CreateBuffer(0, inline_index_size),
+        };
+        return;
+    }
     const GPUVAddr gpu_addr_begin = index_array.StartAddress();
     const GPUVAddr gpu_addr_end = index_array.EndAddress();
     const std::optional<VAddr> cpu_addr = gpu_memory->GpuToCpuAddress(gpu_addr_begin);
@@ -1491,6 +1512,14 @@ typename BufferCache<P>::OverlapResult BufferCache<P>::ResolveOverlaps(VAddr cpu
     VAddr end = cpu_addr + wanted_size;
     int stream_score = 0;
     bool has_stream_leap = false;
+    if (begin == 0) {
+        return OverlapResult{
+            .ids = std::move(overlap_ids),
+            .begin = begin,
+            .end = end,
+            .has_stream_leap = has_stream_leap,
+        };
+    }
     for (; cpu_addr >> YUZU_PAGEBITS < Common::DivCeil(end, YUZU_PAGESIZE);
          cpu_addr += YUZU_PAGESIZE) {
         const BufferId overlap_id = page_table[cpu_addr >> YUZU_PAGEBITS];
@@ -1713,12 +1742,12 @@ bool BufferCache<P>::InlineMemory(VAddr dest_address, size_t copy_size,
     SynchronizeBuffer(buffer, dest_address, static_cast<u32>(copy_size));
 
     if constexpr (USE_MEMORY_MAPS) {
+        auto upload_staging = runtime.UploadStagingBuffer(copy_size);
         std::array copies{BufferCopy{
-            .src_offset = 0,
+            .src_offset = upload_staging.offset,
             .dst_offset = buffer.Offset(dest_address),
             .size = copy_size,
         }};
-        auto upload_staging = runtime.UploadStagingBuffer(copy_size);
         u8* const src_pointer = upload_staging.mapped_span.data();
         std::memcpy(src_pointer, inlined_buffer.data(), copy_size);
         runtime.CopyBuffer(buffer, upload_staging.buffer, copies);

src/video_core/control/channel_state.cpp

@@ -20,7 +20,7 @@ void ChannelState::Init(Core::System& system, GPU& gpu) {
     ASSERT(memory_manager);
     dma_pusher = std::make_unique<Tegra::DmaPusher>(system, gpu, *memory_manager, *this);
     maxwell_3d = std::make_unique<Engines::Maxwell3D>(system, *memory_manager);
-    fermi_2d = std::make_unique<Engines::Fermi2D>();
+    fermi_2d = std::make_unique<Engines::Fermi2D>(*memory_manager);
     kepler_compute = std::make_unique<Engines::KeplerCompute>(system, *memory_manager);
     maxwell_dma = std::make_unique<Engines::MaxwellDMA>(system, *memory_manager);
     kepler_memory = std::make_unique<Engines::KeplerMemory>(system, *memory_manager);

src/video_core/engines/engine_upload.cpp

@@ -51,11 +51,11 @@ void State::ProcessData(std::span<const u8> read_buffer) {
         } else {
             for (u32 line = 0; line < regs.line_count; ++line) {
                 const GPUVAddr dest_line = address + static_cast<size_t>(line) * regs.dest.pitch;
-                memory_manager.WriteBlockUnsafe(
-                    dest_line, read_buffer.data() + static_cast<size_t>(line) * regs.line_length_in,
-                    regs.line_length_in);
+                std::span<const u8> buffer(read_buffer.data() +
+                                               static_cast<size_t>(line) * regs.line_length_in,
+                                           regs.line_length_in);
+                rasterizer->AccelerateInlineToMemory(dest_line, regs.line_length_in, buffer);
             }
-            memory_manager.InvalidateRegion(address, regs.dest.pitch * regs.line_count);
         }
     } else {
         u32 width = regs.dest.width;

src/video_core/engines/fermi_2d.cpp

@@ -3,17 +3,25 @@
 
 #include "common/assert.h"
 #include "common/logging/log.h"
+#include "common/microprofile.h"
 #include "video_core/engines/fermi_2d.h"
-#include "video_core/memory_manager.h"
+#include "video_core/engines/sw_blitter/blitter.h"
 #include "video_core/rasterizer_interface.h"
 #include "video_core/surface.h"
+#include "video_core/textures/decoders.h"
+
+MICROPROFILE_DECLARE(GPU_BlitEngine);
+MICROPROFILE_DEFINE(GPU_BlitEngine, "GPU", "Blit Engine", MP_RGB(224, 224, 128));
 
 using VideoCore::Surface::BytesPerBlock;
 using VideoCore::Surface::PixelFormatFromRenderTargetFormat;
 
 namespace Tegra::Engines {
 
-Fermi2D::Fermi2D() {
+using namespace Texture;
+
+Fermi2D::Fermi2D(MemoryManager& memory_manager_) {
+    sw_blitter = std::make_unique<Blitter::SoftwareBlitEngine>(memory_manager_);
     // Nvidia's OpenGL driver seems to assume these values
     regs.src.depth = 1;
     regs.dst.depth = 1;
@@ -42,6 +50,7 @@ void Fermi2D::CallMultiMethod(u32 method, const u32* base_start, u32 amount, u32
 }
 
 void Fermi2D::Blit() {
+    MICROPROFILE_SCOPE(GPU_BlitEngine);
     LOG_DEBUG(HW_GPU, "called. source address=0x{:x}, destination address=0x{:x}",
               regs.src.Address(), regs.dst.Address());
 
@@ -52,9 +61,16 @@ void Fermi2D::Blit() {
     UNIMPLEMENTED_IF_MSG(regs.clip_enable != 0, "Clipped blit enabled");
 
     const auto& args = regs.pixels_from_memory;
+    constexpr s64 null_derivate = 1ULL << 32;
+    Surface src = regs.src;
+    const auto bytes_per_pixel = BytesPerBlock(PixelFormatFromRenderTargetFormat(src.format));
+    const bool delegate_to_gpu = src.width > 512 && src.height > 512 && bytes_per_pixel <= 8 &&
+                                 src.format != regs.dst.format;
     Config config{
         .operation = regs.operation,
         .filter = args.sample_mode.filter,
+        .must_accelerate =
+            args.du_dx != null_derivate || args.dv_dy != null_derivate || delegate_to_gpu,
         .dst_x0 = args.dst_x0,
         .dst_y0 = args.dst_y0,
         .dst_x1 = args.dst_x0 + args.dst_width,
@@ -64,8 +80,7 @@ void Fermi2D::Blit() {
         .src_x1 = static_cast<s32>((args.du_dx * args.dst_width + args.src_x0) >> 32),
         .src_y1 = static_cast<s32>((args.dv_dy * args.dst_height + args.src_y0) >> 32),
     };
-    Surface src = regs.src;
-    const auto bytes_per_pixel = BytesPerBlock(PixelFormatFromRenderTargetFormat(src.format));
+
     const auto need_align_to_pitch =
         src.linear == Tegra::Engines::Fermi2D::MemoryLayout::Pitch &&
         static_cast<s32>(src.width) == config.src_x1 &&
@@ -78,8 +93,9 @@ void Fermi2D::Blit() {
         config.src_x1 -= config.src_x0;
         config.src_x0 = 0;
     }
+
     if (!rasterizer->AccelerateSurfaceCopy(src, regs.dst, config)) {
-        UNIMPLEMENTED();
+        sw_blitter->Blit(src, regs.dst, config);
     }
 }
 

src/video_core/engines/fermi_2d.h

@@ -5,6 +5,7 @@
 
 #include <array>
 #include <cstddef>
+#include <memory>
 #include "common/bit_field.h"
 #include "common/common_funcs.h"
 #include "common/common_types.h"
@@ -21,6 +22,10 @@ class RasterizerInterface;
 
 namespace Tegra::Engines {
 
+namespace Blitter {
+class SoftwareBlitEngine;
+}
+
 /**
  * This Engine is known as G80_2D. Documentation can be found in:
  * https://github.com/envytools/envytools/blob/master/rnndb/graph/g80_2d.xml
@@ -32,7 +37,7 @@ namespace Tegra::Engines {
 
 class Fermi2D final : public EngineInterface {
 public:
-    explicit Fermi2D();
+    explicit Fermi2D(MemoryManager& memory_manager_);
     ~Fermi2D() override;
 
     /// Binds a rasterizer to this engine.
@@ -286,6 +291,7 @@ public:
     struct Config {
         Operation operation;
         Filter filter;
+        bool must_accelerate;
         s32 dst_x0;
         s32 dst_y0;
         s32 dst_x1;
@@ -298,6 +304,7 @@ public:
 
 private:
     VideoCore::RasterizerInterface* rasterizer = nullptr;
+    std::unique_ptr<Blitter::SoftwareBlitEngine> sw_blitter;
 
     /// Performs the copy from the source surface to the destination surface as configured in the
     /// registers.

src/video_core/engines/maxwell_3d.cpp

@@ -232,7 +232,7 @@ void Maxwell3D::ProcessMethodCall(u32 method, u32 argument, u32 nonshadow_argume
         use_topology_override = true;
         return;
     case MAXWELL3D_REG_INDEX(clear_surface):
-        return ProcessClearBuffers();
+        return ProcessClearBuffers(1);
     case MAXWELL3D_REG_INDEX(report_semaphore.query):
         return ProcessQueryGet();
     case MAXWELL3D_REG_INDEX(render_enable.mode):
@@ -249,9 +249,6 @@ void Maxwell3D::ProcessMethodCall(u32 method, u32 argument, u32 nonshadow_argume
         return;
     case MAXWELL3D_REG_INDEX(fragment_barrier):
         return rasterizer->FragmentBarrier();
-    case MAXWELL3D_REG_INDEX(invalidate_texture_data_cache):
-        rasterizer->InvalidateGPUCache();
-        return rasterizer->WaitForIdle();
     case MAXWELL3D_REG_INDEX(tiled_cache_barrier):
         return rasterizer->TiledCacheBarrier();
     }
@@ -511,10 +508,7 @@ void Maxwell3D::ProcessCounterReset() {
 
 void Maxwell3D::ProcessSyncPoint() {
     const u32 sync_point = regs.sync_info.sync_point.Value();
-    const u32 cache_flush = regs.sync_info.clean_l2.Value();
-    if (cache_flush != 0) {
-        rasterizer->InvalidateGPUCache();
-    }
+    [[maybe_unused]] const u32 cache_flush = regs.sync_info.clean_l2.Value();
     rasterizer->SignalSyncPoint(sync_point);
 }
 
@@ -596,8 +590,8 @@ u32 Maxwell3D::GetRegisterValue(u32 method) const {
     return regs.reg_array[method];
 }
 
-void Maxwell3D::ProcessClearBuffers() {
-    rasterizer->Clear();
+void Maxwell3D::ProcessClearBuffers(u32 layer_count) {
+    rasterizer->Clear(layer_count);
 }
 
 void Maxwell3D::ProcessDraw(u32 instance_count) {

src/video_core/engines/maxwell_3d.h

@@ -3086,6 +3086,9 @@ public:
 
     std::vector<u8> inline_index_draw_indexes;
 
+    /// Handles a write to the CLEAR_BUFFERS register.
+    void ProcessClearBuffers(u32 layer_count);
+
 private:
     void InitializeRegisterDefaults();
 
@@ -3120,9 +3123,6 @@ private:
     /// Handles firmware blob 4
     void ProcessFirmwareCall4();
 
-    /// Handles a write to the CLEAR_BUFFERS register.
-    void ProcessClearBuffers();
-
     /// Handles a write to the QUERY_GET register.
     void ProcessQueryGet();
 

src/video_core/engines/maxwell_dma.cpp

@@ -62,7 +62,8 @@ void MaxwellDMA::Launch() {
 
         if (!is_src_pitch && !is_dst_pitch) {
             // If both the source and the destination are in block layout, assert.
-            UNIMPLEMENTED_MSG("Tiled->Tiled DMA transfers are not yet implemented");
+            CopyBlockLinearToBlockLinear();
+            ReleaseSemaphore();
             return;
         }
 
@@ -291,6 +292,70 @@ void MaxwellDMA::FastCopyBlockLinearToPitch() {
     memory_manager.WriteBlock(regs.offset_out, write_buffer.data(), dst_size);
 }
 
+void MaxwellDMA::CopyBlockLinearToBlockLinear() {
+    UNIMPLEMENTED_IF(regs.src_params.block_size.width != 0);
+
+    const bool is_remapping = regs.launch_dma.remap_enable != 0;
+
+    // Deswizzle the input and copy it over.
+    const Parameters& src = regs.src_params;
+    const Parameters& dst = regs.dst_params;
+
+    const u32 num_remap_components = regs.remap_const.num_dst_components_minus_one + 1;
+    const u32 remap_components_size = regs.remap_const.component_size_minus_one + 1;
+
+    const u32 base_bpp = !is_remapping ? 1U : num_remap_components * remap_components_size;
+
+    u32 src_width = src.width;
+    u32 dst_width = dst.width;
+    u32 x_elements = regs.line_length_in;
+    u32 src_x_offset = src.origin.x;
+    u32 dst_x_offset = dst.origin.x;
+    u32 bpp_shift = 0U;
+    if (!is_remapping) {
+        bpp_shift = Common::FoldRight(
+            4U, [](u32 x, u32 y) { return std::min(x, static_cast<u32>(std::countr_zero(y))); },
+            src_width, dst_width, x_elements, src_x_offset, dst_x_offset,
+            static_cast<u32>(regs.offset_in), static_cast<u32>(regs.offset_out));
+        src_width >>= bpp_shift;
+        dst_width >>= bpp_shift;
+        x_elements >>= bpp_shift;
+        src_x_offset >>= bpp_shift;
+        dst_x_offset >>= bpp_shift;
+    }
+
+    const u32 bytes_per_pixel = base_bpp << bpp_shift;
+    const size_t src_size = CalculateSize(true, bytes_per_pixel, src_width, src.height, src.depth,
+                                          src.block_size.height, src.block_size.depth);
+    const size_t dst_size = CalculateSize(true, bytes_per_pixel, dst_width, dst.height, dst.depth,
+                                          dst.block_size.height, dst.block_size.depth);
+
+    const u32 pitch = x_elements * bytes_per_pixel;
+    const size_t mid_buffer_size = pitch * regs.line_count;
+
+    if (read_buffer.size() < src_size) {
+        read_buffer.resize(src_size);
+    }
+    if (write_buffer.size() < dst_size) {
+        write_buffer.resize(dst_size);
+    }
+
+    intermediate_buffer.resize(mid_buffer_size);
+
+    memory_manager.ReadBlock(regs.offset_in, read_buffer.data(), src_size);
+    memory_manager.ReadBlock(regs.offset_out, write_buffer.data(), dst_size);
+
+    UnswizzleSubrect(intermediate_buffer, read_buffer, bytes_per_pixel, src_width, src.height,
+                     src.depth, src_x_offset, src.origin.y, x_elements, regs.line_count,
+                     src.block_size.height, src.block_size.depth, pitch);
+
+    SwizzleSubrect(write_buffer, intermediate_buffer, bytes_per_pixel, dst_width, dst.height,
+                   dst.depth, dst_x_offset, dst.origin.y, x_elements, regs.line_count,
+                   dst.block_size.height, dst.block_size.depth, pitch);
+
+    memory_manager.WriteBlock(regs.offset_out, write_buffer.data(), dst_size);
+}
+
 void MaxwellDMA::ReleaseSemaphore() {
     const auto type = regs.launch_dma.semaphore_type;
     const GPUVAddr address = regs.semaphore.address;
@@ -314,6 +379,7 @@ void MaxwellDMA::ReleaseSemaphore() {
     }
     default:
         ASSERT_MSG(false, "Unknown semaphore type: {}", static_cast<u32>(type.Value()));
+        break;
     }
 }
 

src/video_core/engines/maxwell_dma.h

@@ -223,6 +223,8 @@ private:
 
     void CopyPitchToBlockLinear();
 
+    void CopyBlockLinearToBlockLinear();
+
     void FastCopyBlockLinearToPitch();
 
     void ReleaseSemaphore();
@@ -234,6 +236,7 @@ private:
 
     std::vector<u8> read_buffer;
     std::vector<u8> write_buffer;
+    std::vector<u8> intermediate_buffer;
 
     static constexpr std::size_t NUM_REGS = 0x800;
     struct Regs {

src/video_core/engines/puller.cpp

@@ -50,6 +50,7 @@ void Puller::ProcessBindMethod(const MethodCall& method_call) {
         break;
     default:
         UNIMPLEMENTED_MSG("Unimplemented engine {:04X}", engine_id);
+        break;
     }
 }
 
@@ -65,6 +66,7 @@ void Puller::ProcessFenceActionMethod() {
         break;
     default:
         UNIMPLEMENTED_MSG("Unimplemented operation {}", regs.fence_action.op.Value());
+        break;
     }
 }
 
@@ -116,7 +118,7 @@ void Puller::ProcessSemaphoreRelease() {
     std::function<void()> operation([this, sequence_address, payload] {
         memory_manager.Write<u32>(sequence_address, payload);
     });
-    rasterizer->SyncOperation(std::move(operation));
+    rasterizer->SignalFence(std::move(operation));
 }
 
 void Puller::ProcessSemaphoreAcquire() {
@@ -149,8 +151,8 @@ void Puller::CallPullerMethod(const MethodCall& method_call) {
     case BufferMethods::SemaphoreAddressLow:
     case BufferMethods::SemaphoreSequencePayload:
     case BufferMethods::SyncpointPayload:
-        break;
     case BufferMethods::WrcacheFlush:
+        break;
     case BufferMethods::RefCnt:
         rasterizer->SignalReference();
         break;
@@ -228,6 +230,7 @@ void Puller::CallEngineMethod(const MethodCall& method_call) {
         break;
     default:
         UNIMPLEMENTED_MSG("Unimplemented engine");
+        break;
     }
 }
 
@@ -254,6 +257,7 @@ void Puller::CallEngineMultiMethod(u32 method, u32 subchannel, const u32* base_s
         break;
     default:
         UNIMPLEMENTED_MSG("Unimplemented engine");
+        break;
     }
 }
 

src/video_core/engines/sw_blitter/blitter.cpp

@@ -0,0 +1,238 @@
+// SPDX-FileCopyrightText: Copyright 2022 yuzu Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
+#include <algorithm>
+#include <cmath>
+#include <vector>
+
+#include "video_core/engines/sw_blitter/blitter.h"
+#include "video_core/engines/sw_blitter/converter.h"
+#include "video_core/memory_manager.h"
+#include "video_core/surface.h"
+#include "video_core/textures/decoders.h"
+
+namespace Tegra {
+class MemoryManager;
+}
+
+using VideoCore::Surface::BytesPerBlock;
+using VideoCore::Surface::PixelFormatFromRenderTargetFormat;
+
+namespace Tegra::Engines::Blitter {
+
+using namespace Texture;
+
+namespace {
+
+constexpr size_t ir_components = 4;
+
+void NearestNeighbor(std::span<const u8> input, std::span<u8> output, u32 src_width, u32 src_height,
+                     u32 dst_width, u32 dst_height, size_t bpp) {
+    const size_t dx_du = std::llround((static_cast<f64>(src_width) / dst_width) * (1ULL << 32));
+    const size_t dy_dv = std::llround((static_cast<f64>(src_height) / dst_height) * (1ULL << 32));
+    size_t src_y = 0;
+    for (u32 y = 0; y < dst_height; y++) {
+        size_t src_x = 0;
+        for (u32 x = 0; x < dst_width; x++) {
+            const size_t read_from = ((src_y * src_width + src_x) >> 32) * bpp;
+            const size_t write_to = (y * dst_width + x) * bpp;
+
+            std::memcpy(&output[write_to], &input[read_from], bpp);
+            src_x += dx_du;
+        }
+        src_y += dy_dv;
+    }
+}
+
+void NearestNeighborFast(std::span<const f32> input, std::span<f32> output, u32 src_width,
+                         u32 src_height, u32 dst_width, u32 dst_height) {
+    const size_t dx_du = std::llround((static_cast<f64>(src_width) / dst_width) * (1ULL << 32));
+    const size_t dy_dv = std::llround((static_cast<f64>(src_height) / dst_height) * (1ULL << 32));
+    size_t src_y = 0;
+    for (u32 y = 0; y < dst_height; y++) {
+        size_t src_x = 0;
+        for (u32 x = 0; x < dst_width; x++) {
+            const size_t read_from = ((src_y * src_width + src_x) >> 32) * ir_components;
+            const size_t write_to = (y * dst_width + x) * ir_components;
+
+            std::memcpy(&output[write_to], &input[read_from], sizeof(f32) * ir_components);
+            src_x += dx_du;
+        }
+        src_y += dy_dv;
+    }
+}
+
+void Bilinear(std::span<const f32> input, std::span<f32> output, size_t src_width,
+              size_t src_height, size_t dst_width, size_t dst_height) {
+    const auto bilinear_sample = [](std::span<const f32> x0_y0, std::span<const f32> x1_y0,
+                                    std::span<const f32> x0_y1, std::span<const f32> x1_y1,
+                                    f32 weight_x, f32 weight_y) {
+        std::array<f32, ir_components> result{};
+        for (size_t i = 0; i < ir_components; i++) {
+            const f32 a = std::lerp(x0_y0[i], x1_y0[i], weight_x);
+            const f32 b = std::lerp(x0_y1[i], x1_y1[i], weight_x);
+            result[i] = std::lerp(a, b, weight_y);
+        }
+        return result;
+    };
+    const f32 dx_du =
+        dst_width > 1 ? static_cast<f32>(src_width - 1) / static_cast<f32>(dst_width - 1) : 0.f;
+    const f32 dy_dv =
+        dst_height > 1 ? static_cast<f32>(src_height - 1) / static_cast<f32>(dst_height - 1) : 0.f;
+    for (u32 y = 0; y < dst_height; y++) {
+        for (u32 x = 0; x < dst_width; x++) {
+            const f32 x_low = std::floor(static_cast<f32>(x) * dx_du);
+            const f32 y_low = std::floor(static_cast<f32>(y) * dy_dv);
+            const f32 x_high = std::ceil(static_cast<f32>(x) * dx_du);
+            const f32 y_high = std::ceil(static_cast<f32>(y) * dy_dv);
+            const f32 weight_x = (static_cast<f32>(x) * dx_du) - x_low;
+            const f32 weight_y = (static_cast<f32>(y) * dy_dv) - y_low;
+
+            const auto read_src = [&](f32 in_x, f32 in_y) {
+                const size_t read_from =
+                    ((static_cast<size_t>(in_x) * src_width + static_cast<size_t>(in_y)) >> 32) *
+                    ir_components;
+                return std::span<const f32>(&input[read_from], ir_components);
+            };
+
+            auto x0_y0 = read_src(x_low, y_low);
+            auto x1_y0 = read_src(x_high, y_low);
+            auto x0_y1 = read_src(x_low, y_high);
+            auto x1_y1 = read_src(x_high, y_high);
+
+            const auto result = bilinear_sample(x0_y0, x1_y0, x0_y1, x1_y1, weight_x, weight_y);
+
+            const size_t write_to = (y * dst_width + x) * ir_components;
+
+            std::memcpy(&output[write_to], &result, sizeof(f32) * ir_components);
+        }
+    }
+}
+
+} // namespace
+
+struct SoftwareBlitEngine::BlitEngineImpl {
+    std::vector<u8> tmp_buffer;
+    std::vector<u8> src_buffer;
+    std::vector<u8> dst_buffer;
+    std::vector<f32> intermediate_src;
+    std::vector<f32> intermediate_dst;
+    ConverterFactory converter_factory;
+};
+
+SoftwareBlitEngine::SoftwareBlitEngine(MemoryManager& memory_manager_)
+    : memory_manager{memory_manager_} {
+    impl = std::make_unique<BlitEngineImpl>();
+}
+
+SoftwareBlitEngine::~SoftwareBlitEngine() = default;
+
+bool SoftwareBlitEngine::Blit(Fermi2D::Surface& src, Fermi2D::Surface& dst,
+                              Fermi2D::Config& config) {
+    const auto get_surface_size = [](Fermi2D::Surface& surface, u32 bytes_per_pixel) {
+        if (surface.linear == Fermi2D::MemoryLayout::BlockLinear) {
+            return CalculateSize(true, bytes_per_pixel, surface.width, surface.height,
+                                 surface.depth, surface.block_height, surface.block_depth);
+        }
+        return static_cast<size_t>(surface.pitch * surface.height);
+    };
+    const auto process_pitch_linear = [](bool unpack, std::span<const u8> input,
+                                         std::span<u8> output, u32 extent_x, u32 extent_y,
+                                         u32 pitch, u32 x0, u32 y0, size_t bpp) {
+        const size_t base_offset = x0 * bpp;
+        const size_t copy_size = extent_x * bpp;
+        for (u32 y = y0; y < extent_y; y++) {
+            const size_t first_offset = y * pitch + base_offset;
+            const size_t second_offset = y * extent_x * bpp;
+            u8* write_to = unpack ? &output[first_offset] : &output[second_offset];
+            const u8* read_from = unpack ? &input[second_offset] : &input[first_offset];
+            std::memcpy(write_to, read_from, copy_size);
+        }
+    };
+
+    const u32 src_extent_x = config.src_x1 - config.src_x0;
+    const u32 src_extent_y = config.src_y1 - config.src_y0;
+
+    const u32 dst_extent_x = config.dst_x1 - config.dst_x0;
+    const u32 dst_extent_y = config.dst_y1 - config.dst_y0;
+    const auto src_bytes_per_pixel = BytesPerBlock(PixelFormatFromRenderTargetFormat(src.format));
+    const auto dst_bytes_per_pixel = BytesPerBlock(PixelFormatFromRenderTargetFormat(dst.format));
+    const size_t src_size = get_surface_size(src, src_bytes_per_pixel);
+    impl->tmp_buffer.resize(src_size);
+    memory_manager.ReadBlock(src.Address(), impl->tmp_buffer.data(), src_size);
+
+    const size_t src_copy_size = src_extent_x * src_extent_y * src_bytes_per_pixel;
+
+    const size_t dst_copy_size = dst_extent_x * dst_extent_y * dst_bytes_per_pixel;
+
+    impl->src_buffer.resize(src_copy_size);
+
+    const bool no_passthrough =
+        src.format != dst.format || src_extent_x != dst_extent_x || src_extent_y != dst_extent_y;
+
+    const auto convertion_phase_same_format = [&]() {
+        NearestNeighbor(impl->src_buffer, impl->dst_buffer, src_extent_x, src_extent_y,
+                        dst_extent_x, dst_extent_y, dst_bytes_per_pixel);
+    };
+
+    const auto convertion_phase_ir = [&]() {
+        auto* input_converter = impl->converter_factory.GetFormatConverter(src.format);
+        impl->intermediate_src.resize((src_copy_size / src_bytes_per_pixel) * ir_components);
+        impl->intermediate_dst.resize((dst_copy_size / dst_bytes_per_pixel) * ir_components);
+        input_converter->ConvertTo(impl->src_buffer, impl->intermediate_src);
+
+        if (config.filter != Fermi2D::Filter::Bilinear) {
+            NearestNeighborFast(impl->intermediate_src, impl->intermediate_dst, src_extent_x,
+                                src_extent_y, dst_extent_x, dst_extent_y);
+        } else {
+            Bilinear(impl->intermediate_src, impl->intermediate_dst, src_extent_x, src_extent_y,
+                     dst_extent_x, dst_extent_y);
+        }
+
+        auto* output_converter = impl->converter_factory.GetFormatConverter(dst.format);
+        output_converter->ConvertFrom(impl->intermediate_dst, impl->dst_buffer);
+    };
+
+    // Do actuall Blit
+
+    impl->dst_buffer.resize(dst_copy_size);
+    if (src.linear == Fermi2D::MemoryLayout::BlockLinear) {
+        UnswizzleSubrect(impl->src_buffer, impl->tmp_buffer, src_bytes_per_pixel, src.width,
+                         src.height, src.depth, config.src_x0, config.src_y0, src_extent_x,
+                         src_extent_y, src.block_height, src.block_depth,
+                         src_extent_x * src_bytes_per_pixel);
+    } else {
+        process_pitch_linear(false, impl->tmp_buffer, impl->src_buffer, src_extent_x, src_extent_y,
+                             src.pitch, config.src_x0, config.src_y0, src_bytes_per_pixel);
+    }
+
+    // Conversion Phase
+    if (no_passthrough) {
+        if (src.format != dst.format || config.filter == Fermi2D::Filter::Bilinear) {
+            convertion_phase_ir();
+        } else {
+            convertion_phase_same_format();
+        }
+    } else {
+        impl->dst_buffer.swap(impl->src_buffer);
+    }
+
+    const size_t dst_size = get_surface_size(dst, dst_bytes_per_pixel);
+    impl->tmp_buffer.resize(dst_size);
+    memory_manager.ReadBlock(dst.Address(), impl->tmp_buffer.data(), dst_size);
+
+    if (dst.linear == Fermi2D::MemoryLayout::BlockLinear) {
+        SwizzleSubrect(impl->tmp_buffer, impl->dst_buffer, dst_bytes_per_pixel, dst.width,
+                       dst.height, dst.depth, config.dst_x0, config.dst_y0, dst_extent_x,
+                       dst_extent_y, dst.block_height, dst.block_depth,
+                       dst_extent_x * dst_bytes_per_pixel);
+    } else {
+        process_pitch_linear(true, impl->dst_buffer, impl->tmp_buffer, dst_extent_x, dst_extent_y,
+                             dst.pitch, config.dst_x0, config.dst_y0,
+                             static_cast<size_t>(dst_bytes_per_pixel));
+    }
+    memory_manager.WriteBlock(dst.Address(), impl->tmp_buffer.data(), dst_size);
+    return true;
+}
+
+} // namespace Tegra::Engines::Blitter

src/video_core/engines/sw_blitter/blitter.h

@@ -0,0 +1,27 @@
+// SPDX-FileCopyrightText: Copyright 2022 yuzu Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
+#pragma once
+
+#include "video_core/engines/fermi_2d.h"
+
+namespace Tegra {
+class MemoryManager;
+}
+
+namespace Tegra::Engines::Blitter {
+
+class SoftwareBlitEngine {
+public:
+    explicit SoftwareBlitEngine(MemoryManager& memory_manager_);
+    ~SoftwareBlitEngine();
+
+    bool Blit(Fermi2D::Surface& src, Fermi2D::Surface& dst, Fermi2D::Config& copy_config);
+
+private:
+    MemoryManager& memory_manager;
+    struct BlitEngineImpl;
+    std::unique_ptr<BlitEngineImpl> impl;
+};
+
+} // namespace Tegra::Engines::Blitter

src/video_core/engines/sw_blitter/converter.h

@@ -0,0 +1,36 @@
+// SPDX-FileCopyrightText: Copyright 2022 yuzu Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
+#pragma once
+
+#include <memory>
+#include <span>
+
+#include "common/common_types.h"
+
+#include "video_core/gpu.h"
+
+namespace Tegra::Engines::Blitter {
+
+class Converter {
+public:
+    virtual void ConvertTo(std::span<const u8> input, std::span<f32> output) = 0;
+    virtual void ConvertFrom(std::span<const f32> input, std::span<u8> output) = 0;
+    virtual ~Converter() = default;
+};
+
+class ConverterFactory {
+public:
+    ConverterFactory();
+    ~ConverterFactory();
+
+    Converter* GetFormatConverter(RenderTargetFormat format);
+
+private:
+    Converter* BuildConverter(RenderTargetFormat format);
+
+    struct ConverterFactoryImpl;
+    std::unique_ptr<ConverterFactoryImpl> impl;
+};
+
+} // namespace Tegra::Engines::Blitter

src/video_core/engines/sw_blitter/generate_converters.py

@@ -0,0 +1,136 @@
+# SPDX-FileCopyrightText: Copyright 2022 yuzu Emulator Project
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+import re
+
+class Format:
+    def __init__(self, string_value):
+        self.name = string_value
+        tmp = string_value.split('_')
+        self.component_type = tmp[1]
+        component_data = re.findall(r"\w\d+", tmp[0])
+        self.num_components = len(component_data)
+        sizes = []
+        swizzle = []
+        for data in component_data:
+            swizzle.append(data[0])
+            sizes.append(int(data[1:]))
+        self.sizes = sizes
+        self.swizzle = swizzle
+
+    def build_component_type_array(self):
+        result = "{ "
+        b = False
+        for i in range(0, self.num_components):
+            if b:
+                result += ", "
+            b = True
+            result += "ComponentType::" + self.component_type
+        result += " }"
+        return result
+
+    def build_component_sizes_array(self):
+        result = "{ "
+        b = False
+        for i in range(0, self.num_components):
+            if b:
+                result += ", "
+            b = True
+            result += str(self.sizes[i])
+        result += " }"
+        return result
+
+    def build_component_swizzle_array(self):
+        result = "{ "
+        b = False
+        for i in range(0, self.num_components):
+            if b:
+                result += ", "
+            b = True
+            swizzle = self.swizzle[i]
+            if swizzle == "X":
+                swizzle = "None"
+            result += "Swizzle::" + swizzle
+        result += " }"
+        return result
+
+    def print_declaration(self):
+        print("struct " + self.name + "Traits {")
+        print("  static constexpr size_t num_components = " + str(self.num_components) + ";")
+        print("  static constexpr std::array<ComponentType, num_components> component_types = " + self.build_component_type_array() + ";")
+        print("  static constexpr std::array<size_t, num_components> component_sizes = " + self.build_component_sizes_array() + ";")
+        print("  static constexpr std::array<Swizzle, num_components> component_swizzle = " + self.build_component_swizzle_array() + ";")
+        print("};\n")
+
+    def print_case(self):
+        print("case RenderTargetFormat::" + self.name + ":")
+        print("  return impl->converters_cache")
+        print("    .emplace(format, std::make_unique<ConverterImpl<" + self.name + "Traits>>())")
+        print("    .first->second.get();")
+        print("  break;")
+
+txt = """
+R32G32B32A32_FLOAT
+R32G32B32A32_SINT
+R32G32B32A32_UINT
+R32G32B32X32_FLOAT
+R32G32B32X32_SINT
+R32G32B32X32_UINT
+R16G16B16A16_UNORM
+R16G16B16A16_SNORM
+R16G16B16A16_SINT
+R16G16B16A16_UINT
+R16G16B16A16_FLOAT
+R32G32_FLOAT
+R32G32_SINT
+R32G32_UINT
+R16G16B16X16_FLOAT
+A8R8G8B8_UNORM
+A8R8G8B8_SRGB
+A2B10G10R10_UNORM
+A2B10G10R10_UINT
+A2R10G10B10_UNORM
+A8B8G8R8_UNORM
+A8B8G8R8_SRGB
+A8B8G8R8_SNORM
+A8B8G8R8_SINT
+A8B8G8R8_UINT
+R16G16_UNORM
+R16G16_SNORM
+R16G16_SINT
+R16G16_UINT
+R16G16_FLOAT
+B10G11R11_FLOAT
+R32_SINT
+R32_UINT
+R32_FLOAT
+X8R8G8B8_UNORM
+X8R8G8B8_SRGB
+R5G6B5_UNORM
+A1R5G5B5_UNORM
+R8G8_UNORM
+R8G8_SNORM
+R8G8_SINT
+R8G8_UINT
+R16_UNORM
+R16_SNORM
+R16_SINT
+R16_UINT
+R16_FLOAT
+R8_UNORM
+R8_SNORM
+R8_SINT
+R8_UINT
+X1R5G5B5_UNORM
+X8B8G8R8_UNORM
+X8B8G8R8_SRGB
+"""
+
+x = txt.split()
+y = list(map(lambda a: Format(a), x))
+formats = list(y)
+for format in formats:
+  format.print_declaration()
+
+for format in formats:
+  format.print_case()

src/video_core/gpu.h

@@ -27,12 +27,12 @@ struct CommandList;
 // TODO: Implement the commented ones
 enum class RenderTargetFormat : u32 {
     NONE = 0x0,
-    R32B32G32A32_FLOAT = 0xC0,
+    R32G32B32A32_FLOAT = 0xC0,
     R32G32B32A32_SINT = 0xC1,
     R32G32B32A32_UINT = 0xC2,
-    // R32G32B32X32_FLOAT = 0xC3,
-    // R32G32B32X32_SINT = 0xC4,
-    // R32G32B32X32_UINT = 0xC5,
+    R32G32B32X32_FLOAT = 0xC3,
+    R32G32B32X32_SINT = 0xC4,
+    R32G32B32X32_UINT = 0xC5,
     R16G16B16A16_UNORM = 0xC6,
     R16G16B16A16_SNORM = 0xC7,
     R16G16B16A16_SINT = 0xC8,
@@ -56,13 +56,13 @@ enum class RenderTargetFormat : u32 {
     R16G16_SINT = 0xDC,
     R16G16_UINT = 0xDD,
     R16G16_FLOAT = 0xDE,
-    // A2R10G10B10_UNORM = 0xDF,
+    A2R10G10B10_UNORM = 0xDF,
     B10G11R11_FLOAT = 0xE0,
     R32_SINT = 0xE3,
     R32_UINT = 0xE4,
     R32_FLOAT = 0xE5,
-    // X8R8G8B8_UNORM = 0xE6,
-    // X8R8G8B8_SRGB = 0xE7,
+    X8R8G8B8_UNORM = 0xE6,
+    X8R8G8B8_SRGB = 0xE7,
     R5G6B5_UNORM = 0xE8,
     A1R5G5B5_UNORM = 0xE9,
     R8G8_UNORM = 0xEA,
@@ -79,11 +79,11 @@ enum class RenderTargetFormat : u32 {
     R8_SINT = 0xF5,
     R8_UINT = 0xF6,
 
-    /*
-    A8_UNORM = 0xF7,
+    // A8_UNORM = 0xF7,
     X1R5G5B5_UNORM = 0xF8,
     X8B8G8R8_UNORM = 0xF9,
     X8B8G8R8_SRGB = 0xFA,
+    /*
     Z1R5G5B5_UNORM = 0xFB,
     O1R5G5B5_UNORM = 0xFC,
     Z8R8G8B8_UNORM = 0xFD,

src/video_core/macro/macro_hle.cpp

@@ -126,11 +126,25 @@ void HLE_3F5E74B9C9A50164(Engines::Maxwell3D& maxwell3d, const std::vector<u32>&
     }
 }
 
-constexpr std::array<std::pair<u64, HLEFunction>, 4> hle_funcs{{
+// Multi-layer Clear
+void HLE_EAD26C3E2109B06B(Engines::Maxwell3D& maxwell3d, const std::vector<u32>& parameters) {
+    ASSERT(parameters.size() == 1);
+
+    const Engines::Maxwell3D::Regs::ClearSurface clear_params{parameters[0]};
+    const u32 rt_index = clear_params.RT;
+    const u32 num_layers = maxwell3d.regs.rt[rt_index].depth;
+    ASSERT(clear_params.layer == 0);
+
+    maxwell3d.regs.clear_surface.raw = clear_params.raw;
+    maxwell3d.ProcessClearBuffers(num_layers);
+}
+
+constexpr std::array<std::pair<u64, HLEFunction>, 5> hle_funcs{{
     {0x771BB18C62444DA0, &HLE_771BB18C62444DA0},
     {0x0D61FC9FAAC9FCAD, &HLE_0D61FC9FAAC9FCAD},
     {0x0217920100488FF7, &HLE_0217920100488FF7},
     {0x3F5E74B9C9A50164, &HLE_3F5E74B9C9A50164},
+    {0xEAD26C3E2109B06B, &HLE_EAD26C3E2109B06B},
 }};
 
 class HLEMacroImpl final : public CachedMacro {

src/video_core/macro/macro_interpreter.cpp

@@ -201,6 +201,7 @@ bool MacroInterpreterImpl::Step(bool is_delay_slot) {
     }
     default:
         UNIMPLEMENTED_MSG("Unimplemented macro operation {}", opcode.operation.Value());
+        break;
     }
 
     // An instruction with the Exit flag will not actually
@@ -297,6 +298,7 @@ void MacroInterpreterImpl::ProcessResult(Macro::ResultOperation operation, u32 r
         break;
     default:
         UNIMPLEMENTED_MSG("Unimplemented result operation {}", operation);
+        break;
     }
 }
 

src/video_core/macro/macro_jit_x64.cpp

@@ -652,6 +652,7 @@ void MacroJITx64Impl::Compile_ProcessResult(Macro::ResultOperation operation, u3
         break;
     default:
         UNIMPLEMENTED_MSG("Unimplemented macro operation {}", operation);
+        break;
     }
 }
 

src/video_core/rasterizer_interface.h

@@ -43,7 +43,7 @@ public:
     virtual void Draw(bool is_indexed, u32 instance_count) = 0;
 
     /// Clear the current framebuffer
-    virtual void Clear() = 0;
+    virtual void Clear(u32 layer_count) = 0;
 
     /// Dispatches a compute shader invocation
     virtual void DispatchCompute() = 0;

src/video_core/renderer_opengl/gl_rasterizer.cpp

@@ -136,7 +136,7 @@ void RasterizerOpenGL::LoadDiskResources(u64 title_id, std::stop_token stop_load
     shader_cache.LoadDiskResources(title_id, stop_loading, callback);
 }
 
-void RasterizerOpenGL::Clear() {
+void RasterizerOpenGL::Clear(u32 layer_count) {
     MICROPROFILE_SCOPE(OpenGL_Clears);
     if (!maxwell3d->ShouldExecute()) {
         return;
@@ -222,8 +222,6 @@ void RasterizerOpenGL::Draw(bool is_indexed, u32 instance_count) {
     pipeline->SetEngine(maxwell3d, gpu_memory);
     pipeline->Configure(is_indexed);
 
-    BindInlineIndexBuffer();
-
     SyncState();
 
     const GLenum primitive_mode = MaxwellToGL::PrimitiveTopology(maxwell3d->regs.draw.topology);
@@ -468,8 +466,7 @@ bool RasterizerOpenGL::AccelerateSurfaceCopy(const Tegra::Engines::Fermi2D::Surf
                                              const Tegra::Engines::Fermi2D::Config& copy_config) {
     MICROPROFILE_SCOPE(OpenGL_Blits);
     std::scoped_lock lock{texture_cache.mutex};
-    texture_cache.BlitImage(dst, src, copy_config);
-    return true;
+    return texture_cache.BlitImage(dst, src, copy_config);
 }
 
 Tegra::Engines::AccelerateDMAInterface& RasterizerOpenGL::AccessAccelerateDMA() {
@@ -1140,16 +1137,6 @@ void RasterizerOpenGL::ReleaseChannel(s32 channel_id) {
     query_cache.EraseChannel(channel_id);
 }
 
-void RasterizerOpenGL::BindInlineIndexBuffer() {
-    if (maxwell3d->inline_index_draw_indexes.empty()) {
-        return;
-    }
-    const auto data_count = static_cast<u32>(maxwell3d->inline_index_draw_indexes.size());
-    auto buffer = Buffer(buffer_cache_runtime, *this, 0, data_count);
-    buffer.ImmediateUpload(0, maxwell3d->inline_index_draw_indexes);
-    buffer_cache_runtime.BindIndexBuffer(buffer, 0, data_count);
-}
-
 AccelerateDMA::AccelerateDMA(BufferCache& buffer_cache_) : buffer_cache{buffer_cache_} {}
 
 bool AccelerateDMA::BufferCopy(GPUVAddr src_address, GPUVAddr dest_address, u64 amount) {

src/video_core/renderer_opengl/gl_rasterizer.h

@@ -69,7 +69,7 @@ public:
     ~RasterizerOpenGL() override;
 
     void Draw(bool is_indexed, u32 instance_count) override;
-    void Clear() override;
+    void Clear(u32 layer_count) override;
     void DispatchCompute() override;
     void ResetCounter(VideoCore::QueryType type) override;
     void Query(GPUVAddr gpu_addr, VideoCore::QueryType type, std::optional<u64> timestamp) override;
@@ -199,8 +199,6 @@ private:
     /// End a transform feedback
     void EndTransformFeedback();
 
-    void BindInlineIndexBuffer();
-
     Tegra::GPU& gpu;
 
     const Device& device;

src/video_core/renderer_opengl/gl_texture_cache.cpp

@@ -891,6 +891,7 @@ void Image::CopyBufferToImage(const VideoCommon::BufferImageCopy& copy, size_t b
         break;
     default:
         ASSERT(false);
+        break;
     }
 }
 
@@ -927,6 +928,7 @@ void Image::CopyImageToBuffer(const VideoCommon::BufferImageCopy& copy, size_t b
         break;
     default:
         ASSERT(false);
+        break;
     }
     // Compressed formats don't have a pixel format or type
     const bool is_compressed = gl_format == GL_NONE;

src/video_core/renderer_opengl/maxwell_to_gl.h

@@ -28,6 +28,7 @@ constexpr std::array<FormatTuple, VideoCore::Surface::MaxPixelFormat> FORMAT_TAB
     {GL_RGB5_A1, GL_BGRA, GL_UNSIGNED_SHORT_1_5_5_5_REV},             // A1R5G5B5_UNORM
     {GL_RGB10_A2, GL_RGBA, GL_UNSIGNED_INT_2_10_10_10_REV},           // A2B10G10R10_UNORM
     {GL_RGB10_A2UI, GL_RGBA_INTEGER, GL_UNSIGNED_INT_2_10_10_10_REV}, // A2B10G10R10_UINT
+    {GL_RGB10_A2, GL_BGRA, GL_UNSIGNED_INT_2_10_10_10_REV},           // A2R10G10B10_UNORM
     {GL_RGB5_A1, GL_RGBA, GL_UNSIGNED_SHORT_1_5_5_5_REV},             // A1B5G5R5_UNORM
     {GL_RGB5_A1, GL_RGBA, GL_UNSIGNED_SHORT_5_5_5_1},                 // A5B5G5R1_UNORM
     {GL_R8, GL_RED, GL_UNSIGNED_BYTE},                                // R8_UNORM