Create python-publish.yml

Replace GLSL as the default OpenGL shader backend

.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

.github/workflows/python-publish.yml

@@ -0,0 +1,39 @@
+# This workflow will upload a Python Package using Twine when a release is created
+# For more information see: https://docs.github.com/en/actions/automating-builds-and-tests/building-and-testing-python#publishing-to-package-registries
+
+# This workflow uses actions that are not certified by GitHub.
+# They are provided by a third-party and are governed by
+# separate terms of service, privacy policy, and support
+# documentation.
+
+name: Upload Python Package
+
+on:
+  release:
+    types: [published]
+
+permissions:
+  contents: read
+
+jobs:
+  deploy:
+
+    runs-on: ubuntu-latest
+
+    steps:
+    - uses: actions/checkout@v3
+    - name: Set up Python
+      uses: actions/setup-python@v3
+      with:
+        python-version: '3.x'
+    - name: Install dependencies
+      run: |
+        python -m pip install --upgrade pip
+        pip install build
+    - name: Build package
+      run: python -m build
+    - name: Publish package
+      uses: pypa/gh-action-pypi-publish@27b31702a0e7fc50959f5ad993c78deac1bdfc29
+      with:
+        user: __token__
+        password: ${{ secrets.PYPI_API_TOKEN }}

.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/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/settings.h

@@ -442,7 +442,7 @@ struct Values {
     SwitchableSetting<NvdecEmulation> nvdec_emulation{NvdecEmulation::GPU, "nvdec_emulation"};
     SwitchableSetting<bool> accelerate_astc{true, "accelerate_astc"};
     SwitchableSetting<bool> use_vsync{true, "use_vsync"};
-    SwitchableSetting<ShaderBackend, true> shader_backend{ShaderBackend::GLASM, ShaderBackend::GLSL,
+    SwitchableSetting<ShaderBackend, true> shader_backend{ShaderBackend::GLSL, ShaderBackend::GLSL,
                                                           ShaderBackend::SPIRV, "shader_backend"};
     SwitchableSetting<bool> use_asynchronous_shaders{false, "use_asynchronous_shaders"};
     SwitchableSetting<bool> use_fast_gpu_time{true, "use_fast_gpu_time"};

src/common/x64/cpu_detect.cpp

@@ -4,14 +4,27 @@
 
 #include <array>
 #include <cstring>
+#include <fstream>
 #include <iterator>
+#include <optional>
 #include <string_view>
+#include <thread>
+#include <vector>
 #include "common/bit_util.h"
 #include "common/common_types.h"
+#include "common/logging/log.h"
 #include "common/x64/cpu_detect.h"
 
+#ifdef _WIN32
+#include <windows.h>
+#endif
+
 #ifdef _MSC_VER
 #include <intrin.h>
+
+static inline u64 xgetbv(u32 index) {
+    return _xgetbv(index);
+}
 #else
 
 #if defined(__DragonFly__) || defined(__FreeBSD__)
@@ -39,12 +52,11 @@ static inline void __cpuid(int info[4], u32 function_id) {
 }
 
 #define _XCR_XFEATURE_ENABLED_MASK 0
-static inline u64 _xgetbv(u32 index) {
+static inline u64 xgetbv(u32 index) {
     u32 eax, edx;
     __asm__ __volatile__("xgetbv" : "=a"(eax), "=d"(edx) : "c"(index));
     return ((u64)edx << 32) | eax;
 }
-
 #endif // _MSC_VER
 
 namespace Common {
@@ -107,7 +119,7 @@ static CPUCaps Detect() {
         //  - Is the XSAVE bit set in CPUID?
         //  - XGETBV result has the XCR bit set.
         if (Common::Bit<28>(cpu_id[2]) && Common::Bit<27>(cpu_id[2])) {
-            if ((_xgetbv(_XCR_XFEATURE_ENABLED_MASK) & 0x6) == 0x6) {
+            if ((xgetbv(_XCR_XFEATURE_ENABLED_MASK) & 0x6) == 0x6) {
                 caps.avx = true;
                 if (Common::Bit<12>(cpu_id[2]))
                     caps.fma = true;
@@ -192,4 +204,45 @@ const CPUCaps& GetCPUCaps() {
     return caps;
 }
 
+std::optional<int> GetProcessorCount() {
+#if defined(_WIN32)
+    // Get the buffer length.
+    DWORD length = 0;
+    GetLogicalProcessorInformation(nullptr, &length);
+    if (GetLastError() != ERROR_INSUFFICIENT_BUFFER) {
+        LOG_ERROR(Frontend, "Failed to query core count.");
+        return std::nullopt;
+    }
+    std::vector<SYSTEM_LOGICAL_PROCESSOR_INFORMATION> buffer(
+        length / sizeof(SYSTEM_LOGICAL_PROCESSOR_INFORMATION));
+    // Now query the core count.
+    if (!GetLogicalProcessorInformation(buffer.data(), &length)) {
+        LOG_ERROR(Frontend, "Failed to query core count.");
+        return std::nullopt;
+    }
+    return static_cast<int>(
+        std::count_if(buffer.cbegin(), buffer.cend(), [](const auto& proc_info) {
+            return proc_info.Relationship == RelationProcessorCore;
+        }));
+#elif defined(__unix__)
+    const int thread_count = std::thread::hardware_concurrency();
+    std::ifstream smt("/sys/devices/system/cpu/smt/active");
+    char state = '0';
+    if (smt) {
+        smt.read(&state, sizeof(state));
+    }
+    switch (state) {
+    case '0':
+        return thread_count;
+    case '1':
+        return thread_count / 2;
+    default:
+        return std::nullopt;
+    }
+#else
+    // Shame on you
+    return std::nullopt;
+#endif
+}
+
 } // namespace Common

src/common/x64/cpu_detect.h

@@ -4,6 +4,7 @@
 
 #pragma once
 
+#include <optional>
 #include <string_view>
 #include "common/common_types.h"
 
@@ -74,4 +75,7 @@ struct CPUCaps {
  */
 const CPUCaps& GetCPUCaps();
 
+/// Detects CPU core count
+std::optional<int> GetProcessorCount();
+
 } // namespace Common

src/core/CMakeLists.txt

@@ -120,6 +120,8 @@ add_library(core STATIC
     file_sys/vfs_vector.h
     file_sys/xts_archive.cpp
     file_sys/xts_archive.h
+    frontend/applets/cabinet.cpp
+    frontend/applets/cabinet.h
     frontend/applets/controller.cpp
     frontend/applets/controller.h
     frontend/applets/error.cpp
@@ -312,6 +314,8 @@ add_library(core STATIC
     hle/service/am/applet_ae.h
     hle/service/am/applet_oe.cpp
     hle/service/am/applet_oe.h
+    hle/service/am/applets/applet_cabinet.cpp
+    hle/service/am/applets/applet_cabinet.h
     hle/service/am/applets/applet_controller.cpp
     hle/service/am/applets/applet_controller.h
     hle/service/am/applets/applet_error.cpp
@@ -526,6 +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/frontend/applets/cabinet.cpp

@@ -0,0 +1,20 @@
+// SPDX-FileCopyrightText: Copyright 2022 yuzu Emulator Project
+// SPDX-License-Identifier: GPL-2.0-or-later
+
+#include "common/logging/log.h"
+#include "core/frontend/applets/cabinet.h"
+
+#include <thread>
+
+namespace Core::Frontend {
+
+CabinetApplet::~CabinetApplet() = default;
+
+void DefaultCabinetApplet::ShowCabinetApplet(
+    const CabinetCallback& callback, const CabinetParameters& parameters,
+    std::shared_ptr<Service::NFP::NfpDevice> nfp_device) const {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
+    callback(false, {});
+}
+
+} // namespace Core::Frontend

src/core/frontend/applets/cabinet.h

@@ -0,0 +1,37 @@
+// SPDX-FileCopyrightText: Copyright 2022 yuzu Emulator Project
+// SPDX-License-Identifier: GPL-2.0-or-later
+
+#pragma once
+
+#include <functional>
+#include "core/hle/service/nfp/nfp_types.h"
+
+namespace Service::NFP {
+class NfpDevice;
+} // namespace Service::NFP
+
+namespace Core::Frontend {
+
+struct CabinetParameters {
+    Service::NFP::TagInfo tag_info;
+    Service::NFP::RegisterInfo register_info;
+    Service::NFP::CabinetMode mode;
+};
+
+using CabinetCallback = std::function<void(bool, const std::string&)>;
+
+class CabinetApplet {
+public:
+    virtual ~CabinetApplet();
+    virtual void ShowCabinetApplet(const CabinetCallback& callback,
+                                   const CabinetParameters& parameters,
+                                   std::shared_ptr<Service::NFP::NfpDevice> nfp_device) const = 0;
+};
+
+class DefaultCabinetApplet final : public CabinetApplet {
+public:
+    void ShowCabinetApplet(const CabinetCallback& callback, const CabinetParameters& parameters,
+                           std::shared_ptr<Service::NFP::NfpDevice> nfp_device) const override;
+};
+
+} // namespace Core::Frontend

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_event.cpp

@@ -20,8 +20,12 @@ void KEvent::Initialize(KProcess* owner) {
     m_readable_event.Initialize(this);
 
     // Set our owner process.
-    m_owner = owner;
-    m_owner->Open();
+    // HACK: this should never be nullptr, but service threads don't have a
+    // proper parent process yet.
+    if (owner != nullptr) {
+        m_owner = owner;
+        m_owner->Open();
+    }
 
     // Mark initialized.
     m_initialized = true;
@@ -50,8 +54,11 @@ Result KEvent::Clear() {
 void KEvent::PostDestroy(uintptr_t arg) {
     // Release the event count resource the owner process holds.
     KProcess* owner = reinterpret_cast<KProcess*>(arg);
-    owner->GetResourceLimit()->Release(LimitableResource::EventCountMax, 1);
-    owner->Close();
+
+    if (owner != nullptr) {
+        owner->GetResourceLimit()->Release(LimitableResource::EventCountMax, 1);
+        owner->Close();
+    }
 }
 
 } // namespace Kernel

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/service_thread.cpp

@@ -40,7 +40,6 @@ private:
     std::mutex m_session_mutex;
     std::map<KServerSession*, std::shared_ptr<SessionRequestManager>> m_sessions;
     KEvent* m_wakeup_event;
-    KProcess* m_process;
     KThread* m_thread;
     std::atomic<bool> m_shutdown_requested;
     const std::string m_service_name;
@@ -180,39 +179,17 @@ ServiceThread::Impl::~Impl() {
 
     // Close thread.
     m_thread->Close();
-
-    // Close process.
-    m_process->Close();
 }
 
 ServiceThread::Impl::Impl(KernelCore& kernel_, const std::string& service_name)
     : kernel{kernel_}, m_service_name{service_name} {
-    // Initialize process.
-    m_process = KProcess::Create(kernel);
-    KProcess::Initialize(m_process, kernel.System(), service_name,
-                         KProcess::ProcessType::KernelInternal, kernel.GetSystemResourceLimit());
-
-    // Reserve a new event from the process resource limit
-    KScopedResourceReservation event_reservation(m_process, LimitableResource::EventCountMax);
-    ASSERT(event_reservation.Succeeded());
-
     // Initialize event.
     m_wakeup_event = KEvent::Create(kernel);
-    m_wakeup_event->Initialize(m_process);
-
-    // Commit the event reservation.
-    event_reservation.Commit();
-
-    // Reserve a new thread from the process resource limit
-    KScopedResourceReservation thread_reservation(m_process, LimitableResource::ThreadCountMax);
-    ASSERT(thread_reservation.Succeeded());
+    m_wakeup_event->Initialize(nullptr);
 
     // Initialize thread.
     m_thread = KThread::Create(kernel);
-    ASSERT(KThread::InitializeDummyThread(m_thread, m_process).IsSuccess());
-
-    // Commit the thread reservation.
-    thread_reservation.Commit();
+    ASSERT(KThread::InitializeDummyThread(m_thread, nullptr).IsSuccess());
 
     // Start thread.
     m_host_thread = std::jthread([this] { LoopProcess(); });

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_cabinet.cpp

@@ -0,0 +1,177 @@
+// SPDX-FileCopyrightText: Copyright 2022 yuzu Emulator Project
+// SPDX-License-Identifier: GPL-2.0-or-later
+
+#include "common/assert.h"
+#include "common/logging/log.h"
+#include "core/core.h"
+#include "core/frontend/applets/cabinet.h"
+#include "core/hid/hid_core.h"
+#include "core/hle/kernel/k_event.h"
+#include "core/hle/kernel/k_readable_event.h"
+#include "core/hle/service/am/am.h"
+#include "core/hle/service/am/applets/applet_cabinet.h"
+#include "core/hle/service/mii/mii_manager.h"
+#include "core/hle/service/nfp/nfp_device.h"
+
+namespace Service::AM::Applets {
+
+Cabinet::Cabinet(Core::System& system_, LibraryAppletMode applet_mode_,
+                 const Core::Frontend::CabinetApplet& frontend_)
+    : Applet{system_, applet_mode_}, frontend{frontend_}, system{system_}, service_context{
+                                                                               system_,
+                                                                               "CabinetApplet"} {
+
+    availability_change_event =
+        service_context.CreateEvent("CabinetApplet:AvailabilityChangeEvent");
+}
+
+Cabinet::~Cabinet() = default;
+
+void Cabinet::Initialize() {
+    Applet::Initialize();
+
+    LOG_INFO(Service_HID, "Initializing Cabinet Applet.");
+
+    LOG_DEBUG(Service_HID,
+              "Initializing Applet with common_args: arg_version={}, lib_version={}, "
+              "play_startup_sound={}, size={}, system_tick={}, theme_color={}",
+              common_args.arguments_version, common_args.library_version,
+              common_args.play_startup_sound, common_args.size, common_args.system_tick,
+              common_args.theme_color);
+
+    const auto storage = broker.PopNormalDataToApplet();
+    ASSERT(storage != nullptr);
+
+    const auto applet_input_data = storage->GetData();
+    ASSERT(applet_input_data.size() >= sizeof(StartParamForAmiiboSettings));
+
+    std::memcpy(&applet_input_common, applet_input_data.data(),
+                sizeof(StartParamForAmiiboSettings));
+}
+
+bool Cabinet::TransactionComplete() const {
+    return is_complete;
+}
+
+Result Cabinet::GetStatus() const {
+    return ResultSuccess;
+}
+
+void Cabinet::ExecuteInteractive() {
+    ASSERT_MSG(false, "Attempted to call interactive execution on non-interactive applet.");
+}
+
+void Cabinet::Execute() {
+    if (is_complete) {
+        return;
+    }
+
+    const auto callback = [this](bool apply_changes, const std::string& amiibo_name) {
+        DisplayCompleted(apply_changes, amiibo_name);
+    };
+
+    // TODO: listen on all controllers
+    if (nfp_device == nullptr) {
+        nfp_device = std::make_shared<Service::NFP::NfpDevice>(
+            system.HIDCore().GetFirstNpadId(), system, service_context, availability_change_event);
+        nfp_device->Initialize();
+        nfp_device->StartDetection(Service::NFP::TagProtocol::All);
+    }
+
+    const Core::Frontend::CabinetParameters parameters{
+        .tag_info = applet_input_common.tag_info,
+        .register_info = applet_input_common.register_info,
+        .mode = applet_input_common.applet_mode,
+    };
+
+    switch (applet_input_common.applet_mode) {
+    case Service::NFP::CabinetMode::StartNicknameAndOwnerSettings:
+    case Service::NFP::CabinetMode::StartGameDataEraser:
+    case Service::NFP::CabinetMode::StartRestorer:
+    case Service::NFP::CabinetMode::StartFormatter:
+        frontend.ShowCabinetApplet(callback, parameters, nfp_device);
+        break;
+    default:
+        UNIMPLEMENTED_MSG("Unknown CabinetMode={}", applet_input_common.applet_mode);
+        DisplayCompleted(false, {});
+        break;
+    }
+}
+
+void Cabinet::DisplayCompleted(bool apply_changes, std::string_view amiibo_name) {
+    Service::Mii::MiiManager manager;
+    ReturnValueForAmiiboSettings applet_output{};
+
+    if (!apply_changes) {
+        Cancel();
+    }
+
+    if (nfp_device->GetCurrentState() != Service::NFP::DeviceState::TagFound &&
+        nfp_device->GetCurrentState() != Service::NFP::DeviceState::TagMounted) {
+        Cancel();
+    }
+
+    if (nfp_device->GetCurrentState() == Service::NFP::DeviceState::TagFound) {
+        nfp_device->Mount(Service::NFP::MountTarget::All);
+    }
+
+    switch (applet_input_common.applet_mode) {
+    case Service::NFP::CabinetMode::StartNicknameAndOwnerSettings: {
+        Service::NFP::AmiiboName name{};
+        std::memcpy(name.data(), amiibo_name.data(), std::min(amiibo_name.size(), name.size() - 1));
+        nfp_device->SetNicknameAndOwner(name);
+        break;
+    }
+    case Service::NFP::CabinetMode::StartGameDataEraser:
+        nfp_device->DeleteApplicationArea();
+        break;
+    case Service::NFP::CabinetMode::StartRestorer:
+        nfp_device->RestoreAmiibo();
+        break;
+    case Service::NFP::CabinetMode::StartFormatter:
+        nfp_device->DeleteAllData();
+        break;
+    default:
+        UNIMPLEMENTED_MSG("Unknown CabinetMode={}", applet_input_common.applet_mode);
+        break;
+    }
+
+    applet_output.device_handle = applet_input_common.device_handle;
+    applet_output.result = CabinetResult::Cancel;
+    const auto reg_result = nfp_device->GetRegisterInfo(applet_output.register_info);
+    const auto tag_result = nfp_device->GetTagInfo(applet_output.tag_info);
+    nfp_device->Finalize();
+
+    if (reg_result.IsSuccess()) {
+        applet_output.result |= CabinetResult::RegisterInfo;
+    }
+
+    if (tag_result.IsSuccess()) {
+        applet_output.result |= CabinetResult::TagInfo;
+    }
+
+    std::vector<u8> out_data(sizeof(ReturnValueForAmiiboSettings));
+    std::memcpy(out_data.data(), &applet_output, sizeof(ReturnValueForAmiiboSettings));
+
+    is_complete = true;
+
+    broker.PushNormalDataFromApplet(std::make_shared<IStorage>(system, std::move(out_data)));
+    broker.SignalStateChanged();
+}
+
+void Cabinet::Cancel() {
+    ReturnValueForAmiiboSettings applet_output{};
+    applet_output.device_handle = applet_input_common.device_handle;
+    applet_output.result = CabinetResult::Cancel;
+    nfp_device->Finalize();
+
+    std::vector<u8> out_data(sizeof(ReturnValueForAmiiboSettings));
+    std::memcpy(out_data.data(), &applet_output, sizeof(ReturnValueForAmiiboSettings));
+
+    is_complete = true;
+
+    broker.PushNormalDataFromApplet(std::make_shared<IStorage>(system, std::move(out_data)));
+    broker.SignalStateChanged();
+}
+
+} // namespace Service::AM::Applets

src/core/hle/service/am/applets/applet_cabinet.h

@@ -0,0 +1,104 @@
+// SPDX-FileCopyrightText: Copyright 2022 yuzu Emulator Project
+// SPDX-License-Identifier: GPL-2.0-or-later
+
+#pragma once
+
+#include <array>
+
+#include "core/hle/result.h"
+#include "core/hle/service/am/applets/applets.h"
+#include "core/hle/service/kernel_helpers.h"
+#include "core/hle/service/nfp/nfp_types.h"
+
+namespace Kernel {
+class KEvent;
+class KReadableEvent;
+} // namespace Kernel
+
+namespace Core {
+class System;
+} // namespace Core
+
+namespace Service::NFP {
+class NfpDevice;
+}
+
+namespace Service::AM::Applets {
+
+enum class CabinetAppletVersion : u32 {
+    Version1 = 0x1,
+};
+
+enum class CabinetResult : u8 {
+    Cancel = 0,
+    TagInfo = 1 << 1,
+    RegisterInfo = 1 << 2,
+    All = TagInfo | RegisterInfo,
+};
+DECLARE_ENUM_FLAG_OPERATORS(CabinetResult)
+
+// This is nn::nfp::AmiiboSettingsStartParam
+struct AmiiboSettingsStartParam {
+    u64 device_handle;
+    std::array<u8, 0x20> param_1;
+    u8 param_2;
+};
+static_assert(sizeof(AmiiboSettingsStartParam) == 0x30,
+              "AmiiboSettingsStartParam is an invalid size");
+
+#pragma pack(push, 1)
+// This is nn::nfp::StartParamForAmiiboSettings
+struct StartParamForAmiiboSettings {
+    u8 param_1;
+    Service::NFP::CabinetMode applet_mode;
+    u8 flags;
+    u8 amiibo_settings_1;
+    u64 device_handle;
+    Service::NFP::TagInfo tag_info;
+    Service::NFP::RegisterInfo register_info;
+    std::array<u8, 0x20> amiibo_settings_3;
+    INSERT_PADDING_BYTES(0x24);
+};
+static_assert(sizeof(StartParamForAmiiboSettings) == 0x1A8,
+              "StartParamForAmiiboSettings is an invalid size");
+
+// This is nn::nfp::ReturnValueForAmiiboSettings
+struct ReturnValueForAmiiboSettings {
+    CabinetResult result;
+    INSERT_PADDING_BYTES(0x3);
+    u64 device_handle;
+    Service::NFP::TagInfo tag_info;
+    Service::NFP::RegisterInfo register_info;
+    INSERT_PADDING_BYTES(0x24);
+};
+static_assert(sizeof(ReturnValueForAmiiboSettings) == 0x188,
+              "ReturnValueForAmiiboSettings is an invalid size");
+#pragma pack(pop)
+
+class Cabinet final : public Applet {
+public:
+    explicit Cabinet(Core::System& system_, LibraryAppletMode applet_mode_,
+                     const Core::Frontend::CabinetApplet& frontend_);
+    ~Cabinet() override;
+
+    void Initialize() override;
+
+    bool TransactionComplete() const override;
+    Result GetStatus() const override;
+    void ExecuteInteractive() override;
+    void Execute() override;
+    void DisplayCompleted(bool apply_changes, std::string_view amiibo_name);
+    void Cancel();
+
+private:
+    const Core::Frontend::CabinetApplet& frontend;
+    Core::System& system;
+
+    bool is_complete{false};
+    std::shared_ptr<Service::NFP::NfpDevice> nfp_device;
+    Kernel::KEvent* availability_change_event;
+    KernelHelpers::ServiceContext service_context;
+    StartParamForAmiiboSettings applet_input_common{};
+};
+
+} // namespace Service::AM::Applets

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/am/applets/applets.cpp

@@ -5,6 +5,7 @@
 
 #include "common/assert.h"
 #include "core/core.h"
+#include "core/frontend/applets/cabinet.h"
 #include "core/frontend/applets/controller.h"
 #include "core/frontend/applets/error.h"
 #include "core/frontend/applets/general_frontend.h"
@@ -16,6 +17,7 @@
 #include "core/hle/service/am/am.h"
 #include "core/hle/service/am/applet_ae.h"
 #include "core/hle/service/am/applet_oe.h"
+#include "core/hle/service/am/applets/applet_cabinet.h"
 #include "core/hle/service/am/applets/applet_controller.h"
 #include "core/hle/service/am/applets/applet_error.h"
 #include "core/hle/service/am/applets/applet_general_backend.h"
@@ -171,13 +173,15 @@ void Applet::Initialize() {
 
 AppletFrontendSet::AppletFrontendSet() = default;
 
-AppletFrontendSet::AppletFrontendSet(ControllerApplet controller_applet, ErrorApplet error_applet,
+AppletFrontendSet::AppletFrontendSet(CabinetApplet cabinet_applet,
+                                     ControllerApplet controller_applet, ErrorApplet error_applet,
                                      MiiEdit mii_edit_,
                                      ParentalControlsApplet parental_controls_applet,
                                      PhotoViewer photo_viewer_, ProfileSelect profile_select_,
                                      SoftwareKeyboard software_keyboard_, WebBrowser web_browser_)
-    : controller{std::move(controller_applet)}, error{std::move(error_applet)},
-      mii_edit{std::move(mii_edit_)}, parental_controls{std::move(parental_controls_applet)},
+    : cabinet{std::move(cabinet_applet)}, controller{std::move(controller_applet)},
+      error{std::move(error_applet)}, mii_edit{std::move(mii_edit_)},
+      parental_controls{std::move(parental_controls_applet)},
       photo_viewer{std::move(photo_viewer_)}, profile_select{std::move(profile_select_)},
       software_keyboard{std::move(software_keyboard_)}, web_browser{std::move(web_browser_)} {}
 
@@ -196,6 +200,10 @@ const AppletFrontendSet& AppletManager::GetAppletFrontendSet() const {
 }
 
 void AppletManager::SetAppletFrontendSet(AppletFrontendSet set) {
+    if (set.cabinet != nullptr) {
+        frontend.cabinet = std::move(set.cabinet);
+    }
+
     if (set.controller != nullptr) {
         frontend.controller = std::move(set.controller);
     }
@@ -235,6 +243,10 @@ void AppletManager::SetDefaultAppletFrontendSet() {
 }
 
 void AppletManager::SetDefaultAppletsIfMissing() {
+    if (frontend.cabinet == nullptr) {
+        frontend.cabinet = std::make_unique<Core::Frontend::DefaultCabinetApplet>();
+    }
+
     if (frontend.controller == nullptr) {
         frontend.controller =
             std::make_unique<Core::Frontend::DefaultControllerApplet>(system.HIDCore());
@@ -279,6 +291,8 @@ std::shared_ptr<Applet> AppletManager::GetApplet(AppletId id, LibraryAppletMode
     switch (id) {
     case AppletId::Auth:
         return std::make_shared<Auth>(system, mode, *frontend.parental_controls);
+    case AppletId::Cabinet:
+        return std::make_shared<Cabinet>(system, mode, *frontend.cabinet);
     case AppletId::Controller:
         return std::make_shared<Controller>(system, mode, *frontend.controller);
     case AppletId::Error:

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

@@ -16,6 +16,7 @@ class System;
 }
 
 namespace Core::Frontend {
+class CabinetApplet;
 class ControllerApplet;
 class ECommerceApplet;
 class ErrorApplet;
@@ -176,6 +177,7 @@ protected:
 };
 
 struct AppletFrontendSet {
+    using CabinetApplet = std::unique_ptr<Core::Frontend::CabinetApplet>;
     using ControllerApplet = std::unique_ptr<Core::Frontend::ControllerApplet>;
     using ErrorApplet = std::unique_ptr<Core::Frontend::ErrorApplet>;
     using MiiEdit = std::unique_ptr<Core::Frontend::MiiEditApplet>;
@@ -186,10 +188,11 @@ struct AppletFrontendSet {
     using WebBrowser = std::unique_ptr<Core::Frontend::WebBrowserApplet>;
 
     AppletFrontendSet();
-    AppletFrontendSet(ControllerApplet controller_applet, ErrorApplet error_applet,
-                      MiiEdit mii_edit_, ParentalControlsApplet parental_controls_applet,
-                      PhotoViewer photo_viewer_, ProfileSelect profile_select_,
-                      SoftwareKeyboard software_keyboard_, WebBrowser web_browser_);
+    AppletFrontendSet(CabinetApplet cabinet_applet, ControllerApplet controller_applet,
+                      ErrorApplet error_applet, MiiEdit mii_edit_,
+                      ParentalControlsApplet parental_controls_applet, PhotoViewer photo_viewer_,
+                      ProfileSelect profile_select_, SoftwareKeyboard software_keyboard_,
+                      WebBrowser web_browser_);
     ~AppletFrontendSet();
 
     AppletFrontendSet(const AppletFrontendSet&) = delete;
@@ -198,6 +201,7 @@ struct AppletFrontendSet {
     AppletFrontendSet(AppletFrontendSet&&) noexcept;
     AppletFrontendSet& operator=(AppletFrontendSet&&) noexcept;
 
+    CabinetApplet cabinet;
     ControllerApplet controller;
     ErrorApplet error;
     MiiEdit mii_edit;

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;
     }
@@ -77,6 +75,9 @@ void NfpDevice::NpadUpdate(Core::HID::ControllerTriggerType type) {
         LoadAmiibo(nfc_status.data);
         break;
     case Common::Input::NfcState::AmiiboRemoved:
+        if (device_state == DeviceState::Initialized || device_state == DeviceState::TagRemoved) {
+            break;
+        }
         if (device_state != DeviceState::SearchingForTag) {
             CloseAmiibo();
         }
@@ -97,6 +98,8 @@ bool NfpDevice::LoadAmiibo(std::span<const u8> data) {
         return false;
     }
 
+    // TODO: Filter by allowed_protocols here
+
     memcpy(&encrypted_tag_data, data.data(), sizeof(EncryptedNTAG215File));
 
     device_state = DeviceState::TagFound;
@@ -143,7 +146,7 @@ void NfpDevice::Finalize() {
     device_state = DeviceState::Unavailable;
 }
 
-Result NfpDevice::StartDetection(s32 protocol_) {
+Result NfpDevice::StartDetection(TagProtocol allowed_protocol) {
     if (device_state != DeviceState::Initialized && device_state != DeviceState::TagRemoved) {
         LOG_ERROR(Service_NFP, "Wrong device state {}", device_state);
         return WrongDeviceState;
@@ -155,7 +158,7 @@ Result NfpDevice::StartDetection(s32 protocol_) {
     }
 
     device_state = DeviceState::SearchingForTag;
-    protocol = protocol_;
+    allowed_protocols = allowed_protocol;
     return ResultSuccess;
 }
 
@@ -469,6 +472,32 @@ Result NfpDevice::OpenApplicationArea(u32 access_id) {
     return ResultSuccess;
 }
 
+Result NfpDevice::GetApplicationAreaId(u32& application_area_id) const {
+    application_area_id = {};
+
+    if (device_state != DeviceState::TagMounted) {
+        LOG_ERROR(Service_NFP, "Wrong device state {}", device_state);
+        if (device_state == DeviceState::TagRemoved) {
+            return TagRemoved;
+        }
+        return WrongDeviceState;
+    }
+
+    if (mount_target == MountTarget::None || mount_target == MountTarget::Rom) {
+        LOG_ERROR(Service_NFP, "Amiibo is read only", device_state);
+        return WrongDeviceState;
+    }
+
+    if (tag_data.settings.settings.appdata_initialized.Value() == 0) {
+        LOG_WARNING(Service_NFP, "Application area is not initialized");
+        return ApplicationAreaIsNotInitialized;
+    }
+
+    application_area_id = tag_data.application_area_id;
+
+    return ResultSuccess;
+}
+
 Result NfpDevice::GetApplicationArea(std::vector<u8>& data) const {
     if (device_state != DeviceState::TagMounted) {
         LOG_ERROR(Service_NFP, "Wrong device state {}", device_state);

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

@@ -3,10 +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"
@@ -37,7 +37,7 @@ public:
     void Initialize();
     void Finalize();
 
-    Result StartDetection(s32 protocol_);
+    Result StartDetection(TagProtocol allowed_protocol);
     Result StopDetection();
     Result Mount(MountTarget mount_target);
     Result Unmount();
@@ -53,6 +53,7 @@ public:
     Result DeleteAllData();
 
     Result OpenApplicationArea(u32 access_id);
+    Result GetApplicationAreaId(u32& application_area_id) const;
     Result GetApplicationArea(std::vector<u8>& data) const;
     Result SetApplicationArea(std::span<const u8> data);
     Result CreateApplicationArea(u32 access_id, std::span<const u8> data);
@@ -88,7 +89,7 @@ private:
 
     bool is_data_moddified{};
     bool is_app_area_open{};
-    s32 protocol{};
+    TagProtocol allowed_protocols{};
     s64 current_posix_time{};
     MountTarget mount_target{MountTarget::None};
     DeviceState device_state{DeviceState::Unavailable};

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

@@ -88,11 +88,22 @@ enum class PackedTagType : u8 {
     Type5, // ISO15693 RW/RO 540 bytes 106kbit/s
 };
 
+// Verify this enum. It might be completely wrong default protocol is 0x48
 enum class TagProtocol : u32 {
     None,
-    TypeA, // ISO14443A
-    TypeB, // ISO14443B
-    TypeF, // Sony Felica
+    TypeA = 1U << 0, // ISO14443A
+    TypeB = 1U << 1, // ISO14443B
+    TypeF = 1U << 2, // Sony Felica
+    Unknown1 = 1U << 3,
+    Unknown2 = 1U << 5,
+    All = 0xFFFFFFFFU,
+};
+
+enum class CabinetMode : u8 {
+    StartNicknameAndOwnerSettings,
+    StartGameDataEraser,
+    StartRestorer,
+    StartFormatter,
 };
 
 using UniqueSerialNumber = std::array<u8, 7>;

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;
@@ -130,7 +131,7 @@ void IUser::ListDevices(Kernel::HLERequestContext& ctx) {
 void IUser::StartDetection(Kernel::HLERequestContext& ctx) {
     IPC::RequestParser rp{ctx};
     const auto device_handle{rp.Pop<u64>()};
-    const auto nfp_protocol{rp.Pop<s32>()};
+    const auto nfp_protocol{rp.PopEnum<TagProtocol>()};
     LOG_INFO(Service_NFP, "called, device_handle={}, nfp_protocol={}", device_handle, nfp_protocol);
 
     if (state == State::NonInitialized) {
@@ -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/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/drivers/virtual_amiibo.cpp

@@ -60,6 +60,8 @@ Common::Input::NfcState VirtualAmiibo::WriteNfcData(
         return Common::Input::NfcState::WriteFailed;
     }
 
+    amiibo_data = data;
+
     return Common::Input::NfcState::Success;
 }
 
@@ -91,6 +93,15 @@ VirtualAmiibo::Info VirtualAmiibo::LoadAmiibo(const std::string& filename) {
     return Info::Success;
 }
 
+VirtualAmiibo::Info VirtualAmiibo::ReloadAmiibo() {
+    if (state == State::AmiiboIsOpen) {
+        SetNfc(identifier, {Common::Input::NfcState::NewAmiibo, amiibo_data});
+        return Info::Success;
+    }
+
+    return LoadAmiibo(file_path);
+}
+
 VirtualAmiibo::Info VirtualAmiibo::CloseAmiibo() {
     state = polling_mode == Common::Input::PollingMode::NFC ? State::WaitingForAmiibo
                                                             : State::Initialized;
@@ -98,4 +109,8 @@ VirtualAmiibo::Info VirtualAmiibo::CloseAmiibo() {
     return Info::Success;
 }
 
+std::string VirtualAmiibo::GetLastFilePath() const {
+    return file_path;
+}
+
 } // namespace InputCommon

src/input_common/drivers/virtual_amiibo.h

@@ -47,8 +47,11 @@ public:
     State GetCurrentState() const;
 
     Info LoadAmiibo(const std::string& amiibo_file);
+    Info ReloadAmiibo();
     Info CloseAmiibo();
 
+    std::string GetLastFilePath() const;
+
 private:
     static constexpr std::size_t amiibo_size = 0x21C;
     static constexpr std::size_t amiibo_size_without_password = amiibo_size - 0x8;

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_engine.h

@@ -133,7 +133,7 @@ public:
         return Common::Input::CameraError::NotSupported;
     }
 
-    // Request nfc data from a controller
+    // Returns success if nfc is supported
     virtual Common::Input::NfcState SupportsNfc(
         [[maybe_unused]] const PadIdentifier& identifier) const {
         return Common::Input::NfcState::NotSupported;

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

@@ -126,6 +126,7 @@ void Maxwell3D::InitializeRegisterDefaults() {
     draw_command[MAXWELL3D_REG_INDEX(draw_inline_index)] = true;
     draw_command[MAXWELL3D_REG_INDEX(inline_index_2x16.even)] = true;
     draw_command[MAXWELL3D_REG_INDEX(inline_index_4x8.index0)] = true;
+    draw_command[MAXWELL3D_REG_INDEX(draw.instance_id)] = true;
 }
 
 void Maxwell3D::ProcessMacro(u32 method, const u32* base_start, u32 amount, bool is_last_call) {
@@ -232,7 +233,7 @@ 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):
@@ -249,6 +250,8 @@ void Maxwell3D::ProcessMethodCall(u32 method, u32 argument, u32 nonshadow_argume
         return;
     case MAXWELL3D_REG_INDEX(fragment_barrier):
         return rasterizer->FragmentBarrier();
+    case MAXWELL3D_REG_INDEX(tiled_cache_barrier):
+        return rasterizer->TiledCacheBarrier();
     }
 }
 
@@ -283,31 +286,58 @@ void Maxwell3D::CallMethod(u32 method, u32 method_argument, bool is_last_call) {
     ASSERT_MSG(method < Regs::NUM_REGS,
                "Invalid Maxwell3D register, increase the size of the Regs structure");
 
+    const u32 argument = ProcessShadowRam(method, method_argument);
+    ProcessDirtyRegisters(method, argument);
+
     if (draw_command[method]) {
         regs.reg_array[method] = method_argument;
         deferred_draw_method.push_back(method);
-        auto u32_to_u8 = [&](const u32 argument) {
-            inline_index_draw_indexes.push_back(static_cast<u8>(argument & 0x000000ff));
-            inline_index_draw_indexes.push_back(static_cast<u8>((argument & 0x0000ff00) >> 8));
-            inline_index_draw_indexes.push_back(static_cast<u8>((argument & 0x00ff0000) >> 16));
-            inline_index_draw_indexes.push_back(static_cast<u8>((argument & 0xff000000) >> 24));
+        auto update_inline_index = [&](const u32 index) {
+            inline_index_draw_indexes.push_back(static_cast<u8>(index & 0x000000ff));
+            inline_index_draw_indexes.push_back(static_cast<u8>((index & 0x0000ff00) >> 8));
+            inline_index_draw_indexes.push_back(static_cast<u8>((index & 0x00ff0000) >> 16));
+            inline_index_draw_indexes.push_back(static_cast<u8>((index & 0xff000000) >> 24));
+            draw_mode = DrawMode::InlineIndex;
         };
-        if (MAXWELL3D_REG_INDEX(draw_inline_index) == method) {
-            u32_to_u8(method_argument);
-        } else if (MAXWELL3D_REG_INDEX(inline_index_2x16.even) == method) {
-            u32_to_u8(regs.inline_index_2x16.even);
-            u32_to_u8(regs.inline_index_2x16.odd);
-        } else if (MAXWELL3D_REG_INDEX(inline_index_4x8.index0) == method) {
-            u32_to_u8(regs.inline_index_4x8.index0);
-            u32_to_u8(regs.inline_index_4x8.index1);
-            u32_to_u8(regs.inline_index_4x8.index2);
-            u32_to_u8(regs.inline_index_4x8.index3);
+        switch (method) {
+        case MAXWELL3D_REG_INDEX(draw.end):
+            switch (draw_mode) {
+            case DrawMode::General:
+                ProcessDraw(1);
+                break;
+            case DrawMode::InlineIndex:
+                regs.index_buffer.count = static_cast<u32>(inline_index_draw_indexes.size() / 4);
+                regs.index_buffer.format = Regs::IndexFormat::UnsignedInt;
+                ProcessDraw(1);
+                inline_index_draw_indexes.clear();
+                break;
+            case DrawMode::Instance:
+                break;
+            }
+            break;
+        case MAXWELL3D_REG_INDEX(draw_inline_index):
+            update_inline_index(method_argument);
+            break;
+        case MAXWELL3D_REG_INDEX(inline_index_2x16.even):
+            update_inline_index(regs.inline_index_2x16.even);
+            update_inline_index(regs.inline_index_2x16.odd);
+            break;
+        case MAXWELL3D_REG_INDEX(inline_index_4x8.index0):
+            update_inline_index(regs.inline_index_4x8.index0);
+            update_inline_index(regs.inline_index_4x8.index1);
+            update_inline_index(regs.inline_index_4x8.index2);
+            update_inline_index(regs.inline_index_4x8.index3);
+            break;
+        case MAXWELL3D_REG_INDEX(draw.instance_id):
+            draw_mode =
+                (regs.draw.instance_id == Maxwell3D::Regs::Draw::InstanceId::Subsequent) ||
+                        (regs.draw.instance_id == Maxwell3D::Regs::Draw::InstanceId::Unchanged)
+                    ? DrawMode::Instance
+                    : DrawMode::General;
+            break;
         }
     } else {
         ProcessDeferredDraw();
-
-        const u32 argument = ProcessShadowRam(method, method_argument);
-        ProcessDirtyRegisters(method, argument);
         ProcessMethodCall(method, argument, method_argument, is_last_call);
     }
 }
@@ -506,10 +536,7 @@ 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);
 }
 
@@ -591,8 +618,8 @@ 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) {
@@ -621,57 +648,27 @@ void Maxwell3D::ProcessDraw(u32 instance_count) {
 }
 
 void Maxwell3D::ProcessDeferredDraw() {
-    if (deferred_draw_method.empty()) {
+    if (draw_mode != DrawMode::Instance || deferred_draw_method.empty()) {
         return;
     }
 
-    enum class DrawMode {
-        Undefined,
-        General,
-        Instance,
-    };
-    DrawMode draw_mode{DrawMode::Undefined};
     u32 method_count = static_cast<u32>(deferred_draw_method.size());
-    u32 method = deferred_draw_method[method_count - 1];
-    if (MAXWELL3D_REG_INDEX(draw.end) != method) {
-        return;
-    }
-    draw_mode = (regs.draw.instance_id == Maxwell3D::Regs::Draw::InstanceId::Subsequent) ||
-                        (regs.draw.instance_id == Maxwell3D::Regs::Draw::InstanceId::Unchanged)
-                    ? DrawMode::Instance
-                    : DrawMode::General;
-    u32 instance_count = 0;
-    if (draw_mode == DrawMode::Instance) {
-        u32 vertex_buffer_count = 0;
-        u32 index_buffer_count = 0;
-        for (u32 index = 0; index < method_count; ++index) {
-            method = deferred_draw_method[index];
-            if (method == MAXWELL3D_REG_INDEX(vertex_buffer.count)) {
-                instance_count = ++vertex_buffer_count;
-            } else if (method == MAXWELL3D_REG_INDEX(index_buffer.count)) {
-                instance_count = ++index_buffer_count;
-            }
-        }
-        ASSERT_MSG(!(vertex_buffer_count && index_buffer_count),
-                   "Instance both indexed and direct?");
-    } else {
-        instance_count = 1;
-        for (u32 index = 0; index < method_count; ++index) {
-            method = deferred_draw_method[index];
-            if (MAXWELL3D_REG_INDEX(draw_inline_index) == method ||
-                MAXWELL3D_REG_INDEX(inline_index_2x16.even) == method ||
-                MAXWELL3D_REG_INDEX(inline_index_4x8.index0) == method) {
-                regs.index_buffer.count = static_cast<u32>(inline_index_draw_indexes.size() / 4);
-                regs.index_buffer.format = Regs::IndexFormat::UnsignedInt;
-                break;
-            }
+    u32 instance_count = 1;
+    u32 vertex_buffer_count = 0;
+    u32 index_buffer_count = 0;
+    for (u32 index = 0; index < method_count; ++index) {
+        u32 method = deferred_draw_method[index];
+        if (method == MAXWELL3D_REG_INDEX(vertex_buffer.count)) {
+            instance_count = ++vertex_buffer_count;
+        } else if (method == MAXWELL3D_REG_INDEX(index_buffer.count)) {
+            instance_count = ++index_buffer_count;
         }
     }
+    ASSERT_MSG(!(vertex_buffer_count && index_buffer_count), "Instance both indexed and direct?");
 
     ProcessDraw(instance_count);
 
     deferred_draw_method.clear();
-    inline_index_draw_indexes.clear();
 }
 
 } // namespace Tegra::Engines

src/video_core/engines/maxwell_3d.h

@@ -707,7 +707,7 @@ public:
                 case Size::Size_A2_B10_G10_R10:
                     return "2_10_10_10";
                 case Size::Size_B10_G11_R11:
-                    return "10_11_12";
+                    return "10_11_11";
                 default:
                     ASSERT(false);
                     return {};
@@ -2639,7 +2639,7 @@ public:
                 L2CacheControl l2_cache_control;                                       ///< 0x0218
                 InvalidateShaderCache invalidate_shader_cache;                         ///< 0x021C
                 INSERT_PADDING_BYTES_NOINIT(0xA8);
-                SyncInfo sync_info; ///< 0x02C8
+                SyncInfo sync_info;                                                    ///< 0x02C8
                 INSERT_PADDING_BYTES_NOINIT(0x4);
                 u32 prim_circular_buffer_throttle;                                     ///< 0x02D0
                 u32 flush_invalidate_rop_mini_cache;                                   ///< 0x02D4
@@ -2731,7 +2731,11 @@ public:
                 s32 stencil_back_ref;                                                  ///< 0x0F54
                 u32 stencil_back_mask;                                                 ///< 0x0F58
                 u32 stencil_back_func_mask;                                            ///< 0x0F5C
-                INSERT_PADDING_BYTES_NOINIT(0x24);
+                INSERT_PADDING_BYTES_NOINIT(0x14);
+                u32 invalidate_texture_data_cache;                                     ///< 0x0F74 Assumed - Not in official docs.
+                INSERT_PADDING_BYTES_NOINIT(0x4);
+                u32 tiled_cache_barrier;                                               ///< 0x0F7C Assumed - Not in official docs.
+                INSERT_PADDING_BYTES_NOINIT(0x4);
                 VertexStreamSubstitute vertex_stream_substitute;                       ///< 0x0F84
                 u32 line_mode_clip_generated_edge_do_not_draw;                         ///< 0x0F8C
                 u32 color_mask_common;                                                 ///< 0x0F90
@@ -2791,7 +2795,8 @@ public:
                 FillViaTriangleMode fill_via_triangle_mode;                            ///< 0x113C
                 u32 blend_per_format_snorm8_unorm16_snorm16_enabled;                   ///< 0x1140
                 u32 flush_pending_writes_sm_gloal_store;                               ///< 0x1144
-                INSERT_PADDING_BYTES_NOINIT(0x18);
+                u32 conservative_raster_enable;                                        ///< 0x1148 Assumed - Not in official docs.
+                INSERT_PADDING_BYTES_NOINIT(0x14);
                 std::array<VertexAttribute, NumVertexAttributes> vertex_attrib_format; ///< 0x1160
                 std::array<MsaaSampleLocation, 4> multisample_sample_locations;        ///< 0x11E0
                 u32 offset_render_target_index_by_viewport_index;                      ///< 0x11F0
@@ -3081,6 +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();
 
@@ -3115,9 +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();
 
@@ -3143,10 +3148,12 @@ private:
     /// Handles use of topology overrides (e.g., to avoid using a topology assigned from a macro)
     void ProcessTopologyOverride();
 
-    void ProcessDraw(u32 instance_count = 1);
-
+    /// Handles deferred draw(e.g., instance draw).
     void ProcessDeferredDraw();
 
+    /// Handles a draw.
+    void ProcessDraw(u32 instance_count = 1);
+
     /// Returns a query's value or an empty object if the value will be deferred through a cache.
     std::optional<u64> GetQueryResult();
 
@@ -3173,6 +3180,8 @@ private:
 
     std::array<bool, Regs::NUM_REGS> draw_command{};
     std::vector<u32> deferred_draw_method;
+    enum class DrawMode : u32 { General = 0, Instance, InlineIndex };
+    DrawMode draw_mode{DrawMode::General};
 };
 
 #define ASSERT_REG_POSITION(field_name, position)                                                  \
@@ -3287,6 +3296,8 @@ ASSERT_REG_POSITION(const_color_rendering, 0x0F40);
 ASSERT_REG_POSITION(stencil_back_ref, 0x0F54);
 ASSERT_REG_POSITION(stencil_back_mask, 0x0F58);
 ASSERT_REG_POSITION(stencil_back_func_mask, 0x0F5C);
+ASSERT_REG_POSITION(invalidate_texture_data_cache, 0x0F74);
+ASSERT_REG_POSITION(tiled_cache_barrier, 0x0F7C);
 ASSERT_REG_POSITION(vertex_stream_substitute, 0x0F84);
 ASSERT_REG_POSITION(line_mode_clip_generated_edge_do_not_draw, 0x0F8C);
 ASSERT_REG_POSITION(color_mask_common, 0x0F90);
@@ -3343,6 +3354,7 @@ ASSERT_REG_POSITION(post_ps_use_pre_ps_coverage, 0x1138);
 ASSERT_REG_POSITION(fill_via_triangle_mode, 0x113C);
 ASSERT_REG_POSITION(blend_per_format_snorm8_unorm16_snorm16_enabled, 0x1140);
 ASSERT_REG_POSITION(flush_pending_writes_sm_gloal_store, 0x1144);
+ASSERT_REG_POSITION(conservative_raster_enable, 0x1148);
 ASSERT_REG_POSITION(vertex_attrib_format, 0x1160);
 ASSERT_REG_POSITION(multisample_sample_locations, 0x11E0);
 ASSERT_REG_POSITION(offset_render_target_index_by_viewport_index, 0x11F0);

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;
         }
 
@@ -102,26 +103,29 @@ void MaxwellDMA::Launch() {
             const bool is_src_pitch = IsPitchKind(static_cast<PTEKind>(src_kind));
             const bool is_dst_pitch = IsPitchKind(static_cast<PTEKind>(dst_kind));
             if (!is_src_pitch && is_dst_pitch) {
-                std::vector<u8> tmp_buffer(regs.line_length_in);
-                std::vector<u8> dst_buffer(regs.line_length_in);
-                memory_manager.ReadBlockUnsafe(regs.offset_in, tmp_buffer.data(),
-                                               regs.line_length_in);
-                for (u32 offset = 0; offset < regs.line_length_in; ++offset) {
-                    dst_buffer[offset] =
-                        tmp_buffer[convert_linear_2_blocklinear_addr(regs.offset_in + offset) -
-                                   regs.offset_in];
+                UNIMPLEMENTED_IF(regs.line_length_in % 16 != 0);
+                UNIMPLEMENTED_IF(regs.offset_in % 16 != 0);
+                UNIMPLEMENTED_IF(regs.offset_out % 16 != 0);
+                std::vector<u8> tmp_buffer(16);
+                for (u32 offset = 0; offset < regs.line_length_in; offset += 16) {
+                    memory_manager.ReadBlockUnsafe(
+                        convert_linear_2_blocklinear_addr(regs.offset_in + offset),
+                        tmp_buffer.data(), tmp_buffer.size());
+                    memory_manager.WriteBlock(regs.offset_out + offset, tmp_buffer.data(),
+                                              tmp_buffer.size());
                 }
-                memory_manager.WriteBlock(regs.offset_out, dst_buffer.data(), regs.line_length_in);
             } else if (is_src_pitch && !is_dst_pitch) {
-                std::vector<u8> tmp_buffer(regs.line_length_in);
-                std::vector<u8> dst_buffer(regs.line_length_in);
-                memory_manager.ReadBlockUnsafe(regs.offset_in, tmp_buffer.data(),
-                                               regs.line_length_in);
-                for (u32 offset = 0; offset < regs.line_length_in; ++offset) {
-                    dst_buffer[convert_linear_2_blocklinear_addr(regs.offset_out + offset) -
-                               regs.offset_out] = tmp_buffer[offset];
+                UNIMPLEMENTED_IF(regs.line_length_in % 16 != 0);
+                UNIMPLEMENTED_IF(regs.offset_in % 16 != 0);
+                UNIMPLEMENTED_IF(regs.offset_out % 16 != 0);
+                std::vector<u8> tmp_buffer(16);
+                for (u32 offset = 0; offset < regs.line_length_in; offset += 16) {
+                    memory_manager.ReadBlockUnsafe(regs.offset_in + offset, tmp_buffer.data(),
+                                                   tmp_buffer.size());
+                    memory_manager.WriteBlock(
+                        convert_linear_2_blocklinear_addr(regs.offset_out + offset),
+                        tmp_buffer.data(), tmp_buffer.size());
                 }
-                memory_manager.WriteBlock(regs.offset_out, dst_buffer.data(), regs.line_length_in);
             } else {
                 if (!accelerate.BufferCopy(regs.offset_in, regs.offset_out, regs.line_length_in)) {
                     std::vector<u8> tmp_buffer(regs.line_length_in);
@@ -288,6 +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;
@@ -311,6 +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()