Initialize polygon_offset in the constructor.

CMakeLists.txt

@@ -377,7 +377,7 @@ if (CLANG_FORMAT)
     set(CCOMMENT "Running clang format against all the .h and .cpp files in src/")
     if (WIN32)
         add_custom_target(clang-format
-            COMMAND powershell.exe -Command "Get-ChildItem ${SRCS}/* -Include *.cpp,*.h -Recurse | Foreach {${CLANG_FORMAT} -i $_.fullname}"
+            COMMAND powershell.exe -Command "${CLANG_FORMAT} -i @(Get-ChildItem -Recurse ${SRCS}/* -Include \'*.h\', \'*.cpp\')"
             COMMENT ${CCOMMENT})
     elseif(MINGW)
         add_custom_target(clang-format

src/common/thread.cpp

@@ -25,6 +25,23 @@
 
 namespace Common {
 
+int CurrentThreadId() {
+#ifdef _MSC_VER
+    return GetCurrentThreadId();
+#elif defined __APPLE__
+    return mach_thread_self();
+#else
+    return 0;
+#endif
+}
+
+#ifdef _WIN32
+// Supporting functions
+void SleepCurrentThread(int ms) {
+    Sleep(ms);
+}
+#endif
+
 #ifdef _MSC_VER
 
 void SetThreadAffinity(std::thread::native_handle_type thread, u32 mask) {
@@ -45,7 +62,7 @@ void SwitchCurrentThread() {
 
 // This is implemented much nicer in upcoming msvc++, see:
 // http://msdn.microsoft.com/en-us/library/xcb2z8hs(VS.100).aspx
-void SetCurrentThreadName(const char* name) {
+void SetCurrentThreadName(const char* szThreadName) {
     static const DWORD MS_VC_EXCEPTION = 0x406D1388;
 
 #pragma pack(push, 8)
@@ -58,7 +75,7 @@ void SetCurrentThreadName(const char* name) {
 #pragma pack(pop)
 
     info.dwType = 0x1000;
-    info.szName = name;
+    info.szName = szThreadName;
     info.dwThreadID = -1; // dwThreadID;
     info.dwFlags = 0;
 
@@ -90,6 +107,10 @@ void SetCurrentThreadAffinity(u32 mask) {
 }
 
 #ifndef _WIN32
+void SleepCurrentThread(int ms) {
+    usleep(1000 * ms);
+}
+
 void SwitchCurrentThread() {
     usleep(1000 * 1);
 }
@@ -97,15 +118,15 @@ void SwitchCurrentThread() {
 
 // MinGW with the POSIX threading model does not support pthread_setname_np
 #if !defined(_WIN32) || defined(_MSC_VER)
-void SetCurrentThreadName(const char* name) {
+void SetCurrentThreadName(const char* szThreadName) {
 #ifdef __APPLE__
-    pthread_setname_np(name);
+    pthread_setname_np(szThreadName);
 #elif defined(__Bitrig__) || defined(__DragonFly__) || defined(__FreeBSD__) || defined(__OpenBSD__)
-    pthread_set_name_np(pthread_self(), name);
+    pthread_set_name_np(pthread_self(), szThreadName);
 #elif defined(__NetBSD__)
-    pthread_setname_np(pthread_self(), "%s", (void*)name);
+    pthread_setname_np(pthread_self(), "%s", (void*)szThreadName);
 #else
-    pthread_setname_np(pthread_self(), name);
+    pthread_setname_np(pthread_self(), szThreadName);
 #endif
 }
 #endif

src/common/thread.h

@@ -13,8 +13,15 @@
 
 namespace Common {
 
+int CurrentThreadId();
+
+void SetThreadAffinity(std::thread::native_handle_type thread, u32 mask);
+void SetCurrentThreadAffinity(u32 mask);
+
 class Event {
 public:
+    Event() : is_set(false) {}
+
     void Set() {
         std::lock_guard<std::mutex> lk(mutex);
         if (!is_set) {
@@ -46,14 +53,14 @@ public:
     }
 
 private:
-    bool is_set = false;
+    bool is_set;
     std::condition_variable condvar;
     std::mutex mutex;
 };
 
 class Barrier {
 public:
-    explicit Barrier(std::size_t count_) : count(count_) {}
+    explicit Barrier(std::size_t count_) : count(count_), waiting(0), generation(0) {}
 
     /// Blocks until all "count" threads have called Sync()
     void Sync() {
@@ -73,13 +80,12 @@ public:
 private:
     std::condition_variable condvar;
     std::mutex mutex;
-    std::size_t count;
-    std::size_t waiting = 0;
-    std::size_t generation = 0; // Incremented once each time the barrier is used
+    const std::size_t count;
+    std::size_t waiting;
+    std::size_t generation; // Incremented once each time the barrier is used
 };
 
-void SetThreadAffinity(std::thread::native_handle_type thread, u32 mask);
-void SetCurrentThreadAffinity(u32 mask);
+void SleepCurrentThread(int ms);
 void SwitchCurrentThread(); // On Linux, this is equal to sleep 1ms
 void SetCurrentThreadName(const char* name);
 

src/core/CMakeLists.txt

@@ -12,8 +12,6 @@ add_library(core STATIC
     core_timing.h
     core_timing_util.cpp
     core_timing_util.h
-    cpu_core_manager.cpp
-    cpu_core_manager.h
     crypto/aes_util.cpp
     crypto/aes_util.h
     crypto/encryption_layer.cpp
@@ -158,8 +156,6 @@ add_library(core STATIC
     hle/service/am/applets/applets.h
     hle/service/am/applets/software_keyboard.cpp
     hle/service/am/applets/software_keyboard.h
-    hle/service/am/applets/stub_applet.cpp
-    hle/service/am/applets/stub_applet.h
     hle/service/am/idle.cpp
     hle/service/am/idle.h
     hle/service/am/omm.cpp

src/core/core.cpp

@@ -14,7 +14,6 @@
 #include "core/core.h"
 #include "core/core_cpu.h"
 #include "core/core_timing.h"
-#include "core/cpu_core_manager.h"
 #include "core/file_sys/mode.h"
 #include "core/file_sys/vfs_concat.h"
 #include "core/file_sys/vfs_real.h"
@@ -29,6 +28,7 @@
 #include "core/hle/service/sm/sm.h"
 #include "core/loader/loader.h"
 #include "core/perf_stats.h"
+#include "core/settings.h"
 #include "core/telemetry_session.h"
 #include "frontend/applets/software_keyboard.h"
 #include "video_core/debug_utils/debug_utils.h"
@@ -71,22 +71,64 @@ FileSys::VirtualFile GetGameFileFromPath(const FileSys::VirtualFilesystem& vfs,
 
     return vfs->OpenFile(path, FileSys::Mode::Read);
 }
+
+/// Runs a CPU core while the system is powered on
+void RunCpuCore(Cpu& cpu_state) {
+    while (Core::System::GetInstance().IsPoweredOn()) {
+        cpu_state.RunLoop(true);
+    }
+}
 } // Anonymous namespace
 
 struct System::Impl {
     Cpu& CurrentCpuCore() {
-        return cpu_core_manager.GetCurrentCore();
+        if (Settings::values.use_multi_core) {
+            const auto& search = thread_to_cpu.find(std::this_thread::get_id());
+            ASSERT(search != thread_to_cpu.end());
+            ASSERT(search->second);
+            return *search->second;
+        }
+
+        // Otherwise, use single-threaded mode active_core variable
+        return *cpu_cores[active_core];
     }
 
     ResultStatus RunLoop(bool tight_loop) {
         status = ResultStatus::Success;
 
-        cpu_core_manager.RunLoop(tight_loop);
+        // Update thread_to_cpu in case Core 0 is run from a different host thread
+        thread_to_cpu[std::this_thread::get_id()] = cpu_cores[0].get();
+
+        if (GDBStub::IsServerEnabled()) {
+            GDBStub::HandlePacket();
+
+            // If the loop is halted and we want to step, use a tiny (1) number of instructions to
+            // execute. Otherwise, get out of the loop function.
+            if (GDBStub::GetCpuHaltFlag()) {
+                if (GDBStub::GetCpuStepFlag()) {
+                    tight_loop = false;
+                } else {
+                    return ResultStatus::Success;
+                }
+            }
+        }
+
+        for (active_core = 0; active_core < NUM_CPU_CORES; ++active_core) {
+            cpu_cores[active_core]->RunLoop(tight_loop);
+            if (Settings::values.use_multi_core) {
+                // Cores 1-3 are run on other threads in this mode
+                break;
+            }
+        }
+
+        if (GDBStub::IsServerEnabled()) {
+            GDBStub::SetCpuStepFlag(false);
+        }
 
         return status;
     }
 
-    ResultStatus Init(System& system, Frontend::EmuWindow& emu_window) {
+    ResultStatus Init(Frontend::EmuWindow& emu_window) {
         LOG_DEBUG(HW_Memory, "initialized OK");
 
         CoreTiming::Init();
@@ -103,6 +145,12 @@ struct System::Impl {
         auto main_process = Kernel::Process::Create(kernel, "main");
         kernel.MakeCurrentProcess(main_process.get());
 
+        cpu_barrier = std::make_unique<CpuBarrier>();
+        cpu_exclusive_monitor = Cpu::MakeExclusiveMonitor(cpu_cores.size());
+        for (std::size_t index = 0; index < cpu_cores.size(); ++index) {
+            cpu_cores[index] = std::make_unique<Cpu>(*cpu_exclusive_monitor, *cpu_barrier, index);
+        }
+
         telemetry_session = std::make_unique<Core::TelemetrySession>();
         service_manager = std::make_shared<Service::SM::ServiceManager>();
 
@@ -116,8 +164,17 @@ struct System::Impl {
 
         gpu_core = std::make_unique<Tegra::GPU>(renderer->Rasterizer());
 
-        cpu_core_manager.Initialize(system);
-        is_powered_on = true;
+        // Create threads for CPU cores 1-3, and build thread_to_cpu map
+        // CPU core 0 is run on the main thread
+        thread_to_cpu[std::this_thread::get_id()] = cpu_cores[0].get();
+        if (Settings::values.use_multi_core) {
+            for (std::size_t index = 0; index < cpu_core_threads.size(); ++index) {
+                cpu_core_threads[index] =
+                    std::make_unique<std::thread>(RunCpuCore, std::ref(*cpu_cores[index + 1]));
+                thread_to_cpu[cpu_core_threads[index]->get_id()] = cpu_cores[index + 1].get();
+            }
+        }
+
         LOG_DEBUG(Core, "Initialized OK");
 
         // Reset counters and set time origin to current frame
@@ -127,8 +184,7 @@ struct System::Impl {
         return ResultStatus::Success;
     }
 
-    ResultStatus Load(System& system, Frontend::EmuWindow& emu_window,
-                      const std::string& filepath) {
+    ResultStatus Load(Frontend::EmuWindow& emu_window, const std::string& filepath) {
         app_loader = Loader::GetLoader(GetGameFileFromPath(virtual_filesystem, filepath));
 
         if (!app_loader) {
@@ -145,7 +201,7 @@ struct System::Impl {
             return ResultStatus::ErrorSystemMode;
         }
 
-        ResultStatus init_result{Init(system, emu_window)};
+        ResultStatus init_result{Init(emu_window)};
         if (init_result != ResultStatus::Success) {
             LOG_CRITICAL(Core, "Failed to initialize system (Error {})!",
                          static_cast<int>(init_result));
@@ -175,8 +231,6 @@ struct System::Impl {
         Telemetry().AddField(Telemetry::FieldType::Performance, "Shutdown_Frametime",
                              perf_results.frametime * 1000.0);
 
-        is_powered_on = false;
-
         // Shutdown emulation session
         renderer.reset();
         GDBStub::Shutdown();
@@ -186,7 +240,19 @@ struct System::Impl {
         gpu_core.reset();
 
         // Close all CPU/threading state
-        cpu_core_manager.Shutdown();
+        cpu_barrier->NotifyEnd();
+        if (Settings::values.use_multi_core) {
+            for (auto& thread : cpu_core_threads) {
+                thread->join();
+                thread.reset();
+            }
+        }
+        thread_to_cpu.clear();
+        for (auto& cpu_core : cpu_cores) {
+            cpu_core.reset();
+        }
+        cpu_exclusive_monitor.reset();
+        cpu_barrier.reset();
 
         // Shutdown kernel and core timing
         kernel.Shutdown();
@@ -223,8 +289,11 @@ struct System::Impl {
     std::unique_ptr<VideoCore::RendererBase> renderer;
     std::unique_ptr<Tegra::GPU> gpu_core;
     std::shared_ptr<Tegra::DebugContext> debug_context;
-    CpuCoreManager cpu_core_manager;
-    bool is_powered_on = false;
+    std::unique_ptr<ExclusiveMonitor> cpu_exclusive_monitor;
+    std::unique_ptr<CpuBarrier> cpu_barrier;
+    std::array<std::unique_ptr<Cpu>, NUM_CPU_CORES> cpu_cores;
+    std::array<std::unique_ptr<std::thread>, NUM_CPU_CORES - 1> cpu_core_threads;
+    std::size_t active_core{}; ///< Active core, only used in single thread mode
 
     /// Frontend applets
     std::unique_ptr<Core::Frontend::SoftwareKeyboardApplet> software_keyboard;
@@ -238,6 +307,9 @@ struct System::Impl {
     ResultStatus status = ResultStatus::Success;
     std::string status_details = "";
 
+    /// Map of guest threads to CPU cores
+    std::map<std::thread::id, Cpu*> thread_to_cpu;
+
     Core::PerfStats perf_stats;
     Core::FrameLimiter frame_limiter;
 };
@@ -262,15 +334,17 @@ System::ResultStatus System::SingleStep() {
 }
 
 void System::InvalidateCpuInstructionCaches() {
-    impl->cpu_core_manager.InvalidateAllInstructionCaches();
+    for (auto& cpu : impl->cpu_cores) {
+        cpu->ArmInterface().ClearInstructionCache();
+    }
 }
 
 System::ResultStatus System::Load(Frontend::EmuWindow& emu_window, const std::string& filepath) {
-    return impl->Load(*this, emu_window, filepath);
+    return impl->Load(emu_window, filepath);
 }
 
 bool System::IsPoweredOn() const {
-    return impl->is_powered_on;
+    return impl->cpu_barrier && impl->cpu_barrier->IsAlive();
 }
 
 void System::PrepareReschedule() {
@@ -334,20 +408,21 @@ const ARM_Interface& System::ArmInterface(std::size_t core_index) const {
 }
 
 Cpu& System::CpuCore(std::size_t core_index) {
-    return impl->cpu_core_manager.GetCore(core_index);
+    ASSERT(core_index < NUM_CPU_CORES);
+    return *impl->cpu_cores[core_index];
 }
 
 const Cpu& System::CpuCore(std::size_t core_index) const {
     ASSERT(core_index < NUM_CPU_CORES);
-    return impl->cpu_core_manager.GetCore(core_index);
+    return *impl->cpu_cores[core_index];
 }
 
 ExclusiveMonitor& System::Monitor() {
-    return impl->cpu_core_manager.GetExclusiveMonitor();
+    return *impl->cpu_exclusive_monitor;
 }
 
 const ExclusiveMonitor& System::Monitor() const {
-    return impl->cpu_core_manager.GetExclusiveMonitor();
+    return *impl->cpu_exclusive_monitor;
 }
 
 Tegra::GPU& System::GPU() {
@@ -431,7 +506,7 @@ const Core::Frontend::SoftwareKeyboardApplet& System::GetSoftwareKeyboard() cons
 }
 
 System::ResultStatus System::Init(Frontend::EmuWindow& emu_window) {
-    return impl->Init(*this, emu_window);
+    return impl->Init(emu_window);
 }
 
 void System::Shutdown() {

src/core/cpu_core_manager.cpp

@@ -1,142 +0,0 @@
-// Copyright 2018 yuzu emulator team
-// Licensed under GPLv2 or any later version
-// Refer to the license.txt file included.
-
-#include "common/assert.h"
-#include "core/arm/exclusive_monitor.h"
-#include "core/core.h"
-#include "core/core_cpu.h"
-#include "core/cpu_core_manager.h"
-#include "core/gdbstub/gdbstub.h"
-#include "core/settings.h"
-
-namespace Core {
-namespace {
-void RunCpuCore(const System& system, Cpu& cpu_state) {
-    while (system.IsPoweredOn()) {
-        cpu_state.RunLoop(true);
-    }
-}
-} // Anonymous namespace
-
-CpuCoreManager::CpuCoreManager() = default;
-CpuCoreManager::~CpuCoreManager() = default;
-
-void CpuCoreManager::Initialize(System& system) {
-    barrier = std::make_unique<CpuBarrier>();
-    exclusive_monitor = Cpu::MakeExclusiveMonitor(cores.size());
-
-    for (std::size_t index = 0; index < cores.size(); ++index) {
-        cores[index] = std::make_unique<Cpu>(*exclusive_monitor, *barrier, index);
-    }
-
-    // Create threads for CPU cores 1-3, and build thread_to_cpu map
-    // CPU core 0 is run on the main thread
-    thread_to_cpu[std::this_thread::get_id()] = cores[0].get();
-    if (!Settings::values.use_multi_core) {
-        return;
-    }
-
-    for (std::size_t index = 0; index < core_threads.size(); ++index) {
-        core_threads[index] = std::make_unique<std::thread>(RunCpuCore, std::cref(system),
-                                                            std::ref(*cores[index + 1]));
-        thread_to_cpu[core_threads[index]->get_id()] = cores[index + 1].get();
-    }
-}
-
-void CpuCoreManager::Shutdown() {
-    barrier->NotifyEnd();
-    if (Settings::values.use_multi_core) {
-        for (auto& thread : core_threads) {
-            thread->join();
-            thread.reset();
-        }
-    }
-
-    thread_to_cpu.clear();
-    for (auto& cpu_core : cores) {
-        cpu_core.reset();
-    }
-
-    exclusive_monitor.reset();
-    barrier.reset();
-}
-
-Cpu& CpuCoreManager::GetCore(std::size_t index) {
-    return *cores.at(index);
-}
-
-const Cpu& CpuCoreManager::GetCore(std::size_t index) const {
-    return *cores.at(index);
-}
-
-ExclusiveMonitor& CpuCoreManager::GetExclusiveMonitor() {
-    return *exclusive_monitor;
-}
-
-const ExclusiveMonitor& CpuCoreManager::GetExclusiveMonitor() const {
-    return *exclusive_monitor;
-}
-
-Cpu& CpuCoreManager::GetCurrentCore() {
-    if (Settings::values.use_multi_core) {
-        const auto& search = thread_to_cpu.find(std::this_thread::get_id());
-        ASSERT(search != thread_to_cpu.end());
-        ASSERT(search->second);
-        return *search->second;
-    }
-
-    // Otherwise, use single-threaded mode active_core variable
-    return *cores[active_core];
-}
-
-const Cpu& CpuCoreManager::GetCurrentCore() const {
-    if (Settings::values.use_multi_core) {
-        const auto& search = thread_to_cpu.find(std::this_thread::get_id());
-        ASSERT(search != thread_to_cpu.end());
-        ASSERT(search->second);
-        return *search->second;
-    }
-
-    // Otherwise, use single-threaded mode active_core variable
-    return *cores[active_core];
-}
-
-void CpuCoreManager::RunLoop(bool tight_loop) {
-    // Update thread_to_cpu in case Core 0 is run from a different host thread
-    thread_to_cpu[std::this_thread::get_id()] = cores[0].get();
-
-    if (GDBStub::IsServerEnabled()) {
-        GDBStub::HandlePacket();
-
-        // If the loop is halted and we want to step, use a tiny (1) number of instructions to
-        // execute. Otherwise, get out of the loop function.
-        if (GDBStub::GetCpuHaltFlag()) {
-            if (GDBStub::GetCpuStepFlag()) {
-                tight_loop = false;
-            } else {
-                return;
-            }
-        }
-    }
-
-    for (active_core = 0; active_core < NUM_CPU_CORES; ++active_core) {
-        cores[active_core]->RunLoop(tight_loop);
-        if (Settings::values.use_multi_core) {
-            // Cores 1-3 are run on other threads in this mode
-            break;
-        }
-    }
-
-    if (GDBStub::IsServerEnabled()) {
-        GDBStub::SetCpuStepFlag(false);
-    }
-}
-
-void CpuCoreManager::InvalidateAllInstructionCaches() {
-    for (auto& cpu : cores) {
-        cpu->ArmInterface().ClearInstructionCache();
-    }
-}
-
-} // namespace Core

src/core/cpu_core_manager.h

@@ -1,59 +0,0 @@
-// Copyright 2018 yuzu emulator team
-// Licensed under GPLv2 or any later version
-// Refer to the license.txt file included.
-
-#pragma once
-
-#include <array>
-#include <map>
-#include <memory>
-#include <thread>
-
-namespace Core {
-
-class Cpu;
-class CpuBarrier;
-class ExclusiveMonitor;
-class System;
-
-class CpuCoreManager {
-public:
-    CpuCoreManager();
-    CpuCoreManager(const CpuCoreManager&) = delete;
-    CpuCoreManager(CpuCoreManager&&) = delete;
-
-    ~CpuCoreManager();
-
-    CpuCoreManager& operator=(const CpuCoreManager&) = delete;
-    CpuCoreManager& operator=(CpuCoreManager&&) = delete;
-
-    void Initialize(System& system);
-    void Shutdown();
-
-    Cpu& GetCore(std::size_t index);
-    const Cpu& GetCore(std::size_t index) const;
-
-    Cpu& GetCurrentCore();
-    const Cpu& GetCurrentCore() const;
-
-    ExclusiveMonitor& GetExclusiveMonitor();
-    const ExclusiveMonitor& GetExclusiveMonitor() const;
-
-    void RunLoop(bool tight_loop);
-
-    void InvalidateAllInstructionCaches();
-
-private:
-    static constexpr std::size_t NUM_CPU_CORES = 4;
-
-    std::unique_ptr<ExclusiveMonitor> exclusive_monitor;
-    std::unique_ptr<CpuBarrier> barrier;
-    std::array<std::unique_ptr<Cpu>, NUM_CPU_CORES> cores;
-    std::array<std::unique_ptr<std::thread>, NUM_CPU_CORES - 1> core_threads;
-    std::size_t active_core{}; ///< Active core, only used in single thread mode
-
-    /// Map of guest threads to CPU cores
-    std::map<std::thread::id, Cpu*> thread_to_cpu;
-};
-
-} // namespace Core

src/core/file_sys/patch_manager.cpp

@@ -26,11 +26,6 @@ namespace FileSys {
 constexpr u64 SINGLE_BYTE_MODULUS = 0x100;
 constexpr u64 DLC_BASE_TITLE_ID_MASK = 0xFFFFFFFFFFFFE000;
 
-constexpr std::array<const char*, 14> EXEFS_FILE_NAMES{
-    "main",    "main.npdm", "rtld",    "sdk",     "subsdk0", "subsdk1", "subsdk2",
-    "subsdk3", "subsdk4",   "subsdk5", "subsdk6", "subsdk7", "subsdk8", "subsdk9",
-};
-
 struct NSOBuildHeader {
     u32_le magic;
     INSERT_PADDING_BYTES(0x3C);
@@ -62,15 +57,6 @@ VirtualDir PatchManager::PatchExeFS(VirtualDir exefs) const {
     if (exefs == nullptr)
         return exefs;
 
-    if (Settings::values.dump_exefs) {
-        LOG_INFO(Loader, "Dumping ExeFS for title_id={:016X}", title_id);
-        const auto dump_dir = Service::FileSystem::GetModificationDumpRoot(title_id);
-        if (dump_dir != nullptr) {
-            const auto exefs_dir = GetOrCreateDirectoryRelative(dump_dir, "/exefs");
-            VfsRawCopyD(exefs, exefs_dir);
-        }
-    }
-
     const auto installed = Service::FileSystem::GetUnionContents();
 
     // Game Updates
@@ -84,30 +70,6 @@ VirtualDir PatchManager::PatchExeFS(VirtualDir exefs) const {
         exefs = update->GetExeFS();
     }
 
-    // LayeredExeFS
-    const auto load_dir = Service::FileSystem::GetModificationLoadRoot(title_id);
-    if (load_dir != nullptr && load_dir->GetSize() > 0) {
-        auto patch_dirs = load_dir->GetSubdirectories();
-        std::sort(
-            patch_dirs.begin(), patch_dirs.end(),
-            [](const VirtualDir& l, const VirtualDir& r) { return l->GetName() < r->GetName(); });
-
-        std::vector<VirtualDir> layers;
-        layers.reserve(patch_dirs.size() + 1);
-        for (const auto& subdir : patch_dirs) {
-            auto exefs_dir = subdir->GetSubdirectory("exefs");
-            if (exefs_dir != nullptr)
-                layers.push_back(std::move(exefs_dir));
-        }
-        layers.push_back(exefs);
-
-        auto layered = LayeredVfsDirectory::MakeLayeredDirectory(std::move(layers));
-        if (layered != nullptr) {
-            LOG_INFO(Loader, "    ExeFS: LayeredExeFS patches applied successfully");
-            exefs = std::move(layered);
-        }
-    }
-
     return exefs;
 }
 
@@ -352,25 +314,18 @@ std::map<std::string, std::string, std::less<>> PatchManager::GetPatchVersionNam
             if (IsDirValidAndNonEmpty(exefs_dir)) {
                 bool ips = false;
                 bool ipswitch = false;
-                bool layeredfs = false;
 
                 for (const auto& file : exefs_dir->GetFiles()) {
-                    if (file->GetExtension() == "ips") {
+                    if (file->GetExtension() == "ips")
                         ips = true;
-                    } else if (file->GetExtension() == "pchtxt") {
+                    else if (file->GetExtension() == "pchtxt")
                         ipswitch = true;
-                    } else if (std::find(EXEFS_FILE_NAMES.begin(), EXEFS_FILE_NAMES.end(),
-                                         file->GetName()) != EXEFS_FILE_NAMES.end()) {
-                        layeredfs = true;
-                    }
                 }
 
                 if (ips)
                     AppendCommaIfNotEmpty(types, "IPS");
                 if (ipswitch)
                     AppendCommaIfNotEmpty(types, "IPSwitch");
-                if (layeredfs)
-                    AppendCommaIfNotEmpty(types, "LayeredExeFS");
             }
             if (IsDirValidAndNonEmpty(mod->GetSubdirectory("romfs")))
                 AppendCommaIfNotEmpty(types, "LayeredFS");

src/core/gdbstub/gdbstub.cpp

@@ -71,6 +71,10 @@ constexpr u32 PSTATE_REGISTER = 33;
 constexpr u32 UC_ARM64_REG_Q0 = 34;
 constexpr u32 FPCR_REGISTER = 66;
 
+// TODO/WiP - Used while working on support for FPU
+constexpr u32 TODO_DUMMY_REG_997 = 997;
+constexpr u32 TODO_DUMMY_REG_998 = 998;
+
 // For sample XML files see the GDB source /gdb/features
 // GDB also wants the l character at the start
 // This XML defines what the registers are for this specific ARM device
@@ -256,36 +260,6 @@ static void RegWrite(std::size_t id, u64 val, Kernel::Thread* thread = nullptr)
     }
 }
 
-static u128 FpuRead(std::size_t id, Kernel::Thread* thread = nullptr) {
-    if (!thread) {
-        return u128{0};
-    }
-
-    auto& thread_context = thread->GetContext();
-
-    if (id >= UC_ARM64_REG_Q0 && id < FPCR_REGISTER) {
-        return thread_context.vector_registers[id - UC_ARM64_REG_Q0];
-    } else if (id == FPCR_REGISTER) {
-        return u128{thread_context.fpcr, 0};
-    } else {
-        return u128{0};
-    }
-}
-
-static void FpuWrite(std::size_t id, u128 val, Kernel::Thread* thread = nullptr) {
-    if (!thread) {
-        return;
-    }
-
-    auto& thread_context = thread->GetContext();
-
-    if (id >= UC_ARM64_REG_Q0 && id < FPCR_REGISTER) {
-        thread_context.vector_registers[id - UC_ARM64_REG_Q0] = val;
-    } else if (id == FPCR_REGISTER) {
-        thread_context.fpcr = val[0];
-    }
-}
-
 /**
  * Turns hex string character into the equivalent byte.
  *
@@ -435,27 +409,6 @@ static u64 GdbHexToLong(const u8* src) {
     return output;
 }
 
-/**
- * Convert a gdb-formatted hex string into a u128.
- *
- * @param src Pointer to hex string.
- */
-static u128 GdbHexToU128(const u8* src) {
-    u128 output;
-
-    for (int i = 0; i < 16; i += 2) {
-        output[0] = (output[0] << 4) | HexCharToValue(src[15 - i - 1]);
-        output[0] = (output[0] << 4) | HexCharToValue(src[15 - i]);
-    }
-
-    for (int i = 0; i < 16; i += 2) {
-        output[1] = (output[1] << 4) | HexCharToValue(src[16 + 15 - i - 1]);
-        output[1] = (output[1] << 4) | HexCharToValue(src[16 + 15 - i]);
-    }
-
-    return output;
-}
-
 /// Read a byte from the gdb client.
 static u8 ReadByte() {
     u8 c;
@@ -646,7 +599,8 @@ static void HandleQuery() {
         for (u32 core = 0; core < Core::NUM_CPU_CORES; core++) {
             const auto& threads = Core::System::GetInstance().Scheduler(core).GetThreadList();
             for (const auto& thread : threads) {
-                val += fmt::format("{:x},", thread->GetThreadID());
+                val += fmt::format("{:x}", thread->GetThreadID());
+                val += ",";
             }
         }
         val.pop_back();
@@ -837,15 +791,11 @@ static void ReadRegister() {
     } else if (id == PSTATE_REGISTER) {
         IntToGdbHex(reply, static_cast<u32>(RegRead(id, current_thread)));
     } else if (id >= UC_ARM64_REG_Q0 && id < FPCR_REGISTER) {
-        u128 r = FpuRead(id, current_thread);
-        LongToGdbHex(reply, r[0]);
-        LongToGdbHex(reply + 16, r[1]);
+        LongToGdbHex(reply, RegRead(id, current_thread));
     } else if (id == FPCR_REGISTER) {
-        u128 r = FpuRead(id, current_thread);
-        IntToGdbHex(reply, static_cast<u32>(r[0]));
-    } else if (id == FPCR_REGISTER + 1) {
-        u128 r = FpuRead(id, current_thread);
-        IntToGdbHex(reply, static_cast<u32>(r[0] >> 32));
+        LongToGdbHex(reply, RegRead(TODO_DUMMY_REG_998, current_thread));
+    } else {
+        LongToGdbHex(reply, RegRead(TODO_DUMMY_REG_997, current_thread));
     }
 
     SendReply(reinterpret_cast<char*>(reply));
@@ -872,18 +822,13 @@ static void ReadRegisters() {
 
     bufptr += 8;
 
-    u128 r;
-
-    for (u32 reg = UC_ARM64_REG_Q0; reg < FPCR_REGISTER; reg++) {
-        r = FpuRead(reg, current_thread);
-        LongToGdbHex(bufptr + reg * 32, r[0]);
-        LongToGdbHex(bufptr + reg * 32 + 16, r[1]);
+    for (u32 reg = UC_ARM64_REG_Q0; reg <= UC_ARM64_REG_Q0 + 31; reg++) {
+        LongToGdbHex(bufptr + reg * 16, RegRead(reg, current_thread));
     }
 
     bufptr += 32 * 32;
 
-    r = FpuRead(FPCR_REGISTER, current_thread);
-    IntToGdbHex(bufptr, static_cast<u32>(r[0]));
+    LongToGdbHex(bufptr, RegRead(TODO_DUMMY_REG_998, current_thread));
 
     bufptr += 8;
 
@@ -908,12 +853,14 @@ static void WriteRegister() {
     } else if (id == PSTATE_REGISTER) {
         RegWrite(id, GdbHexToInt(buffer_ptr), current_thread);
     } else if (id >= UC_ARM64_REG_Q0 && id < FPCR_REGISTER) {
-        FpuWrite(id, GdbHexToU128(buffer_ptr), current_thread);
+        RegWrite(id, GdbHexToLong(buffer_ptr), current_thread);
     } else if (id == FPCR_REGISTER) {
-    } else if (id == FPCR_REGISTER + 1) {
+        RegWrite(TODO_DUMMY_REG_998, GdbHexToLong(buffer_ptr), current_thread);
+    } else {
+        RegWrite(TODO_DUMMY_REG_997, GdbHexToLong(buffer_ptr), current_thread);
     }
 
-    // Update ARM context, skipping scheduler - no running threads at this point
+    // Update Unicorn context skipping scheduler, no running threads at this point
     Core::System::GetInstance()
         .ArmInterface(current_core)
         .LoadContext(current_thread->GetContext());
@@ -938,13 +885,13 @@ static void WriteRegisters() {
         } else if (reg >= UC_ARM64_REG_Q0 && reg < FPCR_REGISTER) {
             RegWrite(reg, GdbHexToLong(buffer_ptr + i * 16), current_thread);
         } else if (reg == FPCR_REGISTER) {
-            RegWrite(FPCR_REGISTER, GdbHexToLong(buffer_ptr + i * 16), current_thread);
-        } else if (reg == FPCR_REGISTER + 1) {
-            RegWrite(FPCR_REGISTER, GdbHexToLong(buffer_ptr + i * 16), current_thread);
+            RegWrite(TODO_DUMMY_REG_998, GdbHexToLong(buffer_ptr + i * 16), current_thread);
+        } else {
+            UNIMPLEMENTED();
         }
     }
 
-    // Update ARM context, skipping scheduler - no running threads at this point
+    // Update Unicorn context skipping scheduler, no running threads at this point
     Core::System::GetInstance()
         .ArmInterface(current_core)
         .LoadContext(current_thread->GetContext());
@@ -970,6 +917,12 @@ static void ReadMemory() {
         SendReply("E01");
     }
 
+    const auto& vm_manager = Core::CurrentProcess()->VMManager();
+    if (addr < vm_manager.GetCodeRegionBaseAddress() ||
+        addr >= vm_manager.GetMapRegionEndAddress()) {
+        return SendReply("E00");
+    }
+
     if (!Memory::IsValidVirtualAddress(addr)) {
         return SendReply("E00");
     }
@@ -1014,7 +967,7 @@ void Break(bool is_memory_break) {
 static void Step() {
     if (command_length > 1) {
         RegWrite(PC_REGISTER, GdbHexToLong(command_buffer + 1), current_thread);
-        // Update ARM context, skipping scheduler - no running threads at this point
+        // Update Unicorn context skipping scheduler, no running threads at this point
         Core::System::GetInstance()
             .ArmInterface(current_core)
             .LoadContext(current_thread->GetContext());
@@ -1057,7 +1010,7 @@ static bool CommitBreakpoint(BreakpointType type, VAddr addr, u64 len) {
     breakpoint.addr = addr;
     breakpoint.len = len;
     Memory::ReadBlock(addr, breakpoint.inst.data(), breakpoint.inst.size());
-    static constexpr std::array<u8, 4> btrap{0x00, 0x7d, 0x20, 0xd4};
+    static constexpr std::array<u8, 4> btrap{{0x00, 0x7d, 0x20, 0xd4}};
     Memory::WriteBlock(addr, btrap.data(), btrap.size());
     Core::System::GetInstance().InvalidateCpuInstructionCaches();
     p.insert({addr, breakpoint});
@@ -1368,15 +1321,13 @@ void SetCpuStepFlag(bool is_step) {
 }
 
 void SendTrap(Kernel::Thread* thread, int trap) {
-    if (!send_trap) {
-        return;
+    if (send_trap) {
+        if (!halt_loop || current_thread == thread) {
+            current_thread = thread;
+            SendSignal(thread, trap);
+        }
+        halt_loop = true;
+        send_trap = false;
     }
-
-    if (!halt_loop || current_thread == thread) {
-        current_thread = thread;
-        SendSignal(thread, trap);
-    }
-    halt_loop = true;
-    send_trap = false;
 }
 }; // namespace GDBStub

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

@@ -6,6 +6,8 @@
 #include <cinttypes>
 #include <cstring>
 #include <stack>
+#include "applets/applets.h"
+#include "applets/software_keyboard.h"
 #include "audio_core/audio_renderer.h"
 #include "core/core.h"
 #include "core/hle/ipc_helpers.h"
@@ -16,9 +18,6 @@
 #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/applets.h"
-#include "core/hle/service/am/applets/software_keyboard.h"
-#include "core/hle/service/am/applets/stub_applet.h"
 #include "core/hle/service/am/idle.h"
 #include "core/hle/service/am/omm.h"
 #include "core/hle/service/am/spsm.h"
@@ -483,15 +482,11 @@ void ICommonStateGetter::GetDefaultDisplayResolution(Kernel::HLERequestContext&
     rb.Push(RESULT_SUCCESS);
 
     if (Settings::values.use_docked_mode) {
-        rb.Push(static_cast<u32>(Service::VI::DisplayResolution::DockedWidth) *
-                static_cast<u32>(Settings::values.resolution_factor));
-        rb.Push(static_cast<u32>(Service::VI::DisplayResolution::DockedHeight) *
-                static_cast<u32>(Settings::values.resolution_factor));
+        rb.Push(static_cast<u32>(Service::VI::DisplayResolution::DockedWidth));
+        rb.Push(static_cast<u32>(Service::VI::DisplayResolution::DockedHeight));
     } else {
-        rb.Push(static_cast<u32>(Service::VI::DisplayResolution::UndockedWidth) *
-                static_cast<u32>(Settings::values.resolution_factor));
-        rb.Push(static_cast<u32>(Service::VI::DisplayResolution::UndockedHeight) *
-                static_cast<u32>(Settings::values.resolution_factor));
+        rb.Push(static_cast<u32>(Service::VI::DisplayResolution::UndockedWidth));
+        rb.Push(static_cast<u32>(Service::VI::DisplayResolution::UndockedHeight));
     }
 
     LOG_DEBUG(Service_AM, "called");
@@ -766,9 +761,8 @@ static std::shared_ptr<Applets::Applet> GetAppletFromId(AppletId id) {
     case AppletId::SoftwareKeyboard:
         return std::make_shared<Applets::SoftwareKeyboard>();
     default:
-        LOG_ERROR(Service_AM, "Unimplemented AppletId [{:08X}]! -- Falling back to stub!",
-                  static_cast<u32>(id));
-        return std::make_shared<Applets::StubApplet>();
+        UNREACHABLE_MSG("Unimplemented AppletId [{:08X}]!", static_cast<u32>(id));
+        return nullptr;
     }
 }
 

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

@@ -1,70 +0,0 @@
-// Copyright 2018 yuzu emulator team
-// Licensed under GPLv2 or any later version
-// Refer to the license.txt file included.
-
-#include <string>
-
-#include "common/hex_util.h"
-#include "common/logging/log.h"
-#include "core/hle/result.h"
-#include "core/hle/service/am/am.h"
-#include "core/hle/service/am/applets/stub_applet.h"
-
-namespace Service::AM::Applets {
-
-static void LogCurrentStorage(AppletDataBroker& broker, std::string prefix) {
-    std::unique_ptr<IStorage> storage = broker.PopNormalDataToApplet();
-    for (; storage != nullptr; storage = broker.PopNormalDataToApplet()) {
-        const auto data = storage->GetData();
-        LOG_INFO(Service_AM,
-                 "called (STUBBED), during {} recieved normal data with size={:08X}, data={}",
-                 prefix, data.size(), Common::HexVectorToString(data));
-    }
-
-    storage = broker.PopInteractiveDataToApplet();
-    for (; storage != nullptr; storage = broker.PopInteractiveDataToApplet()) {
-        const auto data = storage->GetData();
-        LOG_INFO(Service_AM,
-                 "called (STUBBED), during {} recieved interactive data with size={:08X}, data={}",
-                 prefix, data.size(), Common::HexVectorToString(data));
-    }
-}
-
-StubApplet::StubApplet() = default;
-
-StubApplet::~StubApplet() = default;
-
-void StubApplet::Initialize() {
-    LOG_WARNING(Service_AM, "called (STUBBED)");
-    Applet::Initialize();
-    LogCurrentStorage(broker, "Initialize");
-}
-
-bool StubApplet::TransactionComplete() const {
-    LOG_WARNING(Service_AM, "called (STUBBED)");
-    return true;
-}
-
-ResultCode StubApplet::GetStatus() const {
-    LOG_WARNING(Service_AM, "called (STUBBED)");
-    return RESULT_SUCCESS;
-}
-
-void StubApplet::ExecuteInteractive() {
-    LOG_WARNING(Service_AM, "called (STUBBED)");
-    LogCurrentStorage(broker, "ExecuteInteractive");
-
-    broker.PushNormalDataFromApplet(IStorage{std::vector<u8>(0x1000)});
-    broker.PushInteractiveDataFromApplet(IStorage{std::vector<u8>(0x1000)});
-    broker.SignalStateChanged();
-}
-
-void StubApplet::Execute() {
-    LOG_WARNING(Service_AM, "called (STUBBED)");
-    LogCurrentStorage(broker, "Execute");
-
-    broker.PushNormalDataFromApplet(IStorage{std::vector<u8>(0x1000)});
-    broker.PushInteractiveDataFromApplet(IStorage{std::vector<u8>(0x1000)});
-    broker.SignalStateChanged();
-}
-} // namespace Service::AM::Applets

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

@@ -1,24 +0,0 @@
-// Copyright 2018 yuzu emulator team
-// Licensed under GPLv2 or any later version
-// Refer to the license.txt file included.
-
-#pragma once
-
-#include "core/hle/service/am/applets/applets.h"
-
-namespace Service::AM::Applets {
-
-class StubApplet final : public Applet {
-public:
-    StubApplet();
-    ~StubApplet() override;
-
-    void Initialize() override;
-
-    bool TransactionComplete() const override;
-    ResultCode GetStatus() const override;
-    void ExecuteInteractive() override;
-    void Execute() override;
-};
-
-} // namespace Service::AM::Applets

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

@@ -341,10 +341,6 @@ std::shared_ptr<FileSys::RegisteredCacheUnion> GetUnionContents() {
     return registered_cache_union;
 }
 
-void ClearUnionContents() {
-    registered_cache_union = nullptr;
-}
-
 FileSys::RegisteredCache* GetSystemNANDContents() {
     LOG_TRACE(Service_FS, "Opening System NAND Contents");
 
@@ -395,7 +391,6 @@ void CreateFactories(FileSys::VfsFilesystem& vfs, bool overwrite) {
         bis_factory = nullptr;
         save_data_factory = nullptr;
         sdmc_factory = nullptr;
-        ClearUnionContents();
     }
 
     auto nand_directory = vfs.OpenDirectory(FileUtil::GetUserPath(FileUtil::UserPath::NANDDir),

src/core/hle/service/filesystem/filesystem.h

@@ -49,7 +49,6 @@ ResultVal<FileSys::VirtualDir> OpenSaveDataSpace(FileSys::SaveDataSpaceId space)
 ResultVal<FileSys::VirtualDir> OpenSDMC();
 
 std::shared_ptr<FileSys::RegisteredCacheUnion> GetUnionContents();
-void ClearUnionContents();
 
 FileSys::RegisteredCache* GetSystemNANDContents();
 FileSys::RegisteredCache* GetUserNANDContents();

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

@@ -5,8 +5,6 @@
 #include <cstring>
 #include "common/assert.h"
 #include "common/logging/log.h"
-#include "core/core_timing.h"
-#include "core/core_timing_util.h"
 #include "core/hle/service/nvdrv/devices/nvhost_ctrl_gpu.h"
 
 namespace Service::Nvidia::Devices {
@@ -35,8 +33,6 @@ u32 nvhost_ctrl_gpu::ioctl(Ioctl command, const std::vector<u8>& input, std::vec
         return ZBCQueryTable(input, output);
     case IoctlCommand::IocFlushL2:
         return FlushL2(input, output);
-    case IoctlCommand::IocGetGpuTime:
-        return GetGpuTime(input, output);
     }
     UNIMPLEMENTED_MSG("Unimplemented ioctl");
     return 0;
@@ -173,13 +169,4 @@ u32 nvhost_ctrl_gpu::FlushL2(const std::vector<u8>& input, std::vector<u8>& outp
     return 0;
 }
 
-u32 nvhost_ctrl_gpu::GetGpuTime(const std::vector<u8>& input, std::vector<u8>& output) {
-    LOG_DEBUG(Service_NVDRV, "called");
-    IoctlGetGpuTime params{};
-    std::memcpy(&params, input.data(), input.size());
-    params.gpu_time = CoreTiming::cyclesToNs(CoreTiming::GetTicks());
-    std::memcpy(output.data(), &params, output.size());
-    return 0;
-}
-
 } // namespace Service::Nvidia::Devices

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

@@ -156,11 +156,6 @@ private:
     };
     static_assert(sizeof(IoctlFlushL2) == 8, "IoctlFlushL2 is incorrect size");
 
-    struct IoctlGetGpuTime {
-        u64_le gpu_time;
-    };
-    static_assert(sizeof(IoctlGetGpuTime) == 8, "IoctlGetGpuTime is incorrect size");
-
     u32 GetCharacteristics(const std::vector<u8>& input, std::vector<u8>& output);
     u32 GetTPCMasks(const std::vector<u8>& input, std::vector<u8>& output);
     u32 GetActiveSlotMask(const std::vector<u8>& input, std::vector<u8>& output);
@@ -169,7 +164,6 @@ private:
     u32 ZBCSetTable(const std::vector<u8>& input, std::vector<u8>& output);
     u32 ZBCQueryTable(const std::vector<u8>& input, std::vector<u8>& output);
     u32 FlushL2(const std::vector<u8>& input, std::vector<u8>& output);
-    u32 GetGpuTime(const std::vector<u8>& input, std::vector<u8>& output);
 };
 
 } // namespace Service::Nvidia::Devices

src/core/hle/service/sm/sm.cpp

@@ -63,17 +63,6 @@ ResultVal<Kernel::SharedPtr<Kernel::ServerPort>> ServiceManager::RegisterService
     return MakeResult<Kernel::SharedPtr<Kernel::ServerPort>>(std::move(server_port));
 }
 
-ResultCode ServiceManager::UnregisterService(const std::string& name) {
-    CASCADE_CODE(ValidateServiceName(name));
-
-    const auto iter = registered_services.find(name);
-    if (iter == registered_services.end())
-        return ERR_SERVICE_NOT_REGISTERED;
-
-    registered_services.erase(iter);
-    return RESULT_SUCCESS;
-}
-
 ResultVal<Kernel::SharedPtr<Kernel::ClientPort>> ServiceManager::GetServicePort(
     const std::string& name) {
 
@@ -138,52 +127,13 @@ void SM::GetService(Kernel::HLERequestContext& ctx) {
     }
 }
 
-void SM::RegisterService(Kernel::HLERequestContext& ctx) {
-    IPC::RequestParser rp{ctx};
-
-    const auto name_buf = rp.PopRaw<std::array<char, 8>>();
-    const auto end = std::find(name_buf.begin(), name_buf.end(), '\0');
-
-    const std::string name(name_buf.begin(), end);
-
-    const auto unk_bool = static_cast<bool>(rp.PopRaw<u32>());
-    const auto session_count = rp.PopRaw<u32>();
-
-    LOG_DEBUG(Service_SM, "called with unk_bool={}", unk_bool);
-
-    auto handle = service_manager->RegisterService(name, session_count);
-    if (handle.Failed()) {
-        LOG_ERROR(Service_SM, "failed to register service with error_code={:08X}",
-                  handle.Code().raw);
-        IPC::ResponseBuilder rb{ctx, 2};
-        rb.Push(handle.Code());
-        return;
-    }
-
-    IPC::ResponseBuilder rb{ctx, 2, 0, 1, IPC::ResponseBuilder::Flags::AlwaysMoveHandles};
-    rb.Push(handle.Code());
-    rb.PushMoveObjects(std::move(handle).Unwrap());
-}
-
-void SM::UnregisterService(Kernel::HLERequestContext& ctx) {
-    IPC::RequestParser rp{ctx};
-
-    const auto name_buf = rp.PopRaw<std::array<char, 8>>();
-    const auto end = std::find(name_buf.begin(), name_buf.end(), '\0');
-
-    const std::string name(name_buf.begin(), end);
-
-    IPC::ResponseBuilder rb{ctx, 2};
-    rb.Push(service_manager->UnregisterService(name));
-}
-
 SM::SM(std::shared_ptr<ServiceManager> service_manager)
     : ServiceFramework("sm:", 4), service_manager(std::move(service_manager)) {
     static const FunctionInfo functions[] = {
         {0x00000000, &SM::Initialize, "Initialize"},
         {0x00000001, &SM::GetService, "GetService"},
-        {0x00000002, &SM::RegisterService, "RegisterService"},
-        {0x00000003, &SM::UnregisterService, "UnregisterService"},
+        {0x00000002, nullptr, "RegisterService"},
+        {0x00000003, nullptr, "UnregisterService"},
     };
     RegisterHandlers(functions);
 }

src/core/hle/service/sm/sm.h

@@ -35,8 +35,6 @@ public:
 private:
     void Initialize(Kernel::HLERequestContext& ctx);
     void GetService(Kernel::HLERequestContext& ctx);
-    void RegisterService(Kernel::HLERequestContext& ctx);
-    void UnregisterService(Kernel::HLERequestContext& ctx);
 
     std::shared_ptr<ServiceManager> service_manager;
 };
@@ -50,7 +48,6 @@ public:
 
     ResultVal<Kernel::SharedPtr<Kernel::ServerPort>> RegisterService(std::string name,
                                                                      unsigned int max_sessions);
-    ResultCode UnregisterService(const std::string& name);
     ResultVal<Kernel::SharedPtr<Kernel::ClientPort>> GetServicePort(const std::string& name);
     ResultVal<Kernel::SharedPtr<Kernel::ClientSession>> ConnectToService(const std::string& name);
 

src/core/hle/service/vi/vi.cpp

@@ -510,11 +510,7 @@ private:
 
         if (transaction == TransactionId::Connect) {
             IGBPConnectRequestParcel request{ctx.ReadBuffer()};
-            IGBPConnectResponseParcel response{
-                static_cast<u32>(static_cast<u32>(DisplayResolution::UndockedWidth) *
-                                 Settings::values.resolution_factor),
-                static_cast<u32>(static_cast<u32>(DisplayResolution::UndockedHeight) *
-                                 Settings::values.resolution_factor)};
+            IGBPConnectResponseParcel response{1280, 720};
             ctx.WriteBuffer(response.Serialize());
         } else if (transaction == TransactionId::SetPreallocatedBuffer) {
             IGBPSetPreallocatedBufferRequestParcel request{ctx.ReadBuffer()};
@@ -696,15 +692,11 @@ private:
         rb.Push(RESULT_SUCCESS);
 
         if (Settings::values.use_docked_mode) {
-            rb.Push(static_cast<u32>(Service::VI::DisplayResolution::DockedWidth) *
-                    static_cast<u32>(Settings::values.resolution_factor));
-            rb.Push(static_cast<u32>(Service::VI::DisplayResolution::DockedHeight) *
-                    static_cast<u32>(Settings::values.resolution_factor));
+            rb.Push(static_cast<u32>(Service::VI::DisplayResolution::DockedWidth));
+            rb.Push(static_cast<u32>(Service::VI::DisplayResolution::DockedHeight));
         } else {
-            rb.Push(static_cast<u32>(Service::VI::DisplayResolution::UndockedWidth) *
-                    static_cast<u32>(Settings::values.resolution_factor));
-            rb.Push(static_cast<u32>(Service::VI::DisplayResolution::UndockedHeight) *
-                    static_cast<u32>(Settings::values.resolution_factor));
+            rb.Push(static_cast<u32>(Service::VI::DisplayResolution::UndockedWidth));
+            rb.Push(static_cast<u32>(Service::VI::DisplayResolution::UndockedHeight));
         }
 
         rb.PushRaw<float>(60.0f);
@@ -909,15 +901,11 @@ private:
         rb.Push(RESULT_SUCCESS);
 
         if (Settings::values.use_docked_mode) {
-            rb.Push(static_cast<u64>(DisplayResolution::DockedWidth) *
-                    static_cast<u32>(Settings::values.resolution_factor));
-            rb.Push(static_cast<u64>(DisplayResolution::DockedHeight) *
-                    static_cast<u32>(Settings::values.resolution_factor));
+            rb.Push(static_cast<u64>(DisplayResolution::DockedWidth));
+            rb.Push(static_cast<u64>(DisplayResolution::DockedHeight));
         } else {
-            rb.Push(static_cast<u64>(DisplayResolution::UndockedWidth) *
-                    static_cast<u32>(Settings::values.resolution_factor));
-            rb.Push(static_cast<u64>(DisplayResolution::UndockedHeight) *
-                    static_cast<u32>(Settings::values.resolution_factor));
+            rb.Push(static_cast<u64>(DisplayResolution::UndockedWidth));
+            rb.Push(static_cast<u64>(DisplayResolution::UndockedHeight));
         }
     }
 
@@ -934,8 +922,6 @@ private:
     void ListDisplays(Kernel::HLERequestContext& ctx) {
         IPC::RequestParser rp{ctx};
         DisplayInfo display_info;
-        display_info.width *= static_cast<u64>(Settings::values.resolution_factor);
-        display_info.height *= static_cast<u64>(Settings::values.resolution_factor);
         ctx.WriteBuffer(&display_info, sizeof(DisplayInfo));
         IPC::ResponseBuilder rb{ctx, 4};
         rb.Push(RESULT_SUCCESS);

src/core/settings.h

@@ -403,7 +403,6 @@ struct Values {
     bool use_gdbstub;
     u16 gdbstub_port;
     std::string program_args;
-    bool dump_exefs;
     bool dump_nso;
 
     // WebService

src/video_core/engines/maxwell_3d.h

@@ -590,10 +590,18 @@ public:
 
                 float clear_color[4];
                 float clear_depth;
+
                 INSERT_PADDING_WORDS(0x3);
+
                 s32 clear_stencil;
 
-                INSERT_PADDING_WORDS(0x17);
+                INSERT_PADDING_WORDS(0x7);
+
+                u32 polygon_offset_point_enable;
+                u32 polygon_offset_line_enable;
+                u32 polygon_offset_fill_enable;
+
+                INSERT_PADDING_WORDS(0xD);
 
                 std::array<ScissorTest, NumViewports> scissor_test;
 
@@ -631,16 +639,7 @@ public:
                     }
                 } zeta;
 
-                INSERT_PADDING_WORDS(0x41);
-
-                union {
-                    BitField<0, 4, u32> stencil;
-                    BitField<4, 4, u32> unknown;
-                    BitField<8, 4, u32> scissor;
-                    BitField<12, 4, u32> viewport;
-                } clear_flags;
-
-                INSERT_PADDING_WORDS(0x19);
+                INSERT_PADDING_WORDS(0x5B);
 
                 std::array<VertexAttribute, NumVertexAttributes> vertex_attrib_format;
 
@@ -761,7 +760,11 @@ public:
                     }
                 } tsc;
 
-                INSERT_PADDING_WORDS(0x3);
+                INSERT_PADDING_WORDS(0x1);
+
+                float polygon_offset_factor;
+
+                INSERT_PADDING_WORDS(0x1);
 
                 struct {
                     u32 tic_address_high;
@@ -786,7 +789,9 @@ public:
 
                 u32 framebuffer_srgb;
 
-                INSERT_PADDING_WORDS(0x12);
+                float polygon_offset_units;
+
+                INSERT_PADDING_WORDS(0x11);
 
                 union {
                     BitField<2, 1, u32> coord_origin;
@@ -863,7 +868,9 @@ public:
 
                 INSERT_PADDING_WORDS(0x7);
 
-                INSERT_PADDING_WORDS(0x20);
+                INSERT_PADDING_WORDS(0x1F);
+
+                float polygon_offset_clamp;
 
                 struct {
                     u32 is_instanced[NumVertexArrays];
@@ -1136,6 +1143,9 @@ ASSERT_REG_POSITION(vertex_buffer, 0x35D);
 ASSERT_REG_POSITION(clear_color[0], 0x360);
 ASSERT_REG_POSITION(clear_depth, 0x364);
 ASSERT_REG_POSITION(clear_stencil, 0x368);
+ASSERT_REG_POSITION(polygon_offset_point_enable, 0x370);
+ASSERT_REG_POSITION(polygon_offset_line_enable, 0x371);
+ASSERT_REG_POSITION(polygon_offset_fill_enable, 0x372);
 ASSERT_REG_POSITION(scissor_test, 0x380);
 ASSERT_REG_POSITION(stencil_back_func_ref, 0x3D5);
 ASSERT_REG_POSITION(stencil_back_mask, 0x3D6);
@@ -1143,7 +1153,6 @@ ASSERT_REG_POSITION(stencil_back_func_mask, 0x3D7);
 ASSERT_REG_POSITION(color_mask_common, 0x3E4);
 ASSERT_REG_POSITION(rt_separate_frag_data, 0x3EB);
 ASSERT_REG_POSITION(zeta, 0x3F8);
-ASSERT_REG_POSITION(clear_flags, 0x43E);
 ASSERT_REG_POSITION(vertex_attrib_format, 0x458);
 ASSERT_REG_POSITION(rt_control, 0x487);
 ASSERT_REG_POSITION(zeta_width, 0x48a);
@@ -1174,6 +1183,7 @@ ASSERT_REG_POSITION(point_size, 0x546);
 ASSERT_REG_POSITION(zeta_enable, 0x54E);
 ASSERT_REG_POSITION(multisample_control, 0x54F);
 ASSERT_REG_POSITION(tsc, 0x557);
+ASSERT_REG_POSITION(polygon_offset_factor, 0x55b);
 ASSERT_REG_POSITION(tic, 0x55D);
 ASSERT_REG_POSITION(stencil_two_side_enable, 0x565);
 ASSERT_REG_POSITION(stencil_back_op_fail, 0x566);
@@ -1181,11 +1191,13 @@ ASSERT_REG_POSITION(stencil_back_op_zfail, 0x567);
 ASSERT_REG_POSITION(stencil_back_op_zpass, 0x568);
 ASSERT_REG_POSITION(stencil_back_func_func, 0x569);
 ASSERT_REG_POSITION(framebuffer_srgb, 0x56E);
+ASSERT_REG_POSITION(polygon_offset_units, 0x56F);
 ASSERT_REG_POSITION(point_coord_replace, 0x581);
 ASSERT_REG_POSITION(code_address, 0x582);
 ASSERT_REG_POSITION(draw, 0x585);
 ASSERT_REG_POSITION(primitive_restart, 0x591);
 ASSERT_REG_POSITION(index_array, 0x5F2);
+ASSERT_REG_POSITION(polygon_offset_clamp, 0x61F);
 ASSERT_REG_POSITION(instanced_arrays, 0x620);
 ASSERT_REG_POSITION(cull, 0x646);
 ASSERT_REG_POSITION(logic_op, 0x671);

src/video_core/engines/shader_bytecode.h

@@ -153,7 +153,6 @@ enum class PredCondition : u64 {
     NotEqual = 5,
     GreaterEqual = 6,
     LessThanWithNan = 9,
-    LessEqualWithNan = 11,
     GreaterThanWithNan = 12,
     NotEqualWithNan = 13,
     GreaterEqualWithNan = 14,
@@ -262,7 +261,7 @@ enum class FlowCondition : u64 {
     Fcsm_Tr = 0x1C, // TODO(bunnei): What is this used for?
 };
 
-enum class ConditionCode : u64 {
+enum class ControlCode : u64 {
     F = 0,
     LT = 1,
     EQ = 2,
@@ -570,6 +569,7 @@ union Instruction {
         BitField<39, 2, u64> tab5cb8_2;
         BitField<41, 3, u64> tab5c68_1;
         BitField<44, 2, u64> tab5c68_0;
+        BitField<47, 1, u64> cc;
         BitField<48, 1, u64> negate_b;
     } fmul;
 
@@ -831,7 +831,7 @@ union Instruction {
     union {
         BitField<0, 3, u64> pred0;
         BitField<3, 3, u64> pred3;
-        BitField<8, 5, ConditionCode> cc; // flag in cc
+        BitField<8, 5, ControlCode> cc; // flag in cc
         BitField<39, 3, u64> pred39;
         BitField<42, 1, u64> neg_pred39;
         BitField<45, 4, PredOperation> op; // op with pred39
@@ -1235,7 +1235,7 @@ union Instruction {
     BitField<60, 1, u64> is_b_gpr;
     BitField<59, 1, u64> is_c_gpr;
     BitField<20, 24, s64> smem_imm;
-    BitField<0, 5, ConditionCode> flow_condition_code;
+    BitField<0, 5, ControlCode> flow_control_code;
 
     Attribute attribute;
     Sampler sampler;

src/video_core/renderer_opengl/gl_rasterizer.cpp

@@ -98,9 +98,14 @@ RasterizerOpenGL::RasterizerOpenGL(Core::Frontend::EmuWindow& window, ScreenInfo
             has_ARB_direct_state_access = true;
         } else if (extension == "GL_ARB_multi_bind") {
             has_ARB_multi_bind = true;
+        } else if (extension == "GL_ARB_separate_shader_objects") {
+            has_ARB_separate_shader_objects = true;
+        } else if (extension == "GL_ARB_vertex_attrib_binding") {
+            has_ARB_vertex_attrib_binding = true;
         }
     }
 
+    ASSERT_MSG(has_ARB_separate_shader_objects, "has_ARB_separate_shader_objects is unsupported");
     OpenGLState::ApplyDefaultState();
 
     // Create render framebuffer
@@ -537,30 +542,6 @@ void RasterizerOpenGL::Clear() {
         ASSERT_MSG(regs.zeta_enable != 0, "Tried to clear stencil but buffer is not enabled!");
         use_stencil = true;
         clear_state.stencil.test_enabled = true;
-        if (regs.clear_flags.stencil) {
-            // Stencil affects the clear so fill it with the used masks
-            clear_state.stencil.front.test_func = GL_ALWAYS;
-            clear_state.stencil.front.test_mask = regs.stencil_front_func_mask;
-            clear_state.stencil.front.action_stencil_fail = GL_KEEP;
-            clear_state.stencil.front.action_depth_fail = GL_KEEP;
-            clear_state.stencil.front.action_depth_pass = GL_KEEP;
-            clear_state.stencil.front.write_mask = regs.stencil_front_mask;
-            if (regs.stencil_two_side_enable) {
-                clear_state.stencil.back.test_func = GL_ALWAYS;
-                clear_state.stencil.back.test_mask = regs.stencil_back_func_mask;
-                clear_state.stencil.back.action_stencil_fail = GL_KEEP;
-                clear_state.stencil.back.action_depth_fail = GL_KEEP;
-                clear_state.stencil.back.action_depth_pass = GL_KEEP;
-                clear_state.stencil.back.write_mask = regs.stencil_back_mask;
-            } else {
-                clear_state.stencil.back.test_func = GL_ALWAYS;
-                clear_state.stencil.back.test_mask = 0xFFFFFFFF;
-                clear_state.stencil.back.write_mask = 0xFFFFFFFF;
-                clear_state.stencil.back.action_stencil_fail = GL_KEEP;
-                clear_state.stencil.back.action_depth_fail = GL_KEEP;
-                clear_state.stencil.back.action_depth_pass = GL_KEEP;
-            }
-        }
     }
 
     if (!use_color && !use_depth && !use_stencil) {
@@ -572,14 +553,6 @@ void RasterizerOpenGL::Clear() {
 
     ConfigureFramebuffers(clear_state, use_color, use_depth || use_stencil, false,
                           regs.clear_buffers.RT.Value());
-    if (regs.clear_flags.scissor) {
-        SyncScissorTest(clear_state);
-    }
-
-    if (regs.clear_flags.viewport) {
-        clear_state.EmulateViewportWithScissor();
-    }
-
     clear_state.Apply();
 
     if (use_color) {
@@ -615,12 +588,12 @@ void RasterizerOpenGL::DrawArrays() {
     SyncLogicOpState();
     SyncCullMode();
     SyncPrimitiveRestart();
-    SyncScissorTest(state);
+    SyncScissorTest();
     // Alpha Testing is synced on shaders.
     SyncTransformFeedback();
     SyncPointState();
     CheckAlphaTests();
-
+    SyncPolygonOffset();
     // TODO(bunnei): Sync framebuffer_scale uniform here
     // TODO(bunnei): Sync scissorbox uniform(s) here
 
@@ -842,7 +815,7 @@ void RasterizerOpenGL::SamplerInfo::SyncWithConfig(const Tegra::Texture::TSCEntr
     }
     const u32 bias = config.mip_lod_bias.Value();
     // Sign extend the 13-bit value.
-    constexpr u32 mask = 1U << (13 - 1);
+    const u32 mask = 1U << (13 - 1);
     const float bias_lod = static_cast<s32>((bias ^ mask) - mask) / 256.f;
     if (lod_bias != bias_lod) {
         lod_bias = bias_lod;
@@ -974,8 +947,8 @@ void RasterizerOpenGL::SyncViewport(OpenGLState& current_state) {
         auto& viewport = current_state.viewports[i];
         viewport.x = viewport_rect.left;
         viewport.y = viewport_rect.bottom;
-        viewport.width = viewport_rect.GetWidth();
-        viewport.height = viewport_rect.GetHeight();
+        viewport.width = static_cast<GLfloat>(viewport_rect.GetWidth());
+        viewport.height = static_cast<GLfloat>(viewport_rect.GetHeight());
         viewport.depth_range_far = regs.viewports[i].depth_range_far;
         viewport.depth_range_near = regs.viewports[i].depth_range_near;
     }
@@ -1147,11 +1120,11 @@ void RasterizerOpenGL::SyncLogicOpState() {
     state.logic_op.operation = MaxwellToGL::LogicOp(regs.logic_op.operation);
 }
 
-void RasterizerOpenGL::SyncScissorTest(OpenGLState& current_state) {
+void RasterizerOpenGL::SyncScissorTest() {
     const auto& regs = Core::System::GetInstance().GPU().Maxwell3D().regs;
     for (std::size_t i = 0; i < Tegra::Engines::Maxwell3D::Regs::NumViewports; i++) {
         const auto& src = regs.scissor_test[i];
-        auto& dst = current_state.viewports[i].scissor;
+        auto& dst = state.viewports[i].scissor;
         dst.enabled = (src.enable != 0);
         if (dst.enabled == 0) {
             return;
@@ -1179,6 +1152,16 @@ void RasterizerOpenGL::SyncPointState() {
     state.point.size = regs.point_size;
 }
 
+void RasterizerOpenGL::SyncPolygonOffset() {
+    const auto& regs = Core::System::GetInstance().GPU().Maxwell3D().regs;
+    state.polygon_offset.fill_enable = regs.polygon_offset_fill_enable != 0;
+    state.polygon_offset.line_enable = regs.polygon_offset_line_enable != 0;
+    state.polygon_offset.point_enable = regs.polygon_offset_point_enable != 0;
+    state.polygon_offset.units = regs.polygon_offset_units;
+    state.polygon_offset.factor = regs.polygon_offset_factor;
+    state.polygon_offset.clamp = regs.polygon_offset_clamp;
+}
+
 void RasterizerOpenGL::CheckAlphaTests() {
     const auto& regs = Core::System::GetInstance().GPU().Maxwell3D().regs;
 

src/video_core/renderer_opengl/gl_rasterizer.h

@@ -91,20 +91,19 @@ private:
         void SyncWithConfig(const Tegra::Texture::TSCEntry& info);
 
     private:
-        Tegra::Texture::TextureFilter mag_filter = Tegra::Texture::TextureFilter::Nearest;
-        Tegra::Texture::TextureFilter min_filter = Tegra::Texture::TextureFilter::Nearest;
-        Tegra::Texture::TextureMipmapFilter mip_filter = Tegra::Texture::TextureMipmapFilter::None;
-        Tegra::Texture::WrapMode wrap_u = Tegra::Texture::WrapMode::ClampToEdge;
-        Tegra::Texture::WrapMode wrap_v = Tegra::Texture::WrapMode::ClampToEdge;
-        Tegra::Texture::WrapMode wrap_p = Tegra::Texture::WrapMode::ClampToEdge;
-        bool uses_depth_compare = false;
-        Tegra::Texture::DepthCompareFunc depth_compare_func =
-            Tegra::Texture::DepthCompareFunc::Always;
-        GLvec4 border_color = {};
-        float min_lod = 0.0f;
-        float max_lod = 16.0f;
-        float lod_bias = 0.0f;
-        float max_anisotropic = 1.0f;
+        Tegra::Texture::TextureFilter mag_filter;
+        Tegra::Texture::TextureFilter min_filter;
+        Tegra::Texture::TextureMipmapFilter mip_filter;
+        Tegra::Texture::WrapMode wrap_u;
+        Tegra::Texture::WrapMode wrap_v;
+        Tegra::Texture::WrapMode wrap_p;
+        bool uses_depth_compare;
+        Tegra::Texture::DepthCompareFunc depth_compare_func;
+        GLvec4 border_color;
+        float min_lod;
+        float max_lod;
+        float lod_bias;
+        float max_anisotropic;
     };
 
     /**
@@ -172,7 +171,7 @@ private:
     void SyncMultiSampleState();
 
     /// Syncs the scissor test state to match the guest state
-    void SyncScissorTest(OpenGLState& current_state);
+    void SyncScissorTest();
 
     /// Syncs the transform feedback state to match the guest state
     void SyncTransformFeedback();
@@ -183,11 +182,16 @@ private:
     /// Syncs Color Mask
     void SyncColorMask();
 
+    /// Syncs the polygon offsets
+    void SyncPolygonOffset();
+
     /// Check asserts for alpha testing.
     void CheckAlphaTests();
 
     bool has_ARB_direct_state_access = false;
     bool has_ARB_multi_bind = false;
+    bool has_ARB_separate_shader_objects = false;
+    bool has_ARB_vertex_attrib_binding = false;
 
     OpenGLState state;
 

src/video_core/renderer_opengl/gl_rasterizer_cache.cpp

@@ -265,11 +265,11 @@ static constexpr std::array<FormatTuple, VideoCore::Surface::MaxPixelFormat> tex
     {GL_COMPRESSED_RG_RGTC2, GL_RG, GL_UNSIGNED_INT_8_8_8_8, ComponentType::UNorm,
      true},                                                                     // DXN2UNORM
     {GL_COMPRESSED_SIGNED_RG_RGTC2, GL_RG, GL_INT, ComponentType::SNorm, true}, // DXN2SNORM
-    {GL_COMPRESSED_RGBA_BPTC_UNORM, GL_RGBA, GL_UNSIGNED_INT_8_8_8_8, ComponentType::UNorm,
+    {GL_COMPRESSED_RGBA_BPTC_UNORM_ARB, GL_RGBA, GL_UNSIGNED_INT_8_8_8_8, ComponentType::UNorm,
      true}, // BC7U
-    {GL_COMPRESSED_RGB_BPTC_UNSIGNED_FLOAT, GL_RGB, GL_UNSIGNED_INT_8_8_8_8, ComponentType::Float,
-     true}, // BC6H_UF16
-    {GL_COMPRESSED_RGB_BPTC_SIGNED_FLOAT, GL_RGB, GL_UNSIGNED_INT_8_8_8_8, ComponentType::Float,
+    {GL_COMPRESSED_RGB_BPTC_UNSIGNED_FLOAT_ARB, GL_RGB, GL_UNSIGNED_INT_8_8_8_8,
+     ComponentType::Float, true}, // BC6H_UF16
+    {GL_COMPRESSED_RGB_BPTC_SIGNED_FLOAT_ARB, GL_RGB, GL_UNSIGNED_INT_8_8_8_8, ComponentType::Float,
      true},                                                                    // BC6H_SF16
     {GL_RGBA8, GL_RGBA, GL_UNSIGNED_BYTE, ComponentType::UNorm, false},        // ASTC_2D_4X4
     {GL_RG8, GL_RG, GL_UNSIGNED_BYTE, ComponentType::UNorm, false},            // G8R8U
@@ -306,8 +306,8 @@ static constexpr std::array<FormatTuple, VideoCore::Surface::MaxPixelFormat> tex
      true}, // DXT23_SRGB
     {GL_COMPRESSED_SRGB_ALPHA_S3TC_DXT5_EXT, GL_RGBA, GL_UNSIGNED_INT_8_8_8_8, ComponentType::UNorm,
      true}, // DXT45_SRGB
-    {GL_COMPRESSED_SRGB_ALPHA_BPTC_UNORM, GL_RGBA, GL_UNSIGNED_INT_8_8_8_8, ComponentType::UNorm,
-     true},                                                                    // BC7U_SRGB
+    {GL_COMPRESSED_SRGB_ALPHA_BPTC_UNORM_ARB, GL_RGBA, GL_UNSIGNED_INT_8_8_8_8,
+     ComponentType::UNorm, true},                                              // BC7U_SRGB
     {GL_SRGB8_ALPHA8, GL_RGBA, GL_UNSIGNED_BYTE, ComponentType::UNorm, false}, // ASTC_2D_4X4_SRGB
     {GL_SRGB8_ALPHA8, GL_RGBA, GL_UNSIGNED_BYTE, ComponentType::UNorm, false}, // ASTC_2D_8X8_SRGB
     {GL_SRGB8_ALPHA8, GL_RGBA, GL_UNSIGNED_BYTE, ComponentType::UNorm, false}, // ASTC_2D_8X5_SRGB
@@ -346,7 +346,7 @@ static GLenum SurfaceTargetToGL(SurfaceTarget target) {
     case SurfaceTarget::TextureCubemap:
         return GL_TEXTURE_CUBE_MAP;
     case SurfaceTarget::TextureCubeArray:
-        return GL_TEXTURE_CUBE_MAP_ARRAY;
+        return GL_TEXTURE_CUBE_MAP_ARRAY_ARB;
     }
     LOG_CRITICAL(Render_OpenGL, "Unimplemented texture target={}", static_cast<u32>(target));
     UNREACHABLE();
@@ -726,7 +726,7 @@ static void CopySurface(const Surface& src_surface, const Surface& dst_surface,
     const std::size_t buffer_size = std::max(src_params.size_in_bytes, dst_params.size_in_bytes);
 
     glBindBuffer(GL_PIXEL_PACK_BUFFER, copy_pbo_handle);
-    glBufferData(GL_PIXEL_PACK_BUFFER, buffer_size, nullptr, GL_STREAM_DRAW);
+    glBufferData(GL_PIXEL_PACK_BUFFER, buffer_size, nullptr, GL_STREAM_DRAW_ARB);
     if (source_format.compressed) {
         glGetCompressedTextureImage(src_surface->Texture().handle, src_attachment,
                                     static_cast<GLsizei>(src_params.size_in_bytes), nullptr);

src/video_core/renderer_opengl/gl_shader_cache.cpp

@@ -84,7 +84,6 @@ CachedShader::CachedShader(VAddr addr, Maxwell::ShaderProgram program_type)
     }
 
     entries = program_result.second;
-    shader_length = entries.shader_length;
 
     if (program_type != Maxwell::ShaderProgram::Geometry) {
         OGLShader shader;

src/video_core/renderer_opengl/gl_shader_cache.h

@@ -30,7 +30,7 @@ public:
     }
 
     std::size_t GetSizeInBytes() const override {
-        return shader_length;
+        return GLShader::MAX_PROGRAM_CODE_LENGTH * sizeof(u64);
     }
 
     // We do not have to flush this cache as things in it are never modified by us.
@@ -82,7 +82,6 @@ private:
                                u32 max_vertices, const std::string& debug_name);
 
     VAddr addr;
-    std::size_t shader_length;
     Maxwell::ShaderProgram program_type;
     GLShader::ShaderSetup setup;
     GLShader::ShaderEntries entries;

src/video_core/renderer_opengl/gl_shader_decompiler.cpp

@@ -34,17 +34,6 @@ constexpr u32 PROGRAM_HEADER_SIZE = sizeof(Tegra::Shader::Header);
 constexpr u32 MAX_GEOMETRY_BUFFERS = 6;
 constexpr u32 MAX_ATTRIBUTES = 0x100; // Size in vec4s, this value is untested
 
-static const char* INTERNAL_FLAG_NAMES[] = {"zero_flag", "sign_flag", "carry_flag",
-                                            "overflow_flag"};
-
-enum class InternalFlag : u64 {
-    ZeroFlag = 0,
-    SignFlag = 1,
-    CarryFlag = 2,
-    OverflowFlag = 3,
-    Amount
-};
-
 class DecompileFail : public std::runtime_error {
 public:
     using std::runtime_error::runtime_error;
@@ -95,8 +84,7 @@ struct Subroutine {
 class ControlFlowAnalyzer {
 public:
     ControlFlowAnalyzer(const ProgramCode& program_code, u32 main_offset, const std::string& suffix)
-        : program_code(program_code), shader_coverage_begin(main_offset),
-          shader_coverage_end(main_offset + 1) {
+        : program_code(program_code) {
 
         // Recursively finds all subroutines.
         const Subroutine& program_main = AddSubroutine(main_offset, PROGRAM_END, suffix);
@@ -108,16 +96,10 @@ public:
         return std::move(subroutines);
     }
 
-    std::size_t GetShaderLength() const {
-        return shader_coverage_end * sizeof(u64);
-    }
-
 private:
     const ProgramCode& program_code;
     std::set<Subroutine> subroutines;
     std::map<std::pair<u32, u32>, ExitMethod> exit_method_map;
-    u32 shader_coverage_begin;
-    u32 shader_coverage_end;
 
     /// Adds and analyzes a new subroutine if it is not added yet.
     const Subroutine& AddSubroutine(u32 begin, u32 end, const std::string& suffix) {
@@ -159,9 +141,6 @@ private:
             return exit_method;
 
         for (u32 offset = begin; offset != end && offset != PROGRAM_END; ++offset) {
-            shader_coverage_begin = std::min(shader_coverage_begin, offset);
-            shader_coverage_end = std::max(shader_coverage_end, offset + 1);
-
             const Instruction instr = {program_code[offset]};
             if (const auto opcode = OpCode::Decode(instr)) {
                 switch (opcode->get().GetId()) {
@@ -278,6 +257,14 @@ private:
     const std::string& suffix;
 };
 
+enum class InternalFlag : u64 {
+    ZeroFlag = 0,
+    CarryFlag = 1,
+    OverflowFlag = 2,
+    NaNFlag = 3,
+    Amount
+};
+
 /**
  * Used to manage shader registers that are emulated with GLSL. This class keeps track of the state
  * of all registers (e.g. whether they are currently being used as Floats or Integers), and
@@ -384,7 +371,7 @@ public:
         if (sets_cc) {
             const std::string zero_condition = "( " + ConvertIntegerSize(value, size) + " == 0 )";
             SetInternalFlag(InternalFlag::ZeroFlag, zero_condition);
-            LOG_WARNING(HW_GPU, "Condition codes implementation is incomplete.");
+            LOG_WARNING(HW_GPU, "Control Codes Imcomplete.");
         }
     }
 
@@ -467,25 +454,23 @@ public:
         shader.AddLine("lmem[" + index + "] = " + func + '(' + value + ");");
     }
 
-    std::string GetConditionCode(const Tegra::Shader::ConditionCode cc) const {
+    std::string GetControlCode(const Tegra::Shader::ControlCode cc) const {
         switch (cc) {
-        case Tegra::Shader::ConditionCode::NEU:
+        case Tegra::Shader::ControlCode::NEU:
             return "!(" + GetInternalFlag(InternalFlag::ZeroFlag) + ')';
         default:
-            UNIMPLEMENTED_MSG("Unimplemented condition code: {}", static_cast<u32>(cc));
+            UNIMPLEMENTED_MSG("Unimplemented Control Code: {}", static_cast<u32>(cc));
             return "false";
         }
     }
 
-    std::string GetInternalFlag(const InternalFlag flag) const {
-        const auto index = static_cast<u32>(flag);
-        ASSERT(index < static_cast<u32>(InternalFlag::Amount));
-
-        return std::string(INTERNAL_FLAG_NAMES[index]) + '_' + suffix;
+    std::string GetInternalFlag(const InternalFlag ii) const {
+        const u32 code = static_cast<u32>(ii);
+        return "internalFlag_" + std::to_string(code) + suffix;
     }
 
-    void SetInternalFlag(const InternalFlag flag, const std::string& value) const {
-        shader.AddLine(GetInternalFlag(flag) + " = " + value + ';');
+    void SetInternalFlag(const InternalFlag ii, const std::string& value) const {
+        shader.AddLine(GetInternalFlag(ii) + " = " + value + ';');
     }
 
     /**
@@ -636,8 +621,8 @@ private:
 
     /// Generates declarations for internal flags.
     void GenerateInternalFlags() {
-        for (u32 flag = 0; flag < static_cast<u32>(InternalFlag::Amount); flag++) {
-            const InternalFlag code = static_cast<InternalFlag>(flag);
+        for (u32 ii = 0; ii < static_cast<u64>(InternalFlag::Amount); ii++) {
+            const InternalFlag code = static_cast<InternalFlag>(ii);
             declarations.AddLine("bool " + GetInternalFlag(code) + " = false;");
         }
         declarations.AddNewLine();
@@ -954,10 +939,9 @@ private:
 class GLSLGenerator {
 public:
     GLSLGenerator(const std::set<Subroutine>& subroutines, const ProgramCode& program_code,
-                  u32 main_offset, Maxwell3D::Regs::ShaderStage stage, const std::string& suffix,
-                  std::size_t shader_length)
+                  u32 main_offset, Maxwell3D::Regs::ShaderStage stage, const std::string& suffix)
         : subroutines(subroutines), program_code(program_code), main_offset(main_offset),
-          stage(stage), suffix(suffix), shader_length(shader_length) {
+          stage(stage), suffix(suffix) {
         std::memcpy(&header, program_code.data(), sizeof(Tegra::Shader::Header));
         local_memory_size = header.GetLocalMemorySize();
         regs.SetLocalMemory(local_memory_size);
@@ -970,7 +954,7 @@ public:
 
     /// Returns entries in the shader that are useful for external functions
     ShaderEntries GetEntries() const {
-        return {regs.GetConstBuffersDeclarations(), regs.GetSamplers(), shader_length};
+        return {regs.GetConstBuffersDeclarations(), regs.GetSamplers()};
     }
 
 private:
@@ -1075,17 +1059,11 @@ private:
                                        const std::string& op_a, const std::string& op_b) const {
         using Tegra::Shader::PredCondition;
         static const std::unordered_map<PredCondition, const char*> PredicateComparisonStrings = {
-            {PredCondition::LessThan, "<"},
-            {PredCondition::Equal, "=="},
-            {PredCondition::LessEqual, "<="},
-            {PredCondition::GreaterThan, ">"},
-            {PredCondition::NotEqual, "!="},
-            {PredCondition::GreaterEqual, ">="},
-            {PredCondition::LessThanWithNan, "<"},
-            {PredCondition::NotEqualWithNan, "!="},
-            {PredCondition::LessEqualWithNan, "<="},
-            {PredCondition::GreaterThanWithNan, ">"},
-            {PredCondition::GreaterEqualWithNan, ">="}};
+            {PredCondition::LessThan, "<"},           {PredCondition::Equal, "=="},
+            {PredCondition::LessEqual, "<="},         {PredCondition::GreaterThan, ">"},
+            {PredCondition::NotEqual, "!="},          {PredCondition::GreaterEqual, ">="},
+            {PredCondition::LessThanWithNan, "<"},    {PredCondition::NotEqualWithNan, "!="},
+            {PredCondition::GreaterThanWithNan, ">"}, {PredCondition::GreaterEqualWithNan, ">="}};
 
         const auto& comparison{PredicateComparisonStrings.find(condition)};
         UNIMPLEMENTED_IF_MSG(comparison == PredicateComparisonStrings.end(),
@@ -1094,7 +1072,6 @@ private:
         std::string predicate{'(' + op_a + ") " + comparison->second + " (" + op_b + ')'};
         if (condition == PredCondition::LessThanWithNan ||
             condition == PredCondition::NotEqualWithNan ||
-            condition == PredCondition::LessEqualWithNan ||
             condition == PredCondition::GreaterThanWithNan ||
             condition == PredCondition::GreaterEqualWithNan) {
             predicate += " || isnan(" + op_a + ") || isnan(" + op_b + ')';
@@ -1266,7 +1243,14 @@ private:
         regs.SetRegisterToInteger(dest, true, 0, result, 1, 1);
     }
 
-    void WriteTexsInstruction(const Instruction& instr, const std::string& texture) {
+    void WriteTexsInstruction(const Instruction& instr, const std::string& coord,
+                              const std::string& texture) {
+        // Add an extra scope and declare the texture coords inside to prevent
+        // overwriting them in case they are used as outputs of the texs instruction.
+        shader.AddLine('{');
+        ++shader.scope;
+        shader.AddLine(coord);
+
         // TEXS has two destination registers and a swizzle. The first two elements in the swizzle
         // go into gpr0+0 and gpr0+1, and the rest goes into gpr28+0 and gpr28+1
 
@@ -1289,6 +1273,9 @@ private:
 
             ++written_components;
         }
+
+        --shader.scope;
+        shader.AddLine('}');
     }
 
     static u32 TextureCoordinates(Tegra::Shader::TextureType texture_type) {
@@ -1521,8 +1508,9 @@ private:
                     instr.fmul.tab5c68_0 != 1, "FMUL tab5cb8_0({}) is not implemented",
                     instr.fmul.tab5c68_0
                         .Value()); // SMO typical sends 1 here which seems to be the default
+                UNIMPLEMENTED_IF_MSG(instr.fmul.cc != 0, "FMUL cc is not implemented");
                 UNIMPLEMENTED_IF_MSG(instr.generates_cc,
-                                     "Condition codes generation in FMUL is not implemented");
+                                     "FMUL Generates an unhandled Control Code");
 
                 op_b = GetOperandAbsNeg(op_b, false, instr.fmul.negate_b);
 
@@ -1534,7 +1522,7 @@ private:
             case OpCode::Id::FADD_R:
             case OpCode::Id::FADD_IMM: {
                 UNIMPLEMENTED_IF_MSG(instr.generates_cc,
-                                     "Condition codes generation in FADD is not implemented");
+                                     "FADD Generates an unhandled Control Code");
 
                 op_a = GetOperandAbsNeg(op_a, instr.alu.abs_a, instr.alu.negate_a);
                 op_b = GetOperandAbsNeg(op_b, instr.alu.abs_b, instr.alu.negate_b);
@@ -1584,7 +1572,7 @@ private:
             case OpCode::Id::FMNMX_R:
             case OpCode::Id::FMNMX_IMM: {
                 UNIMPLEMENTED_IF_MSG(instr.generates_cc,
-                                     "Condition codes generation in FMNMX is not implemented");
+                                     "FMNMX Generates an unhandled Control Code");
 
                 op_a = GetOperandAbsNeg(op_a, instr.alu.abs_a, instr.alu.negate_a);
                 op_b = GetOperandAbsNeg(op_b, instr.alu.abs_b, instr.alu.negate_b);
@@ -1621,7 +1609,7 @@ private:
             }
             case OpCode::Id::FMUL32_IMM: {
                 UNIMPLEMENTED_IF_MSG(instr.op_32.generates_cc,
-                                     "Condition codes generation in FMUL32 is not implemented");
+                                     "FMUL32 Generates an unhandled Control Code");
 
                 regs.SetRegisterToFloat(instr.gpr0, 0,
                                         regs.GetRegisterAsFloat(instr.gpr8) + " * " +
@@ -1631,7 +1619,7 @@ private:
             }
             case OpCode::Id::FADD32I: {
                 UNIMPLEMENTED_IF_MSG(instr.op_32.generates_cc,
-                                     "Condition codes generation in FADD32I is not implemented");
+                                     "FADD32 Generates an unhandled Control Code");
 
                 std::string op_a = regs.GetRegisterAsFloat(instr.gpr8);
                 std::string op_b = GetImmediate32(instr);
@@ -1666,8 +1654,7 @@ private:
 
             switch (opcode->get().GetId()) {
             case OpCode::Id::BFE_IMM: {
-                UNIMPLEMENTED_IF_MSG(instr.generates_cc,
-                                     "Condition codes generation in BFE is not implemented");
+                UNIMPLEMENTED_IF_MSG(instr.generates_cc, "BFE Generates an unhandled Control Code");
 
                 std::string inner_shift =
                     '(' + op_a + " << " + std::to_string(instr.bfe.GetLeftShiftValue()) + ')';
@@ -1704,8 +1691,7 @@ private:
             case OpCode::Id::SHR_C:
             case OpCode::Id::SHR_R:
             case OpCode::Id::SHR_IMM: {
-                UNIMPLEMENTED_IF_MSG(instr.generates_cc,
-                                     "Condition codes generation in SHR is not implemented");
+                UNIMPLEMENTED_IF_MSG(instr.generates_cc, "SHR Generates an unhandled Control Code");
 
                 if (!instr.shift.is_signed) {
                     // Logical shift right
@@ -1720,8 +1706,8 @@ private:
             case OpCode::Id::SHL_C:
             case OpCode::Id::SHL_R:
             case OpCode::Id::SHL_IMM:
-                UNIMPLEMENTED_IF_MSG(instr.generates_cc,
-                                     "Condition codes generation in SHL is not implemented");
+                UNIMPLEMENTED_IF_MSG(instr.generates_cc, "SHL Generates an unhandled Control Code");
+
                 regs.SetRegisterToInteger(instr.gpr0, true, 0, op_a + " << " + op_b, 1, 1);
                 break;
             default: {
@@ -1737,7 +1723,7 @@ private:
             switch (opcode->get().GetId()) {
             case OpCode::Id::IADD32I:
                 UNIMPLEMENTED_IF_MSG(instr.op_32.generates_cc,
-                                     "Condition codes generation in IADD32I is not implemented");
+                                     "IADD32 Generates an unhandled Control Code");
 
                 if (instr.iadd32i.negate_a)
                     op_a = "-(" + op_a + ')';
@@ -1747,7 +1733,7 @@ private:
                 break;
             case OpCode::Id::LOP32I: {
                 UNIMPLEMENTED_IF_MSG(instr.op_32.generates_cc,
-                                     "Condition codes generation in LOP32I is not implemented");
+                                     "LOP32I Generates an unhandled Control Code");
 
                 if (instr.alu.lop32i.invert_a)
                     op_a = "~(" + op_a + ')';
@@ -1786,7 +1772,7 @@ private:
             case OpCode::Id::IADD_R:
             case OpCode::Id::IADD_IMM: {
                 UNIMPLEMENTED_IF_MSG(instr.generates_cc,
-                                     "Condition codes generation in IADD is not implemented");
+                                     "IADD Generates an unhandled Control Code");
 
                 if (instr.alu_integer.negate_a)
                     op_a = "-(" + op_a + ')';
@@ -1802,7 +1788,7 @@ private:
             case OpCode::Id::IADD3_R:
             case OpCode::Id::IADD3_IMM: {
                 UNIMPLEMENTED_IF_MSG(instr.generates_cc,
-                                     "Condition codes generation in IADD3 is not implemented");
+                                     "IADD3 Generates an unhandled Control Code");
 
                 std::string op_c = regs.GetRegisterAsInteger(instr.gpr39);
 
@@ -1865,7 +1851,7 @@ private:
             case OpCode::Id::ISCADD_R:
             case OpCode::Id::ISCADD_IMM: {
                 UNIMPLEMENTED_IF_MSG(instr.generates_cc,
-                                     "Condition codes generation in ISCADD is not implemented");
+                                     "ISCADD Generates an unhandled Control Code");
 
                 if (instr.alu_integer.negate_a)
                     op_a = "-(" + op_a + ')';
@@ -1900,8 +1886,7 @@ private:
             case OpCode::Id::LOP_C:
             case OpCode::Id::LOP_R:
             case OpCode::Id::LOP_IMM: {
-                UNIMPLEMENTED_IF_MSG(instr.generates_cc,
-                                     "Condition codes generation in LOP is not implemented");
+                UNIMPLEMENTED_IF_MSG(instr.generates_cc, "LOP Generates an unhandled Control Code");
 
                 if (instr.alu.lop.invert_a)
                     op_a = "~(" + op_a + ')';
@@ -1917,7 +1902,7 @@ private:
             case OpCode::Id::LOP3_R:
             case OpCode::Id::LOP3_IMM: {
                 UNIMPLEMENTED_IF_MSG(instr.generates_cc,
-                                     "Condition codes generation in LOP3 is not implemented");
+                                     "LOP3 Generates an unhandled Control Code");
 
                 const std::string op_c = regs.GetRegisterAsInteger(instr.gpr39);
                 std::string lut;
@@ -1936,7 +1921,7 @@ private:
             case OpCode::Id::IMNMX_IMM: {
                 UNIMPLEMENTED_IF(instr.imnmx.exchange != Tegra::Shader::IMinMaxExchange::None);
                 UNIMPLEMENTED_IF_MSG(instr.generates_cc,
-                                     "Condition codes generation in IMNMX is not implemented");
+                                     "IMNMX Generates an unhandled Control Code");
 
                 const std::string condition =
                     GetPredicateCondition(instr.imnmx.pred, instr.imnmx.negate_pred != 0);
@@ -2109,8 +2094,7 @@ private:
                 instr.ffma.tab5980_0.Value()); // Seems to be 1 by default based on SMO
             UNIMPLEMENTED_IF_MSG(instr.ffma.tab5980_1 != 0, "FFMA tab5980_1({}) not implemented",
                                  instr.ffma.tab5980_1.Value());
-            UNIMPLEMENTED_IF_MSG(instr.generates_cc,
-                                 "Condition codes generation in FFMA is not implemented");
+            UNIMPLEMENTED_IF_MSG(instr.generates_cc, "FFMA Generates an unhandled Control Code");
 
             switch (opcode->get().GetId()) {
             case OpCode::Id::FFMA_CR: {
@@ -2220,8 +2204,7 @@ private:
             case OpCode::Id::I2F_C: {
                 UNIMPLEMENTED_IF(instr.conversion.dest_size != Register::Size::Word);
                 UNIMPLEMENTED_IF(instr.conversion.selector);
-                UNIMPLEMENTED_IF_MSG(instr.generates_cc,
-                                     "Condition codes generation in I2F is not implemented");
+                UNIMPLEMENTED_IF_MSG(instr.generates_cc, "I2F Generates an unhandled Control Code");
 
                 std::string op_a{};
 
@@ -2251,8 +2234,7 @@ private:
             case OpCode::Id::F2F_R: {
                 UNIMPLEMENTED_IF(instr.conversion.dest_size != Register::Size::Word);
                 UNIMPLEMENTED_IF(instr.conversion.src_size != Register::Size::Word);
-                UNIMPLEMENTED_IF_MSG(instr.generates_cc,
-                                     "Condition codes generation in F2F is not implemented");
+                UNIMPLEMENTED_IF_MSG(instr.generates_cc, "F2F Generates an unhandled Control Code");
                 std::string op_a = regs.GetRegisterAsFloat(instr.gpr20);
 
                 if (instr.conversion.abs_a) {
@@ -2290,8 +2272,7 @@ private:
             case OpCode::Id::F2I_R:
             case OpCode::Id::F2I_C: {
                 UNIMPLEMENTED_IF(instr.conversion.src_size != Register::Size::Word);
-                UNIMPLEMENTED_IF_MSG(instr.generates_cc,
-                                     "Condition codes generation in F2I is not implemented");
+                UNIMPLEMENTED_IF_MSG(instr.generates_cc, "F2I Generates an unhandled Control Code");
                 std::string op_a{};
 
                 if (instr.is_b_gpr) {
@@ -2582,6 +2563,7 @@ private:
                 }
                 // TODO: make sure coordinates are always indexed to gpr8 and gpr20 is always bias
                 // or lod.
+                std::string op_c;
 
                 const std::string sampler =
                     GetSampler(instr.sampler, texture_type, is_array, depth_compare);
@@ -2604,35 +2586,27 @@ private:
                 }
                 case Tegra::Shader::TextureProcessMode::LB:
                 case Tegra::Shader::TextureProcessMode::LBA: {
-                    const std::string bias = [&]() {
-                        if (depth_compare) {
-                            if (is_array)
-                                return regs.GetRegisterAsFloat(instr.gpr20.Value() + 2);
-                            else
-                                return regs.GetRegisterAsFloat(instr.gpr20.Value() + 1);
-                        } else {
-                            return regs.GetRegisterAsFloat(instr.gpr20);
-                        }
-                    }();
-                    shader.AddLine("float bias = " + bias + ';');
-
+                    if (depth_compare) {
+                        if (is_array)
+                            op_c = regs.GetRegisterAsFloat(instr.gpr20.Value() + 2);
+                        else
+                            op_c = regs.GetRegisterAsFloat(instr.gpr20.Value() + 1);
+                    } else {
+                        op_c = regs.GetRegisterAsFloat(instr.gpr20);
+                    }
                     // TODO: Figure if A suffix changes the equation at all.
-                    texture = "texture(" + sampler + ", coords, bias)";
+                    texture = "texture(" + sampler + ", coords, " + op_c + ')';
                     break;
                 }
                 case Tegra::Shader::TextureProcessMode::LL:
                 case Tegra::Shader::TextureProcessMode::LLA: {
-                    const std::string lod = [&]() {
-                        if (num_coordinates <= 2) {
-                            return regs.GetRegisterAsFloat(instr.gpr20);
-                        } else {
-                            return regs.GetRegisterAsFloat(instr.gpr20.Value() + 1);
-                        }
-                    }();
-                    shader.AddLine("float lod = " + lod + ';');
-
+                    if (num_coordinates <= 2) {
+                        op_c = regs.GetRegisterAsFloat(instr.gpr20);
+                    } else {
+                        op_c = regs.GetRegisterAsFloat(instr.gpr20.Value() + 1);
+                    }
                     // TODO: Figure if A suffix changes the equation at all.
-                    texture = "textureLod(" + sampler + ", coords, lod)";
+                    texture = "textureLod(" + sampler + ", coords, " + op_c + ')';
                     break;
                 }
                 default: {
@@ -2659,6 +2633,7 @@ private:
                 break;
             }
             case OpCode::Id::TEXS: {
+                std::string coord;
                 Tegra::Shader::TextureType texture_type{instr.texs.GetTextureType()};
                 bool is_array{instr.texs.IsArrayTexture()};
 
@@ -2671,21 +2646,17 @@ private:
                 if (depth_compare)
                     num_coordinates += 1;
 
-                // Scope to avoid variable name overlaps.
-                shader.AddLine('{');
-                ++shader.scope;
-
                 switch (num_coordinates) {
                 case 2: {
                     if (is_array) {
                         const std::string index = regs.GetRegisterAsInteger(instr.gpr8);
                         const std::string x = regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
                         const std::string y = regs.GetRegisterAsFloat(instr.gpr20);
-                        shader.AddLine("vec3 coords = vec3(" + x + ", " + y + ", " + index + ");");
+                        coord = "vec3 coords = vec3(" + x + ", " + y + ", " + index + ");";
                     } else {
                         const std::string x = regs.GetRegisterAsFloat(instr.gpr8);
                         const std::string y = regs.GetRegisterAsFloat(instr.gpr20);
-                        shader.AddLine("vec2 coords = vec2(" + x + ", " + y + ");");
+                        coord = "vec2 coords = vec2(" + x + ", " + y + ");";
                     }
                     break;
                 }
@@ -2695,13 +2666,13 @@ private:
                         const std::string x = regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
                         const std::string y = regs.GetRegisterAsFloat(instr.gpr8.Value() + 2);
                         const std::string z = regs.GetRegisterAsFloat(instr.gpr20);
-                        shader.AddLine("vec4 coords = vec4(" + x + ", " + y + ", " + z + ", " +
-                                       index + ");");
+                        coord =
+                            "vec4 coords = vec4(" + x + ", " + y + ", " + z + ", " + index + ");";
                     } else {
                         const std::string x = regs.GetRegisterAsFloat(instr.gpr8);
                         const std::string y = regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
                         const std::string z = regs.GetRegisterAsFloat(instr.gpr20);
-                        shader.AddLine("vec3 coords = vec3(" + x + ", " + y + ", " + z + ");");
+                        coord = "vec3 coords = vec3(" + x + ", " + y + ", " + z + ");";
                     }
                     break;
                 }
@@ -2712,7 +2683,7 @@ private:
                     // Fallback to interpreting as a 2D texture for now
                     const std::string x = regs.GetRegisterAsFloat(instr.gpr8);
                     const std::string y = regs.GetRegisterAsFloat(instr.gpr20);
-                    shader.AddLine("vec2 coords = vec2(" + x + ", " + y + ");");
+                    coord = "vec2 coords = vec2(" + x + ", " + y + ");";
                     texture_type = Tegra::Shader::TextureType::Texture2D;
                     is_array = false;
                 }
@@ -2744,16 +2715,14 @@ private:
                 }
                 }
                 if (!depth_compare) {
-                    WriteTexsInstruction(instr, texture);
+                    WriteTexsInstruction(instr, coord, texture);
                 } else {
-                    WriteTexsInstruction(instr, "vec4(" + texture + ')');
+                    WriteTexsInstruction(instr, coord, "vec4(" + texture + ')');
                 }
-
-                shader.AddLine('}');
-                --shader.scope;
                 break;
             }
             case OpCode::Id::TLDS: {
+                std::string coord;
                 const Tegra::Shader::TextureType texture_type{instr.tlds.GetTextureType()};
                 const bool is_array{instr.tlds.IsArrayTexture()};
 
@@ -2767,16 +2736,12 @@ private:
                 UNIMPLEMENTED_IF_MSG(instr.tlds.UsesMiscMode(Tegra::Shader::TextureMiscMode::MZ),
                                      "MZ is not implemented");
 
-                u32 extra_op_offset = 0;
-
-                // Scope to avoid variable name overlaps.
-                shader.AddLine('{');
-                ++shader.scope;
+                u32 op_c_offset = 0;
 
                 switch (texture_type) {
                 case Tegra::Shader::TextureType::Texture1D: {
                     const std::string x = regs.GetRegisterAsInteger(instr.gpr8);
-                    shader.AddLine("int coords = " + x + ';');
+                    coord = "int coords = " + x + ';';
                     break;
                 }
                 case Tegra::Shader::TextureType::Texture2D: {
@@ -2784,8 +2749,8 @@ private:
 
                     const std::string x = regs.GetRegisterAsInteger(instr.gpr8);
                     const std::string y = regs.GetRegisterAsInteger(instr.gpr20);
-                    shader.AddLine("ivec2 coords = ivec2(" + x + ", " + y + ");");
-                    extra_op_offset = 1;
+                    coord = "ivec2 coords = ivec2(" + x + ", " + y + ");";
+                    op_c_offset = 1;
                     break;
                 }
                 default:
@@ -2800,10 +2765,9 @@ private:
                     break;
                 }
                 case Tegra::Shader::TextureProcessMode::LL: {
-                    shader.AddLine(
-                        "float lod = " +
-                        regs.GetRegisterAsInteger(instr.gpr20.Value() + extra_op_offset) + ';');
-                    texture = "texelFetch(" + sampler + ", coords, lod)";
+                    const std::string op_c =
+                        regs.GetRegisterAsInteger(instr.gpr20.Value() + op_c_offset);
+                    texture = "texelFetch(" + sampler + ", coords, " + op_c + ')';
                     break;
                 }
                 default: {
@@ -2812,10 +2776,7 @@ private:
                                       static_cast<u32>(instr.tlds.GetTextureProcessMode()));
                 }
                 }
-                WriteTexsInstruction(instr, texture);
-
-                --shader.scope;
-                shader.AddLine('}');
+                WriteTexsInstruction(instr, coord, texture);
                 break;
             }
             case OpCode::Id::TLD4: {
@@ -2838,23 +2799,18 @@ private:
                 if (depth_compare)
                     num_coordinates += 1;
 
-                // Add an extra scope and declare the texture coords inside to prevent
-                // overwriting them in case they are used as outputs of the texs instruction.
-                shader.AddLine('{');
-                ++shader.scope;
-
                 switch (num_coordinates) {
                 case 2: {
                     const std::string x = regs.GetRegisterAsFloat(instr.gpr8);
                     const std::string y = regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
-                    shader.AddLine("vec2 coords = vec2(" + x + ", " + y + ");");
+                    coord = "vec2 coords = vec2(" + x + ", " + y + ");";
                     break;
                 }
                 case 3: {
                     const std::string x = regs.GetRegisterAsFloat(instr.gpr8);
                     const std::string y = regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
                     const std::string z = regs.GetRegisterAsFloat(instr.gpr8.Value() + 2);
-                    shader.AddLine("vec3 coords = vec3(" + x + ", " + y + ", " + z + ");");
+                    coord = "vec3 coords = vec3(" + x + ", " + y + ", " + z + ");";
                     break;
                 }
                 default:
@@ -2862,13 +2818,17 @@ private:
                                       static_cast<u32>(num_coordinates));
                     const std::string x = regs.GetRegisterAsFloat(instr.gpr8);
                     const std::string y = regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
-                    shader.AddLine("vec2 coords = vec2(" + x + ", " + y + ");");
+                    coord = "vec2 coords = vec2(" + x + ", " + y + ");";
                     texture_type = Tegra::Shader::TextureType::Texture2D;
                 }
 
                 const std::string sampler =
                     GetSampler(instr.sampler, texture_type, false, depth_compare);
-
+                // Add an extra scope and declare the texture coords inside to prevent
+                // overwriting them in case they are used as outputs of the texs instruction.
+                shader.AddLine("{");
+                ++shader.scope;
+                shader.AddLine(coord);
                 const std::string texture = "textureGather(" + sampler + ", coords, " +
                                             std::to_string(instr.tld4.component) + ')';
                 if (!depth_compare) {
@@ -2885,7 +2845,7 @@ private:
                     regs.SetRegisterToFloat(instr.gpr0, 0, texture, 1, 1, false);
                 }
                 --shader.scope;
-                shader.AddLine('}');
+                shader.AddLine("}");
                 break;
             }
             case OpCode::Id::TLD4S: {
@@ -2896,10 +2856,6 @@ private:
                     instr.tld4s.UsesMiscMode(Tegra::Shader::TextureMiscMode::AOFFI),
                     "AOFFI is not implemented");
 
-                // Scope to avoid variable name overlaps.
-                shader.AddLine('{');
-                ++shader.scope;
-
                 const bool depth_compare =
                     instr.tld4s.UsesMiscMode(Tegra::Shader::TextureMiscMode::DC);
                 const std::string op_a = regs.GetRegisterAsFloat(instr.gpr8);
@@ -2907,33 +2863,28 @@ private:
                 // TODO(Subv): Figure out how the sampler type is encoded in the TLD4S instruction.
                 const std::string sampler = GetSampler(
                     instr.sampler, Tegra::Shader::TextureType::Texture2D, false, depth_compare);
+                std::string coord;
                 if (!depth_compare) {
-                    shader.AddLine("vec2 coords = vec2(" + op_a + ", " + op_b + ");");
+                    coord = "vec2 coords = vec2(" + op_a + ", " + op_b + ");";
                 } else {
                     // Note: TLD4S coordinate encoding works just like TEXS's
-                    shader.AddLine(
-                        "float op_y = " + regs.GetRegisterAsFloat(instr.gpr8.Value() + 1) + ';');
-                    shader.AddLine("vec3 coords = vec3(" + op_a + ", op_y, " + op_b + ");");
+                    const std::string op_c = regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
+                    coord = "vec3 coords = vec3(" + op_a + ", " + op_c + ", " + op_b + ");";
                 }
                 const std::string texture = "textureGather(" + sampler + ", coords, " +
                                             std::to_string(instr.tld4s.component) + ')';
 
                 if (!depth_compare) {
-                    WriteTexsInstruction(instr, texture);
+                    WriteTexsInstruction(instr, coord, texture);
                 } else {
-                    WriteTexsInstruction(instr, "vec4(" + texture + ')');
+                    WriteTexsInstruction(instr, coord, "vec4(" + texture + ')');
                 }
-
-                --shader.scope;
-                shader.AddLine('}');
                 break;
             }
             case OpCode::Id::TXQ: {
                 UNIMPLEMENTED_IF_MSG(instr.txq.UsesMiscMode(Tegra::Shader::TextureMiscMode::NODEP),
                                      "NODEP is not implemented");
 
-                ++shader.scope;
-                shader.AddLine('{');
                 // TODO: the new commits on the texture refactor, change the way samplers work.
                 // Sadly, not all texture instructions specify the type of texture their sampler
                 // uses. This must be fixed at a later instance.
@@ -2941,14 +2892,8 @@ private:
                     GetSampler(instr.sampler, Tegra::Shader::TextureType::Texture2D, false, false);
                 switch (instr.txq.query_type) {
                 case Tegra::Shader::TextureQueryType::Dimension: {
-                    const std::string texture = "textureSize(" + sampler + ", " +
-                                                regs.GetRegisterAsInteger(instr.gpr8) + ')';
-                    const std::string mip_level = "textureQueryLevels(" + sampler + ')';
-                    shader.AddLine("ivec2 sizes = " + texture + ';');
-                    regs.SetRegisterToInteger(instr.gpr0, true, 0, "sizes.x", 1, 1);
-                    regs.SetRegisterToInteger(instr.gpr0.Value() + 1, true, 0, "sizes.y", 1, 1);
-                    regs.SetRegisterToInteger(instr.gpr0.Value() + 2, true, 0, "0", 1, 1);
-                    regs.SetRegisterToInteger(instr.gpr0.Value() + 3, true, 0, mip_level, 1, 1);
+                    const std::string texture = "textureQueryLevels(" + sampler + ')';
+                    regs.SetRegisterToInteger(instr.gpr0, true, 0, texture, 1, 1);
                     break;
                 }
                 default: {
@@ -2956,8 +2901,6 @@ private:
                                       static_cast<u32>(instr.txq.query_type.Value()));
                 }
                 }
-                --shader.scope;
-                shader.AddLine('}');
                 break;
             }
             case OpCode::Id::TMML: {
@@ -3140,8 +3083,7 @@ private:
             break;
         }
         case OpCode::Type::PredicateSetRegister: {
-            UNIMPLEMENTED_IF_MSG(instr.generates_cc,
-                                 "Condition codes generation in PSET is not implemented");
+            UNIMPLEMENTED_IF_MSG(instr.generates_cc, "PSET Generates an unhandled Control Code");
 
             const std::string op_a =
                 GetPredicateCondition(instr.pset.pred12, instr.pset.neg_pred12 != 0);
@@ -3200,14 +3142,14 @@ private:
                 const std::string pred =
                     GetPredicateCondition(instr.csetp.pred39, instr.csetp.neg_pred39 != 0);
                 const std::string combiner = GetPredicateCombiner(instr.csetp.op);
-                const std::string condition_code = regs.GetConditionCode(instr.csetp.cc);
+                const std::string control_code = regs.GetControlCode(instr.csetp.cc);
                 if (instr.csetp.pred3 != static_cast<u64>(Pred::UnusedIndex)) {
                     SetPredicate(instr.csetp.pred3,
-                                 '(' + condition_code + ") " + combiner + " (" + pred + ')');
+                                 '(' + control_code + ") " + combiner + " (" + pred + ')');
                 }
                 if (instr.csetp.pred0 != static_cast<u64>(Pred::UnusedIndex)) {
                     SetPredicate(instr.csetp.pred0,
-                                 "!(" + condition_code + ") " + combiner + " (" + pred + ')');
+                                 "!(" + control_code + ") " + combiner + " (" + pred + ')');
                 }
                 break;
             }
@@ -3338,8 +3280,7 @@ private:
         case OpCode::Type::Xmad: {
             UNIMPLEMENTED_IF(instr.xmad.sign_a);
             UNIMPLEMENTED_IF(instr.xmad.sign_b);
-            UNIMPLEMENTED_IF_MSG(instr.generates_cc,
-                                 "Condition codes generation in XMAD is not implemented");
+            UNIMPLEMENTED_IF_MSG(instr.generates_cc, "XMAD Generates an unhandled Control Code");
 
             std::string op_a{regs.GetRegisterAsInteger(instr.gpr8, 0, instr.xmad.sign_a)};
             std::string op_b;
@@ -3431,9 +3372,9 @@ private:
         default: {
             switch (opcode->get().GetId()) {
             case OpCode::Id::EXIT: {
-                const Tegra::Shader::ConditionCode cc = instr.flow_condition_code;
-                UNIMPLEMENTED_IF_MSG(cc != Tegra::Shader::ConditionCode::T,
-                                     "EXIT condition code used: {}", static_cast<u32>(cc));
+                const Tegra::Shader::ControlCode cc = instr.flow_control_code;
+                UNIMPLEMENTED_IF_MSG(cc != Tegra::Shader::ControlCode::T,
+                                     "EXIT Control Code used: {}", static_cast<u32>(cc));
 
                 if (stage == Maxwell3D::Regs::ShaderStage::Fragment) {
                     EmitFragmentOutputsWrite();
@@ -3465,9 +3406,9 @@ private:
             case OpCode::Id::KIL: {
                 UNIMPLEMENTED_IF(instr.flow.cond != Tegra::Shader::FlowCondition::Always);
 
-                const Tegra::Shader::ConditionCode cc = instr.flow_condition_code;
-                UNIMPLEMENTED_IF_MSG(cc != Tegra::Shader::ConditionCode::T,
-                                     "KIL condition code used: {}", static_cast<u32>(cc));
+                const Tegra::Shader::ControlCode cc = instr.flow_control_code;
+                UNIMPLEMENTED_IF_MSG(cc != Tegra::Shader::ControlCode::T,
+                                     "KIL Control Code used: {}", static_cast<u32>(cc));
 
                 // Enclose "discard" in a conditional, so that GLSL compilation does not complain
                 // about unexecuted instructions that may follow this.
@@ -3529,9 +3470,9 @@ private:
                 UNIMPLEMENTED_IF_MSG(instr.bra.constant_buffer != 0,
                                      "BRA with constant buffers are not implemented");
 
-                const Tegra::Shader::ConditionCode cc = instr.flow_condition_code;
-                UNIMPLEMENTED_IF_MSG(cc != Tegra::Shader::ConditionCode::T,
-                                     "BRA condition code used: {}", static_cast<u32>(cc));
+                const Tegra::Shader::ControlCode cc = instr.flow_control_code;
+                UNIMPLEMENTED_IF_MSG(cc != Tegra::Shader::ControlCode::T,
+                                     "BRA Control Code used: {}", static_cast<u32>(cc));
 
                 const u32 target = offset + instr.bra.GetBranchTarget();
                 shader.AddLine("{ jmp_to = " + std::to_string(target) + "u; break; }");
@@ -3574,9 +3515,9 @@ private:
                 break;
             }
             case OpCode::Id::SYNC: {
-                const Tegra::Shader::ConditionCode cc = instr.flow_condition_code;
-                UNIMPLEMENTED_IF_MSG(cc != Tegra::Shader::ConditionCode::T,
-                                     "SYNC condition code used: {}", static_cast<u32>(cc));
+                const Tegra::Shader::ControlCode cc = instr.flow_control_code;
+                UNIMPLEMENTED_IF_MSG(cc != Tegra::Shader::ControlCode::T,
+                                     "SYNC Control Code used: {}", static_cast<u32>(cc));
 
                 // The SYNC opcode jumps to the address previously set by the SSY opcode
                 EmitPopFromFlowStack();
@@ -3584,10 +3525,10 @@ private:
             }
             case OpCode::Id::BRK: {
                 // The BRK opcode jumps to the address previously set by the PBK opcode
-                const Tegra::Shader::ConditionCode cc = instr.flow_condition_code;
-                UNIMPLEMENTED_IF_MSG(cc != Tegra::Shader::ConditionCode::T,
-                                     "BRK condition code used: {}", static_cast<u32>(cc));
-
+                const Tegra::Shader::ControlCode cc = instr.flow_control_code;
+                if (cc != Tegra::Shader::ControlCode::T) {
+                    UNIMPLEMENTED_MSG("BRK Control Code used: {}", static_cast<u32>(cc));
+                }
                 EmitPopFromFlowStack();
                 break;
             }
@@ -3598,9 +3539,6 @@ private:
                 break;
             }
             case OpCode::Id::VMAD: {
-                UNIMPLEMENTED_IF_MSG(instr.generates_cc,
-                                     "Condition codes generation in VMAD is not implemented");
-
                 const bool result_signed = instr.video.signed_a == 1 || instr.video.signed_b == 1;
                 const std::string op_a = GetVideoOperandA(instr);
                 const std::string op_b = GetVideoOperandB(instr);
@@ -3620,6 +3558,10 @@ private:
                 regs.SetRegisterToInteger(instr.gpr0, result_signed, 1, result, 1, 1,
                                           instr.vmad.saturate == 1, 0, Register::Size::Word,
                                           instr.vmad.cc);
+                if (instr.generates_cc) {
+                    UNIMPLEMENTED_MSG("VMAD Generates an unhandled Control Code");
+                }
+
                 break;
             }
             case OpCode::Id::VSETP: {
@@ -3771,7 +3713,6 @@ private:
     Maxwell3D::Regs::ShaderStage stage;
     const std::string& suffix;
     u64 local_memory_size;
-    std::size_t shader_length;
 
     ShaderWriter shader;
     ShaderWriter declarations;
@@ -3790,10 +3731,9 @@ std::optional<ProgramResult> DecompileProgram(const ProgramCode& program_code, u
                                               Maxwell3D::Regs::ShaderStage stage,
                                               const std::string& suffix) {
     try {
-        ControlFlowAnalyzer analyzer(program_code, main_offset, suffix);
-        const auto subroutines = analyzer.GetSubroutines();
-        GLSLGenerator generator(subroutines, program_code, main_offset, stage, suffix,
-                                analyzer.GetShaderLength());
+        const auto subroutines =
+            ControlFlowAnalyzer(program_code, main_offset, suffix).GetSubroutines();
+        GLSLGenerator generator(subroutines, program_code, main_offset, stage, suffix);
         return ProgramResult{generator.GetShaderCode(), generator.GetEntries()};
     } catch (const DecompileFail& exception) {
         LOG_ERROR(HW_GPU, "Shader decompilation failed: {}", exception.what());

src/video_core/renderer_opengl/gl_shader_gen.h

@@ -163,7 +163,6 @@ private:
 struct ShaderEntries {
     std::vector<ConstBufferEntry> const_buffer_entries;
     std::vector<SamplerEntry> texture_samplers;
-    std::size_t shader_length;
 };
 
 using ProgramResult = std::pair<std::string, ShaderEntries>;

src/video_core/renderer_opengl/gl_state.cpp

@@ -92,6 +92,13 @@ OpenGLState::OpenGLState() {
 
     point.size = 1;
     fragment_color_clamp.enabled = false;
+
+    polygon_offset.fill_enable = false;
+    polygon_offset.line_enable = false;
+    polygon_offset.point_enable = false;
+    polygon_offset.factor = 0.0f;
+    polygon_offset.units = 0.0f;
+    polygon_offset.clamp = 0.0f;
 }
 
 void OpenGLState::ApplyDefaultState() {
@@ -233,28 +240,6 @@ void OpenGLState::ApplyStencilTest() const {
         config_stencil(GL_BACK, stencil.back, cur_state.stencil.back);
     }
 }
-// Viewport does not affects glClearBuffer so emulate viewport using scissor test
-void OpenGLState::EmulateViewportWithScissor() {
-    auto& current = viewports[0];
-    if (current.scissor.enabled) {
-        const GLint left = std::max(current.x, current.scissor.x);
-        const GLint right =
-            std::max(current.x + current.width, current.scissor.x + current.scissor.width);
-        const GLint bottom = std::max(current.y, current.scissor.y);
-        const GLint top =
-            std::max(current.y + current.height, current.scissor.y + current.scissor.height);
-        current.scissor.x = std::max(left, 0);
-        current.scissor.y = std::max(bottom, 0);
-        current.scissor.width = std::max(right - left, 0);
-        current.scissor.height = std::max(top - bottom, 0);
-    } else {
-        current.scissor.enabled = true;
-        current.scissor.x = current.x;
-        current.scissor.y = current.y;
-        current.scissor.width = current.width;
-        current.scissor.height = current.height;
-    }
-}
 
 void OpenGLState::ApplyViewport() const {
     if (GLAD_GL_ARB_viewport_array && geometry_shaders.enabled) {
@@ -264,9 +249,7 @@ void OpenGLState::ApplyViewport() const {
             const auto& updated = viewports[i];
             if (updated.x != current.x || updated.y != current.y ||
                 updated.width != current.width || updated.height != current.height) {
-                glViewportIndexedf(
-                    i, static_cast<GLfloat>(updated.x), static_cast<GLfloat>(updated.y),
-                    static_cast<GLfloat>(updated.width), static_cast<GLfloat>(updated.height));
+                glViewportIndexedf(i, updated.x, updated.y, updated.width, updated.height);
             }
             if (updated.depth_range_near != current.depth_range_near ||
                 updated.depth_range_far != current.depth_range_far) {
@@ -294,7 +277,8 @@ void OpenGLState::ApplyViewport() const {
         const auto& updated = viewports[0];
         if (updated.x != current.x || updated.y != current.y || updated.width != current.width ||
             updated.height != current.height) {
-            glViewport(updated.x, updated.y, updated.width, updated.height);
+            glViewport(static_cast<GLint>(updated.x), static_cast<GLint>(updated.y),
+                       static_cast<GLsizei>(updated.width), static_cast<GLsizei>(updated.height));
         }
         if (updated.depth_range_near != current.depth_range_near ||
             updated.depth_range_far != current.depth_range_far) {
@@ -362,14 +346,14 @@ void OpenGLState::ApplyTargetBlending(std::size_t target, bool force) const {
     if (blend_changed || updated.src_rgb_func != current.src_rgb_func ||
         updated.dst_rgb_func != current.dst_rgb_func || updated.src_a_func != current.src_a_func ||
         updated.dst_a_func != current.dst_a_func) {
-        glBlendFuncSeparatei(static_cast<GLuint>(target), updated.src_rgb_func,
-                             updated.dst_rgb_func, updated.src_a_func, updated.dst_a_func);
+        glBlendFuncSeparateiARB(static_cast<GLuint>(target), updated.src_rgb_func,
+                                updated.dst_rgb_func, updated.src_a_func, updated.dst_a_func);
     }
 
     if (blend_changed || updated.rgb_equation != current.rgb_equation ||
         updated.a_equation != current.a_equation) {
-        glBlendEquationSeparatei(static_cast<GLuint>(target), updated.rgb_equation,
-                                 updated.a_equation);
+        glBlendEquationSeparateiARB(static_cast<GLuint>(target), updated.rgb_equation,
+                                    updated.a_equation);
     }
 }
 
@@ -406,6 +390,55 @@ void OpenGLState::ApplyLogicOp() const {
     }
 }
 
+void OpenGLState::ApplyPolygonOffset() const {
+
+    const bool fill_enable_changed =
+        polygon_offset.fill_enable != cur_state.polygon_offset.fill_enable;
+    const bool line_enable_changed =
+        polygon_offset.line_enable != cur_state.polygon_offset.line_enable;
+    const bool point_enable_changed =
+        polygon_offset.point_enable != cur_state.polygon_offset.point_enable;
+    const bool factor_changed = polygon_offset.factor != cur_state.polygon_offset.factor;
+    const bool units_changed = polygon_offset.units != cur_state.polygon_offset.units;
+    const bool clamp_changed = polygon_offset.clamp != cur_state.polygon_offset.clamp;
+
+    if (fill_enable_changed) {
+        if (polygon_offset.fill_enable) {
+            glEnable(GL_POLYGON_OFFSET_FILL);
+        } else {
+            glDisable(GL_POLYGON_OFFSET_FILL);
+        }
+    }
+
+    if (line_enable_changed) {
+        if (polygon_offset.line_enable) {
+            glEnable(GL_POLYGON_OFFSET_LINE);
+        } else {
+            glDisable(GL_POLYGON_OFFSET_LINE);
+        }
+    }
+
+    if (point_enable_changed) {
+        if (polygon_offset.point_enable) {
+            glEnable(GL_POLYGON_OFFSET_POINT);
+        } else {
+            glDisable(GL_POLYGON_OFFSET_POINT);
+        }
+    }
+
+    if ((polygon_offset.fill_enable || polygon_offset.line_enable || polygon_offset.point_enable) &&
+        (factor_changed || units_changed || clamp_changed)) {
+
+        if (GLAD_GL_EXT_polygon_offset_clamp && polygon_offset.clamp != 0) {
+            glPolygonOffsetClamp(polygon_offset.factor, polygon_offset.units, polygon_offset.clamp);
+        } else {
+            glPolygonOffset(polygon_offset.factor, polygon_offset.units);
+            UNIMPLEMENTED_IF_MSG(polygon_offset.clamp != 0,
+                                 "Unimplemented Depth polygon offset clamp.");
+        }
+    }
+}
+
 void OpenGLState::ApplyTextures() const {
     for (std::size_t i = 0; i < std::size(texture_units); ++i) {
         const auto& texture_unit = texture_units[i];
@@ -532,6 +565,7 @@ void OpenGLState::Apply() const {
     ApplyLogicOp();
     ApplyTextures();
     ApplySamplers();
+    ApplyPolygonOffset();
     cur_state = *this;
 }
 

src/video_core/renderer_opengl/gl_state.h

@@ -156,10 +156,10 @@ public:
     } draw;
 
     struct viewport {
-        GLint x;
-        GLint y;
-        GLint width;
-        GLint height;
+        GLfloat x;
+        GLfloat y;
+        GLfloat width;
+        GLfloat height;
         GLfloat depth_range_near; // GL_DEPTH_RANGE
         GLfloat depth_range_far;  // GL_DEPTH_RANGE
         struct {
@@ -176,6 +176,15 @@ public:
         float size; // GL_POINT_SIZE
     } point;
 
+    struct {
+        bool point_enable;
+        bool line_enable;
+        bool fill_enable;
+        GLfloat units;
+        GLfloat factor;
+        GLfloat clamp;
+    } polygon_offset;
+
     std::array<bool, 2> clip_distance; // GL_CLIP_DISTANCE
 
     OpenGLState();
@@ -206,7 +215,6 @@ public:
     OpenGLState& ResetBuffer(GLuint handle);
     OpenGLState& ResetVertexArray(GLuint handle);
     OpenGLState& ResetFramebuffer(GLuint handle);
-    void EmulateViewportWithScissor();
 
 private:
     static OpenGLState cur_state;
@@ -226,6 +234,7 @@ private:
     void ApplyLogicOp() const;
     void ApplyTextures() const;
     void ApplySamplers() const;
+    void ApplyPolygonOffset() const;
 };
 
 } // namespace OpenGL

src/video_core/renderer_opengl/renderer_opengl.cpp

@@ -490,7 +490,7 @@ bool RendererOpenGL::Init() {
     Core::Telemetry().AddField(Telemetry::FieldType::UserSystem, "GPU_Model", gpu_model);
     Core::Telemetry().AddField(Telemetry::FieldType::UserSystem, "GPU_OpenGL_Version", gl_version);
 
-    if (!GLAD_GL_VERSION_4_3) {
+    if (!GLAD_GL_VERSION_3_3) {
         return false;
     }
 

src/yuzu/applets/software_keyboard.cpp

@@ -82,8 +82,8 @@ QtSoftwareKeyboardDialog::QtSoftwareKeyboardDialog(
                            : QString::fromStdU16String(parameters.submit_text),
                        QDialogButtonBox::AcceptRole);
 
-    connect(buttons, &QDialogButtonBox::accepted, this, &QtSoftwareKeyboardDialog::accept);
-    connect(buttons, &QDialogButtonBox::rejected, this, &QtSoftwareKeyboardDialog::reject);
+    connect(buttons, &QDialogButtonBox::accepted, this, &QtSoftwareKeyboardDialog::Submit);
+    connect(buttons, &QDialogButtonBox::rejected, this, &QtSoftwareKeyboardDialog::Reject);
     layout->addWidget(header_label);
     layout->addWidget(sub_label);
     layout->addWidget(guide_label);
@@ -96,16 +96,16 @@ QtSoftwareKeyboardDialog::QtSoftwareKeyboardDialog(
 
 QtSoftwareKeyboardDialog::~QtSoftwareKeyboardDialog() = default;
 
-void QtSoftwareKeyboardDialog::accept() {
+void QtSoftwareKeyboardDialog::Submit() {
     ok = true;
     text = line_edit->text().toStdU16String();
-    QDialog::accept();
+    accept();
 }
 
-void QtSoftwareKeyboardDialog::reject() {
+void QtSoftwareKeyboardDialog::Reject() {
     ok = false;
     text.clear();
-    QDialog::reject();
+    accept();
 }
 
 std::u16string QtSoftwareKeyboardDialog::GetText() const {
@@ -129,13 +129,13 @@ QtSoftwareKeyboard::~QtSoftwareKeyboard() = default;
 
 void QtSoftwareKeyboard::RequestText(std::function<void(std::optional<std::u16string>)> out,
                                      Core::Frontend::SoftwareKeyboardParameters parameters) const {
-    text_output = std::move(out);
+    text_output = out;
     emit MainWindowGetText(parameters);
 }
 
 void QtSoftwareKeyboard::SendTextCheckDialog(std::u16string error_message,
                                              std::function<void()> finished_check) const {
-    this->finished_check = std::move(finished_check);
+    this->finished_check = finished_check;
     emit MainWindowTextCheckDialog(error_message);
 }
 

src/yuzu/applets/software_keyboard.h

@@ -33,8 +33,8 @@ public:
                              Core::Frontend::SoftwareKeyboardParameters parameters);
     ~QtSoftwareKeyboardDialog() override;
 
-    void accept() override;
-    void reject() override;
+    void Submit();
+    void Reject();
 
     std::u16string GetText() const;
     bool GetStatus() const;
@@ -70,10 +70,11 @@ signals:
     void MainWindowGetText(Core::Frontend::SoftwareKeyboardParameters parameters) const;
     void MainWindowTextCheckDialog(std::u16string error_message) const;
 
-private:
+public slots:
     void MainWindowFinishedText(std::optional<std::u16string> text);
     void MainWindowFinishedCheckDialog();
 
+private:
     mutable std::function<void(std::optional<std::u16string>)> text_output;
     mutable std::function<void()> finished_check;
 };

src/yuzu/bootmanager.cpp

@@ -310,7 +310,7 @@ void GRenderWindow::InitRenderTarget() {
     // TODO: One of these flags might be interesting: WA_OpaquePaintEvent, WA_NoBackground,
     // WA_DontShowOnScreen, WA_DeleteOnClose
     QGLFormat fmt;
-    fmt.setVersion(4, 3);
+    fmt.setVersion(3, 3);
     fmt.setProfile(QGLFormat::CoreProfile);
     fmt.setSwapInterval(false);
 

src/yuzu/configuration/config.cpp

@@ -432,7 +432,6 @@ void Config::ReadValues() {
     Settings::values.use_gdbstub = qt_config->value("use_gdbstub", false).toBool();
     Settings::values.gdbstub_port = qt_config->value("gdbstub_port", 24689).toInt();
     Settings::values.program_args = qt_config->value("program_args", "").toString().toStdString();
-    Settings::values.dump_exefs = qt_config->value("dump_exefs", false).toBool();
     Settings::values.dump_nso = qt_config->value("dump_nso", false).toBool();
     qt_config->endGroup();
 
@@ -639,7 +638,6 @@ void Config::SaveValues() {
     qt_config->setValue("use_gdbstub", Settings::values.use_gdbstub);
     qt_config->setValue("gdbstub_port", Settings::values.gdbstub_port);
     qt_config->setValue("program_args", QString::fromStdString(Settings::values.program_args));
-    qt_config->setValue("dump_exefs", Settings::values.dump_exefs);
     qt_config->setValue("dump_nso", Settings::values.dump_nso);
     qt_config->endGroup();
 

src/yuzu/configuration/configure_debug.cpp

@@ -34,7 +34,6 @@ void ConfigureDebug::setConfiguration() {
     ui->toggle_console->setChecked(UISettings::values.show_console);
     ui->log_filter_edit->setText(QString::fromStdString(Settings::values.log_filter));
     ui->homebrew_args_edit->setText(QString::fromStdString(Settings::values.program_args));
-    ui->dump_exefs->setChecked(Settings::values.dump_exefs);
     ui->dump_decompressed_nso->setChecked(Settings::values.dump_nso);
 }
 
@@ -44,7 +43,6 @@ void ConfigureDebug::applyConfiguration() {
     UISettings::values.show_console = ui->toggle_console->isChecked();
     Settings::values.log_filter = ui->log_filter_edit->text().toStdString();
     Settings::values.program_args = ui->homebrew_args_edit->text().toStdString();
-    Settings::values.dump_exefs = ui->dump_exefs->isChecked();
     Settings::values.dump_nso = ui->dump_decompressed_nso->isChecked();
     Debugger::ToggleConsole();
     Log::Filter filter;

src/yuzu/configuration/configure_debug.ui

@@ -145,16 +145,6 @@
         </property>
        </widget>
       </item>
-      <item>
-       <widget class="QCheckBox" name="dump_exefs">
-        <property name="whatsThis">
-         <string>When checked, any game that yuzu loads will have its ExeFS dumped to the yuzu/dump directory.</string>
-        </property>
-        <property name="text">
-         <string>Dump ExeFS</string>
-        </property>
-       </widget>
-      </item>
      </layout>
     </widget>
    </item>

src/yuzu/configuration/configure_graphics.ui

@@ -23,31 +23,31 @@
        </property>
        <layout class="QVBoxLayout" name="verticalLayout_2">
         <item>
-         <layout class="QHBoxLayout" name="horizontalLayout_2">
-          <item>
-           <widget class="QCheckBox" name="toggle_frame_limit">
-            <property name="text">
-             <string>Limit Speed Percent</string>
-            </property>
-           </widget>
-          </item>
-          <item>
-           <widget class="QSpinBox" name="frame_limit">
-            <property name="suffix">
-             <string>%</string>
-            </property>
-            <property name="minimum">
-             <number>1</number>
-            </property>
-            <property name="maximum">
-             <number>9999</number>
-            </property>
-            <property name="value">
-             <number>100</number>
-            </property>
-           </widget>
-          </item>
-         </layout>
+          <layout class="QHBoxLayout" name="horizontalLayout_2">
+            <item>
+              <widget class="QCheckBox" name="toggle_frame_limit">
+                <property name="text">
+                  <string>Limit Speed Percent</string>
+                </property>
+              </widget>
+            </item>
+            <item>
+              <widget class="QSpinBox" name="frame_limit">
+                <property name="suffix">
+                  <string>%</string>
+                </property>
+                <property name="minimum">
+                  <number>1</number>
+                </property>
+                <property name="maximum">
+                  <number>9999</number>
+                </property>
+                <property name="value">
+                  <number>100</number>
+                </property>
+              </widget>
+            </item>
+          </layout>
         </item>
         <item>
          <widget class="QCheckBox" name="use_accurate_gpu_emulation">
@@ -61,7 +61,7 @@
           <item>
            <widget class="QLabel" name="label">
             <property name="text">
-             <string>Internal Resolution</string>
+             <string>Internal Resolution:(Currently does nothing.)</string>
             </property>
            </widget>
           </item>
@@ -96,27 +96,27 @@
           </item>
          </layout>
         </item>
-        <item>
-         <layout class="QHBoxLayout" name="horizontalLayout_6">
-          <item>
-           <widget class="QLabel" name="bg_label">
-            <property name="text">
-             <string>Background Color:</string>
-            </property>
-           </widget>
-          </item>
-          <item>
-           <widget class="QPushButton" name="bg_button">
-            <property name="maximumSize">
-             <size>
-              <width>40</width>
-              <height>16777215</height>
-             </size>
-            </property>
-           </widget>
-          </item>
-         </layout>
-        </item>
+         <item>
+           <layout class="QHBoxLayout" name="horizontalLayout_6">
+             <item>
+               <widget class="QLabel" name="bg_label">
+                 <property name="text">
+                   <string>Background Color:</string>
+                 </property>
+               </widget>
+             </item>
+             <item>
+               <widget class="QPushButton" name="bg_button">
+                 <property name="maximumSize">
+                   <size>
+                     <width>40</width>
+                     <height>16777215</height>
+                   </size>
+                 </property>
+               </widget>
+             </item>
+           </layout>
+         </item>
        </layout>
       </widget>
      </item>

src/yuzu/main.cpp

@@ -518,18 +518,32 @@ void GMainWindow::OnDisplayTitleBars(bool show) {
 QStringList GMainWindow::GetUnsupportedGLExtensions() {
     QStringList unsupported_ext;
 
+    if (!GLAD_GL_ARB_program_interface_query)
+        unsupported_ext.append("ARB_program_interface_query");
+    if (!GLAD_GL_ARB_separate_shader_objects)
+        unsupported_ext.append("ARB_separate_shader_objects");
+    if (!GLAD_GL_ARB_vertex_attrib_binding)
+        unsupported_ext.append("ARB_vertex_attrib_binding");
     if (!GLAD_GL_ARB_vertex_type_10f_11f_11f_rev)
         unsupported_ext.append("ARB_vertex_type_10f_11f_11f_rev");
     if (!GLAD_GL_ARB_texture_mirror_clamp_to_edge)
         unsupported_ext.append("ARB_texture_mirror_clamp_to_edge");
+    if (!GLAD_GL_ARB_base_instance)
+        unsupported_ext.append("ARB_base_instance");
+    if (!GLAD_GL_ARB_texture_storage)
+        unsupported_ext.append("ARB_texture_storage");
     if (!GLAD_GL_ARB_multi_bind)
         unsupported_ext.append("ARB_multi_bind");
+    if (!GLAD_GL_ARB_copy_image)
+        unsupported_ext.append("ARB_copy_image");
 
     // Extensions required to support some texture formats.
     if (!GLAD_GL_EXT_texture_compression_s3tc)
         unsupported_ext.append("EXT_texture_compression_s3tc");
     if (!GLAD_GL_ARB_texture_compression_rgtc)
         unsupported_ext.append("ARB_texture_compression_rgtc");
+    if (!GLAD_GL_ARB_texture_compression_bptc)
+        unsupported_ext.append("ARB_texture_compression_bptc");
     if (!GLAD_GL_ARB_depth_buffer_float)
         unsupported_ext.append("ARB_depth_buffer_float");
 
@@ -548,8 +562,8 @@ bool GMainWindow::LoadROM(const QString& filename) {
     render_window->MakeCurrent();
 
     if (!gladLoadGL()) {
-        QMessageBox::critical(this, tr("Error while initializing OpenGL 4.3 Core!"),
-                              tr("Your GPU may not support OpenGL 4.3, or you do not "
+        QMessageBox::critical(this, tr("Error while initializing OpenGL 3.3 Core!"),
+                              tr("Your GPU may not support OpenGL 3.3, or you do not "
                                  "have the latest graphics driver."));
         return false;
     }

src/yuzu_cmd/config.cpp

@@ -366,7 +366,6 @@ void Config::ReadValues() {
     Settings::values.gdbstub_port =
         static_cast<u16>(sdl2_config->GetInteger("Debugging", "gdbstub_port", 24689));
     Settings::values.program_args = sdl2_config->Get("Debugging", "program_args", "");
-    Settings::values.dump_exefs = sdl2_config->GetBoolean("Debugging", "dump_exefs", false);
     Settings::values.dump_nso = sdl2_config->GetBoolean("Debugging", "dump_nso", false);
 
     // Web Service

src/yuzu_cmd/default_ini.h

@@ -206,8 +206,6 @@ log_filter = *:Trace
 # Port for listening to GDB connections.
 use_gdbstub=false
 gdbstub_port=24689
-# Determines whether or not yuzu will dump the ExeFS of all games it attempts to load while loading them
-dump_exefs=false
 # Determines whether or not yuzu will dump all NSOs it attempts to load while loading them
 dump_nso=false
 

src/yuzu_cmd/emu_window/emu_window_sdl2.cpp

@@ -111,18 +111,32 @@ void EmuWindow_SDL2::Fullscreen() {
 bool EmuWindow_SDL2::SupportsRequiredGLExtensions() {
     std::vector<std::string> unsupported_ext;
 
+    if (!GLAD_GL_ARB_program_interface_query)
+        unsupported_ext.push_back("ARB_program_interface_query");
+    if (!GLAD_GL_ARB_separate_shader_objects)
+        unsupported_ext.push_back("ARB_separate_shader_objects");
+    if (!GLAD_GL_ARB_vertex_attrib_binding)
+        unsupported_ext.push_back("ARB_vertex_attrib_binding");
     if (!GLAD_GL_ARB_vertex_type_10f_11f_11f_rev)
         unsupported_ext.push_back("ARB_vertex_type_10f_11f_11f_rev");
     if (!GLAD_GL_ARB_texture_mirror_clamp_to_edge)
         unsupported_ext.push_back("ARB_texture_mirror_clamp_to_edge");
+    if (!GLAD_GL_ARB_base_instance)
+        unsupported_ext.push_back("ARB_base_instance");
+    if (!GLAD_GL_ARB_texture_storage)
+        unsupported_ext.push_back("ARB_texture_storage");
     if (!GLAD_GL_ARB_multi_bind)
         unsupported_ext.push_back("ARB_multi_bind");
+    if (!GLAD_GL_ARB_copy_image)
+        unsupported_ext.push_back("ARB_copy_image");
 
     // Extensions required to support some texture formats.
     if (!GLAD_GL_EXT_texture_compression_s3tc)
         unsupported_ext.push_back("EXT_texture_compression_s3tc");
     if (!GLAD_GL_ARB_texture_compression_rgtc)
         unsupported_ext.push_back("ARB_texture_compression_rgtc");
+    if (!GLAD_GL_ARB_texture_compression_bptc)
+        unsupported_ext.push_back("ARB_texture_compression_bptc");
     if (!GLAD_GL_ARB_depth_buffer_float)
         unsupported_ext.push_back("ARB_depth_buffer_float");
 
@@ -143,7 +157,7 @@ EmuWindow_SDL2::EmuWindow_SDL2(bool fullscreen) {
         exit(1);
     }
 
-    SDL_GL_SetAttribute(SDL_GL_CONTEXT_MAJOR_VERSION, 4);
+    SDL_GL_SetAttribute(SDL_GL_CONTEXT_MAJOR_VERSION, 3);
     SDL_GL_SetAttribute(SDL_GL_CONTEXT_MINOR_VERSION, 3);
     SDL_GL_SetAttribute(SDL_GL_CONTEXT_PROFILE_MASK, SDL_GL_CONTEXT_PROFILE_CORE);
     SDL_GL_SetAttribute(SDL_GL_DOUBLEBUFFER, 1);